Quality and Risk Management in the IVF Laboratory

Quality and Risk Management in the IVF Laboratory

This essential survival guide for successfully managing the modern-day IVF

clinic condenses a wealth of expertise and experience from the authors in trou-

bleshooting and implementing quality management in the IVF laboratory. With

high-profile media coverage of mistakes at IVF clinics and escalating regula-

tory scrutiny, there is increasing pressure for professional accreditation. Modern

accreditation schemes, which are largely based on the principles of ISO 9001 and

related standards, require quality systems. Yet quality management beyond basic

assay quality control is often poorly understood by biomedical scientists outside

clinical chemistry laboratories. Quality and risk management are thus becoming

hot topics for those working in IVF clinics and this book brings together, for the

first time in one place, the basics of these essential aspects of laboratory manage-

ment. The focus on taking a holistic approach to “prophylactic management” –

prevention rather than cure – will be welcomed by all scientists working in IVF.

Quality and RiskManagement in theIVF Laboratory

David Mortimer, Ph.D. and

Sharon T. Mortimer, Ph.D.Oozoa Biomedical Inc. West VancouverBritish Columbia, Canada

p u b l i s h e d by t h e p re s s s y n d i c ate o f t h e u n ive r s i t y o f c a m b r i d g eThe Pitt Building, Trumpington Street, Cambridge, United Kingdom

c a m b r i d g e u n ive r s i t y p re s sThe Edinburgh Building, Cambridge CB2 2RU, UK40 West 20th Street, New York, NY 10011–4211, USA477 Williamstown Road, Port Melbourne, VIC 3207, AustraliaRuiz de Alarcon 13, 28014 Madrid, SpainDock House, The Waterfront, Cape Town 8001, South Africa

http://www.cambridge.org

C© D. Mortimer and S. T. Mortimer 2005

This book is in copyright. Subject to statutory exceptionand to the provisions of relevant collective licensing agreements,no reproduction of any part may take place withoutthe written permission of Cambridge University Press.

First published 2005

Printed in the United Kingdom at the University Press, Cambridge

Typefaces Minion 10.5/15 pt. and Formata System LATEX 2ε [tb]

A catalogue record for this book is available from the British Library

Library of Congress Cataloguing in Publication dataMortimer, David.Quality and risk management in the IVF laboratory / David Mortimer and Sharon T. Mortimer.

p. cm.Includes bibliographical references.ISBN 0 521 84349 9 (hardback: alk. paper)1. Fertility clinics – Quality control. 2. Fertility clinics – Risk management. 3. Fertilization invitro, Human – Standards. I. Mortimer, Sharon T. (Sharon Tracey), 1961– II. Title.[DNLM: 1. Fertilization in Vitro – standards. 2. Laboratory Techniques and Procedures –standards. 3. Quality Control. 4. Risk Management – methods. WQ 208 M888q 2005]RG135.M67 2005362.198′1780599 – dc22 2004054535

ISBN 0 521 84349 9 hardback

The publisher has used its best endeavors to ensure that the URLs for external websites referred to inthis book are correct and active at the time of going to press. However, the publisher has noresponsibility for the websites and can make no guarantee that a site will remain live or that thecontent is or will remain appropriate.

Every effort has been made in preparing this book to provide accurate and up-to-dateinformation that is in accord with accepted standards and practice at the time of publication.Nevertheless, the authors, editors, and publisher can make no warranties that the informationcontained herein is totally free from error, not least because clinical standards are constantlychanging through research and regulation. The authors, editors, and publisher therefore disclaim allliability for direct or consequential damages resulting from the use of material contained in thisbook. Readers are strongly advised to pay careful attention to information provided by themanufacturer of any drugs or equipment that they plan to use.

Contents

1 Introduction page 1

2 Regulation, licensing and accreditation 8

3 Quality and quality management 24

4 What is risk? 45

5 Process and systems 54

6 Making it work 86

7 Quality and risk management tools 118

8 What’s gone wrong? Troubleshooting 135

9 Risk management: being proactive 145

10 How are we doing? Benchmarking 169

11 Specifying systems 178

12 Human resources: finding (and keeping) the right staff 201

13 The well-run lab 210

14 References and recommended reading 218

Index 225

v

1

Introduction

It seems that we hear news reports of disasters in IVF clinics almostweekly. Public concern over these reports has resulted in governmentsintroducing regulation of IVF labs around the world, and within ourprofession there is a growing recognition of the need for accreditationof IVF labs to ensure that the potential for such errors occurring isminimized.

Quality systems, which have an inherent role in all modern accredi-tation schemes, are essentially based on the principles of ISO 9000 andrelated standards. Yet quality management beyond basic assay qualitycontrol is often poorly understood by biomedical scientists, especiallyoutside clinical chemistry and pathology laboratories. In particular,risk analysis and minimization are being demanded of IVF labs, butmany IVF scientists have only limited understanding of how to go aboutthese tasks. Perhaps this is because the majority of scientists working inclinical IVF labs have come from academic/research backgrounds and,as a consequence, many have limited experience of the practicalities oflaboratory management – and even fewer have any formal training in it.Certainly IVF has evolved rapidly over the last two-and-a-half decadesor so: from its beginnings as a highly experimental procedure in thelate 1970s, culminating in the birth of Louise Brown on 25 July 1978(Edwards and Steptoe, 1980), to a rapidly expanding field of researchand clinical practice that swept the world in the 1980s and was consol-idated as a routine clinical service in the 1990s. From the mid-1980swe also saw the rapid growth in commercial IVF clinics, to the extentthat IVF is often now described as an “industry” and IVF treatment(even intracytoplasmic sperm injection [ICSI]) is increasingly seen bymany as a commodity product, especially in the developed world.

1

2 Quality and Risk Management in the IVF Laboratory

As a result of this global expansion and commercialization, qualitymanagement and risk management are becoming increasingly impor-tant to those responsible for running IVF clinics, and consequentlythey are fast becoming “hot topics” for scientists working in them.

But quality management and risk management cannot be appliedin isolation; they must be integrated within the holistic frameworkof total quality management, itself essentially synonymous with thegoal of “best practice.” In this way quality and risk management willnot be seen as just additional annoying, expensive regulatory require-ments that “don’t help the patients get pregnant.” The provision ofeffective and safe IVF treatment depends on achieving improved stan-dards of technical services and medical care. Healthcare is slowly learn-ing the lessons that have transformed the manufacturing industriessince World War II, and have done the same for service industriesmore recently. Within this context, calls for IVF Centers to operateaccording to international standards such as ISO 9001 (Alper et al.,2002; International Standards Organization, 2000) reflect modernawareness of our professional – and commercial – environment, andshould be embraced by all Centers that truly care for their patients andemployees.

The structure and organization of IVF Centers varies widely betweensmall, “sole practitioner”-size clinics and large corporate IVF orga-nizations which typically operate multiple sites. Figure 1.1 shows ageneric concept for viewing the organization of an IVF Center by dis-ciplines, which is applicable to all clinics, regardless of size. The internalmanagement of an IVF Center is illustrated in Figure 1.2, establishingthe appropriate levels of control necessary to operate a multidisci-plinary organization that expresses mutual respect for all professionsinvolved. IVF labs vary in size between a single scientist (we abhorthe word “tech” or “technician” since we believe ardently that any-one performing IVF lab procedures must function as an autonomousprofessional scientist, but more of that later) and a large team that isoften sub-divided by functions and responsibilities. These extremes

3 Introduction

Figure 1.1 Diagrammatic representation of the organization of an IVF Center showing the“core team” that must have effective administration, finance and support teamsworking alongside it.

Board of Directors

Executive CommitteeProfessional Advisory

BoardBest Practice Committee

(= Quality Committee)

Lab ManagersCommittee

FinanceCommittee

NursingManagement

GeneticsLab Meeting

EmbryologyLab Meeting

AndrologyLab Meeting

Patient Complaints

Safety & InfectionControl Committee

Medical RecordsCommittee

Physicians' ClinicalMeeting

(e.g. proficiency, privileges,credentialing)

Ethics Committee

Figure 1.2 Organization chart showing the committee structure that might be required torun a large IVF Center according to the principles of Total Quality Management –or a generic accreditation scheme.


MedicalDirector

Lab Director(part-time, off-site)

LaboratoryManager

PatientCoordinator

OfficeManager

Nurse(s) Scientist(s)

Technician(s)

OfficeStaff

Cleaner(s)

Figure 1.3 Organization chart for a small (”boutique”) IVF Center.

are illustrated in the organization charts shown in Figures 1.3 and 1.4.A full understanding of organizational structure, the hierarchies ofauthority and responsibility, and lines of communication is an essentialprerequisite for anyone embarking upon implementing programs ofquality management and risk management.

Fortunately, each Center does not need to reinvent the disciplinesof quality management and risk management. Not only have severalIVF Centers around the world already achieved ISO 9001 certification,but the basic processes of managing quality improvement and riskmanagement in IVF are not fundamentally different from other areasof business. There are many resources available to Centers embarkingupon this journey, ranging from “self help” and reference books at all

5 Introduction

Scientific Director

AndrologyLab Director

EmbryologyLab Director

GeneticsLab Director

AndrologyLab Manager

EmbryologyLab Manager

EndocrineLab Manager

FISH LabManager

PCR LabManager

Senior LabAndrologists

LaboratoryAndrologists

Trainee LabAndrologists

SeniorEmbryologists

Embryologists

TraineeEmbryologists

Trainee AssayScientists

TechniciansTechnicians

Trainee FISHScientists

Trainee PCRScientists

Technicians

AssayScientists

FISHScientists

PCRScientists

CytogeneticsLab Manager

SeniorCytogeneticists

CertifiedCytogeneticists

TraineeCytogeneticists

Technicians

Figure 1.4 Organization chart for the laboratory operations of a large IVF Center.

levels (e.g. Dale and McQuater, 1998; Heller and Hindle, 2003) to prac-tical advice from friendly Centers based upon their own experiences,to expert advice and assistance from commercially orientated Centers,management companies or individual consultants.

We have written this book to bring together the basics of these essen-tial aspects of laboratory management in the context of IVF labs. Thebook is aimed at scientists who know their own technical field, butto whom the concepts of process and systems management are lessfamiliar – if not actually alien. We see education as the foundation forbringing about any change or improvement. Simply teaching peoplehow to do a task is not enough: unless people understand the “whys”(and the “why nots”) they are not truly competent to perform a job ascomplex and responsible as IVF. Therefore, the early chapters provide


basic definitions (unfortunately sometimes didactic and boring, butessential nonetheless) and explanations of the concepts and terminol-ogy that are used in quality and risk management. Later chapters thengo on to demonstrate how quality and risk management are tightlyintegrated in achieving optimum success rates, avoiding mistakes, andrunning an efficient – and successful – laboratory service. Finally, thereare chapters that provide basic advice and examples on the use of thevarious quality and risk management tools and techniques for devel-oping and implementing management systems in your own lab.

Throughout the book we have used illustrative examples from thegeneral world as well as ones specific to the IVF lab. The latter oftenrepresent examples of what happens when things do go wrong: issuessuch as mis-matched sperm and eggs, transferring the wrong embryos,losing samples from the cryobank, letting cryobank tanks go dry andso on – painful as they might be for any of us to think about. Of course,there are examples of where things went right for us or our colleaguesas well.

What happens when an IVF lab is “out of control”? The effects canbe very varied, and not all aspects will appear at the same time (orever), but some, many or all of the following features will be revealed.

• Unpredictable and inexplicable variations in outcomes (and indica-tors, if they’re being followed), with a likely general downward trendin results. In extreme cases things might deteriorate to such a statethat the best description is that “the wheels have fallen off.”

• That generalized perception that the feeling of “comfort” that youhad when things were running smoothly fades, and ultimately a senseof panic (controlled or not) might eventuate.

• Everyone starts to get “defensive” and this can deteriorate into fault-finding, “finger-pointing” and blame. If this is not checked then ageneral culture of fear, blame and retribution can develop and thelab (and, by then, probably the whole clinic) can become a “toxicworkplace.”

7 Introduction

If you recognize any of these symptoms in your lab or clinic then youshould definitely read on!

When we were asked to sum up what this book would be about,what its focus would be, we synthesized our concepts and ideas, ourbeliefs and attitudes, as well as summarizing our combined 40 years ofpractical experience in the field into the simple statement of “taking aholistic approach to prophylactic management” – achieving preventionrather than cure.

2

Regulation, licensing and accreditation

What’s the difference?

Regulation, licensing and accreditation are often confused with eachother, or seen as alternative viewpoints on how IVF labs are governed. Infact, they are different concepts and all three must work together withinan integrated system of governance. Let’s start with some definitions.

Regulations These are legal requirements∗ to which an organiza-tion or individual must conform in order to operate. Compli-ance is often verified by inspection (examination for individuals)and confirmed by the issuance of a license. Regulations are typi-cally highly prescriptive as to what an organization or individualmust/must not do in order to be compliant.

∗A requirement is a need or expectation that can be either statedexplicitly, customarily implied, or obligatory (i.e. a regulation).

Accreditation This is a collegial process based on self- andpeer-assessment whereby an authoritative body (usually a non-government organization) gives formal recognition that an orga-nization is in voluntary compliance with one or more Standardsset by the authoritative body. Unlike licensing, accreditation isbased upon process rather than procedure, and the principles ofquality improvement rather than strict obedience of regulations,so that it is not prescriptive in relation to technical procedures orrules. The end result of an accreditation process (being “accred-ited”) is often termed certification or registration by the author-itative body.

Licensing This is the process whereby an organization (or indi-vidual) is identified as being compliant with required regulations.

8

9 Regulation, licensing and accreditation

Usually, licensing is a legal requirement under government regu-lations in order for an organization to be allowed to operate [cf.certification]. For individuals, licensing is conferred to denotetheir competence to perform a given activity (e.g. driving a motorvehicle) in compliance with regulations.

Other specific terms that are often confused and misused as synonymswhen discussing regulation, licensing, and accreditation include: “cer-tification” [cf. “credentialing” and also “licensing”], “standards” ascompared to “regulations,” and “inspection” as opposed to “survey.”Again, some more definitions:

Certification This is the process whereby an organization (or indi-vidual) is identified as meeting one or more selected standards.The term is essentially synonymous with “registration” in the ISOsystem. A Certification Report will typically highlight any areasof nonconformance and require changes that “must” be madein order to achieve certification, as well as recommendations orsuggestions of changes that the organization “should” or “could”make to improve its operations. [cf. licensing]

Credentialing This is a process for assigning specific responsibil-ities (or “scope of practice”) to individual professionals basedon their training, qualifications, experience and current practice(actual expertise) within an organizational framework. It is anemployer’s responsibility, with a professional development focus,that commences upon appointment and continues through-out each individual’s employment. Credentialing is designed toensure quality of practice and management of risk, in medicineit is sometimes referred to as “clinical governance.”

Inspection This is a process carried out by one or more autho-rized inspectors, to determine whether an organization or facilityconforms to a defined set of Regulations. Inspection is typicallya requirement for licensing under Regulations.

Standards These are published documents that contain techni-cal specifications or criteria to be used consistently as rules,


guidelines, or definitions of characteristics to ensure that mate-rials, products, processes and services are fit for their purpose.Unlike a Regulation, a Standard is a “living document” thatdescribes a voluntary agreement between all stakeholders relevantto the product or service and encompasses everything that canhave a profound influence on the product or service, especiallyits safety, reliability and efficiency. Compliance with Standardsis ascertained through a process of assessment or accreditation,rather than inspection. These Standards are not synonymous with“minimum standards” which, while they define the minimumtechnical requirements for a process to be performed or under-taken, do not usually consider anything beyond basic quality con-trol (i.e. do not consider quality improvement or the quality cycle,see Chapter 3).

Survey This is the preferred term for the visit to a facility or orga-nization that is being assessed for accreditation. A Survey typi-cally follows a self-assessment process by the organization and isperformed by a (typically) multidisciplinary survey team whichevaluates the organization’s progress towards the goals describedin the Standards. (See “A Generic Accreditation Process,” below).

Regulation and licensing of IVF

Regulation and licensing are systems that are imposed on an organi-zation, such as a clinical laboratory or an IVF Center. These systems,which are not optional, are usually created and enforced via legislationand consequently vary widely between countries, and even betweenstates in countries such as Australia and the USA. Licensing bodies(e.g. the Human Fertilisation and Embryology Authority, the HFEA,in the UK) typically issue a licence after an inspection process to con-firm that an organization is, indeed, operating in accordance with thelaw. While this process does create some sort of minimum standardsto which the facility or organization will operate, there is often no con-sideration of performance standards or quality within the terms of thelicensing process.


Regulation and licensing are therefore not particularly relevant tothe focus of this book, and will be left for other authors to explore.Instead, our focus will be on the setting of – and compliance with –Standards that go beyond meeting minimum standards, an approachthat can be described simplistically as seeking to achieve best prac-tice. The formalization of such an approach is usually referred to asCertification or Accreditation. As defined above, “Certification” is typ-ically used when referring to standards such ISO Standards (see below),while “Accreditation” is a more broad-based approach founded upona perpetual process of quality improvement.

As a final word, we must all be aware of other regulations that weare obliged to follow in any workplace:• Regulations that affect the employer/employee relationship, such as

those that create statutory requirements pertaining to maximumwork hours, statutory holidays, annual leave, etc. Labor relations ingeneral is an area that no employer can ignore – if for no other reasonthan a disgruntled employee will be sure to remind him/her of them!

• Regulations that concern the handling and use of hazardous materi-als such as flammable solvents, strong acids and alkalis, liquid nitro-gen, radioactive materials, etc. All materials used in the IVF lab mustbe stored, handled and used correctly for the safety of everyone –and the facility. For example, in Canada and the USA the WorkplaceHazardous Materials Information System (WHMIS) is designed toreduce the risk from hazardous products in the workplace at all lev-els (i.e. suppliers, workers and employers) through proper trainingand the requirement that a Material Safety Data Sheet (MSDS) foreach product must be available to anyone who comes into contactwith it.

• General occupational health and safety regulations.• Fire regulations.• Building codes.Add to this such things as European Directives and there is a veritableminefield of regulation that affects almost everything we do, fromdesigning a lab to how high a fire extinguisher can be placed abovethe floor! Just because someone works in a (small) private IVF lab


and, in their opinion, “such-and-such doesn’t matter here,” does notgive them any right to break such regulations. Ignore them at yourperil!

Accreditation

As defined already, accreditation is a voluntary, collegial process basedon self- and peer-assessment whereby an authoritative body (usuallya non-governmental organization) gives formal recognition that anorganization is complying to an acceptable degree with one or moreStandards set by the non-governmental body. Accreditation is based onprocess rather than procedure, and the principles of quality improve-ment rather than strict obedience of regulations. An accreditationscheme is not prescriptive in relation to any technical procedures orrules.

Accreditation standards are most definitely not “minimumstandards.” Minimum standards only define the essential technicalrequirements for a process to be performed or undertaken, includingthe basic quality control procedures necessary to ensure that it has beendone correctly; they do not usually consider quality improvement orthe quality cycle (see Chapter 3).

Accreditation standards contain the technical specifications or cri-teria that must be applied consistently – whether as rules, guidelines,or definitions of characteristics – to ensure that materials, products,processes and services are fit for their purpose. Moreover, an Accred-itation Standard describes a voluntary agreement between all partiesinvolved in the product or service, and it encompasses every componentor factor that can influence the product or service, especially its safety,reliability and efficiency. Because our understanding of the processesby which we create a product or provide a service grows with experi-ence, it is vital that an Accreditation Standard be a “living document.”Processes are dynamic and therefore Standards cannot be embodiedwithin legislation that will probably take years to modify or reform.

Determining whether an organization is complying with an agreedset of Accreditation Standards involves a process of assessment and


evaluation that typically includes a self-assessment exercise in advanceof a survey (not an “inspection” or “assessment” site visit) by a multi-disciplinary team of surveyors who have received specialized trainingin reviewing an organization’s systems and processes – both as gener-alized concepts and with specialist, industry-specific knowledge andexperience. The organization seeking accreditation is supplied witha set of descriptive Standards against which it can evaluate itself andthen submit a preliminary self-assessment. After review of this doc-ument, a Survey Team is sent out to review the organization and itsoperations and assess their compliance with the Standards and theirprogress towards achieving their goals.

The following are examples of accreditation schemes:

Australia The Reproductive Technology Accreditation Commit-tee or “RTAC,” which operates under the aegis of the FertilitySociety of Australia. RTAC accreditation is required for all IVFunits in Australia in order for their patients to receive Medicarerebates for IVF treatment and to access gonadotrophins under theGovernment’s Pharmaceutical Benefits Scheme. IVF centers inNew Zealand also participate in the RTAC accreditation scheme.

Australia The National Association of Testing Authorities or“NATA” accredits all testing facilities including medical laborato-ries. Although IVF units are not required to have NATA accredi-tation, several have sought this independent accreditation. How-ever, any laboratory performing diagnostic testing (e.g. andrologyor endocrine) must be NATA accredited.

Australia Australian Council on Healthcare Standards or “ACHS”is a non-governmental organization that accredits hospitals andhealthcare organizations.

Canada Canadian Council on Health Services Accreditation or“CCHSA” is a non-governmental organization that accredits hos-pitals and healthcare organizations. An accreditation scheme forIVF clinics has recently been introduced as a joint venture betweenthe CCHSA and the Canadian Fertility and Andrology Society(CFAS).


UK Clinical Pathology Accreditation (UK) Ltd. or “CPA” is a non-governmental organization that accredits medical laboratories. Italso operates several External Quality Assurance (EQA) schemes.

USA The College of American Pathologists or “CAP” operatesa voluntary Reproductive Laboratory Accreditation Program(RLAP) that was developed in conjunction with the AmericanSociety for Reproductive Medicine (ASRM) but this programonly applies to laboratories performing andrology tests regu-lated by Clinical Laboratory Improvement Amendments of 1988(USDHHS, 1992). IVF Centers are not accredited by the CAPRLAP.

USA The Joint Commission on Accreditation of Healthcare Orga-nizations or “JCAHO” is an independent, not-for-profit orga-nization that considers itself to be the nation’s predominantstandards-setting and accrediting body in healthcare. JCAHOaccredits all types of laboratories and healthcare organizations,including IVF labs.

Beyond these national accreditation schemes there is internationalaccreditation by the International Organization for Standardization,commonly known as “ISO,” whose Standards are being increasinglyseen as the “gold standard” for IVF clinics.

ISO standards

The International Organization for Standardization (www.iso.ch) or“ISO” is based in Geneva and develops standards according to theessential principles of:

consensus the views of all interested parties are taken into account:manufacturers, vendors and users, consumer groups, testinglaboratories, governments, engineering professions and researchorganizations;

industry-wide they are global solutions intended to satisfy indus-tries and customers worldwide; and


voluntary international standardization is market-driven andtherefore based on the voluntary involvement of all interests inthe marketplace.

The following ISO standards are relevant to IVF Centers and theirlaboratories.

The ISO 9000 family of standards

The first edition of the ISO 9000 series of standards for quality man-agement and quality assurance was released in 1987, at which time theywere known in the various member countries by their own designation(e.g. BS 5750 in the UK). The second edition was introduced in 1994when most countries made their numbering compatible with the ISOsystem:

ISO 9001:1994 Quality Systems – Model for Quality Assurance inDesign, Development, Production, Installation andServicing. This Standard was essentially directedtowards manufacturers.

ISO 9002:1994 Quality Systems – Model for Quality Assurance in Pro-duction, Installation and Servicing. This standard wasvery similar to ISO 9001:1994 but had no require-ments for design control, being aimed essentially atservice organizations.

ISO 9003:1994 Quality Systems – Model for Quality Assurance in FinalInspection and Test. This standard was intended forquality testing organizations.

For the third (2000) edition, ISO 9002 and ISO 9003 were with-drawn leaving just one standard for certification: ISO 9001:2000. Thissingle quality management system requirement standard replaces thethree quality assurance requirement standards ISO 9001:1994, ISO9002:1994, and ISO 9003:1994. ISO 9001:2000 was developed to assistorganizations of all types and sizes to implement and operate an


effective Quality Management System (QMS) based on a more process-based approach, including an expectation of processes for ensuringcontinuous improvement. Therefore, ISO 9001:2000 specifies require-ments for a user-defined QMS that will allow an organization todemonstrate its ability to provide products that meet customer require-ments and applicable regulatory requirements and aims to enhancecustomer satisfaction. Organizations can exclude certain requirementsof the standard if some of its clauses are not relevant to their qualitysystems.

Consequently, the ISO 9000:2000 family now comprises four corestandards that form a coherent set of QMS standards facilitating mutualunderstanding in national and international trade:

ISO 9000:2000 Quality Management Systems – Fundamentals andVocabulary.

ISO 9001:2000 Quality Management Systems – Requirements.ISO 9004:2000 Quality Management Systems – Guidelines for Per-

formance Improvements. This document providesguidelines for both the effectiveness and efficiency ofa QMS based upon the fundamental aim of improv-ing the performance of an organization and the sat-isfaction of customers and other interested parties.

ISO 19011:2002 Guidelines for Quality and/or Environmental Man-agement Systems Auditing.

ISO standards for laboratories

There are other specific ISO standards that affect laboratories. Untilrecently, this was the ISO standard applicable to all laboratories(ISO/IEC 17025:2000 General Requirements for the Competence ofTesting and Calibration Laboratories), but for medical laboratories ithas now been superseded by ISO 15189:2003 Medical Laboratories –Particular Requirements for Quality and Competence. This new stan-dard specifically considers the provision of laboratory-based medical


services and is, therefore, the ISO standard most relevant to andrologyand IVF labs.

Therefore, while IVF Centers might choose to be accredited accord-ing to ISO 9001:2000, their laboratory activities must (also) complywith ISO 15189:2003.

The European Tissues and Cells Directive

On 31 March 2004 the European Union passed Directive 2004/23/ECOn setting standards of quality and safety for the donation, procurement,testing, processing, preservation, storage and distribution of human tissuesand cells – the “Tissues and Cells Directive” – which specifically includesreproductive cells and stem cells for human transplantation (EuropeanUnion, 2004). The extent of the Directive’s application is defined inArticle 6, which requires that all “Tissue Establishments” where any ofthe aforementioned activities are undertaken will be accredited, desig-nated, authorized or licensed by a competent authority of the MemberStates for the purpose of those activities, with legal compliance requirednot later than 7 April 2006. Therefore, it seems that not only spermbanks and IVF labs, but also any lab that processes sperm for artificialinsemination (e.g. intra-uterine insemination [IUI]) will be consideredas “Tissues Establishments” and will be subject to regulation under thisDirective. Article 16 goes on to state that all necessary measures shouldbe undertaken to ensure that each Tissue Establishment puts in placeand updates a quality system based on the principles of good practice.The creation of one or more Committees that will define the tech-nical requirements for implementing the provisions of the Directive,including the preparation, processing and storage of sperm, eggs andembryos, as well as the quality system, is provided for in Articles 28and 29. It is expected that “tissue-specific” technical requirements willbe promulgated in a series of further Directives to be enacted in thenear future.

Clearly this Directive will expand the requirement for qualitymanagement and accreditation of IVF labs throughout the EU,


necessitating a great deal of effort for hundreds of centers, but effortthat will contribute substantially towards the guiding principle of theDirective, ensuring high standards of quality and safety with respect tothe therapeutic use of human tissues and cells and thereby promotingthe highest possible level of protection to safeguard public health inthat regard.

A generic accreditation process

Accreditation can be viewed as a structured means of achievingpositive organizational change, rather than change being enforcedthrough an adversarial process. Usually, the accrediting authority isa non-governmental organization or not-for-profit company that hasdeveloped, in consultation with the professional bodies and otherstakeholders involved in the particular field, a set of Standards thatrepresent the consensus opinion as to operational standards and per-formance in the field. Effective accreditation schemes around the worldshare the same three basic characteristics (see Figure 2.1):

1. Self-study/evaluation/assessment;2. External assessment via a survey by peers; and3. Recommendations.

Self-assessment

An initial self-assessment by the organization is at the heart ofaccreditation. The organization undertakes a comprehensive exami-nation of all aspects of its mission, programs and services, a processthat necessarily involves individuals from every area and level of theorganization, as well as the organization’s customers (patients) and,ideally, the public. Input from all these “stakeholders” is used to cre-ate a detailed Self-Assessment Report documenting the organization’scurrent status quo.

Sometimes, during preparation for the self-assessment phase, on-site focus group consultations might be held that allow surveyors (not


Figure 2.1 A generic accreditation process.

necessarily the ones who will undertake the actual formal survey ofthe organization) to meet with staff, patients and the organization’scommunity stakeholders. The goal of these meetings is to help increasecommunication and collaboration throughout the organization, andthereby improve the validity of the self-assessment exercise’s findings.The self-assessment process has the following goals:

• To determine compliance with established accreditation criteria or“Standards”;

• To assess the organization’s alignment with its own stated philoso-phies and goals, as well as those that might be imposed by any regu-latory authority, in terms of patient care and the delivery of service;

• To evaluate outcomes and effectiveness; and• To identify, and prioritize, areas for improvement.


A major benefit that flows from the self-assessment process is the “buy-in” to the process from everyone involved (“Building a Process Map”in Chapter 5).

Typical accreditation standards cover the following operationalareas of the organization, across which the concepts of “educationfor life” and “achieving best results” are over-arching philosophicalprinciples.

• Governance: Including management structure and responsibility,leadership issues, partnerships with and accountability towards otherstakeholders, ethical issues, risk management, and perhaps even theorganization’s financial soundness.

• Human resources: Identifying and addressing needs, attracting andkeeping the right people (including career development issues),creating and maintaining good working relationships as well as ahealthy work environment.

• Information management: Collecting and keeping data, data secu-rity and confidentiality, and the use of data in benchmarking,decision-making and research.

• Information technology: The application and use of informationtechnology not only in relation to information management, but alsooperational efficiency, continuing education and career developmentfor managers and staff, and educational material for patients.

• Environmental: Providing a suitable environment for staff andpatients, as well as for the procedures and services being performed,minimizing the occurrence of adverse events, and respect for theenvironment in general.

• Clinical services: The provision of patient care, including diagnos-tic work-up, the delivery of therapeutic services and subsequentfollow-up (often termed the “continuum of care”); competent andresponsive clinical practices that meet the needs of the patients in par-ticular and the community (society) in general, including the con-sent process and respecting patients’ rights; continuing education


for medical and related staff (nurse coordinators, counselors, etc)and also patients.

• Laboratory services: General compliance with good laboratorypractice (GLP), adequate and appropriately-designed space, having,operating and maintaining adequate and suitable equipment, usingappropriate and suitable reagents and other products, as well as thegeneral technical procedures involved in the handling, culturing andcryopreservation of gametes and embryos.

External assessment

After the self-assessment exercise, a site visit by a team of surveyorsfrom outside the organization seeking accreditation is scheduled toassess the strengths and weaknesses of the organization. The SurveyTeam comprises a group of objective professionals who have receivedspecial training in performing such surveys. During the survey they willview the premises, meet with management, conduct interviews withmembers of staff and (willing) patients, and examine data to deter-mine whether the organization is in compliance with the accreditingauthority’s established criteria or standards. The Survey Team wouldtypically conduct an “exit interview” to present its findings and mightoffer advice that will be included in its written report. A draft reportis then submitted by the Team to the accrediting authority’s board ormanagement, who will then pass on the approved official report andrecommendations to the organization’s management.

Assessment results and recommendations

The findings from the survey, including their analysis in relation to theself-assessment document, are summarized in a written report whosepurpose is to focus on the organization’s strengths and weaknesses.Recommendations are made to help the organization develop plansnot only to improve areas where they are weak, but also to maintain


and expand areas where they are strong. Recommendations in a sur-vey report follow a standardized code of expression that allows theorganization seeking accreditation to interpret them:

Must or shall denote recommendations that are considered neces-sary for the organization to become compliant against a particularStandard, or alleviate a recognized problem.

Should denotes recommendations which, in the light of the survey-ors’ experience, will improve the organization’s rating accordingto a particular Standard, or are likely to provide significant ben-efit to the organization’s operational standards or performance(although they might be subject to prioritization by the organi-zation’s management or its governing body).

Could identifies suggestions for changes that might, in light ofthe surveyors’ experience and/or that of other similar organiza-tions, improve the organization’s operational standards or per-formance, and would be expected to generate an improvement inthe organization’s rating according to a particular Standard.

Afterwards

Accreditation is not a cyclical process. While the surveys might wellrun on a 3-year cycle, quality does not. Quality must be ongoing, aperpetual process built upon the Quality Cycle (see Chapter 3). Cer-tainly, everyone deserves a break after completing all the exhaustingpreparations for a survey visit and surviving the survey visit itself –but just for a few hours or a day (just enough time for a party – anda hangover, perhaps). But there should be no need for anyone to “getback” to the quality management program – because by then the wholething should be tightly integrated into the daily functioning of everypart of the clinic (not just the lab). Quality must be integral, it can’t bean add-on, and any organization that relaxes its commitment to qualityand considers that it doesn’t have to worry about “accreditation stuff”until the time comes around to prepare their next self-assessment, has


simply failed to see the point of accreditation, and will not reap the fullbenefit of everyone’s hard work.

The need to use your own processes

When developing a Quality Management System, it is vital that youdevelop and use your own processes rather than try to follow somebodyelse’s rules. Every IVF Lab or IVF Center is different. There will be localpermutations or variations in practice that will preclude you blindlyfollowing a system or process that you have “borrowed” from anotherclinic. Certainly you can base your methods and systems on those fromanother IVF lab, but unless you intend to copy that entire Center in themost minute detail, you will always have to adapt them. This subject isdiscussed in more detail in Chapters 5 and 6.

3

Quality and quality management

What is quality?

Traditionally, quality was seen as an expression of the superiority of aproduct, meaning that it might work better, or last longer, or just looknicer. Usually higher grade materials were used and more care was putinto its manufacture and finish. Usually a quality product cost moremoney, but was worth it if one took into account its attributes – and,of course, if one could afford the extra cost. But quality is not the sameas luxury, which represents opulence, i.e. products that are better thanthey need to be to serve their primary purpose, and is typically moreexpensive. After all, even if you are a successful sales rep with a largeterritory you don’t need a Lexus for driving around the country – butyou do need a car that doesn’t break down, is preferably not too fuelhungry, and is comfortable (because you’ll spend several hours eachday sitting in it).

From a basic product manufacturing perspective, quality can bedefined as conformance to specifications – specifications that are setby the manufacturer, based on the manufacturer’s experience of whatthe customer wants. This can, of course, be discovered by carryingout customer surveys or by looking at sales figures. But in serviceindustries the concept of quality is rather more difficult to define.Again, businesses have tried to define their best quality efforts (i.e. thequality of their services) according to certain specifications, usuallyones which management have defined and then refined – hopefully asa result of seeking the opinions and approval of their customers.

A major advance in understanding the concept of quality comeswhen it is defined as fitness for use – with the focus being orientated

24

25 Quality and quality management

completely towards the customers’ perceptions and opinions. Some-times this change in perspective is described as a switch from being a“product-out” company to a “market-in” company. Another descrip-tion of this approach is conformance to customer requirements asopposed to the earlier conformance to [manufacturer] specifications.

Quality management

Quality management is the integration of quality activities, whichinclude quality control, quality assurance and quality improvement,into a management philosophy. Historically, quality management hasits roots in the rebirth of the Japanese manufacturing industry afterWorld War II, when Japan had a reputation for producing some ofthe worst manufactured goods in the world. However, by the 1960sJapanese products were often the highest quality ones in various mar-ket areas. This philosophy of Total Quality Control was taught in Japanby W. Edwards Deming, PhD and Joseph M. Juran, BSEE, JD, and whenit was embraced by Western companies and organizations in the 1980sand 1990s it became known as Total Quality Management, or TQM.For many experts, TQM is simply the scientific way of doing business –so it is ideally suited to running an IVF lab.

Deming made a very important statement about quality and itsmodern application: Good quality does not necessarily mean high qual-ity. It means a predictable degree of uniformity and dependability with aquality suited to the market. But in medicine we must expand our hori-zons and go beyond just applying concepts that relate to manufacturingindustries.

In medicine, quality can also be defined as duty of care and hasbeen equated to the achievement of best practice. For an IVF lab, thesedefinitions can be combined with conformance to customer require-ments to establish a framework that embraces the provision of qualityservices that not only meet the customers’ needs, but also their expecta-tions. From a holistic perspective, these services must also be effective,efficient and safe, while protecting the rights and dignity of all parties


involved – including the children who will result from successful treat-ment. It is also worth noting that “customers” include not only thepatients but also other referring doctors and health care providers (e.g.NHS Trusts in the UK, or HMOs in the United States).

Terminology

As in all specialized areas of expert knowledge, TQM has a wealth ofterminology that must be used correctly to communicate one’s ideasand intentions clearly to others, and also to avoid the confusion whicharises from the incorrect use of terms. The most commonly-used termsin TQM are defined below:

Quality Assurance “QA” is the entirety of systematic activitiesimplemented within a quality system (i.e. including QC) thatare necessary to provide adequate confidence that a product orservice will satisfy its required quality characteristics.

Quality Control “QC” is the establishment of quality specifica-tions for each quality characteristic, assessment of proceduresused to determine conformance to these specifications, and tak-ing any necessary corrective actions to bring them into confor-mance. For example, ensuring that an assay procedure has beenperformed correctly and that its result is within the (pre-defined)acceptable limits of uncertainty of measurement (see Chapter 6).

Quality Cycle This is a repeated cycle, often expressed graphically,of quality improvement in relation to a single product or service(a.k.a. “Continual Quality Improvement”).

Quality Improvement “QI” is that part of a quality systemfocussed on continually increasing effectiveness and efficiency.“Continual Quality Improvement” is when QI is progressive andthe organization actively seeks and pursues quality improvementopportunities.

Quality Management This describes the sum of all activities of theoverall management function that determine the quality policy,


objectives and responsibilities, and implementation of them bymeans such as quality planning, quality control, quality assuranceand quality improvement within a defined quality system.

Quality Management System “QMS” describes the entire systemdeveloped by an organization involving the establishment of aquality policy and quality objectives and the processes to achievethose objectives.

Quality Manager This is the individual within an organizationcharged with achieving quality, and who is given the authority(and resources) to pursue it; the senior manager responsible forthe organization’s quality system.

Quality Manual This document describes an organization’s qual-ity management system.

Quality Objective This is something that is sought, or aimed for,in relation to creating or defining quality.

Quality Planning The part of quality management that isfocussed on setting quality objectives and specifying the neces-sary operational processes and related resources to fulfill qualityobjectives.

Quality Policy The overall intentions and directions of an orga-nization, as specified by management, related to the fulfilment ofquality requirements.

Quality System The organizational structure, procedures, pro-cesses and resources for implementing quality management.

Total Quality Management The management approach of anorganization, centered on quality and based on the participa-tion of all its members, that aims at long-term success throughcustomer satisfaction and creating benefits to all its members andto society.

An example of quality in action

Consider a company that manufactures TVs. The company receivesmany complaints from its customers that the TV sets they buy do


not work when they get them home. The company realizes that threeout of every ten sets that it makes are apparently faulty – and is deeplyconcerned that very soon word of the customer complaints will spreadand no-one will buy their product any more. So they install a man atthe end of the production line whose job is to plug each TV set intothe mains and check that it works before packing it into its carton:a quality control (QC) inspector. Very soon the complaints stop andcustomer satisfaction is at 100% – problem solved!

But, the company’s financial controller soon realizes that 30% oftheir raw materials costs, and 30% of their manufacturing costs aregoing entirely to waste, being dumped into a skip out behind the factoryand sent to the garbage dump. Such waste represents a very large part ofthe company’s profit margin – and it can’t go on. So, the non-workingTVs are taken apart and the faults are identified and tabulated. Themost common fault is found, perhaps a problem with soldering on themain circuit board, and the problem rectified. Now the QC inspectoronly has to reject three sets out of every twenty – the proportion of“good” TVs is now 85% instead of 70% – a major saving in costs.

However, the financial controller is still not happy because profitsare still not great and therefore insists that the next most common faultalso be remedied, which, in due course, it is. Now only one in twentyof the TV sets coming off the production line doesn’t work: the wasteis down to 5% and the proportion of “good” TVs is 95% – wonderfulnews! But the cost of employing the QC inspector is now greater thanthe cost of the wasted parts and labor and so the financial controllerrecommends that he be made redundant and the 5% wastage be writtenoff.

Fortunately for the QC inspector, the owner of the company believesthat 5% of his customers being angry is still too many – after all, theymight tell their friends about their bad experience and then they’ll buytheir next TVs from another company. The owner decides to keep onthe QC inspector until the manufacturing problems have been reducedto less than 1% – and so the engineers keep dealing with the less and lessfrequent manufacturing problems until they have reached the owner’s


target. Coincidentally, just at that time there happens to be a vacancy onthe production line so the QC inspector gets a better job, making surethat all the components that are delivered to the assemblers are correctand that they never run out, which would slow down production.

End result: the company now has a much better quality product,more than 99% of their customers are satisfied with their new TV sets,the company is more profitable than ever, the owner is happy, the finan-cial controller gets a bonus and the ex-QC inspector has a better, moresecure job. And everyone is absolutely convinced that “quality pays.”

Going beyond QC and QA: the quality cycle

In industry quality control focuses on inspection and checking, itspurpose is to reduce waste, and uses inspectors to check the work ofothers. In a laboratory, QC typically equates to making sure an assay wasrun properly: calibrators are used to ensure instruments were workingproperly, reference standards are used to verify that the results comeout close to where they should. QC is about making sure each task isdone correctly.

In all areas of endeavour, quality assurance focuses on proceduresand systems. Quality is designed into a process thereby increasing thelikelihood that when the particular method is followed the process willgo exactly as planned, increasing consistency and overall performance.In other words, QA relates to the way in which work is done.

The quality cycle (see Figure 3.1) is a process whereby an issue orproblem is recognized, a solution identified and put into effect, and theoutcome checked to ensure that the issue or problem has been resolved.The cycle can be repeated if the issue is a complex one and there areseveral solutions to its component problems.

Continual quality improvement: the ultimate goal

Quality describes the goal of satisfying requirements. But requirementschange as customers’ expectations rise, a perfectly normal situationthat results from the essential spirit of competition that is inherent


Figure 3.1 The Quality Cycle.

in the human psyche. In order to attract customers a business has tooffer more of something, or charge less. Charging less is an asymp-totic process due to such realities as raw materials costs, labor costs,wholesale prices, etc, and once economies of scale have been realizedthere is no room to create further competitive edge. So, businessesmust offer more for the same price – or at least a price that their cus-tomers perceive as having good value to them and being worth the pricedifference.

In IVF we have decidedly fixed costs in terms of consumables suchas plastic-ware, culture media, oocyte retrieval needles and embryotransfer catheters, and given the shortage of skilled embryologists thereis essentially a “sellers market” for labor, so salaries must be competitiveand hence are more likely to go up than be amenable to savings. So wemust offer “more” to our patients, for example:

• more zygotes for a given number of oocytes;• more embryos that are suitable for transfer or freezing in each treat-

ment cycle;• higher implantation potential for each fresh embryo transferred;• higher cryosurvival of the embryos frozen; and• higher implantation potential for each frozen embryo transferred.

In other words we must optimize our IVF/ICSI fertilization rates,our embryo culture systems and our cryopreservation techniques.Then, within the context of the IVF Center, we must make those


services more easily available, provide them in a more pleasant envi-ronment and with more personalized attention. Finally, our moreeffective services must be provided in a more efficient manner: i.e.in the private sector we must maximize our profit margins, in the pub-lic sector we must control costs so more treatment can be providedwithin a given budget. This is what total quality is about: continuouslyimproving customer satisfaction levels and simultaneously improvingmargins.

With “buy-in” from the Team our goal becomes the achievementof total quality through everyone’s commitment and involvement – acommon definition for Total Quality Management.

Total quality management

While there are many definitions of TQM, they all share the commonperspective of it being a philosophy rather than a simple manage-ment procedure. It must be seen as a process of improvement beyondthe status quo that then extends into an all-encompassing programof developing, and fostering, the desire for change and improvementthroughout the entire organization. TQM is an all-encompassing qual-ity system, it includes QC, QA and QI within a perpetual reiterativeprocess, but it is still based upon the foundations of inspection andaudit (see Figure 3.2). TQM must be seen as a long-term goal, there areno short cuts or quick fixes in implementing TQM. There is no tool ortechnique that can be seen as a panacea for all the problems and woesof an organization, no turn-key systems that can be plugged into anorganization’s pre-existing management structure. Achieving a systemof continuous improvement can take years, depending on the natureand size of the organization: time frames of 8 to 10 years have beensuggested for big corporations, although from experience the morelimited nature of even a big IVF Center can allow success within 2 or3 years.

TQM in IVF can be seen as encompassing the following areas of aCenter’s operation, all of which impinge upon the laboratory.


Figure 3.2 A diagrammatic overview of Total Quality Management.

• Medical and scientific standardsObviously the latest and best techniques and protocols are requiredto enable the medical and laboratory staff to provide the highestquality services to patients. This can be summed up as striving toachieve “best practice.”

• ResponsibilityEveryone involved in IVF must have a sense of responsibility for theiractions. The fundamental principle of medicine, primum non nocereor “first do no harm” must always be uppermost in everyone’s minds.

• Duty of careThere is a clear, and inescapable, duty of care towards not just thepatients being treated, but also to the future children who will becreated by successful treatment.

• EthicsA great deal has been written and said about ethics in assistedreproduction technology, which ranges from simple artificial insemi-nation on the one hand to reproductive cloning on the other. The vastmajority of practitioners, both medical and scientific, have extremelyhigh ethical standards, but it only takes the odd person who is deter-mined to challenge society, or “push the envelope,” to create headlines


and bring everyone else under suspicion. Ethics must be seen asexisting at several levels:• Personal ethics, which can be moral or religious (although these

can be considered more as “beliefs” that do not necessarily haveany foundation in science);

• Professional ethics, often consisting of codes of conduct or recom-mendations for good practice produced by professional bodies; and

• Society also has ethical perspectives, and these are embodied in therequirement for ethics committees or institutional review boards –which can be seen as either a normal obligation stemming from ourexistence within a developed society or as a requirement imposedby regulatory bodies.

• Customer expectationsIn IVF, as in any service industry, our customers – both the patientsto whom we provide treatment and the other professionals who referthose patients to our clinics – have certain expectations, which are thefoundation for modern concepts of quality (see “What is Quality?,”above).

• Legal obligationsThere are many laws that affect the practice of IVF, and their enforce-ment is increasingly being achieved via government regulatory agen-cies – bodies who develop “Codes of Practice.” However, they can alsoderive from guidelines developed by professional bodies that thenbecome regulations by virtue of being referenced in legislation – anevent that might not have been anticipated, and can occur withoutwarning. A good example of this was when the Canadian Fertility andAndrology Society’s Guidelines for Therapeutic Donor Inseminationwere suddenly elevated to the status of regulations by their referenc-ing in the Processing and Distribution of Semen for Assisted ConceptionRegulations (“Semen Regulations”) under the authority of the Foodand Drugs Act, which led to the Health Canada Directive TechnicalRequirements for Therapeutic Donor Insemination. There is also thewhole area of contract law surrounding the provision of services inreturn for financial considerations.


• LiabilityWhere there are legal obligations and considerations of responsibil-ity, best practice and duty of care, then there are also issues of liability.However, such issues are best left to those qualified in the law, andwill not be considered further here!

Implementing TQM

TQM requires a broad approach and skills in many areas, and itssuccessful implementation is utterly dependent on planning andorganization. For example, it has often been said that running an IVFCenter “requires 10% clinical skills, 30% scientific skills and 60% sheerorganization.” The important thing here is not to argue over the break-down in relative professional contributions but to note the importanceattributed to organizational skills.

From a conceptual perspective, implementing TQM requires manythings to happen, some of which can be seen as sequential processeswhile others will perforce be in parallel. In summary:

• Developing a clear, long-term approach that is integrated with all theorganization’s other business plans and strategies (e.g. operations,human resources, facility development, information managementand technology, fiscal planning).

• Creating a comprehensive collection of policies that address the needsof all areas within the organization. These policies form the skeletonof how TQM will be implemented within the organization and willinclude goals, objectives, targets, specific projects and resources. Thelatter must be developed in full consultation with those individualswho will have the responsibility for translating the policies intoachievements.

• Deployment of these policies through all levels of the organization’shierarchy and through all areas of the organization’s activities.

• Systems analysis and the integration of quality into all processes atthe most fundamental levels.


• Developing prevention-based activities. This includes risk analy-sis and management and often a cultural shift from fault-findingand blame to recognizing “genuine” mistakes as opportunities forimprovement (while not trivializing poor performance or ignoringincompetence).

• Educating everyone in the organization – from the CEO to the “low-est” position – so that they embrace change. Creating a sense of“ownership” is essential in order to achieve “buy-in” from the orga-nization’s greatest asset – its people.

• Developing and introducing targeted quality assurance processes sothat quality improvement can take place. This is essential for theQuality Cycle.

• Developing the organization’s management and infrastructure tosupport quality and quality improvement activities. While some newpositions will be essential to achieve this (e.g. the Quality Manager)this should not be seen as a separate part of the organization’s man-agement, it must be tightly integrated into the organization’s normalbusiness management structure.

• Continuing to pursue standardization, systematization and simpli-fication of all work instructions, procedures and systems. Processmapping and systems analysis are essential for this.

The essential steps towards implementing TQM listed above all requirefocussed effort in the following general areas of management.

Leadership

While it is an integral principle that quality is everyone’s business, andthat everyone must be actively involved, there must be strong leadershipfrom the top. Even though the CEO of a business, or the MedicalDirector of an IVF clinic, might not be deeply involved in all the hands-on aspects of implementing a TQM program, they must have somedirect participation and provide total support for the process. Such


people cannot see themselves as being “special” or have the perspectivethat quality is “what I employ you all for.”

A vital role of senior management is to ensure that the process ofimplementing TQM does not stall, because this will cause a major lossof faith in the process by the affected personnel – and getting them tobuy back in will be far more difficult than it was the first time around.

There is a great deal of excellent advice on how to be a good managerin the Essential Manager’s Manual (Heller and Hindle, 2003) and werecommend it highly.

Education and training

Everyone in the organization – the directors, managers and all employ-ees – must be provided with the education to ensure that their generalawareness and understanding of quality, its concepts and the necessaryattitudes and skills, is sufficient for them to understand the philosophyof continuous quality improvement and to allow them to participateactively in the process. What might seem like a load of jargon at the out-set must become a language of quality that is understood by the entireorganization. This requires that it be made relevant to running an IVFCenter, with explanation given in terms that illustrate its relevance andvalue.

People must also be taught the tools and techniques that they willneed as the process of implementing TQM proceeds. For scientistswho have received proper training in scientific method and investiga-tion much of this will be easy, if not second nature. But not everyonewill have had the benefit of the highest calibre teachers, and will not,therefore, all be at the same level.

Education and training of its workforce is a large, yet vital, invest-ment that every organization must be prepared – and able – to make.Without such skills and knowledge it will be difficult for many peopleto solve problems, and education is a prerequisite for attitude andbehavioural changes (see Chapter 12 for a more detailed discussion


of Human Resources issues). Becoming a “learning organization” is afundamental principle of accreditation schemes.

Using tools and techniques

In addition to knowing the basic biology and other related scientificdisciplines that affect gametes and embryos, effective employment oftools and techniques is necessary for systems analysis, problem solving,technology development and quality improvement. Without adequateknowledge of such tools and techniques the speed of change, and henceof TQM implementation, will be slowed. In the case of insufficientknowledge, or inadequately widespread knowledge, the entire processcould fail.

Involvement and commitment

Employee interest and active participation in quality improvementare vital not just throughout the period of TQM implementation butmust be a normal part of everyday activities in a “quality” organization.Through this involvement employees develop a sense of “ownership”in their work, which leads to an ongoing sense of commitment toensuring that everyone, and hence the organization as an entity, doestheir best. With this increased involvement comes a desire for moreactive participation and a feeling of satisfaction in the organization’sachievements, in other words “buy-in.”

Enhanced job satisfaction is very important for employee reten-tion, and also facilitates management. Indeed, along with these sortsof changes, and the seeking of – and listening to – employees’ opinionson the organization’s activities by managers, comes what many con-sider to be the greatest benefit of all: increased morale. As a corollaryto this closer involvement of employees in the business comes the needfor managers to share some of their powers (real or perceived!) andresponsibilities. But rather than being the “thin end of the wedge”or “the start of the slippery slope” a good manager will see this, and


promote it, as a mechanism for getting the most out of his/her staff.The bonus is that this is all achieved with the implicit participation ofthe staff, there is no need for coercion or for short-term financial orartificial rewards.

As a result of this involvement, there is increased recognition that anorganization’s employees are its greatest asset – one that will appreciateover time. Concern for, and support of, professional development, aswell as providing a career structure, are too often ignored in manyindustries, including IVF labs. We all know that it takes a long timeto properly train a new IVF scientist, yet how many of us have evertried to put a value on this investment? Helping people realize theirpotential is rewarding on many levels. Staff development and retentionare integral to accreditation schemes and merely reflect one of the mostbasic business principles: don’t allow your competitor to acquire yourassets (see Chapter 12 for further discussion of these topics).

Teamwork

Everyone working in IVF knows that it can only be successful if it isbased on sound teamwork. It is a multidisciplinary field and while suc-cessful outcomes might be achieved if everyone does their job properly,the highest success rates seem to come (usually) from those Centerswhere there is well-defined teamwork – a spirit of community amongthe staff.

Teams are created for many different reasons but, as discussedalready, creating effective teams is probably the key factor in achiev-ing successful TQM. Teams are a very effective means of maximizingthe rate of change in systems because they allow access to the full reper-toire of skills of all the members – combining the best features of eachwhile minimizing the impact of any particular individual’s weakness.The true strength and potential for achievement of a successful teamis not merely the sum of the complementary skills and strengths ofits members, a dynamic is created that generates an output that issynergistic, not just additive.


Creating effective teams requires a structured approach, not just asimple process of delegating to people based on their apparent availabil-ity (remember that the most effective people are usually the busiest),or being the next-in-line. It involves:

• analyzing the purpose of the team: definition of its goals andtasks;

• knowing the skills and strengths of all available staff – as well as theirweaknesses and short-comings;

• identifying the person who is best-suited to take the lead in theparticular task;

• ensuring that the team membership includes the spectrum of nec-essary talent;

• providing the correct reporting structure and necessary externalresources; and

• setting an appropriate and adequate timetable for the project.

Beyond recognizable, probably task-orientated or purpose-specific,teams there is also the requirement for everyone to work together.We have already considered this in some depth, and won’t labor thepoint here any further than to say that implementing TQM dependson mutual respect (both professional and personal) and appreciationfor everyone in the organization, as well as a recognition that honesty,sincerity and care, in addition to competence, must be integral to dailywork life. There must also be an acceptance that no-one is perfect, andthat a mistake is an opportunity for improvement.

Measurement and feedback

In TQM, as in life, you need to be able to measure each task in termsof an output or result. If you can’t, then it will seem like some never-ending chore, and you won’t have any way of knowing whether you’veachieved a goal or not. There has to be an endpoint, as well as a “feelgood” factor. We discuss Indicators in detail in Chapter 10, but forthe moment it’s sufficient to know that you must identify or devise


some way of measuring the progress, outcome or result of any actionor process that is undertaken.

An ongoing process

TQM is not a short-term project. It is not “finished” once the initial cer-tification or whatever has been awarded. Like an accreditation scheme,TQM is a never-ending process of continually seeking improvement.In both TQM and accreditation there is a continuum of quality cycleevents, although in accreditation the continuum is punctuated bysurveys.

Why does TQM fail?

Actually, TQM per se does not fail – but its implementation might.Common causes for an organization not being successful in imple-menting an effective TQM program include the following problems.

• Insufficient or inappropriate human and/or financial resources.• Lack of commitment by and/or support from the management

(“hollow words”).• Resistance to change, either active or passive (see “Resistance to

change,” below).• Insufficient knowledge and/or understanding of what was required.• Inadequate information management resources and/or systems

(includes documentation and data).• Wrong attitudes or an inappropriate environment (e.g. a culture of

fear, fault-finding, blame and retribution, see “The toxic workplace,”below).

Resistance to change

This is expressed most commonly by the epithet illustrated inFigure 3.3, but we must also recognize that it is an intrinsic feature of


Figure 3.3 A common expression of human nature with regard to the perceived need forchange.

human nature to be frightened of change. But confident, competentpeople can learn that change is a good thing, that the challenges (andperhaps a little residual fear) that it brings lead to major rewards, notjust professional but personal and financial. Resistance to change canbe either passive or active.

Passive resistance to change is best characterized as inertia. Howoften do you hear the complaint “We don’t have enough time!”? Pro-vided that you have established that there are, indeed, sufficient humanresources for the task or project, then this can be a red flag to identifypassive resistance. The individual(s) concerned need to be listenedto and then counseled, encouraged, and supported in adapting tothe (new) culture of change. If such educational efforts do not workthen you might well have to ask whether such people are in the rightjob, and whether they might find life easier working somewhere else.Unfortunately team building is not always about being supportiveand doing positive things, sometimes immovable obstacles need to beremoved.

Active resistance is where someone takes positive action to block,undermine or destroy changes that have been introduced. Again agood manager will be considerate and make every reasonable effort tohelp the unfortunate person adapt to the modern world, but here theemphasis is definitely on “reasonable.”


Figure 3.4 An illustration of the “detection-based mentality.”

The textbook by Heller and Hindle (2003) provides a wealth of adviceon managing change.

The toxic workplace

If people lack confidence or are insecure in their competence, employ-ment situation or personal life, this insecurity is often displayed inbehavior that has been called the “detection-based mentality” (seeFigure 3.4). An integral part of TQM is providing support to all staffmembers to allow them to develop as professionals and rise above suchshort-comings. However, if the managers (or owners) suffer the sameissues themselves, then there is often an abuse of power that leads to amassive escalation of the problem until the organization becomes whathas been described as a “toxic workplace” (“a place where people come towork so they can make enough money so they can leave it”: Jeffrey Pfeffer,cited in Coombs, 2001). By constantly criticizing or disrespecting thestaff, these managers feel better able to “control” them and, at the same


time, prevent anyone questioning their authority (which is not basedon respect) or their competence.

Ann Coombs (2001) gives the example of a tree to describe the effectof a toxic workplace. She explains that while trees are able to withstandstorms and drought, if you poison the soil around the tree, the poisonwill be drawn into the tree through its roots. The poison will collectslowly in the tree, but its effect won’t be noticed until the leaves start toturn yellow, and the tree withers and dies, ready to be knocked over bythe smallest force. Because of this hidden damage caused by workplacetoxicity, she suggests that it is easier to recognize a toxic workplace inretrospect, but gives the following indicators:

• no support for workers from management;• no support between the workers themselves;• lethargy;• absenteeism;• verbal and physical intimidation;• an increased level of complaints;• changes in employees’ behavior, e.g. loss of confidence or initiative,

declining interpersonal relationships, development of turf wars, inci-dences of “work rage,” avoidance of company social functions; and

• a culture of fear.

If you find yourself in a toxic workplace then TQM will never be success-ful in the long-term – although there might be some short-term benefitsbefore everyone gets ground down again. An IVF Center that suffersfrom this syndrome might well be successful in passing an accreditationexercise, but if will never reap the true long-term rewards of everyone’sefforts because as soon as the survey is over things will drift back totheir previous distasteful state. If the culprit is a manager then hope-fully the system will identify him/her and remove this human obstaclefrom the path towards excellence, but if the source of the problem isthe owner then there is only one solution to save your sanity – find anew job! Take your professional skills, and your experience with TQM,


to an organization that will value them. Being a bully only works ifpeople accept the bullying.

Quality itself is not the goal

The process of achieving quality is not a self-serving goal, the qualitysought has to be useful. Quality must also be real, anyone can attachthe word “quality” to their activities or systems, but unless there is truecommitment and achievement the supposed quality program will notsurvive. Quality management cannot stand alone, it has to be integratedinto all levels of the organization’s operations and embraced on a dailybasis.

In the summer of 2003 we were in Aberdeen for a conference andnear the guest house where we were staying there was a sad, derelictbuilding that was a poignant reminder of this to us (Figure 3.5). Hope-fully, Quality Assurance Management Ltd. moved to bigger and betterpremises – but either way they should have removed the name plate!

Figure 3.5 Achieving quality is a process, not a label.

4

What is risk?

Reducing medical errors and enhancing patient and staff safety is aprime focus in modern medicine. But what is “risk”?

Put simply, a risk is any uncertainty about a future event that mightthreaten an organization’s ability to accomplish its mission. It is thechance of something happening that will have a negative impact onan organization’s objectives. In particular, risk is the possibility of suf-fering “loss”: loss of quality of outcome, loss of professional regard orprofile, loss of referrals, loss of patient/staff health (or even loss of life),loss of profitability, loss of success. It is said that failure is a key partof learning, and that in business risk and opportunity often go hand-in-hand, with risk per se not only being not bad, but even essentialto progress. Clearly such a perception of risk “as a good thing” is notacceptable in IVF Centers.

Continued developments in reproductive biomedicine, combinedwith heightened regulatory requirements, have led to more, unex-pected, and often complex, risk issues for IVF Centers. It is now moreimportant than ever to be proactive in identifying risk and takingappropriate preventative measures, hence IVF Centers must embracerisk analysis and risk minimization. Together these constitute riskmanagement which is an integral part of total quality managementor “TQM” (see Chapter 3) and any ISO 9000 family-compliant qual-ity system or laboratory accreditation scheme. Much has been said inrecent years about “clinical governance,” which is a process essentiallysynonymous with TQM for medical practice, and a lot of attention hasbeen focussed on iatrogenic illness (Sharpe and Faden, 1998). RichardKennedy has recently reviewed risk management in IVF from the clin-ical perspective (Kennedy, 2004).

45


Risk management was originally an engineering discipline deal-ing with the possibility that some future event might cause harm.It includes strategies and techniques for recognizing and confrontingany such threat and provides a disciplined environment for proactivedecision-making for the purpose of:

• assessing on a continuous basis what can go wrong;• determining which risks need to be dealt with; and• implementing strategies to deal with these risks.

Risk management can be summarized as asking – and answering –three basic questions:

1. “What can go wrong?”;2. “What will we do?” (both to prevent the harm from occurring and

in the aftermath of an incident); and3. “If something happens, how will we resolve it, put things right

and/or pay for it?”

The most widely-used standard for managing risk (e.g. by the UKNational Health Service) is the Australia and New Zealand Standard“Risk Management” AS/NZS4360:1999 (Standards Australia, 1999). Itis currently under review and a revised version is anticipated soon.

An excellent example of how a corporation might deal with potentialfuture risk was the prohibition of the use of Percoll for any clini-cal application by its manufacturer, Pharmacia Biotech AB of Upp-sala, Sweden, effective 1 January 1997 (Mortimer, 2000). Percoll isa general-purpose laboratory reagent not manufactured to the stan-dards required for a medical device. To do so would have made itprohibitively expensive for the vast majority of its users, who employedit for research purposes only – as illustrated by the higher cost ofreplacement products that are manufactured and registered as med-ical devices, e.g. PureSperm (Nidacon International, AB, Goteborg,Sweden).

But the value of an effective risk management program is not lim-ited just to reducing the economic impacts of indemnity (insurance)

47 What is risk?

and claims (litigation). Fears of criminal prosecution within thebiomedical community are not without foundation, and the economicimpact of effective risk management should also be measured by thedevelopment of goodwill and the increased level of satisfaction experi-enced by everyone involved (patients, staff and management). Becausepatient satisfaction is closely associated with quality, a successful riskmanagement program is also a powerful marketing tool. Managers innon-healthcare industries do not rely on price alone to sell their prod-ucts (either goods or services) to customers, and modern-thinkingemployers and employees want to know about safety. This all adds upto a better quality service, and in the field of assisted conception, that iswhat the increasingly better-educated and more sophisticated patientsand referrers (an IVF Center’s customers) are looking for.

A major benefit of an effective risk management program, in whichrisks are continuously identified, analyzed and minimized, mitigatedor eliminated, is that problems are prevented before they occur: there isa cultural shift from “fire-fighting” and “crisis management” to proac-tive decision-making and planning. Anticipating what might go wrongbecomes inherent in everyday operations. While the implementationof risk management is no “magic bullet” – it does not guarantee suc-cess (because there are many aspects to achieving success in an IVFCenter) – it does improve decision making, help avoid surprises andimprove the chances of succeeding.

The consequences of not pursuing risk management are that:

• more resources will be expended to correct problems that could havebeen avoided;

• catastrophic problems will occur without warning;• there will be no ability to respond rapidly to such “surprises” and

recovery will be very difficult and/or costly, or even impossible;• decisions will be made with incomplete information or inadequate

knowledge of their possible future consequences;• the overall probability of success will be reduced; and• the organization will always be in a state of crisis.


Identifying “high risk” IVF laboratories

IVF is an area of rapidly advancing technology, meaning that continualtraining and proficiency testing are imperative. But these tactics inisolation will not prevent the type of problems that keep appearing inthe mass media, and we must recognize that the overall environmentin which people work can either support or obstruct both technicaltraining and improvements in operational systems. Is it possible fora hospital or Fertility Center to identify whether their IVF laboratoryis at particular risk for adverse events, especially of the “high profile”kind? From experience, the following are the most likely areas wheresuch risk factors might be identified (but it must be stressed that thisis by no means an exhaustive risk analysis of an IVF Lab).

Staffing issues

• Insufficient staff. A recent UK-based workload analysis revealed thatover 70% of IVF labs there are understaffed (Harbottle, 2003). Thisstudy confirmed our own earlier (unpublished) calculations that inan IVF laboratory operating to established quality standards, onefull-time embryologist is required for each 125 stimulated treatmentcycles per annum.

• Overworking. In order to perform all aspects of their jobs accu-rately and reliably, with the lowest possible risk of making mistakes,embryologists must be alert and not distracted by tiredness. Anycircumstances that contribute to over-tiredness or exhaustion rep-resent serious risk factors. Staffing levels should reflect the maximumcaseload – some slack must be available within the system so thatstaff are not constantly working at their maximum capacity. Riskfactors might include regularly working more than 48 hours perweek or more than 6 consecutive days without at least one full day’srest.

• Inexperienced staff. Even with effective training programs, a highstaff turnover increases the number of people who are less sure of

49 What is risk?

the laboratory’s systems, standard operating procedures (SOPs) andusual practices. Laboratories which have a higher proportion of rela-tively inexperienced embryologists will be less equipped to recognizeand deal with operational problems as they arise.

• Poorly trained staff. Comprehensive, formal programs are essen-tial not only for training all embryologists in new techniques andprocedures (i.e. novices and more junior staff), but also for theorientation, and re-training as necessary, of embryologists com-ing from other laboratories. If someone unintentionally fails tocomplete all aspects of their assigned tasks then this is an expres-sion of inadequate or incomplete training, and should be easilyremedied.

• Not accepting professional responsibility. This occurs when one ormore members of a team do not take enough care to ensure that theyhave performed – and completed – all the tasks that were assigned tothem. This can be either intentional or unintentional (see above), butin both cases is unprofessional. The intentional omission of parts ofone’s job can only continue without jeopardizing outcomes if thereare others who take the time (and/or are prepared to make the time)to ensure that the whole process is completed: in effect they are the“safety net.” All professionals must be prepared to work withouta safety net; if someone cannot do this then (s)he should not beworking in an IVF lab.

Resource issues

• The need for slack. Slack must be present in any organization toallow not only for the differential between the average and busiestactivity levels, but also to cope with “unpredictable” events such asan influenza epidemic. An allowance for slack time is essential foran organization to evolve (DeMarco, 2001).

• Inadequate resources. Centers that insist on running with the min-imum possible resources (physical, human or financial) will be atgreater risk. There must be sufficient capacity in critical equipment


to deal with the busiest of times: e.g. incubator space (so incubatordoors are not opened too often), or controlled rate freezer capacity(so all patients’ embryos can be frozen at the right time).

• Equipment failure. All equipment must be included in a preven-tative maintenance program and all “mission critical” equipmentmust be monitored on a continual basis, e.g. cryostorage tanks,incubators, CO2 supply, liquid nitrogen supply. In addition, thereshould be an out-of-hours alarm system that can call or page alist of contact persons, any of whom is capable of resolving theissue.

• Power failure. Provisions must be in place to ensure the continuityof electrical power to critical equipment. Items sensitive to powerfluctuations need to be protected by a line conditioner or, better, anuninterruptible power supply or “UPS.”

Organizational issues

• Not double-checking. Every time that something is labeled, or mate-rial (i.e. gametes or embryos) is moved from one container to another,identity checks must be followed strictly and verified either by acompetent witness or some valid technological solution. In the UK,the HFEA introduced a requirement for “double-witnessing” of allsuch events in October 2002 (see Brison, 2004).

• Inadequate SOPs. Incomplete, or poorly written, SOPs create oppor-tunities for embryologists to make mistakes.

• Omissions. Lack of (i) a comprehensive, documented system ofnotifications and (ii) task lists to ensure that the laboratory staffknow exactly what has to be done each day, will increase therisk of things “falling through the cracks” and critical tasks beingforgotten.

• Unauthorized “improvements” in methods. All staff must followthe laboratory’s SOPs exactly. Any changes to documented SOPsmust be authorized by the Lab Director to ensure that the changes

51 What is risk?

would not be detrimental. Introduction of variations, short-cutsor perceived “improvements” without authorization should be adisciplinary offence, because of the enormous increase in risk.

Risk management issues

• Poor documentation. Great care must be taken when completingall documentation (and everything must be documented) in orderto ensure that all records (laboratory, medical and government) arecomplete and accurate. For example, a Center that makes repeatedmistakes in its submissions of data to an external regulatory agencycould be considered suspect in its ability to complete its own paper-work. Other indicators of carelessness or lack of attention to detail(i.e. risk) include personnel not being aware of patients’ appoint-ments/procedures/management plan, mistakes in patient accounts,telephone calls being made to the wrong patient, etc.

• Unrecognized incidents. Unless a comprehensive system of IncidentReports for all adverse events is in place, enforced, and employedconstructively, many mistakes will never be recognized or remem-bered. In this context, an “adverse event” can be defined as any eventthat potentially or actually affects staff safety, patient safety or theprovision of treatment according to the patient’s care plan or theexpected outcome of their treatment.

• Use of non-approved products or devices. Some IVF Centers useproducts or devices that are either intended for veterinary use orthat have not been approved for medical use by the appropriateregulatory authorities, e.g. The Food and Drug Administration inthe USA, CE marking in Europe, etc. A common example of this isthe Tomcat catheter, a veterinary product that is still used in someCenters for intra-uterine insemination and even embryo transfer.Studies reported at conferences, and in peer-reviewed journals haveclearly demonstrated the detrimental effect of this catheter whenused for embryo transfer (see Chapter 11). Consequently, not only


is such usage a significant risk factor for decreased success rates, butit must also be considered a risk in terms of potential liability.

Disaster planning

On a larger scale, risk management also includes “disaster planning,”three examples of which are summarized here.

• Earthquake preparedness. Obviously the consequence of an earth-quake could be catastrophic, possibly resulting in the loss of multiplelives and potentially in the total destruction of the facility. If an IVFCenter is located in, say, the San Francisco area, then the risk level is“likely,” and clearly disaster planning must be undertaken. Obviouslynothing can be done to reduce the likelihood, since an earthquake isa natural disaster, but the Center’s disaster plan must take all reason-able steps to mitigate the potential impact of an earthquake uponstaff and patients, the embryos in culture and the contents of the cry-obank. However, for a Center located in a historically geologicallystable part of the world the earthquake risk would be rated as “veryunlikely,” and hence would not warrant serious attention.

• Possible meteor strike. Again the consequence of such a naturaldisaster would be rated as “catastrophic,” but the likelihood mustsurely be considered to be “very unlikely,” meaning that such anevent need not be included in a Center’s disaster planning.

• Anthrax bioterrorism. Given our heightened awareness of terroristthreats, such risks must be included in a Center’s disaster planning.Certainly many, if not all, US and Canadian IVF Centers under-took such planning subsequent to the anthrax attacks in the USA inOctober 2001.

Tools for risk management

There are two main tools used in risk management, one proactive, theother reactive. The proactive tool is called Failure Modes and Effects

53 What is risk?

Analysis or “FMEA,” and the reactive tool is Root Cause Analysis or“RCA” (see Chapter 7). While FMEA works towards the prevention ofrisk, RCA is used to deal with actual adverse events and troubleshoot-ing. Because both tools are based upon analyzing systems and pro-cesses, understanding the concepts and principles of process mappingare prerequisite to effective risk management (see Chapter 5).

5

Process and systems

We have used the terms “process” and “system” several times alreadyin the preceding chapters. To many they are synonymous, but this isnot true.

A process is defined as a whole series of continuous actions or tasks,or a method by which something is done.

A system is defined as a group of objects related or interacting soas to form a unity, or a methodically arranged set of ideas, principles,methods, procedures, etc.

It can therefore be seen that a system is on a more macro scale than aprocess and, indeed, typically comprises a collection of processes, someof which might occur sequentially while others might occur simulta-neously or in parallel with one or more other processes. At the mostbasic level, a process can be defined as a single, simple sequence, asillustrated in Figure 5.1.

Systems analysis

A system typically comprises several processes, some of which mightrun in parallel, but many of which usually operate serially or insequence, i.e. the output of one is an input to the next.

Systems analysis can be defined as the diagnosis, formulation andsolution of problems which arise from the complex forms of interac-tion in any system (e.g. from computer hardware to corporations) thatexist or are conceived to accomplish one or more specific objectives.The typical use of systems analysis in the IVF Lab is to guide decisions

54

55 Process and systems

Figure 5.1 A generic process.

on issues such as planning laboratory operations, resource use andstaff/environment protection, research and development, implement-ing new methods, educational needs, and clinical service provision.

Before a whole system can be understood, there must an understand-ing of both the individual processes and the environment of extrinsicfactors within which each process occurs. The technique whereby asystem is best analyzed so it can be understood – or at least representedgraphically – is called process mapping.

Process mapping

Process mapping is a means by which you can draw out a process andunderstand it, as a first step towards either simplifying the processbased on that understanding, improving the process or eliminatingunnecessary steps in the process (e.g. recursions). Process mappingrequires that any system or “complex process” be drawn as a flowchart,identifying every step or individual component process in the sys-tem. In our case the system of interest is, on the most macro scale,an IVF treatment cycle. But it is essential that the system be reducedto its most fundamental steps in order for the factors that act uponany (indeed, every) component process to be identified and analyzed.A simple guide to achieving this is to consider each process step asa generic system, one which has inputs to which “something hap-pens” in order to generate the output(s); there must be no lower-level


derivative process(es). Unless this level of simplicity is reached thelikelihood of success of the analysis will be greatly diminished, if notforedoomed.

A component process will probably have both intrinsic and extrinsicfactors affecting its operation:

Intrinsic factors are those which are inherent to the process, i.e.they are the effects that cause or control the process. The mostcommon intrinsic factors in the IVF laboratory are systems toregulate temperature, pH and humidity.

Extrinsic factors are ones that are not inherently involved inthe process, e.g. uncontrolled sources of cooling, toxic vapors,biological variation.

Before we look at the specific tools that are available to perform processmapping, let’s look at a “real world” event that will help illustrate acomplex process: the preparation of a dinner party (see Figure 5.2).A complex menu requires not only a very careful shopping trip butalso the design of an intricate timetable to complete each course at theright time; some courses can be prepared ahead of time (e.g. the sorbet),others are best served immediately (the filled profiteroles), while othersrequire very careful timing so they are ready at just the right time to beserved (e.g. the onion soup and the beef Wellington). Each course hasa very precise series of steps, making up a serial process to complete thedish, but preparation of the dishes requires some processes to run inparallel and processes for different courses to be intercalated betweenothers within the overall timetable.

With the proper organization, so long as the recipes are followedthe meal will be a success. The starter is a wonderful dish, but takes along time to prepare just before serving – at a time when the recentlyarrived guests might want to chat, or while preparation work for othercourses is still to be done. But what if the grill is set too high, orleft too long? – the cheese on the onion soup will burn. If the oven


Figure 5.2 One of our very favourite dinner party menus (see text for explanation).

is malfunctioning and running too hot then the profiteroles will getburned and the beef Wellington will be overdone. If the cook has aheavy hand then several dishes might be too salty for the guests’ tastes.Putting the sorbet out early could result in it melting. Preparing thecheese plate too soon could result in the cheese sweating or going brownat the edges. Just remember the tightly integrated teams that work inmajor league restaurants.


In effect, this is just how an IVF Lab operates on a daily basis: prepar-ing for various procedures while others are in process, many of whichmust be timed according to their specific biological requirements (asopposed to culinary requirements), and not the embryologists’ – orclinicians’ – convenience.

Having now considered the integration of complex systems, let’sreturn to the IVF Lab and look at how processes make up our opera-tional systems.

sperm eggs

"IVF"

embryos

spermpreparation

IVFinsemination

embryoculture

semenanalysis

spermpreparation

oocyteculture

IVFinsemination

spermsample

oocyteretrieval

semencollection

embryoculture

embryoculture

embryotransfer

embryofreeze

embryotransfer

oocyteretrieval

(a) (b) (c)

fertcheck

Day 2assessment

Day 3assessment

Figure 5.3 Diagrammatic representations of IVF as single process. (a) shows it at its sim-plest level, while (b) and (c) reveal progressively greater depths of detail in thesequence of generalized stages in the process.


IVF laboratory systems

As we have already said, an IVF treatment cycle is a system, but it is toocomplex to be analyzed as a single system. In order to be able to analyzethe operation of an IVF Lab we must “zoom in” on what’s going on.Figure 5.3 shows how an IVF cycle can be broken down from a single“black box” process into a series of generalized stages, while Figure 5.4reveals how the laboratory component of an IVF treatment can besplit into a series of procedural steps. Both of these figures illustratehow a more detailed examination of a “system” exposes its lower ordercomponent processes.

However, there are also other dimensions of complexity that arisewhen one goes beyond a simple IVF treatment cycle to include, forexample, the use of previously cryopreserved gametes or embryos(Figure 5.5) or the incorporation of preimplantation genetic diagnosis(Figure 5.6).

Process mapping tools

We greatly benefited from an excellent primer on process mappingfound on the website of John Grout (Grout, 2004); other recom-mended textbooks include Damelio (1996) and Jacka and Keller (2001).The examples shown in Figures 5.3 through 5.6 were illustrated usingthe well-known tool of flowcharts, but there are more sophisticatedapproaches than this available to perform process mapping. Some ofthe more common approaches are listed in Table 5.1, which showshow they differ in their functional attributes. Where temporal anal-yses are important (i.e. the time taken by processes, e.g. the timeto reach certain steps in the process or to complete the process),a more specialized technique called value stream mapping can beused, a tool important in the business development process of “leanmanufacturing.”


oocytes(from OPU)

washedsperm hyaluron-

idase

IVFinsemination

ICSI

mature?

IVM?

IVM culture

DISCARD

3

overnight culture

culture

strip zygotes

fertilized?rescueICSI?

DISCARD

culture

freeze zygotetransfer

Day 2 assessment DISCARD

freeze embryotransfer

Day 3 assessment

freeze culture embryotransfer

DISCARD

Day 5 assessment

freeze blastocysttransfer

DISCARD

2

1

4cryobank

NO NO

YESYES

YES

YES

NONO

ARRESTED / DEGENERATE



Figure 5.4 A map of the ”standard” IVF laboratory process. Off-page connectors relate to:1© the actual oocyte retrieval process; 2© obtaining, analyzing and preparing the

sperm sample; 3© the ancillary process of in-vitro maturation of oocytes; and 4©the process whereby specimens are transferred into the cryobank.


oocytes(from OPU)

washedsperm

hyaluron-idase

IVFinsemination

ICSI

mature? IVM?

IVM culture

DISCARD

3

overnight culture

culture

strip zygotes

fertilized?rescueICSI?

DISCARD

culture

freeze zygotetransfer

Day 2 assessment

freeze embryotransfer

Day 3 assessment

freeze culture embryotransfer

DISCARD

Day 5 assessment


2

1

4cryobank

NO NO

YESYES

YES

YES

NONO

ARRESTEDDEGENERATE


frozenoocytes

cultureoocytes

thaw oocytes 5

5

frozenzygotes

thaw zygotes

5

frozenembryos

thaw embryos

assesssurvival

DISCARD

YESNO

YESNO

assesssurvival

ARRESTEDDEGENERATE

Figure 5.5 A map of the ”standard” IVF laboratory process including the additional complex-ity of inputs of previously cryopreserved oocytes, zygotes or embryos. Off-pageconnectors relate to: 1© the actual oocyte retrieval process; 2© obtaining, ana-lyzing and preparing the sperm sample; 3© the ancillary process of in-vitro mat-uration of oocytes; 4© the process whereby specimens are transferred into thecryobank; and 5© the process whereby cryopreserved specimens are retrievedfrom the cryobank.

oocytes(from OPU)

washedsperm

hyaluron idase

ICSI

overnight culture

culture

fertilized?DISCARD

culture

Day 2 assessmentDISCARD

embryo transfer(for PB only testing)

Day 3 assessment

freeze

culture

embryotransfer

DISCARD

Day 5 assessment


DISCARD

2

1

4

YESNO

ARRESTEDDEGENERATE

1st PBbiopsy

PB forgenetic analysis

6

2nd PBbiopsy

PB forgenetic analysis

6

ARRESTEDDEGENERATE

blastomerebiopsy

blastomere(s) forgenetic analysis 6

genetics OK?YES

DISCARD

freezeNO

genetics OK?YES

DISCARD

freezeNO

culture

ARRESTEDDEGENERATE

trophectodermbiopsy

cells forgenetic analysis

6

4

4

culture

genetics OK?YES

DISCARD

freezeNO

4

4

IF NOBIOPSY

IF NOBIOPSY

Figure 5.6 A map of the “standard” IVF laboratory process including the additional complex-ity of pre-implantation genetic diagnosis/screening. For clarity, rescue ICSI andIVM have been omitted, as well as the inputs of previously cryopreserved oocytes,zygotes or embryos (see Figs. 5.3 and 5.5). Off-page connectors relate to: 1© theactual oocyte retrieval process; 2© obtaining, analyzing and preparing the spermsample; 4© the process whereby specimens are transferred into the cryobank;and 6© the processes involved in actually performing the genetic testing on thebiopsied material.


Table 5.1 Comparison of the functional attributes of some of the morecommon approaches to process mapping. “Strong” denotes that the attribute isa strong feature of, or explicit in, the approach; “weak” indicates that theattribute is provided in an implicit or weak way; and a blank denotes that theattribute is not provided by that approach. See text for further discussion of thevarious approaches. Modified from Grout (2004)

Attribute Flowcharting Top–down Swim lanes IDEF0

Level of detail Strong Strong StrongHierarchical linking

between mapsWeak Weak Weak Strong

Multiple types of flow Weak Weak StrongOrganizational structure Weak Weak StrongUse of icons Strong StrongUse of logical operators Strong Strong Weak

Flowcharts

Flowcharting is the most widely known approach to process map-ping and has probably been used by everyone working in IVF labs atsome time or another. While flowcharts can be accomplished usingtext only, with various statements interconnected by arrows, formalflowcharting uses a series of symbols that have specific meanings (seeFigure 5.7), although not everyone is aware of these conventions. Theamount of detail included in a flowchart is completely at the authors’discretion, and large flowcharts can be split over multiple pages usingthe off-page connector symbols. While there is no specific conven-tion in flowcharting to represent any hierarchical structure, greaterdetail of sub-processes can easily be illustrated using subsidiary “child”flowcharts on separate sheets provided that a logical naming or num-bering scheme is used to define the linkages (see “IDEF0,” below).Multiple flows can obviously be shown in a flowchart, but explanationof what the different flows mean is the author’s responsibility.

Various specialized software packages are available for drawingprocess maps, and flowchart symbols are available in many popu-lar word processing and spreadsheet programs. Probably the mostwidely acclaimed – and economically priced – flowcharting software


process data / input

alternate process document

predefined process manual input

decision terminator

storage direct accessstorage

punched tape magnetic disk

microfilm punched card

internal storage sequentialaccess storage

off-pageconnector

sort

collate

orconnector

summingconnector

connector

off-line storage

manual operation

on-line storage

preparation

delay

Figure 5.7 Standard symbols used in drawing flowcharts.

package is SmartDraw (SmartDraw.com, San Diego, CA, USA:www.smartdraw.com).

Top–down process maps

These process maps list the main process steps horizontally, with eachset of sub-processes listed vertically below the main process steps (seeFigure 5.8). Because this approach has only very minimal graphi-cal content it is readily usable in text-only situations, as shown inFigure 5.9. From this illustration it is obvious that a properly-writtenSOP (standard operating procedure) in a lab manual is, in reality, nomore – or less – than a top–down process map.


Decision touse ICSI

Obtainoocytes

Preparesperm

PerformICSI

Fertilizationcheck

Documentoutcome

1 2 3 4 5 6

Subprocesses(a) Select treatment plan

(b) Record in chart

(c) Advise lab

Subprocesses(a) Oocyte retrieval

(b) "Egg search"(c) Wash oocytes(d) Culture oocytes

Subprocesses(a) Obtain sperm(b) Semen analysis

(c) Prepare sperm

Subprocesses(a) Prepare ICSI dish

(b) Prepare sperm drops

(c) Strip oocytes(d) Inject sperm

Subprocesses(a) Examine under stereo microscope

(b) Examine on inverted microscope

(c) Put zygotes into culture

Subprocesses(a) Complete lab paperwork

(b) Tell results to nurse / doctor

(c) Enter data into computer

(d) Advise patient

Figure 5.8 The use of ICSI as an example of a top–down process map.

Swim lanes

A swim lane process map is similar to a standard flowchart, exceptthat a grid is superimposed over the flowchart. The analogy is that ofswimmers in their lanes in a pool. While the vertical axis or columnsof the grid shows the chronological sequence of tasks, the grid rowsrepresent the various “participants” (i.e. organizations, departments,functional areas, locations, or individuals) involved in the process. Amajor strength of swim lane process mapping is that it shows who isinvolved at each stage in the process; team or joint activities can beindicated by drawing boxes around multiple tasks (see Figure 5.10).Because the participants can be defined at a macro or micro scale,on the basis of organizational, functional or human participants, e.g.roles or actual people, the process can be mapped at whatever level ofdetail might be desired. If a swim lane process map is constructed usinglocations then one can also map a process in “geographic” terms, forpurposes such as looking at the way biological material – or people,or paperwork – move around (hopefully through) the clinic or thelaboratory.

A great benefit of a swim lane process map is that inter-participantexchanges – events that cause the process flow to change “lanes” – canbe explicitly stated, e.g. “hand-offs” between the lab and nursing, orlab and physician. This feature alone makes the swim lane approachvery useful in IVF where mistake-proofing is such an important issue.


PRE-TREATMENT ARRANGEMENTS

POST-TREATMENT PROCEDURES

LABORATORY PROCEDURES1. Obtain oocytes.

2. Prepare sperm.

3. Perform ICSI.

4. Fertilization check.

(a) Select treatment plan for patient.(b) Record decision in patient’s medical record.(c) Advise IVF Lab that ICSI will be used.

(a) Complete lab paperwork.(b) Advise nurse / doctor of results.(c) Enter data into the computer.(d) Advise the patient.

(a) Oocyte retrieval procedure.

(b) Perform “egg search” on follicular aspirates.

(c) Wash oocytes.

(d) Place oocytes into culture.

Details of technical sub-processes go here




(a) Obtain sperm sample.

(b) Analyze sperm sample.

(c) Prepare sperm.




(a) Prepare ICSI dish.

(b) Prepare sperm drops.

(c) Strip oocytes.

Inject sperm.




Details of technical sub-processes go here(d)

(a) Examine under stereo microscope.

(b) Examine on inverted microscope.

(c) Place zygotes into culture.




Figure 5.9 An alternate textual view of the top–down process map shown in Fig. 5.8. Thesimilarity between this illustration and the format of a laboratory standard oper-ating procedure document is striking.


Figure 5.10 The IVF procedure, from oocyte retrieval to embryo transfer, in the form of aswim lane process map that illustrates the movement of the biological materialand related information through the organization. Swim lane process maps fora process can be drawn in various ways to emphasize different aspects of theprocess, entirely according to the user’s needs and intentions.

IDEF0

Integration Definition (IDEF) is a functional modeling methodol-ogy originally developed by the US Air Force to help programmersdevelop complex software systems. It comprises five separate model-ing methods with IDEF0 (pronounced eye-deff-zero) being for functionmodeling. Detailed specifications for IDEF0 were published by the USGovernment’s National Institute of Standards and Technology in 1993as FIPS publication 183, by which name the source document has sincebeen known (National Institute of Standards and Technology, 1993).Although FIPS PUB 183 is a 128-page document, it is quite (surpris-ingly?) readable and allows anyone to begin using the technique. The


ACTIVITY#

Mechanism(s)

Input(s) Output(s)

Control(s)

Figure 5.11 A generic illustration of a single element in an IDEF0 process map (NationalInstitute of Standards and Technology, 1993).

US Standard was taken as the IEEE Standard for Function Modelingin 1998, and that Standard (IEEE 1320.1-1998) has officially replacedFIPS PUB 183 as the definitive reference for IDEF0 flowcharting (IEEEStandards Association, 1998).

IDEF0 is a totally hierarchical approach to detailed flowchart-basedprocess mapping that uses a completely defined graphical scheme andsyntax. While the details of the IDEF0 standard are beyond the scopeof this book, the basic element used in IDEF0 flowcharts is shown inFigure 5.11. Starting with a single parent map that has only one ele-ment, the technique uses multiple, hierarchically-linked process mapsto display multiple flows in great detail. The IDEF0 specification limitsthe number of tasks that can be displayed on any single map to beingbetween four and six, and if more steps need to be shown then this isthe indication that the process needs to be broken down further andmore linked child maps added. Precise numbering of process steps ortasks is at the heart of IDEF0 and allows ever-increasing detail to beincluded in the subordinate levels of child maps. An IDEF0 process mapis usually accompanied by a table of its levels, known as a “context tree.”

An example of an IDEF0 process map is shown in Chapter 7(Figure 7.5).

Building a process map

Irrespective of the process mapping technique employed, the follow-ing outcomes can be anticipated from a successful process mappingexercise:


• increased understanding of the process;• increased consensus among those who created the map about what

the process is and how it operates;• increased “buy-in” to the need for change; and• an environment where ideas for process improvement can be gener-

ated more rapidly.

When a process mapping exercise is complete, knowledge that hasbeen hidden away inside the heads of many individuals within theorganization has been unlocked and is now available throughout theorganization. Everyone feels empowered and more confident in theirwork. Corporate knowledge can be passed on to the next generationwithout any gaps or misconceptions.

Process mapping usually involves creating a map of the system asit currently operates: the “as-is” map. However, this map should notbe created in isolation of the actual process, for example, by the LabDirector sitting in an office working from the Lab Manual and otherclinic documentation such as the Clinical Policies and Procedures Man-ual, the Nursing Procedures Manual and the Administrative SystemsManual. The real power of process mapping comes from mapping theprocess as it is actually performed, by the people who are actually doingit. Another concept that we must consider here is whether the pro-cess mapping exercise should be performed in a “top–down” manneror using the “bottom–up” approach. This use of the expression “topdown” should not be confused with the specific technique of top–downprocess mapping.

Here, the “top–down” approach is where the Lab Director, or per-haps a group of managers, draws up the process map, or brings inan external process mapping consultant to do it for them. While thismight seem a logical approach, it has several drawbacks.

• As an outsider, the process mapping consultant can only movethrough the activity logically in a linear manner and, being a non-expert in the technical aspects of the process, the total amount oftime required for capturing all the information about the processbeing mapped is greatly increased.


• By relying on an outsider the IVF Center’s staff are effectively side-lined and their willingness to “buy-in” fully to the findings andrecommendations is consequently diminished.

• The people actually performing the tasks in the process being mappedare denied the opportunity to assess how they, as individuals, aredoing their jobs, and hence the chance for them to improve as indi-viduals. A valuable opportunity for self-education is lost.

• Because external consultants are only on-site periodically, the entireprocess can become fragmented, repeatedly stopping and starting.This makes it difficult to create and build momentum for continuousimprovement (see Chapter 3).

• If the organization is pursuing a program of total quality improve-ment, then there will be many processes requiring mapping,and the continual use of external consultants becomes not onlyvery expensive, but can become the rate-limiting step in theprogram.

The bottom–up approach involves staff at all levels directly in the map-ping of processes, thereby preventing the adverse effects described forthe top–down approach. It will also allow for faster completion ofprojects which then encourages their repetition, and in turn fostersthe culture of continual change. The end result is therefore a regularseries of improvements for the Center, rather than a future of con-tinued major effort – and a stream of outside management experts –punctuated by an occasional successful achievement. However, manypeople working in IVF Centers have no experience of process analy-sis, much less process mapping, and the resultant lack of confidencecan be a major hurdle to instituting this essential part of quality andrisk management programs. So how should an IVF center go aboutdeveloping process maps for its activities? Clearly expert assistance isessential, which means that external consultants are extremely useful –if not essential – in the early stages. But they should be brought in asresource people to help staff work through the first few process mappingexercises, and as educators to train staff in performing process


mapping, so that the organization will be able to undertake a con-tinual series of process mapping exercises using its own staff.

Therefore, a process mapping exercise should proceed as follows.

1. Convene a process mapping team that includes everyone who isinvolved in the process, or at least a representative of each area orgroup of workers involved in the process.

2. Have everyone who is involved in the process independently reviewa draft map to identify points of disagreement.

3. The list of points of disagreement is then reviewed by the mappingteam to identify errors in the draft map or, more likely, areas whereoperator-dependent variations have arisen.

Completing the whole mapping process is extremely importantbecause incremental improvements (genuine or otherwise), “copingstrategies” or “work-arounds” have often been introduced, frequentlywithout following the proper review and approval procedure, so thatoften no one person really knows how the entire process is actuallyperformed. Anyone who has worked in an IVF lab (or any other lab,for that matter) will know of examples where unauthorized changes(“improvements”) were incorporated by individuals to facilitate theirown work, that were not optimal – and sometimes distinctly deleteri-ous – to the overall process. All too often the SOP in the Lab Manualis not, in fact, the standard operating procedure for a process, a sit-uation that can lead not only to sub-optimal outcomes, but also toincreased opportunities for mistakes to occur. Furthermore, processmapping, perhaps especially swim lane mapping, plays a vital role in acomprehensive risk management strategy.

For process improvement, the three steps described above should befollowed by:

4. Identify those events in the process map that are the source of eithersub-optimal operational performance or outcome, or risk.


5. Create a “to-be” map, showing the team’s consensus as to how theprocess should actually operate after the improvements have beenimplemented.

Breaking down “silos”

In business management jargon, a “silo” is a single location functionalsystem that is hard to integrate. It might be an operational silo (e.g.the finance department) or a location silo (e.g. a subsidiary facility).Business processes are streams of activity that flow across functionalboundaries, e.g. sales, marketing, manufacturing, distribution, opera-tions, finance, legal, etc (see Table 5.2). Such processes can be describedas being fragmented, scattered across so-called functional silos, wherepeople in a given silo rarely, if ever, have occasion or the opportunityto study their work in the context of the larger business process thattheir function supports. For many people, their company’s businessprocesses are literally unknown quantities. It is often said that the onlysilos that should exist are those for storing grain (since ICBMs are also“user unfriendly”!).

In the IVF world, how many people working at reception, or as nursecoordinators – or even as physicians – actually know what happens inthe IVF Lab? The lack of awareness, caused by lack of knowledge, isone of the greatest sources of operational difficulties that the authorsencounter when reviewing IVF labs for accreditation preparedness orwhen a TQM program is being contemplated. In one Center, one of uswas responsible for initiating a program whereby all employees workingin reception, finance and nursing were introduced to what went on inthe IVF Lab. It was found to be such a powerful tool in understandinghow the Center’s systems worked and could be improved, and where“hand-over” errors could occur – essential aspects of preparing for theinitial survey for a new accreditation scheme – that this education hassubsequently been formalized into the orientation procedure for all


Table 5.2 Departments or disciplines that can create “silos” that control anorganization’s operational activities, including explanations of these sources ofcontrolling factors in the generic industrial / commercial and IVF Center contexts.Within each of these functional areas the people will generally have a good graspof that part of the organization’s processes and systems to which their workcontributes, but often they will not have much knowledge (and lessunderstanding) of what happens in other departments. Achieving understandingor how what they do fits within the company’s operations, and how their activitiesinterface with those of other departments is vital to creating a qualityorganization – the “breaking down of silos”

Department/discipline Generic explanation In the IVF Center context

Manufacturing Where the company’s“product” is created.

The actual provision of diagnostic andtherapeutic services, includingmedical, scientific, nursing andcounseling activities.

Marketing and sales Making the company’s(potential) customersaware of its products andselling them to thosepeople.

Usually only exists in large “corporate”IVF organizations, but all IVF Centersneed to be aware of how they get theirreferrals.

Distribution Getting the products to thecustomers.

Not really an issue since the patients cometo the clinic. However, in largecountries with a low overall populationdensity, e.g. Australia, “outreach”programs have been used verysuccessfully to provide services topatients living outside majormetropolitan areas. “Transport IVF”would also come under this area ofactivity.

Operations The day-to-day managementof the company.

Coordination of the IVF Center’sactivities, including patientappointments, nurse coordinators fortreatment cycles, medical records, andscheduling procedures.

Human resources Recruiting, hiring, trainingand supporting thecompany’s personnel.

Of vital importance to creating andmaintaining a quality organization –but these activities must be closelyintegrated into building amultidisciplinary team.

(cont.)


Table 5.2 (cont.)

Department/discipline Generic explanation In the IVF Center context

Information technology Computer systems andsoftware to support allareas of the company’sactivities.

Just the same functions, although perhapsmore important than usual for mostcommercial businesses in order toachieve efficient management ofpatients and data handling and analysisfor QC/QA, as well as regulatory,purposes.

Finance Accounts receivable,accounts payable andfinancial planning for thecompany.

Just the same, although perhaps not aswell understood as they should be insmaller IVF Centers (especially in thepublic sector).

Legal / regulatory Self-explanatory. Self-explanatory, but perhaps of evengreater importance than in many areasof business due to the growingregulation of IVF by governmentagencies and the expansion ofaccreditation.

new employees there. We were so impressed by the success of this pilotscheme that we now recommend it to all our clients.

Process control and process analysis

To analyze a process for either quality management or risk manage-ment purposes, one must have knowledge of the normal parameters ofits operation. This requires both reliable information on the historicalperformance of the process and knowledge of its inherent variability.It is important to think like a statistician – scary as that might seem!Indeed, TQM (of which risk management is one element) requiresthat scientific method and statistical thought pervade all consider-ations of process control, performance, improvement and deliverysystems.

The technical aspects of undertaking process control will be providedand discussed in Chapter 7. However, examples of process control in the


IVF laboratory have been described previously (Mortimer et al., 1995;Mortimer, 1999; Murray and Mortimer, 1999) and are slowly beingadopted by more and more labs (Wikland and Sjoblom, 2000; Mayeret al., 2003). To decide if a process is sub-optimal, and hence a can-didate for improvement, we need knowledge of its capabilities – inother words we need one or more benchmarks for its performance (seeChapter 10).

Identifying controlling factors

Process analysis allows the identification of all the factors that affect theprocess being analyzed. When we write our lab manuals, the SOPs mustinclude sufficient detail for anyone with basic biology lab competenceto perform the procedure. The detail must include the precise mannerin which the technique is to be carried out. This is especially importantin procedures where variations in the technical method can allow eitherreduced control over factors that can affect the process, or even allow theincursion of extrinsic factors that would otherwise have been excluded.Knowing exactly what must be specified in an SOP to achieve thenecessary level of correct technique and standardization, as opposedto being unnecessarily picky, therefore requires that the author of theSOP must understand not just how the process in question is regulatedby biology, chemistry and physics, but also how it might be impactedby extrinsic factors or the lab environment, or ergonomics.

Therefore, we are obliged to consider other disciplines such as engi-neering. Indeed, if we are to be able to identify all the factors that affecthow our processes and systems operate in the IVF lab we must considerall of the following.

• Biology• Biochemistry• Chemistry• Physics• Engineering


• Ergonomics• Architecture• Process mapping

Knowing “why,” not just “how”

Clearly, in order to write a good SOP that will include the necessarytechnical/procedural detail, plus identifying and controlling the otherfactors that can impinge upon the process, one must know far morethan just how to do the procedure. Knowing why we do things incertain ways – and also perhaps why we don’t do them in other ways –requires more than a simple technical ability to do the task in question.This is why a proper training program for anyone who is to work in anIVF lab must include both the “how” and the “why.” Proper trainingis vital and must be provided within an encompassing framework ofeducation.

Of course innovations come from people trying different ways ofdoing things, but the material we handle in an IVF lab is far too preciousfor empiricism: “suck it and see” has no place in the IVF lab. Learn-ing from history is paramount. If someone has already tried doingsomething in a particular way and found it to be unsuccessful, or sub-optimal, then there is no value in it being tried again without someother change having been introduced that might alter the course ofhistory the second time around. For example, we know that successfulICSI requires not only that the sperm midpiece must be damaged priorto injection but also that the oolemma must be broken by sucking backon the injection pipette prior to expelling the sperm from the injectionpipette. This information came from the many unsuccessful attemptsby many of the ICSI pioneers, and these are technical details that everyICSI practitioner was taught at the outset of their training since earliestdays of ICSI’s rapid spread outside Brussels. Only someone lacking thenecessary knowledge and training, or without any access to the litera-ture (a difficult scenario to imagine if the lab has the resources to buyan ICSI work station) would perform ICSI without these two “tricks.”


Unauthorized variation of methods is a persistent problem in somelabs, and is due to either poor training of new scientists or a failureto maintain control over one’s staff. Consequently, the problem ariseseither from a Lab Director who fails to (re)train new staff, or a failure torecognize – and eradicate – sloppy or lazy behavior by a staff member.And any scientist who will not respect and follow SOPs has no place inan IVF lab.

Tools vs. solutions

A common problem one sees when people are trying to develop newor better processes or systems is confusion between something that isa tool, and something that represents a solution. To illustrate this, let’sconsider the whole issue of sample identification and verification inthe IVF lab, a vital process that extends throughout the entirety of thelab contribution to an IVF treatment cycle.

Obviously each specimen of gametes or embryos must be identi-fied by a label of some description. But because we cannot label thebiological material directly we must label the vessel that it is in, e.g. asemen jar, a centrifuge tube or a culture dish, using perhaps somethingthat is written directly onto the tube or dish, or onto a self-adhesivelabel that is affixed to the tube or dish (see Figure 5.12). This labelneeds to identify the source of the biological material inside the tubeor dish (its “owner” or “provenance”) as well as, most probably, thestatus of the sample or its stage in the IVF process (e.g. fertilizationdish or cleavage dish). The former employs alphanumeric text, andthe latter typically either uses the same form of expression or perhapsa colour code. But the “label” is only a tool, the reliability of its usedepends on the system within which it is used. For example, if thepatient’s name is Smith (or Jones, Dubois, Patel, Cheng, Ng, etc) thenit is quite likely that specimens from two different people of the samename might be in the lab at the same time. This is why we must usemore than just a family name, and why accreditation schemes typi-cally require that any container for a specimen, or preparation made


Figure 5.12 Illustrations of the use of labeling in the IVF Lab. Note that each vessel that willcontain patient gametes is identified by at least two specific pieces of informa-tion, e.g. name and sample/procedure number. Dates are routinely written inYRMODA format to avoid any possible confusion between ”/” and the number 1or the differences between British and American ordering of the date and month.For the aliquot of ”Sperm Buffer” (Fig. 5.12A) the labeling includes the productname, its lot number and expiry date, as well as storage and usage conditions, asper accreditation standards. On the label of the sperm preparation tube (B) thereference numbers indicate both the lab reference for the andrology specimen(A04-0682) and the oocyte retrieval case number (R04-0179). For the dishesshown as C and D the bottom of the dish must also be labeled, for exampleeither on its edge for C or by backward writing on the base for D. The labels onthe dishes shown in E and F are actually attached to the dish bottom but arereadable through the side of the lid. All four dishes are labeled with the name,the oocyte retrieval case number, the date and the purpose of the dish.


from a specimen, must be labeled with at least two “identifiers” sothat the likelihood of ambiguity is eliminated. A family name and firstname are obviously inadequate, so a name + date of birth, or name +sample/cycle reference number are needed. The date is clearly not usefulfor unique identification purposes either, but it is invariably added asfurther information to facilitate identification of information withinthe specific patient’s record.

With such labeling in place, a sample can be identified unambigu-ously – and so the only logical conclusion that can be drawn when sam-ples are misidentified is failure of the system within which the label wasused. The most common cause of such failures is human error, eithermisreading hand-written text or actually failing to perform the IDcheck. The need for diligence and vigilance is compounded by the factthat gametes and embryos must be transferred from one container toanother throughout the IVF process. Figure 5.13 illustrates the num-ber of times that the biological materials involved in an IVF treatmentare transferred between vessels (at least 9 times during a basic IVFtreatment cycle), with each change representing an opportunity forerror. While everyone working in an IVF lab is aware of the need toperform all these ID verification checks, the system has been known tobreak down, sometimes with terrible consequences. Occasional errorsare made in matching IDs, but the major source of problems here isthe failure to verify IDs when the operator is busy. For such reasons,and in response to some events that had very high public profile, theBritish HFEA instituted a requirement for a second person to verifythe ID check and “witness” the operator performing the process.

But “double-witnessing” is a tool, not a solution. Daniel Brisonsummed up the essence of good laboratory practice very succinctly as“concentration and responsibility” and has presented cogent argumentsquestioning the efficacy of double-witnessing as a safety net for IVF labs(see below) and concluded that it has the potential to do more harmthan good (Brison, 2004). He contends that double-witnessing createstwo major problems that are in direct conflict with the principles andaims of good laboratory practice (GLP) and TQM in the IVF Lab.


Semen Jar OPU tubes

Gradient tube(s) “Egg search” dishes

Wash tube Wash dish

Hyaluronidase dish

Wash dish

ICSI dish

Sperm prep tube

SP

ER

M P

RE

P

FE

RT

CH

EC

KIV

F

ICS

I

OO

CY

TE

RE

TR

IEV

AL

Fertilization dish

Fertilization dish

Fertilization dish

Stripping dish

Wash dish

Cleavage dish D1–D3

Container

ET dish

ET dish

Freezing solutiondishes

Thawing solutiondishes

Culture dishStraws

Catheter

Catheter

PATIENT*

PATIENT*

KEYdenotes a change of container between processesdenotes a change of container within a processdenotes patient identity verification and witnessing

Movement between each requires verification.

Day 0

Day 1

LaterDay 2et seq.

PATIENT*

CRYOBANK

FR

EE

ZIN

G

TR

AN

SF

ER

TR

AN

SF

ER

TH

AW

ING

PATIENT* PATIENT*

Extended culture dish (D3–D5)

Wash dish

same

container

Figure 5.13 A map of the IVF process illustrating all the occasions when gametes or embryosare moved from one container to another. Each container is defined by a genericname and a change of vessel action is shown by an arrow. Heavy arrows showchanges of container between general processes and light arrows show changesof container between steps within a process. PATIENT* denotes an event wherea patient’s identity must be verified and witnessed.


1. The sheer number of times that an embryologist must interrupttheir own work to go over and witness someone else’s has the effectof destroying the embryologist’s concentration on the task-at-hand,increasing the risk that (s)he will make other mistakes not relatedto specimen identity, thereby compromising the quality of care forthe oocytes or embryos. The use of an unskilled dedicated witnessprobably creates more issues than it solves because the witness needsto comprehend the task being witnessed – as well as the potentialsources of error and the possibilities for fraud.

2. Devolving responsibility for the task being performed from theembryologist performing a task (the “primary operator”) to thewitness of the task can have significant psychological impacts uponthose performing laboratory tasks. By requiring double-witnessing,the primary operators (regardless of their experience and seniority)are being denied their usual responsibility for accurate specimenhandling, much of which now rests with the witness. This perceivedloss of responsibility for laboratory paperwork can cause a formof risk compensation behavior, a well known phenomenon in thefield of injury prevention, whereby the introduction of preventativemeasures causes a reduction in the sense of personal responsibilityfor one’s actions or the tasks being performed.

While double-witnessing seemed an excellent idea at the time, it mustbe remembered that “good” IVF labs have in fact, always focussed onspecimen identity verification, and that the simple requirement for asecond person to sign a lab form does not ensure that positive ID hasbeen verified. Certainly, in some IVF labs, especially ones that haveperhaps a sole embryologist, or where the work hours are protractedin order to handle a high caseload, such a witnessing procedure is notalways possible – or is just not always done. Absolute enforcement –and documentation – of such ID checks will require an objective andobligatory technological solution.

In recent years there has been much talk of the use of bar codes(Figure 5.14A) as a “more foolproof” form of identification for


Figure 5.14 Examples of advanced labeling tools. A shows a Brady Systems TLS 2200® Ther-mal Labeling System used to print self-adhesive labels that can be attached to avisotube (B), or an ID rod for the CBS High Security Straw system (C) or wrappedaround a cryovial (D); note that the last type of label is self-laminating. E showsseveral radio frequency identification (RFID) devices, smaller ones of which canbe attached to objects such as cryovials, while F shows an RFID encased in glassof the sort used to “tag’’ pets, although similar devices can also be incorporatedinto ID plugs for traditional cryo straws (G). Images A through D are courtesyof Brady Worldwide Inc (Milwaukee, WI, USA) while images E through G weregenerously provided by EPCoT Systems Ltd. (Pinner, Middlesex, UK).


samples – but a bar code is just another tool, it is not a solution. Peoplecannot read bar codes, which requires that they be used in conjunc-tion with some device, and while the device might be infallible in itsreading of the bar code, the actual act of scanning the bar code mustbe performed – which is open to the same risk of failure as someonenot reading any other sort of label. A bar code can be transformedfrom a tool into a solution when the need to perform the scan is inte-grated into some other expanded system, for example, where the barcodes must be scanned in order to enter specimens into a process (e.g.samples being entered into an autoanalyzer machine). Now one cansee how including a bar code on the label can help us build a systemthat is less prone to failure – but it is hard to imagine anyone beingwilling to rely on a bar code label as the sole form of identification for aspecimen, and so bar code labels will inevitably be used in conjunctionwith human-readable text.

The use of radio frequency identification devices (RFIDs) is anemerging technology that has seen widespread use in many indus-tries (see Figure 5.14B). The use of such devices as tools to identifyvessels containing gametes or embryos during the IVF process hasenormous potential, especially in regard to the auditing of cryostoragetanks.

A further dimension of using such technological tools is the abilityto extend the monitoring of processes to their control and regulation.If a bar code and scanner, or an RFID device and reader were integratedinto a software system that not only logged the operator’s ID, along withthe date and time of the event, but also verified the time window of theevent in relation to the biological stage of the IVF treatment as well asthe “authority” of the operator (e.g. only allowing an ICSI process to beperformed by a scientist who had been authorized by the Lab Directoras having the required competence), then we would have a truly use-ful technological solution to ensuring that the IVF lab was operatingproperly. Such a system is, in fact, being developed by the UK companyEPCoT Systems Ltd. (Pinner, Middlesex, UK; www.epcot.co.uk), andhas great promise for the near future.


Who needs to understand all this?

While it might seem that we are obsessed with details, the need foreveryone in the IVF lab to understand processes and systems is integralto how every IVF lab operates. This is not for management control orconvenience, but is a matter of having to respect the biology that con-trols the entire IVF process. All embryologists know and understandwhy an oocyte retrieval must take place 36 hours after the hCG trigger,and understand how that determines the time window for giving trig-ger injections – although how many remember or have experience ofhaving to do oocyte retrievals in the middle of the night as a result ofa patient’s spontaneous LH surge in the days before GnRH agonists?Similarly, the fertilization check has to be performed within a spe-cific window on Day 1 that is 17–20 hours post-insemination, whichimpacts the timing of inseminations on Day 0.

But going beyond this, there are processes such as the design – andperformance – of procedures such as embryo freezing and thawing toadd or remove cryoprotectant while avoiding exceeding the embryos’critical volume limits, or exposing cleavage stage embryos (or oocytesor zygotes) to the common permeating cryoprotectant, propanediol,at 37 ◦C, at which temperature it is toxic to them.

More illustrations and worked examples of how understanding sys-tems is integral to the development or selection of methods will begiven in Chapter 11.

The benefits of process mapping

The technique of process mapping is fundamental to both trou-bleshooting and risk analysis. Its application does not require any elab-orate software, just clear thought and a pencil and paper. Without adetailed process map it can be difficult for someone who is less experi-enced, or less knowledgeable, to troubleshoot a problem, or completea comprehensive risk analysis. But, by constructing a proper processmap, factors that are intuitive to a more expert or knowledgeable person


can be revealed – since it can be considered that the more expert orknowledgeable person recognized the factor(s) as important by virtueof having created a mental process map.

This use of as-is and to-be process maps is integral to the Plan–Do–Check–Act or PDCA Cycle and implicit in performing both a FailureModes and Effects Analysis or a Root Cause Analysis (see Chapter 7).

6

Making it work

So how do we go about using the principles, techniques and toolsdescribed in the preceding chapters to set up quality managementand/or risk management systems in an IVF lab? As we have alreadysaid, in truth these goals cannot be limited to the IVF lab, they mustinvolve the entire IVF Center, all its operations and all its personnel.However, for the purposes of this book, we can consider some specificareas that are pertinent to the lab that will illustrate how they areinherent to proper lab management.

Methods design and selection

The same principles apply whether we are designing a new (or revised)method ourselves, or selecting one of several variant methods thatexist in the literature. When someone in Industry wants to have some-one make or build something, or perform a task, or provide a servicefor them, they establish a comprehensive set of criteria specifying allaspects of what is to be done or provided. These specifications areoften described as the “user requirement specification” or URS, andestablish the detailed framework within which the work will be done.Creating such specifications is, in reality, a universal principle that can –and should – be applied whenever one individual or organization issupplied a product or service by another.

Particular matters relating to the provision of services by anotherorganization, e.g. estradiol assays by an external endocrine assay lab,are discussed in the following section on “Third-Party Services.” Inthe present section we will consider the creation of URSs for internalprocedures, particularly as they relate to the design or selection of

86

87 Making it work

the methods used to perform the component procedures in the IVFprocess. Consideration of the details of how a procedure (or “task”)will be performed is discussed later in this Chapter.

There are many facets of a method that must be considered to ensurethat the “best” one to meet particular needs is selected. At the sim-plest level, we need to consider the whole procedure for which we aredesigning or selecting the method as a process which comprises inputsand actions and generates one or more outputs. Therefore in creat-ing a URS we must include the parameters listed in Table 6.1 in ouranalysis.

It is only after all these parameters have been identified, defined andunderstood that you will be in a position to make a sound judgementon the suitability of a method, or whether one technique or protocolis better than another. Then, before implementing the method in yourlab, you need to write a comprehensive protocol for the method, cover-ing every aspect, to ensure that the desired outcome will be achieved –and that undesired adverse factors are eliminated as far as techni-cally and/or humanly possible. Writing, and using, lab protocols – or“standard operating procedures” (SOPs), as they are becoming morecommonly known – is considered later in this chapter.

“Third party” services

Probably the most common service that an IVF lab obtains fromanother organization or “third party” is estradiol assays for monitoringstimulations. While the supplier – usually a hospital or private pathol-ogy lab – will typically be a well-established organization providing awide range of services to many customers, any accreditation or qualitymanagement system will require the establishment of a formal servicecontract between the IVF lab and the service provider. It is not suffi-cient for the IVF lab to just assume that, since the pathology provideris such a big organization, they “must be doing things right.” The samerelationship is true for whoever services the equipment in the IVF lab,whether it be laminar flow or Class II cabinets, microscopes, or other


Table 6.1 Parameters to be considered when creating a “user requirementspecification” or URS for a lab method that is to be either designed from scratch,modified from an existing technique, or selected from among a number ofalternate techniques described in the literature

Parameter Comments

Define the overall concept of themethod.

Look at the method as a process: define its inputs, actions andoutputs.

What is the purpose of the method? Define the intended purpose of the method as well as those output(s)that can be used as indicators of its performance: e.g. for ICSI theexpected proportions of 2PN zygotes that are produced, as well asthe rate of oocyte damage during the microinjection procedure.

What is the desired (or required)level of performance of themethod?

Establish minimum acceptable levels of performance as well asbenchmark levels: e.g. fertilization rate by IVF or ICSI; rapidcleavage rate; proportion of zygotes that reach blastocyst on Day 5.

What are the factors upon whichthe method is dependent?

These are the factors that control how the method operates, theyinclude technical aspects as well as the underlying biology,chemistry, physics, etc, of the process: e.g. temperature control, pHcontrol, centrifugation force.

What are the necessarycharacteristics of equipmentused to achieve the process?

Define the features of any equipment that will be used to perform theprocess, including safety as well as operational performanceaspects: e.g. Class II cabinets compared to laminar flow cabinets;the temperature and gas (CO2) stability of an incubator; the needfor swing-out rather than fixed-angle centrifuge rotors – or theneed for sealed buckets to avoid aerosol contamination should atube break during centrifugation.

What are the necessarycharacteristics of any reagentsrequired for the process?

Define the required features of any reagents, including culture media,that are needed to perform the process, e.g. HEPES vs. bicarbonatebuffering; minimizing the risk of xenologous contamination.

What are the potential sources ofadverse outcome of the method?

These are any factors that can impede the achievement of theintended output(s), including consideration of the quality of theoutcome: e.g. failure to control temperature or pH withinnecessary tolerance ranges.

What are the potential sources oferror in the method?

These are primarily the opportunities for human or technical errorto impact upon the correct performance of the method and theachievement of the desired outcome (and its quality).

What are the technicalrequirements for anyobservations to be made duringthe method or in determining itsoutcome?

What instrumentation is required for making the observations, andwhat are the operational and performance characteristics that arerequired of that instrumentation: e.g. type of microscope optics ormagnification (Hoffman or Nomarski optics vs. phase contrast forICSI).

89 Making it work

Table 6.1 (cont.)

Parameter Comments

What is the “uncertainty ofmeasurement” for anyobservations that are to be madeduring the method or indetermining its outcome?

Understand the inherent reliability of observations, counts, etc.Define the required accuracy and precision of observations thatare to be made. For example, the numbers of sperm counted in amotility or morphology assessment; or the need for replicatedeterminations.

What special competence isrequired for the method?

Identify if any special skills or competence are required(a) to perform one or more of the technical components of the

procedure: e.g. expertise in ICSI or embryo biopsy; or(b) to perform an observational assessment correctly: e.g. accuracy

and precision of sperm concentration, motility or morphologycounts.

Are there any special training oreducational requirementsnecessary for correctperformance of the method?

Identify if any special licensing is required to perform one or more ofthe technical components of the procedure: e.g. needing an ICSIlicense from the HFEA in the UK.

What quality control proceduresare required for correctperformance of the method?

Having established the techniques whereby the operationalperformance or outcome of the method will be established ormonitored, how do you ensure that, for example, all your scientistswill do it the same? For example: do all your embryologists assessembryo grades according to the same criteria and scales; do allyour andrology lab personnel assess sperm morphology in just thesame way?

equipment, or the company that supplies your culture medium, or thepeople who remove the biohazard waste, and so on.

A formal service contract describes the exact nature of the relation-ship between the customer (i.e. you, the IVF lab) and the service orproduct provider. An example of the points that need to be coveredin a service contract with an endocrine assay service is provided inTable 6.2 and the requirements of Section 4.4 of ISO 15189:2003 aresummarized in Table 6.3.

If no service contract has been agreed then there is no guarantee ofthe level or quality of service that will be provided. Without such guar-antees the IVF lab has no recourse should problems with late deliveryof results, or aberrant results, or even unexpected changes in pricing,occur – or be perceived to occur. This is why all IVF accreditation


Table 6.2 Example list of factors that should be considered in acontract with an external endocrine assay service for estradiol levels

Service aspect Details

Specimen collection Location(s) and hours of serviceWho pays for the phlebotomy suppliesRequirement for test requisitionsArrangements for transportation to the testing labCut-off time for getting samples to testing lab

Sample identification Required informationAssay methodology Method and instrumentation

Backup in case of instrument failureHandling out-of-range results

Quality management Inter- and intra-assay coefficients of variationSatisfactory participation in EQA scheme

Results and reporting Reporting time: expected/maximum/guarantee(?)Uncertainty of measurementNormal ranges and/or reference values

schemes, and quality management systems in general (e.g. ISO 9001)expect to see documentary evidence of the existence of current serviceagreements or contracts for all services that are supplied to the IVFCenter by organizations or individuals that are not under its directcontrol or authority. Gentlemen’s agreements, “hand-shake” deals, orarrangements that might have been in place for many years, are notsufficient. In this case, if it’s not written down it doesn’t exist – and thearrangements have no standing in law.

Standard operating procedures or SOPs

Writing SOPs can be quite a daunting task, especially for a new IVFlab or one that has taken the decision to move from its traditionalmode of operation (perhaps as a university-based lab) to a “proper”mode of operation including formal accreditation. There is no absolutestandard for what an SOP must contain or look like. However, in theUSA the National Council for Clinical Laboratory Standards (NCCLS)has a set of guidelines on how to prepare SOPs (NCCLS, 2002) thatare acceptable to accrediting bodies such as the College of American

91 Making it work

Table 6.3 Summary of the requirements for an external servicecontract as defined in clause 4.4 of ISO 15189:2003

Clause Part Requirement

4.4.1 Where a laboratory enters into a contract to provide medical laboratoryservices, it shall establish and maintain procedures for the creation of,review of and changes to said contracts. These policies and proceduresmust ensure that:

(a) Requirements, including the methods to be used, are adequately defined,documented and understood (as per clause 5.5 of ISO 15189:2003).

(b) The contracting laboratory has the technical ability and both physicaland human resources (including confirmation/verification of theiradequate skills) to meet the requirements of the contract. This mightalso include review of the contractor’s adequate participation inexternal quality assurance (EQA) schemes, e.g. to determineuncertainties of measurement, limits of detection, confidence limits,etc.

(c) The technical procedures selected are able to meet the requirements ofthe contract and the client’s clinical needs (as per clause 5.5 of ISO15189:2003).

4.4.2 Records must be maintained of contract and/or performance reviews,including any significant changes to the contracted services andpertinent discussions.

4.4.3 Reviews shall also cover any work that might be referred by thecontractor to another laboratory.

4.4.4 All clients affected by the contract (e.g. clinicians, health careorganizations, health insurance companies) must be informed of anydeviation from the contract.

4.4.5 If a contract is amended after work has commenced, the same contractreview process must be repeated and any amendments communicatedto all affected parties.

Pathologists (CAP) or the Joint Commission on Accreditation ofHealthcare Organizations (JCAHO). For medical laboratories in gen-eral, the requirements described in section 5.5 of the ISO 15189:2003standard (summarized in Table 6.4) are the “gold standard” for docu-menting a laboratory procedure or test.

It is the responsibility of the Lab Director to ensure that the contentsof each SOP (or other lab document) are complete, current and havebeen properly and comprehensively reviewed. Electronic manuals areacceptable provided that all the required information is included, and


Table 6.4 Summary of the particular requirements for a standard operatingprocedure or lab test as defined in ISO 15189:2003 clause 5.5, “Examinationprocedures”


5.5.1 The laboratory shall use appropriate procedures that meet the needs of the users of thelaboratory’s services. Preferred procedures are those that have been published inestablished or authoritative textbooks, peer-reviewed papers, or in international, nationalor regional guidelines. If “in-house” procedures are used, they must be appropriatelyvalidated for their intended use and fully documented.

5.5.2 When confirming that the chosen procedures are suitable for the intended use, thelaboratory shall use only validated procedures and these validations shall be as extensiveas necessary to meet the needs of the given application or field of application.

Records of the procedure used for the validation, and the results obtained, must be kept.Methods selected for use must be evaluated and determined to give satisfactory results

before being used for clinical procedures.Procedures must be reviewed by the laboratory director (or an appropriate designated

person) both initially and at defined intervals; with such reviews normally being carriedout annually. These reviews must be documented.

5.5.3 All procedures must be documented and be available, in a language commonly understoodby the staff in the laboratory, at the work station for relevant staff. In addition todocument control identifiers, documentation should include, when applicable, thefollowing additional information:

(a) the purpose of the procedure(b) the principle of the procedure(c) any necessary performance specifications such as precision, accuracy (expressed as

uncertainty of measurement), detection limit, measuring interval, trueness ofmeasurement, sensitivity, specificity, etc.

(d) the primary sample(s) for the procedure(e) the particular type of container to be used (including any additive that might be required)(f) any necessary equipment and reagents(g) any necessary calibration procedures(h) detailed description of the procedural steps(i) any necessary quality control procedures(j) warning of any sources of interference in the procedure(k) the principles for calculating results of the procedure, including measurement uncertainty(l) relevant biological reference intervals(m) the reportable interval for patient examination results(n) any alert or critical values, if/when appropriate(o) information on laboratory interpretation of the procedure(p) any safety precautions(q) any potential sources of variability

5.5.4 Performance specifications for each procedure must relate to the intended use of theprocedure.

5.5.5 Biological reference intervals (and control limits) must be reviewed periodically. A review ofbiological reference intervals must take place when a procedure is changed.

93 Making it work

they are subject to the same document control requirements as hardcopy manuals.

Summary protocols, e.g. as index cards, or any “cheat sheets” pre-pared by any member of the lab staff, are acceptable for use as a quickreference at the workbench, but they too must all comply with all thefollowing requirements:

(a) they must correspond to the complete manual;(b) they must be part of the document control system;(c) they must be authorized, and be determined as current, by the Lab

Director; and(d) the complete manual must be available for reference.

It is important that no-one has the authority to prepare any abbreviatedprotocol for any use outside these requirements, because that woulddestroy the integrity of the document control system.

Technical procedures can be based in whole or in part on Instruc-tions for Use (e.g. package insert) written by the manufacturer, pro-vided that these instructions: (a) conform to the provisions of clauses5.5.1 and 5.5.2 of ISO 15189:2003; (b) describe the procedure as it isperformed in the laboratory; and (c) are written in the language com-monly understood by the staff of the laboratory. Any deviations fromsuch manufacturer’s instructions must be reviewed and documented,and any additional information that might be required for the staffto perform the procedure must be documented. Each new version ofa “kit” that includes major changes in reagents or procedure must bechecked for performance and suitability for its intended use by the IVFlab. Any such changes must be dated and authorized as for other labprocedures.

Having a complete set of written SOPs to show the accreditationsurveyors is not the purpose of writing them. SOPs are created inorder to be used in the everyday operation of the lab. Their valuetranscends being a simple method description and includes a widerange of management and educational features. Each properly writtenSOP will:


• ensure that the selected method is being performed or used correctlyin a technical sense;

• ensure that potential adverse or deleterious factors are being con-trolled;

• provide a framework for the proper quality control of the procedure;• detail how the results of the procedure are to be recorded and com-

municated to everyone who has been deemed by management tohave need of them;

• serve as the primary technical resource for new members of staffduring training;

• serve as the operational and procedural reference for the proper per-formance of the method, e.g. to avoid unauthorized “improvements”;and

• serve as evidence of what was done in case of future dispute or legalaction.

As has been explained previously, developing and writing a good SOPrequires a sound understanding not just of the technical procedurebeing performed, but of everything that affects its performance – anunderstanding that is based on understanding process. Using, andrespecting, SOPs is no less than an obligation of everyone workingin the lab – and quality management depends entirely upon this (seeChapter 3). Moreover – and perhaps more importantly in the modernworld – effective risk management is also dependent on the absoluteuse and respect of SOPs (see Chapter 9).

An example of “good” and “poor” versions of an SOP

At its most basic level, an SOP gives sufficient information to ensurethat any person with relevant training will be able to perform thetask. However, whether the SOP is written to act simply as an aide-memoire or as a proper resource and training document will deter-mine its relative usefulness and its value in keeping the lab systemsunder control. To illustrate how different SOPs can be, we have writ-ten two versions of an SOP for sperm washing (Figures 6.1 and 6.2).

95 Making it work

Any Fertil ity Centre: Laboratory Procedures Manual: SOP Lab 013 rev.2004a

SPERM PREPARATIONPURESPERM GRADIENTS

PRINCIPLESperm must be separated from seminal plasma quickly and efficiently for use in IVF.

SPECIMENLiquefied semen

REAGENTSSperm BufferSperm MediumPureSperm from Nidacon International AB (Göteborg, Sweden).

Preparing GradientsUpper layer : 40% v/v PureSperm, prepared by mixing 4 ml of stock PureSperm and 6 ml of Sperm

Buffer.Lower layer : 80% v/v PureSperm, prepared by mixing 8 ml of stock PureSperm and 2 ml of Sperm

Buffer.

PROCEDURE1. Place the upper layers (1.5 ml of 40% v/v PureSperm) into two 15 ml conical tubes. Then, inject the

lower layers (1.5 ml of 80% v/v PureSperm) underneath the upper layers. A clear interface shouldbe visible between the two layers.

2. Layer liquefied semen onto a pair of PureSperm gradients (up to 2 ml of semen per gradient) andcentrifuge at 300 g for 20 minutes.

3. Recover the soft pellet from the bottom of each gradient.4. Resuspend in either Sperm Buffer if processing for IUI or DI, or Sperm Medium if processing for IVF

or ICSI.5. Centrifuge at 500 g for 10 minutes.6. Aspirate the supernatant and resuspend the pellet in 1 to 3 ml of fresh Sperm Buffer or Sperm

Medium.7. Assess the concentration and motility of the washed sperm preparation.

REFERENCESWorld Health Organization (1999) WHO Laboratory Manual for the Examination of Human Semen and

Sperm-Cervical Mucus Interaction, 4th edition. Cambridge University Press, Cambridge, 128pp.

Appended documents : PureSperm package insert (in sheet protector).

APPROVED FOR USE SOP--Lab013--2004a PureSperm gradients

REVISION SCIENTIFIC DIRECTOR SIGNATURE DATE

2004a

Figure 6.1 An example of a ”poor” SOP for the process of sperm preparation using Pure-Sperm density gradients.


Any Fertility Centre: Laboratory Procedures Manual : SOP Lab 013 rev.2004a Page 1 of 4

SPERM PREPARATIONPURESPERM GRADIENTS

PRINCIPLE

To be potentially functional, ejaculate spermatozoa must be separated from seminal plasma quickly andefficiently. Prolonged exposure to seminal plasma results in marked declines in both sperm motility andvitality and can permanently diminish the fertilizing capacity of human spermatozoa. Therefore, it is essentialthat spermatozoa for clinical ART procedures must be separated from seminal plasma as soon as possibleafter ejaculation.

Background

For a detailed review see Mortimer (2000). In general terms, four basic approaches exist for separatingspermatozoa from semen: (1) simple dilution and washing; (2) sperm migration; (3) selective washingprocedures; and (4) adherence methods for the elimination of debris and dead spermatozoa. However, ithas been established that simple dilution and washing can induce severe damage to the spermatozoa asa result of free radicals generated during the centrifugal washing steps. Furthermore, recent evidence hasdemonstrated conclusively that the population of spermatozoa separated using a density gradient are notonly more functional in terms of their fertilizing ability, but also have fewer nicks in their DNA, and hence willcontribute better quality chromatin to the embryo. Consequently, the density gradient method of spermpreparation must be used for all clinical applications, including ICSI.

The success of sperm preparation methods is often assessed in terms of their yield of motile spermatozoa.Obviously, the fertilizing capacity of a sperm population after processing is another significant factor whichmight be particularly important when working with compromised sperm populations such as those recoveredfrom post-retrograde ejaculation urine or cryopreserved spermatozoa.

SPECIMEN

See SOP Lab007 Sperm Collection -- Ejaculated Semen and also the Notes listed below.• Thoroughly mixed liquefied semen is used as soon as possible after the completion of liquefaction.

Liquefaction occurs optimally when semen is incubated at 37°C, and takes 10 to 30 minutes.• Semen must NOT be mixed using a vortex mixer.• Samples with increased viscosity must NOT be “needled” (i.e. passed through an 18G needle to reduce

viscosity by shear force).

REAGENTS

Density gradients that will perform optimally with the vast majority of clinical semen samples encounteredin clinical ART are two-step discontinuous gradients with usually 1.5 ml layers of 40% (v/v) and 80% (v/v)PureSperm as the upper and lower layers respectively. Gradients must be prepared on the day of use,although the two layers can be prepared in bulk and stored at +4°C for several weeks if prepared underaseptic conditions (they can also be sterilized using 0.22 µm Millipore Millex-GV filters).

Sperm Buffer This is a Hepes-buffered culture medium which contains 10 mg/ml of human serumalbumin to protect the spermatozoa and can be used under an air atmospherewithout compromising the pH. (See SOP Lab005 Culture Media).

Sperm Medium This is a bicarbonate-buffered medium used to resuspend washed spermatozoathat are to be incubated under capacitating conditions, e.g. for IVF. It must be usedunder a CO2-enriched atmosphere to maintain the correct pH. (See SOP Lab005Culture Media).

PureSperm Stock PureSperm (100%) is manufactured by Nidacon International AB (Göteborg,Sweden).

Preparing Gradients

Upper layer : 40% v/v PureSperm, prepared by mixing 4 ml of stock PureSperm and 6 ml ofSperm Buffer.

Lower layer : 80% v/v PureSperm, prepared by mixing 8 ml of stock PureSperm and 2 ml ofSperm Buffer.

N.B. For efficiency, prepare 15 ml conical Falcon tubes (#2095) with one of the 1.5 ml layersalready in them and Falcon 2003 tubes with 3 ml of the other layer (see Method step #1,below). Store these sets at +4 °C and place in the 37 °C incubator the night before use toallow temperature equilibration. Do not add the second layer until the day of use.

Figure 6.2 An example of a ”good” SOP for the process of sperm preparation using Pure-Sperm density gradients.

97 Making it work

Any Fertility Centre: Laboratory Procedures Manual: SOP Lab 013 rev.2004a Page 2 of 4

CALIBRATIONThe centrifuge must be calibrated. To convert its “rpm” settings to g use the equation:

g = 0.0000112 × r × N2 where : g = desired relative centrifugal forcer = radius of the rotor to the bottom of the tube (in cm)N = rpm

QUALITY CONTROLNo special procedures required.

PROCEDURE

N.B. a) Sterile, conical bottom tubes (e.g. Falcon #2095) must always be used for centrifugingspermatozoa – in a swing-out rotor – in order to maximize recovery of the spermatozoa.

b) Never change the centrifugation speeds to try and increase the yield, it will only result inrecovering either poorer quality sperm or a dirty preparation.

c) It is highly recommended that gradients always be prepared and run as pairs to maximize yieldand guard against problems that might arise as a result of careless handling of the gradients orsampling after centrifugation.

d) See below for advice on handling atypical specimens.

1. Using a sterile Pasteur pipette place the upper layers (each 1.5 ml of 40% v/v PureSperm) into two15 ml conical tubes. Then, carefully add the lower layers (each 1.5 ml of 80% v/v PureSperm)underneath the upper layers. A clear interface should be visible between the two layers. See Figure“Before centrifugation” (below).

Alternatively, the lower layers can be placed in the tubes first and the upper layers added ontop of them. This is particularly useful if one has problems achieving a sharp interfacebetween the two layers using the under-layering method.

2. Layer liquefied semen onto a pair of PureSperm gradients (up to 2 ml of semen per gradient) andcentrifuge at 300 g for 20 minutes in a swing-out rotor.

3. For each gradient, using a sterile, long-form Pasteur pipette carefully remove the seminal plasma,upper interface, upper (40%) PureSperm layer and the lower interface and discard; leave most of thelower (80%) PureSperm layer in place. Then, using another clean, sterile, long-form Pasteur pipetteremove the soft pellet from the bottom of each gradient by direct aspiration (maximum 0.5 ml) fromthe bottom of the tube beneath the lower (80%) PureSperm layer. See Figure “After centrifugation”(below).

4. Transfer both pellets to a single clean conical tube and resuspend in either Sperm Buffer ifprocessing for IUI or DI, or Sperm Medium if processing for IVF or ICSI.

5. Centrifuge at 500 g for 10 minutes.

6. Aspirate the supernatant with a sterile Pasteur pipette and resuspend the pellet in 1 to 3 ml of freshSperm Buffer or Sperm Medium as appropriate (see step #4). For IVF or in-vivo inseminationtransfer to a small culture tube (Falcon #2058), for ICSI leave in the large culture tube (Falcon#2057). Then:

a) For IUI or DI leave the sample at ambient temperature until it is collected by the nurse forinsemination; or

b) For IVF, gas the tube with 6% CO2-in-air, cap it tightly and place it in the dark at roomtemperature (i.e. in a cupboard in the Embryology Laboratory); or

c) For ICSI, leave the tube loose-capped and place in a 37 °C incubator (i.e. Forma A).

7. Assess the concentration and motility of the washed sperm preparation using a Makler chamber (seeSOP Lab012). If available, computer-aided sperm analysis (CASA) may be used.

N.B. a) Even though there will be some (perhaps even 10%) immotile or dead spermatozoa in thefinal preparation, this is not a problem and there is NO need to perform any furtherpreparation (e.g. swim-up) as it will have no benefit and could compromise sperm functionor survival.

b) On occasions a washed sperm preparation in culture medium might contain a highproportion of hyperactivated cells that, although very vigorous, may present little progression.In such cases the motility report should state the percentages of progressive andhyperactivated cells, as well as the total motility.

ALL slides or tubes prepared from a semen specimen MUST ALWAYS be labelled with atleast two unique identifers

(e.g. Lab Reference Number and the patientís name).

Figure 6.2 (cont.)


Any Fertil ity Centre: Laboratory Procedures Manual: SOP Lab 013 rev.2004a Page 3 of 4

NOTES : Handling atypical specimens

1. “Dirty” specimens : If a semen sample contains high numbers of other cells, or is heavilycontaminated with particulate debris, the “rafts” formed at the interfaces between either the seminalplasma and 40% PureSperm layers or the 40 and 80% PureSperm layers might be too dense and“clog” the gradient, drastically reducing the yield of spermatozoa. There are several simple stepsthat can be taken pro-actively to avoid this problem if the semen sample is seen to be very “dirty”on initial microscopic examination:a) Only process part of the ejaculate; and/orb) Load less semen onto each gradient, perhaps dividing the sample over four gradients

instead of two; and/orc) Use longer columns of PureSperm, e.g. 2 or 3 ml per layer; and/ord) When loading the semen onto the gradient, mix it gently with the upper one-fifth of the upper

layer; and/ore) Prepare a three-step gradient using layers of 40%, 60% and 80% PureSperm (or 30 / 50

/ 80%).N.B. Never use a lower layer of 90% (v/v) PureSperm as this will likely reduce the sperm

yield.

2. Highly viscous samples : It can be extremely difficult to obtain good yields of motile spermatozoafrom highly viscous samples. To reduce the viscosity and so maximize yield, add an equal volumeof Sperm Buffer to the semen sample and mix gently using a sterile Pasteur pipette. If the sampledoes not disperse within 2 minutes of pipetting, incubate at 37°C for 10 minutes and then mixfurther. Once the sample has been successfully diluted it can be loaded onto the gradients as usual.An alternative technique for men who are known to have consistently high viscosity is to have themcollect into chymotrypsin-coated MARQ™ Liquefaction Cups available from EmbryotechLaboratories (Wilmington, MA, USA).

3. Cryopreserved semen : Because of the high osmolarity of semen cryoprotectants, cryopreservedspermatozoa will swell greatly upon entering the 40% PureSperm layer and hence decrease theirspecific gravity. This will not only cause them to be too buoyant to pass through the densitygradient, but can also cause impaired sperm function or even survival. To avoid this problem,cryopreserved semen must be diluted with a large volume of “isotonic” medium prior to loading ontothe gradients:a) Remove the straws or cryovials from the liquid nitrogen. To thaw:

(i) Straws: Place in a 37°C incubator for a 10 minutes.(ii) Cryovials: Allow to stand at room temperature for 10 minutes then unscrew the cap

slightly before placing in a 37°C incubator for a further 10 minutes.b) Wipe the condensation from outside of the straws or cryovials and then wipe their outsides

using 7X detergent.c) Straws: Cut off the Critoseal-plugged end of the straw, at the position of the air

space, using sterile disposable scissors (Rocket Medical, R50000). Placethe open end in a Falcon #2003 tube and cut the upper cotton-plugged endbelow the lower wadding to allow the contents to expel into the tube.

Cryovials: Unscrew the cap and transfer the semen to a round bottom Falcon #2001tube. If the volume of the sample is >1.5 ml then split it between two tubes.

d) Slowly dilute the semen with a 10× volume of Sperm Buffer, adding it drop-wise withconstant gentle mixing over a period of at least 10 minutes.

e) Layer the diluted specimen onto a pair of PureSperm gradients in conical tubes (Falcon2095). Up to 3 ml of the diluted sample, or 1.5 ml of fresh semen, can be loaded safely ontoeach gradient. If necessary, split a diluted sample over 2 pairs of gradients to maximize theyield.Alternatively, but sub-optimally, if the total volume exceeds 6 ml the diluted semen can becentrifuged at 500 g for 5 minutes (this is safe because the cells that generate free radicalsduring centrifugation do not survive the freezing & thawing process). Remove thesupernatant and resuspend the pellet using 1 to 2 ml of Sperm Buffer. Load this spermsuspension onto the gradients as usual.

4. Poor quality specimens : An increased yield can be obtained from most samples, but especiallylow concentration ones, if slightly less dense layers of PureSperm are used, e.g. 72% and 36%dilutions of PureSperm. However, this is only achieved by recovering less dense, and hence lessgood quality, spermatozoa. It should only be used as a last resort.

5. Retrograde ejaculate urine specimens : After the retrograde ejaculate urine specimen has beenconcentrated by centrifugation and resuspended into a small volume of Sperm Buffer it is layeredover two or four PureSperm gradients and processed as per a normal semen sample.

Figure 6.2 (cont.)

99 Making it work

Any Fertil ity Centre: Laboratory Procedures Manual: SOP Lab 013 rev.2004a Page 4 of 4

FIGURE

REFERENCES

Mortimer, D. (1994) Practical Laboratory Andrology. Oxford University Press, New York, 393pp.Mortimer, D. (2000) Sperm preparation methods. J. Androl., 21: 357–366.Mortimer, D. and Mortimer, S.T. (1999) Laboratory investigation of the infertile male. In: A Textbook of

In Vitro Fertilization and Assisted Reproduction, 2nd edition, ed. P.R. Brinsden. Parthenon Publishing,Carnforth, Ch.4, pp.53–81.

World Health Organization (1999) WHO Laboratory Manual for the Examination of Human Semen andSperm-Cervical Mucus Interaction, 4th edition. Cambridge University Press, Cambridge, 128pp.

Appended documents : PureSperm package insert (in sheet protector).

APPROVED FOR USE SOP–Lab013–2004a PureSperm gradients

REVISION SCIENTIFIC DIRECTOR SIGNATURE DATE

2004a

ISSUED MEDICAL DIRECTOR SIGNATURE DATE

Figure 6.2 (cont.)

Figure 6.1 shows an SOP which is sufficient to remind a trained scientistof the reagents and steps in the procedure. However, it does not offerany background information, nor any rules for dealing with atypicalspecimens. Figure 6.2 is a much more comprehensive SOP, reflectingmany years’ experience and observation in using this procedure, and


illustrating the value of the SOP as a living document. While it gives thesame basic information, it includes sufficient background information,procedural details and rules for dealing with atypical specimens to coverthe situations that may be encountered on any given day in an IVF lab.This is important to ensure that all of the laboratory staff membersdeal with any given situation in a uniform way. This not only helpsin troubleshooting, it is also useful when reviewing cycle outcomes,because the procedure used can be removed from the list of potentialconfounding factors.

The “big nightmare”: did we use the right sperm?

Avoiding this situation is not related to the technical competence ofwhoever was performing the procedure, but depends on the operatorobeying the organizational aspects of a properly controlled process. Ithas nothing to do with the technical procedure of washing the sperm,or performing the IVF insemination or ICSI, but everything to dowith clearing the work area between cases, not processing two casessimultaneously in a way that will create a risk of inadvertent mixing oruse of the wrong sperm sample, and of performing the identity checksand verifications that are inherent to the procedure – and which mustalso be included in the SOP.

Understanding the “hand-offs” or “hand-overs” between stages ina process, or between operators (see the discussion of “Swim Lanes”in Chapter 5 under “Process mapping tools”) – and detailing them inthe SOP – is key to ensuring that an SOP contains a complete descrip-tion of a method that will achieve all its desired outcomes and avoidknown adverse outcomes. “Outcomes” includes organizational aspectsof the process, not just its technical endpoint. Therefore, an SOP mustinclude exact descriptions not just of the series of steps that make up thetechnical procedure, but also of data recording, results reporting and,perhaps most importantly of all, the points where operators interact,both within the lab and with other departments, e.g. nursing. Naturallythis also then requires verification that each of the other departments’SOPs mirror the described “interface” events.

101 Making it work

Process control

Again, this is a technique taken from industry, where it was developedto monitor the performance of manufacturing processes. However,rather than compare the performance of a process in relation to someexternal reference or benchmark, we are monitoring performance rel-ative to the historical performance of our own implementation of aprocess over time. Basically, by using process control techniques we areable to state whether, at any point in time, our systems are “under con-trol,” or not, in comparison to our recent and historical performancelevels.

The most frequently used tool is the process control chart orShewhart chart, named after one of the technique’s early proponents,Walter A. Shewhart.

Control charts

Figure 6.3 is an illustrative example of a control chart for a genericprocess whose outcome is measured using an Indicator that has a range

0

Jan

Feb

Mar

Apr

May Jun

Jul

Aug

Sep Oct

Nov

Dec

10

20

30

40

50

60

70

80

90

100

Control mean

Upper warning limit

Ind

icat

or

(0–1

00)

Lower warning limit

Upper control limit

Control limits= mean ± 3SD

Warning limits= mean ± 2SD

Lower control limit

Figure 6.3 A generic control chart (a.k.a. Shewhart chart).


of 0 to 100 (i.e. percentage values). Baseline data on the performanceof the process are required for a representative period of operation, e.g.monthly average values for the Indicator over the preceding 6 months.The mean and standard deviation (SD) of these 6 values is calculatedand these values used to establish the “control mean” and two types ofoperational limit: (a) “warning limits,” defined as the mean +/– 2SD;and (b) “control limits,” defined as the mean +/– 3SD. The numberof prior data periods required to calculate the control limits for anIndicator is not predetermined arbitrarily. The number used must besufficient to give a good indication of the variability of the Indicator,but at the same time not be so many that the standard deviation isreduced to the extent that the control limits become too narrow, withthe result that apparent control deviations frequently turn out to berandom fluctuations in the Indicator. We have used MedCalc softwarevery successfully to produce control charts for the past several years(MedCalc Software, Mariakerke, Belgium; www.medcalc.be).

It is also true that as a lab becomes better organized, implementsmore robust methods, or performs more cases per unit time, therewill be a reduction in the inherent variability in its Indicators. Undersuch circumstances the control limits must be recalculated from morerecent data (see below for an example of this).

So long as subsequent periodic mean values remain within the con-trol limits, the process is considered to be “in control.” However, thereare four principal scenarios that require further action.

1. The Indicator crosses its control limit in the adverse direction.Immediate action is required to determine whether there is a gen-uine problem and, if verified, to seek its resolution.

2. The Indicator crosses its warning limit in the adverse direction.Action is required to determine whether a problem might exist orbe developing.

3. The Indicator shows three consecutive changes in the adverse direc-tion but does not cross the warning (or control) limit. Action isrequired to determine whether a problem might be developing.

103 Making it work

4. The Indicator crosses its control limit in the beneficial direction.The system should be reviewed to see why it occurred and whetherthe improvement is sustained. If the improvement is real, then thecontrol limits must be redefined.

The example shown in Figure 6.4 is based on data for zygote grade(based on a score out of 20) from Mortimer (1999). Monthly aver-age values for the 6 months preceding the period shown in chart A(i.e. February to July 1997) were used to establish the control meanand the warning and control limits. However, the introduction of animproved IVF and embryo culture system in mid-August 1997 (seeD. Mortimer et al., 2002) was the explanation for all the values forsubsequent months except one being above the control mean, and this

18.4 18.4

18.2 18.2

18.0 18.0

17.8 17.8

17.6 17.6

17.4 17.4

17.2 17.2

17.0 17.0

16.8 16.8

16.6 16.6

16.4 16.4

16.2 16.2

16.0 16.0

A S O N D J F M A M J J A S O F M A M J J A S O1997 1998

KEY: Upper & lower control limits

Upper & lower warning limits

Control mean (reference value)

A

Zyg

ote

grad

e (m

ean,

0–2

0)

B

1998

17.15

16.52

17.78

16.20

18.10

17.48

16.95

18.01

16.69

18.27

Figure 6.4 A control chart displaying monthly averages for zygote grade (data taken fromMortimer, 1999: see text for explanation). Panel A shows control limits based onthe 6 months preceding the period graphed while panel B shows recalculatedcontrol limits following the introduction of an improved embryo culture systemin mid-August 1997.


significant change necessitated recalculation of the control values,using the August 1997 to January 1998 monthly average values, asshown in chart B. The increased value for the control mean indicatesa systematic improvement in zygote grade while the narrower controllimits express the improved stability (i.e. reproducibility) of the newculture system.

A further example, shown in Figure 6.5, illustrates the use of controlcharts to investigate whether the major reconstruction of an IVF lab hadany detrimental impact upon the Indicators used to monitor the lab’sperformance. Since the Indicators during the periods preceding andafter the renovations all remained within the previously-establishedcontrol limits, the Lab Director was able to state with confidence thatthe renovations had not had any detrimental effect on fertilization rateor embryo quality (Mortimer et al., 2001a).

Previously we have termed the sort of Indicators that are beingconsidered here as “Laboratory Performance Measures” or LPMs(Mortimer, 1999; Mortimer et al., 2001a), but have more recently har-monized the term as “Laboratory Performance Indicators” or LPIs.Naturally there are other Indicators that can be used to monitor aspectsof clinical performance or even of the overall program where theendpoint of interest depends on other sources of variability rather

Figure 6.5 Control charts displaying monthly averages for three aspects of laboratory per-formance before and after a period of major laboratory renovation. Data fromMortimer et al. (2001a), see text for explanation.

105 Making it work

than just the lab methods, conditions and staff (e.g. pregnancy rate orimplantation rate). Using the same convention as above, the latter canbe termed “Programme Performance Indicators” or PPIs. An extensivelist of example LPIs and PPIs is provided in Chapter 10.

However, it must be mentioned here that the use of control charts tomonitor lab processes via LPIs is very much an activity that will havemeaning to the Lab Manager or Director rather than to the physiciansor other team members in the vast majority of IVF Centres. Even ifthese other team members do understand what the PPIs mean forthe IVF Center, there is often a perceived “disconnect” between laboperations and pregnancy rates. But when you stop and think aboutwhat is actually involved in providing IVF treatment this is not anunexpected – or abnormal – circumstance: after all, the greatest sourceof variation between treatment cycles is the patients. What LPIs tell usis that the levels of performance of the various lab processes that arebeing monitored by these Indicators are not straying outside what havebeen, presumably acceptable, historic ranges of variation.

Consequently, monitoring a comprehensive panel of LPIs will allowthe lab to provide a rapid response to the seemingly perpetual ques-tion of “what’s wrong in the lab?” If all the LPIs are within their controllimits then the lab can conclusively – and immediately – state thateverything is still the same as before, nothing has changed, and hencethe questioner should seek an explanation elsewhere for their perceiveddecrease in whatever endpoint they’re looking at. Of course, if a prob-lem does arise then the lab will see it early on – presumably earlier thanit will be manifest in any clinical endpoints. The lab will then be able toinvestigate it and either disregard it as a fluctuation that is not relatedto any change in operational performance or adverse factor(s) actingon one or more lab processes, or identify the source of the problem anddeal with it. In the ideal world it will be the lab that says to everyone elsesomething like “there was a temporary problem with such-and-such,but we’ve identified the source of the problem and already correctedit.”


Table 6.5 Sequence of steps in selecting, implementing and validatinga method

Process Component activity

Method design/selection Define what you want to do or measure.Review the described/available methods.Determine the feasibility of using the method in your lab

(equipment, reagents, time, complexity, training).Perform a cost analysis.Identify the factors that might affect the method’s accuracy and

precision (uncertainty of measurement).Method implementation 1. Review all possible sources of error and bias.

2. Design a method that controls for these problems.3. Establish a standard protocol.4. Train your staff.5. Verify staff performance (proficiency testing).6. If necessary, apply corrective action/re-training and reassess

staff performance.7. Implement a Quality Control programme for the method.8. Participate in an External Quality Assurance Programme

(“EQAP”) or Scheme for the method.Method validation Remember: Calibration �= Quality Control.

Select appropriate reference materials and calibrators.Quality Control must be a continual process. Consider process

control analysis.Participation in an EQAP cannot replace Quality Control.To be useful, an EQAP must include both Quality Assurance

and a Quality Improvement capability.

Implementing and validating methods

Practical application of your knowledge about the uncertainty of mea-surement must be employed throughout the process of method designor selection, method implementation, and method validation. Theseprocesses have been summarized in Table 6.5.

Reference materials vs. calibrators

These materials are valuable in many aspects of method development,implementation and validation. They can help verify that the method

107 Making it work

is being used correctly and in determining its measurement uncer-tainty, as well as in calibrating instruments, establishing and verifyingstaff proficiency, and in ongoing quality control and quality assurance.However, while many people see them as the same thing, they are notentirely synonymous.

A reference material must be the same as what is being measured,e.g. a panel of serum samples containing known concentrationsof a hormone.

A calibrator, however, does not need to be the same as what isbeing measured, it can be a surrogate that is sufficient for certaintechnical aspects of the method.

A good example here are Accu-BeadsR (Hamilton Thorne Biosciences,Beverly, MA, USA). The Accu-Bead products are aqueous suspensionsof latex particles at known concentrations. In many situations theycan be used as surrogates for human spermatozoa, especially whenverifying that a counting procedure is being performed correctly. ButAccu-Beads do not look exactly the same as spermatozoa, and thereforetheir counting in a specialized counting chamber might well verify thecalibration of the counting chamber, but this cannot verify that eitherthe human eye–brain or automated image analysis system recognitioncomponent of the procedure of identifying and counting spermato-zoa would be performed correctly. Accu-Beads are excellent calibra-tors, and have many uses in calibration and quality control, but thereare some aspects of the procedures that they cannot verify, and moreappropriate materials might have to be used to ensure full technicalcompetence and proficiency of staff.

Uncertainty of measurement

It is a fundamental principle of the science of mensuration that everymeasurement has an error associated with it – and that, without aquantitative statement of that error, a measurement lacks worth, evencredibility. The parameter that describes, in quantitative terms, the


boundaries of measurement error is called the uncertainty of mea-surement. While “accuracy” is a generally-used term and is subject tointerpretation, “uncertainty” has a specific meaning, being defined asthat parameter, associated with the result of a measurement, that char-acterizes the dispersion of values that would reasonably be attributed tothe measurand (the “measurand” is that particular quantity subject tothe measurement). A general background on the uncertainty of mea-surement can be found in Cook (1999) and the principles governingit are published in the ISO 1993 Guide to the Expression of Uncertaintyin Measurement – often referred to as the “ISO GUM” (ISO, 1993).

To be meaningful, the uncertainty statement must have an associ-ated confidence level: i.e. the probability that the true value lies withina given range. The most common range used is the 95% confidenceinterval (or “95% CI”), which means that there is a 95% probabil-ity that the true value lies within the stated range, which is centredaround the measurement value. For the mathematically inclined, the95% CI can be obtained by multiplying the method’s combined stan-dard uncertainty of the measurement by 2, although that must havebeen calculated previously. According to the ISO GUM document, thecombined standard uncertainty of a measurement is the square rootof the combined variance of the factors creating error in the measure-ment. However, for the present purpose we only need to understandthat no measurement is absolutely accurate, and that we must alwaysendeavour to understand, and be able to provide a good estimate of,the possible dispersion of measured values.

Good laboratory practice requires that all reports express results inparticular ways that have meaning and are useful to the intended recip-ient(s). If this includes a quantitative result, then it should be accompa-nied by a statement of uncertainty. Determination of that uncertaintytypically requires the construction of a model of the measurement sys-tem followed by a list of all the factors that can contribute errors tothe final result. Clearly this requires a sound knowledge and under-standing of the measurement system, as well as the equipment andthe environment in which the measurements are made. Yet again, weare confronted with a need to understand process and systems. While

109 Making it work

Table 6.6 List of sources of uncertainty in measurement (see ISO, 1993)

Area of uncertainty Description

What is being measured Incomplete definition of the measurand.Incomplete realization of the definition of the measurand.

Sampling Non-representative sampling.Environmental effects Inadequate knowledge of the effects of environmental conditions on

the measurand (or imperfect measurement of those conditions).Instrumentation and reading Observer bias in reading analog instruments or in making subjective

assessments.Finite instrument resolution or discrimination threshold.

Standards and reference materials Inexact values of measurement standards and reference materials.External factors Inexact values of constants and other parameters obtained from

external sources.Methodology Approximations and assumptions incorporated in the measurement

method and procedure.Variations in repeated observations of the measurand under

apparently identical conditions (“repeatability”).

detailed considerations of statistical theory and technique are outsidethe scope of this book, the principles of uncertainty of measurementmust be borne in mind when developing or selecting methods, inter-preting results or making decisions based on results (see Table 6.6 fora summary of these points).

Two points of statistical theory are, however, important to note here,especially in relation to such assessments as sperm morphology. Theseare considerations of sampling error and the reliability of results beingrelated back to a population when they are based on small numbers ofobservations. Table 6.7 shows the magnitude of simple counting errorwhen making determinations on small numbers of objects counted.Table 6.8 shows the limits of the expectation when a small proportionis determined by counting only a small number of objects, e.g. per-cent normal forms based on examining just 100 spermatozoa. Basedon the Poisson distribution, statistics therefore says that a value of 4%normal forms based on a count of 100 sperm might actually be any-thing from 1% to 10% – and that values of 4% and 7% so-derived arenot statistically different. See also the discussion in Kuist & Bjorndahl(2002).


Table 6.7 Relative magnitude of the countingerror when basing determinations on smallnumbers of objects

Number counted Counting error

40 ± 16%60 ± 13%

100 ± 10%200 ± 7%400 ± 5%

Table 6.8 Limits of the expectation (Motulsky, 1995)

Counts 95% confidence interval

4/100 1.09–10.245/100 1.62–11.676/100 2.20–13.067/100 2.81–14.428/100 3.45–15.769/100 4.12–17.08

10/100 4.80–18.39

There are also two main types of error: “random” and “systematic.”

Random errors cause lack of precision and arise from chance differ-ences in sampling or reading. These are the errors that we try andminimize by using repeated sampling, or repeated measurementsby the same observer, or by the same piece of equipment.

Systematic errors (sometimes referred to as “bias”) are far moreinsidious because they arise from factors that alter the result inonly one direction. These shifts in the reported values cannot bedetected by repeated measurements.

Minimization of both types of error is achieved by careful design ofthe method, analytical instrument selection, adequate sampling andcounting, thorough training of the operators plus internal qualitycontrol and external quality assurance components.

111 Making it work

All scientists are familiar with results that include references to thestandard deviation or the standard error of the mean that establishtheir place as being representative of a (sample) population. But whatabout individual results that do not have such qualification? Obviouslysomething like a number of oocytes or embryos is not subject to anuncertainty (beyond the question of whether some might have beenmissed or lost), but sperm counts or motility assessments can often bevery imprecise numbers (see Mortimer et al., 1986, 1989; Mortimer,1994; World Health Organization, 1999).

How accurate do we need to be?

A final consideration here is “how accurate do the results we quoteactually need to be?” There are clearly trade-offs between the costs intime and money of making more accurate quantitative assessments andwhat the results will be used for. But there is also the obligation that,under the requirements of a quality management system, we must beable to keep the measurement system or assay “under control.” If themethod has a large uncertainty of measurement then there can be hugefluctuations in results entirely due to technical error and variation, andwe have no way of knowing whether a result is genuinely unusual orjust comes from the extreme end of the distribution.

Many embryologists (and gynaecologists) do not care about theaccuracy of sperm counts and related assessments, and will tolerate theuse of commonly-used, but hopelessly inaccurate, andrology methodson the grounds that the results “make no difference” and/or “don’t helpget the patients pregnant.” An exhaustive discussion and analysis of theproper management of couples with male factor infertility is clearly notrelevant to the present book. However, as scientists we cannot ignorethe alternate perception that perhaps had semen analysis results beendetermined more accurately, it might have been possible to establishthem as having greater value in clinical management (there is alsosome further discussion of this in Chapter 11). As a corollary to this,estradiol values during a controlled ovarian hyperstimulation cycle


are only used to establish that estradiol levels are rising comparablyto follicular growth – but the assays used must still operate withinthe standard requirements for accuracy and precision, and expectedtolerances of inter- and intra-assay coefficients of variation, in orderfor the endocrine lab to maintain its accreditation.

A simple answer to the question is, therefore, that any resultsreported by an IVF lab must be obtained using methods that can becontrolled within the constraints of a proper quality management pro-gramme. As accreditation spreads internationally then more and moreIVF labs will have to operate to standards of accuracy and precision forall counts and assessments. Each measurement procedure will have tobe validated and its uncertainty established.

Document control

An integral part of any quality management or accreditation systemis the need for an effective system of document control. A DocumentControl System will establish and maintain documented proceduresfor the following requirements:

• Review and approve documents by a competent authority beforeissue;

• establish a master document list, and make it available where neededthroughout the organization;

• ensure that all relevant documents are available where and whenneeded;

• ensure that obsolete documents are not used accidentally.• clearly identify obsolete documents that are retained for any reason

(e.g. archived for possible future medico-legal purposes); and• control the mechanisms for revision of documents.

Documents subject to an organization’s document control system aretermed “controlled documents.” A more detailed list of the require-ments for controlled documents as laid down in section 4.3 of ISO

113 Making it work

Table 6.9 Summary of the procedural requirements of ISO15189:2003 for a Document Control System


4.3.2 (a) All documents must be reviewed and approved by authorized personnelprior to issue.

(b) A list or “document control log” that identifies the current valid versionsof all documents, as well as their distribution, must be maintained.

(c) Only currently authorized versions of appropriate documents can beavailable for active use.

(d) Documents must be reviewed periodically, revised when necessary, andapproved by authorized personnel before reissue.

(e) Invalid or obsolete documents must be removed promptly from all pointsof use (or other steps taken to ensure they cannot be usedinadvertently).

(f) Superseded documents must be appropriately identified to prevent theirinadvertent use.

(g) If the laboratory’s document control system allows for the amendment ofdocuments by hand pending their reissue, then procedures andauthorities for making any such amendments must be defined.Amendments must be clearly marked, initialled and dated, and a reviseddocument reissued formally as soon as practicable.

(h) There must be procedures in place to describe how changes to documentsthat are maintained electronically are made and controlled.

4.3.3 All documents must be identified uniquely, including the followinginformation:

(a) the title of the document(b) the edition or current revision date, or revision number, or all these(c) the number of pages(d) the authority for issue(e) references for relevant source materials

15189:2003 has been summarized in Table 6.9. Under this Standard,all documents are considered to comprise part of the organization’squality management system, and a “document” is considered to be“any information or instructions, including policy statements, text books,procedures, specifications, calibration tables, biological reference intervalsand their origins, charts, posters, notices, memoranda, software, drawings,plans, and documents of external origin such as regulations, standards orexamination procedures.”


A copy of each controlled document must be archived for later ref-erence, and this can be achieved using any “appropriate medium”(which might, or might not, include printed hard copy). While it isleft to the Lab Director to define the appropriate retention period,national, regional and local regulations concerning document reten-tion will often be applicable.

Many accreditation schemes, as well as good practice, require thatSOPs be reviewed and reissued annually. They must be signed off byboth the Laboratory Director and the Medical Director. This is impor-tant because everything that is done in the IVF lab constitutes partof patient treatment, which must be directed by a clinician. Essen-tially, the SOPs establish that everything that is done within the labhas been agreed to by the senior physician and so is performed underhis/her authority. A corollary of this is, therefore, that so long as thescientists follow the SOPs they ought to be indemnified against anyadverse events (see also Chapter 9: “Protecting IVF laboratory stafffrom unfair litigation”). Even if a problem were to occur as a resultof an oversight in a process defined in an SOP, the responsibility ulti-mately rests with the lead physician who signed off on the SOP, becauseby signing the SOP they agreed that it would be used in the treatmentof their patients. Annual review and reissue ensure that SOPs are cur-rent, an aspect that is perhaps even more important in an IVF lab thanmany other clinical labs due to the rapidly evolving assisted conceptionfield.

However, forms are not “signed off” for use in the way that SOPs are,so we have typically taken a slightly different approach for forms in thatthey are deemed to remain current unless a new version is created. Theidentity of the current forms is stated by referencing them in all SOPsin which they are used, and they are therefore effectively reviewed andreauthorized at the same time as the SOPs are reissued.

Maintaining all the electronic versions of each controlled documentrequires adherence to strict naming conventions for the electronic files.The following conventions show examples of how we have been doingthis recently for several types of document.

115 Making it work

ForSOPs: SOP-Lab001-2004a Short name.EXTforms: FRM-Lab001-20040317 Short name.EXTequipment

instructions: EQP-Lab001-2004a Short name.EXTjob descriptions: JOB-Lab001-2004a Short name.EXT

WhereSOP denotes that this document is an SOP or protocolFRM denotes that this document is a formEQP denotes that this document is a set of instructions

for a particular piece of equipmentJOB denotes that this document is a job descriptionLab identifies this as a laboratory SOP (rather than a

clinical or nursing SOP or protocol)001 identifies which SOP, form, equipment instruc-

tions, or job description it is, using a sim-ple sequential numbering system from 1 to 999(including the leading zeroes makes a list of filenames easier to read through)

2004a denotes the sequence of revisions or issues withinthe year

20040317 denotes the effective date for the form in YEAR-MODA (year, month, day) format

Short name is an abbreviated version of the name of the docu-ment to facilitate identifying the correct documentwhen reviewing directory listings in file folders.

EXT identifies the file type, and hence the software usedto create, edit, open or print it.

The hyphens are used to separate the various components of the filename for clarity.

Other documents that must be included in a document control systeminclude policies and procedures, standard or template letters or reports,patient information sheets or handouts, and perhaps also newsletters.


Figure 6.6 Approaches for classifying the purpose of documents.

In deciding whether a document is policy, a protocol, an objective ora guideline, the principles shown in Figure 6.6 often resolve the debate.Try applying these two approaches to classifying the document: doesit describe an outcome or a process (i.e. does it relate to an “end”or to “ways” or “means”) or can it be differentiated according to itsnecessity: A “gottabe” is something that must be, while a “wannabe” issomething that should, or could, happen. Of course, a document canalso have a composite function. For example, it might be:

• a Policy and a Protocol: describing what must be done and the wayit must be done; or

• a Policy and a Guideline: describing what must be done and the wayit should/could be done; or

• an Objective and a Protocol: describing what should/could be doneand the way it must be done; or

• an Objective and a Guideline: describing what should/could be doneand the way it should/could be done.

While implementing and maintaining a document control systemtakes quite a bit of effort, it does have the practical benefit of mak-ing sure that changes are not accidentally – and unpredictably – lost

117 Making it work

because someone inadvertently uses an outdated document that issomehow still in circulation.

External quality assurance

Although we have mentioned external quality assurance (EQA) severaltimes in this book, there are limited options for IVF labs in terms ofEQA programmes (EQAPs) at this time. However, one resource thatdoes exist is a web-based commercial service called FertAid run by JimStanger of Newcastle, Australia (www.fertaid.com) that offers not onlyEQA but also operates as a training tool and an ongoing educationalscheme. FertAid covers both embryology and andrology, and resultscan be viewed not only in comparison with a peer group, but alsowithin a particular enrolled Center, effectively giving it an internalQC dimension as well. The major drawback that we have seen withthis service is that, for now, unless you’re a very patient individual, abroadband Internet connection is the least that should be available!

A variety of andrology EQA schemes exist, and it is extremely impor-tant to choose one that includes not only sound quality assurance, buta real quality improvement functionality. A scheme that only reportsyour results in comparison with the other participating labs, with nosuggestion as to what the “right answer” was, clearly has limited value.Reference values, even if determined as “consensus” values by averag-ing the results from a select sub-group of internationally-recognizedhigh calibre labs that participate in the scheme, are essential – with-out them the scheme is “directionless,” and there can be no qualityimprovement. The ESHRE Andrology Special Interest Group operatesan international EQA scheme for andrology; contact Lars Bjorndahlby e-mail at [email protected] for more information. For morelocal resources, check with your national andrology society.

7

Quality and risk management tools

There are many tools available to support quality and risk managementin the IVF Lab. However, they are not specific to our field – they are allvery well-established generic tools and techniques that have been usedfor many years in all areas of business.

Inspection and audit are observational tools that establish what ishappening and whether defined practices are being followed. More in-depth investigations where a process must be analyzed and improved,or risks identified and managed, might need to be undertaken eitherproactively or retrospectively, for which the most commonly used toolsare Failure Modes and Effects Analysis and Root Cause Analysisrespectively.

Inspection

Inspection is simply the careful examination of what goes on in theIVF Lab:

• what the environmental conditions are in the lab;• is the lab equipment working properly;• what happens in the lab in terms of material and people movement;• are the products used in the lab appropriate and suitable for use;• how tasks are performed;• how information is recorded; and• how data are analyzed.

It involves the collection, collation and analysis of data, as well as theexamination of processes, which is best accomplished using process

118

119 Quality and risk management tools

mapping. The daily equipment logs maintained by IVF labs followingGLP come under this heading, as does the filing of Certificates of Anal-ysis for each batch of culture media and other reagents and routine QCchecks on equipment. Unless such information is carefully recordedand/or filed it will not be available if required in a future troubleshoot-ing exercise. People often complain about the amount of paper theyhave to keep, or that equipment monitoring is “make work” that doesn’thelp get the patients pregnant. But the Certificates of Assay and dailylog sheets, for example, represent information that could hold the keyto solving a problem, perhaps one that was stopping the patients get-ting pregnant. Meticulous attention to detail and record keeping isvital.

Audit

An audit is a formal examination and verification of an organization’ssystems or records and supporting documents by a properly-qualifiedprofessional. It must be an objective activity designed to add valueand improve an organization’s operations. It helps an organizationaccomplish its objectives by using a systematic, disciplined approachto evaluate and improve the effectiveness of risk management, control,and governance processes.

An internal audit is carried out by someone within the organiza-tion who has received formal training in performing audits, and isusually from a different department to the one where the audit is beingcarried out.

An external audit is one carried out by a qualified professional audi-tor (or a consultant who is properly trained in auditing but has neces-sary specialist professional expertise) who is completely independentof the organization where the audit is being performed.

Audits are undertaken to establish compliance with systems, pro-cesses and SOPs. Any activity in the IVF Center can be the subject ofan audit, as can the Center’s success rates.


Figure 7.1 A diagrammatic overview of Failure Modes and Effects Analysis (FMEA).

Failure Modes and Effects Analysis

Failure Modes and Effects Analysis or FMEA is a simple yet power-ful engineering quality management technique that helps identify andcounter weak points in the design or manufacture of products or thedesign and execution of processes. Its structured approach (summa-rized in Figure 7.1) has made it one of the most widely-used tools fordeveloping quality designs, systems and services and it can be usedto improve processes in any organization. Application of FMEA inhealthcare typically identifies process components for improvementactions based upon relative ratings of their anticipated frequency andthe severity of adverse effects or events.

Conducting an FMEA involves the team following a sequence ofgeneric steps as summarized in Figure 7.2 and Table 7.1. The first stepis to establish the context of the issue and this involves mapping theprocess so that the details of the process are readily apparent (processmapping has been discussed already in Chapter 5). Possible FailureModes can then be identified within the process, each one of whichrepresents a specific risk that is to be considered. The rating schemesfor the likelihood and severity of a Failure Mode or risk are not stan-dardized beyond the organization employing them. In the illustrativeexamples shown in Tables 7.2 and 7.3, we have used scales that go upto 10. Even though some organizations employ scales that only go up


Define the Objectivesand Scope of the FMEA

Select the Team

Gather and Preparethe Information for Analysis

Conduct the FMEA(might include one or more RCAs)

Evaluate and Prioritizeall the Recommendations

Document the Analysis &Recommendations: The Action Plan

Present the Action Plan for Approval

Implement the Action Plan

Track the RecommendedChanges to Closure

Determine the ways (Modes) in which the system can fail

Failure Mode 1 Failure Mode 2 Failure Mode 3 Failure Mode 4

What are the consequences (Effects) of failure in this Mode?

PossibleEffect 2

PossibleEffect 1

PossibleEffect 3

Rate the Risk (R) orConsequence of this Effect

Assess the Likelihood (L)of this Effect occurring

Calculate the Criticality of this Effect(C = R x L )

Is the Riskacceptable?

ENDDevelop a Remedial Action

YES NO

Repeat foreach Modeand Effect

combination

Figure 7.2 A flow chart for performing a Failure Modes and Effects Analysis (FMEA).

to five, we believe that the greater dynamic range in assessments thatthese scales permit is advantageous in deciding which risks are to betackled first, and in their prioritization. Multiplying these two rankingstogether gives the Criticality scores, which are the actual values usedwhen discriminating between risks of greater or lesser overall impor-tance. Because of this step of calculating Criticality scores, FMEA issometimes referred to as Failure Modes, Effects and Criticality Analy-sis or FMECA.


Table 7.1 The generic steps involved in performing a Failure Modes and EffectsAnalysis (FMEA)

Procedural step Explanation

Examine and map the process Identify all the Functions that are expected to occur.Identify Failure Modes Identify any ways in which any of the Functions might go wrong.Determine the Effects Establish the consequences of each Failure Mode.Identify Contributory Factors Identify the underlying causes for each Failure Mode (one or more

RCAs might be required for this).Rate the likelihood and severity of

each Failure ModeEstimate, using standardized rating schemes (see Tables 7.2 and

7.3), the frequency or likelihood of occurrence of each FailureMode or Contributory Factor, and rank each Effect in terms ofthe possible severity of its consequences.

Calculate the Criticality of eachFailure Mode (i.e. each risk)

These values are calculated by multiplying the likelihood andseverity ranks together.

Identify any existing Controls Analyze the process map, identify any monitoring or detectionsystems, mitigation systems, etc, and assess their impact on theassigned Criticality scores.

Prepare an Action Plan Identify courses of action and establish how these actions will beassessed for impact upon the process.

Table 7.2 Suggested list of likelihood ratings for FMEA Failure Modes(“risks”). The real range that might be encountered is 1 to 9 becauseanything rated as impossible is not a real risk, and therefore irrelevantto the FMEA, and anything rated as 10 should never be encountered ina real-world situation

Likelihood Rating Comments

0 Impossible Can never happen, hence it is not a real risk.1 Very unlikely2 Unlikely4 Possible6 Likely8 Very likely

10 Certain This circumstance should never exist in a real world situation.


Table 7.3 Suggested list of consequences of Failure Effects (“risks”)in an FMEA. The real range that might be encountered is 1 to 10,although the ranking of a risk as 9 or 10 should only refer to FailureModes that are assessed as very unlikely (see Table 7.2)

Consequence Rating Comments

0 None Hence it cannot be considered a real risk.1 Trivial In reality, there is no measurable adverse risk.2 Minimal The effect is, in reality, more of a nuisance or

inconvenience with no identifiable effect onpatient care.

4 Minor e.g. an adverse effect upon efficiency, but without anymeasurable effect on treatment outcome.

6 Serious e.g. significant adverse effect upon a patient’streatment outcome.

8 Major e.g. loss of embryos, OHSS,a infection of patients orstaff.

9 Extreme e.g. loss of life, damage to facility.10 Catastrophic e.g. loss of multiple lives, destruction of facility.

a OHSS Ovarian Hyperstimulation Syndrome

A Risk Matrix is a tabulation of the scores for all the risks identifiedby an organization in a way that allows them to be deemed, accordingto a partitioning of the Criticality scores, as (for example) no risk, lowrisk, medium risk, significant risk or high risk. These classificationscan then be used in prioritizing the identified risks for management.

In engineering, Pareto methodology is commonly used to identifythe 20% of “significant risk” issues that cause 80% of the process vari-ability (Hutchison, 1994), i.e. the most important Failure Modes toaddress. However, in biology this sort of relationship is not alwaysapparent or easy to determine, and decisions are often based on expe-rience; this is a significant part of the reason why an IVF lab needs tohave a senior, experienced scientist acting as its Director.

The endpoint of the FMEA is the creation of an Action Plan whichmust then be implemented. It is vital that for each change that is insti-tuted within the Action Plan there is a system for monitoring its effec-tiveness and efficacy – without this there will be no evidence of positive


Figure 7.3 The Plan–Do–Check–Act or ”PDCA” cycle.

benefit. This overall sequence can also be described as the “Plan–Do–Check–Act” or PDCA Cycle (see Figure 7.3), sometimes called theShewhart or Deming cycle (Hutchison, 1994). PDCA-derived knowl-edge of how a process is currently performing is used to identify andtest process changes (improvements). Scientists will immediately rec-ognize the PDCA Cycle as a simple expression of fundamental scientificmethod.

Root cause analysis

Root Cause Analysis (RCA) is a sequence of steps by which the under-lying causes (“Contributory Factors”) of adverse outcomes are identi-fied with the goal of preventing the recurrence of such events. Unlessan RCA is undertaken with the full support of upper management itmight well be performed in a perfunctory manner for the sole purposeof meeting some regulatory requirement.

There is increasing application of RCA in healthcare due to a growingrecognition that the complexity of medicine and its delivery is drivingthe incidence of adverse events (Bogner, 1994; Sharpe and Faden, 1998).Again it must be emphasized that most errors result from faulty systems


rather than human error: poorly-designed processes put people insituations where errors are more likely occur (Reason, 1994). Riskmanagement experts in all industries emphasize system failures andsystem-driven errors over direct human error.

The outcome of an RCA will identify three philosophically differenttypes of issue:

(a) blame, responsibility and emphasis on human error;(b) contributory vs. causative factors; and(c) the degree of efficacy of corrective actions or solutions.

To be effective, it must be accepted throughout the organization thatRCAs are for improvement purposes and not to assign blame – inkeeping with the principles of continuous improvement intrinsic toTQM philosophy.

Because of its history in manufacturing industries, RCA is morereadily accepted there by management and employees compared tohealthcare organizations where it is still typically seen as just anotherexpensive regulatory requirement that does not add value. An addi-tional problem among healthcare workers is a perception that RCAsemphasize human error, raising the spectre of blame, litigation and per-sonal liability. Consequently, there is still resistance to learning aboutRCAs, resistance to their performance, and lack of support at all levelsfor their effective application in healthcare.

Undertaking an RCA (see Figure 7.4) is essentially the same asan after-the-fact FMEA and has six stages (see Table 7.4). First andforemost it is essential that everyone involved understands that thegoal is to discover everything possible about the incident, with thefocus on the systems and processes that could have contributed tothe event happening and on the prevention of future recurrence.Every “Contributory Factor” that has no lower-level derivative is con-sidered to be a root cause. However, when preparing reports oneshould use the expression “Contributory Factor” instead of the word“cause”; it is better-received psychologically and has less blatant legalimplications.


Understand the issue or problem

Identify all possible Contributory Factors

By asking"why" and "how",

classify the relevanceof each possible

Contributory Factor

Insufficient dataContributory Non-contributory

Obtain data andreclassify

Determine Corrective Actionsand monitoring mechanisms

Generate an Action Plan

Implement the Action Plan

Follow-up each Corrective Action to closure

END

Figure 7.4 A flow chart for performing a Root Cause Analysis (RCA).


Table 7.4 The six basic stages of performing a Root Cause Analysis (RCA)

Step Procedure or action Explanation

1 Understand the issue. Discover everything possible about the incident, with the focus onthe systems and processes that could have contributed to theevent happening.

2 Develop a diagram of theContributory Factors.

Ask “Why?” or “How?” for each Contributory Factor so that it canbe classified as either “Insufficient Data,” “Non-Contributory”or “Contributory.”

3 Resolve items classified as“Insufficient Data.”

Obtain data, either by collation, observation, or perhaps evenexperimentation, for all real or potential Contributory Factors,identified as having “Insufficient Data,” and then reclassify eachof these factors, as appropriate.

4 Generate an Action Plan. The Action Plan should include at least one corrective action orimprovement for each Contributory Factor that was identified.An RCA reporting table is developed for the Action Plan (seetext).

5 Implement the Action Plan. Implement both the planned Corrective Actions and theirmonitoring processes.

6 Follow-up. Assess the effectiveness of the Corrective Actions via themonitoring processes.

The goal of an RCA is to develop an Action Plan that includes at leastone corrective action or improvement for each Contributory Factorthat was identified. Before the Action Plan can be implemented anRCA reporting table is developed, with the following columns for eachContributory Factor.

• Corrective Action(s).• Person(s) responsible for implementing the Corrective Action(s).

Unless someone is specifically assigned to oversee each CorrectiveAction they will not all be pursued, and the expectations of successfuloutcome will be greatly reduced.

• Action due date. This sets the timetable for each Corrective Actionwhich avoids things getting put off indefinitely, as well as givingeveryone involved a sense of this being a finite process and a date bywhich the problem will be resolved.


• Measurement technique. There must be some way of determiningthat the Corrective Action has, indeed, had an effect.

• Person(s) responsible for monitoring each Corrective Action. Again,unless someone is specifically assigned to oversee each CorrectiveAction, follow-through can drag on.

• Follow-up date. This is the date by which everyone involved canexpect significant progress to have been made.

Depending on the outcome of the RCA, it might be necessary to reviseand repeat the RCA process if the problem was not fully resolved.

Using Root Cause Analysis

To illustrate the use of Root Cause Analysis, let us consider an exam-ple of troubleshooting where an IVF Lab Director is concerned aboutthe poor quality of the sperm preparations that are being producedby the laboratory (see Table 7.5). The pertinent technical backgroundon sperm preparation methods has been extensively reviewed previ-ously (Mortimer, 2000; Mortimer and Mortimer, 1992) and will notbe repeated here.

Issue Even with normal semen samples our sperm preparationmethod provides a low relative yield (about 15% of the motilespermatozoa initially applied to the gradient) with only 65–70%progressively motile spermatozoa and frequent contamination ofthe post-gradient sperm population with other cells and debris.A swim-up is needed to improve the motility, but that takes moretime and further reduces the yield.

Method The SOP for sperm preparation in our imaginary IVF labcan be summarized as follows. Apply 1 ml of liquefied semen to agradient comprised of 0.5 ml layers of 80% and 40% PureSperm(Nidacon International AB, Goteborg, Sweden) in a Falcon 2003tube. Spin at 2000 rpm for 15 minutes in a Centra CL2 cen-trifuge fitted with a model 809 rotor. Remove and discard thesupernatant (seminal plasma and gradient layers) and resuspend

Table 7.5 An example of troubleshooting using RCA to address the issue of “Why do our sperm preparationshave variable, and often low, proportions of progressively motile spermatozoa, and are contaminated by othercellular elements and debris from the original semen sample?”

Contributory factor Classification Effect Action Notes

Gradient colloidconcentrations

Non-contributory Colloid concentrations of 40% and 80% arethe recommended layers for PureSpermgradients.

None required.

Gradient layer volumes Contributory Only 0.5 ml layers, this will lead to morerapid “raft” creation and blockage ofsperm passage to the lower layer(s) –hence reducing the yield.

Use larger volume layers of 1.5 or2.0 ml.

1

Centrifuge tube diameter Contributory The tube diameter is relatively small, hencethe layer interface area is reduced. Thisleads to more rapid “raft” creation andblockage of sperm passage to the lowerlayer(s) – hence reducing the yield.

Use a larger diameter tube, e.g.Falcon 2095.

1

Centrifuge tube shape Contributory A round bottom tube provides a lessdiscrete pellet than a conical tube.

Use a conical bottom tube, e.g.Falcon 2095.

1

Fixed-angle centrifuge rotor Contributory A fixed-angle rotor means that the pelletwill not be deposited in the very bottomof the tube but “smeared” over a largerarea of the bottom and one side.

Change to a centrifuge that has aswing-out or swinging buckettype rotor.

1

Centrifuge buckets are notsealed

Non-contributory Risk of aerosol contamination in thelaboratory if a tube were to break duringcentrifugation.

None in this regard. 2

(cont.)

Table 7.5 (cont.)

Contributory factor Classification Effect Action Notes

Centrifugation speed of thefirst spin

Contributory The stated rotor has a radius of 12.7 cm,giving a centrifugation force ofapproximately 570 g which is higher thanrequired.

Decrease the speed to give acentrifugation force of 300 g.

1

Centrifugation time of thefirst spin

Contributory Only 15 minutes might not be sufficient forthe spermatozoa to reach their isopycnicpoints on the gradient.

Increase to 20 minutes. 1

Technique for harvestingthe pellet from thegradient

Contributory Removal of all the layers above the pelletexposes the pellet to contamination byresidual material from the upper layersthat contaminates the inner surface of thetube.

Remove only the seminal plasmalayer, the upper (semen/40%colloid) “raft,” 40% colloid layerand the 40/80 “raft” to leavemost of the 80% colloid layerintact to protect the pellet fromcontamination. Then harvest thepellet by aspiration through theremaining 80% colloid layer.

1

Resuspending the pellet ingradient tube

Contributory The pellet will be contaminated withresidual material from the seminalplasma and upper layers (and “rafts” ofpoor/dead sperm, other cells and debris)that coats the inner surface of the tube.

After recovering the pellet fromunderneath the remainder of the80% colloid layer, transfer it to aclean conical tube beforeresuspending the cells in freshculture medium.

1

Centrifugation speed of thesecond spin

Non-contributory Although 570 g is above the recommended500 g, centrifugation force does notbecome harmful until 800 g.

None required, but thecentrifugation speed should bechanged to 500 g so as toconform to the standardprotocols.

1

Centrifugation time of thesecond spin

Non-contributory 15 minutes is the usual duration for thisspin.

None required.

Harvesting of the washedsperm pellet

Contributory Because a round bottom tube was used in afixed-angle rotor the pellet will be“smeared” over a larger area than if aconical tube had been used in a swing-outrotor, rendering complete harvestingmore difficult.

Change to a centrifuge that has aswing-out or swinging buckettype rotor and use a conicalbottom tube, e.g. Falcon 2095.

1

Resuspension of the washedsperm pellet

Non-contributory None. None required.

Performing a second washstep

Contributory Each wash cycle causes the loss of somespermatozoa, and this could contribute tothe perceived low yield.

Since this step is not necessary itshould be omitted from theprocedure.

1

Swim-up from theresuspended washedsperm suspension

Contributory Although many labs perform such aprocedure it is not necessary if the densitygradient procedure is performedcorrectly. The presence of even 15%immotile sperm in the final preparationhas never been shown to be harmful toeither the fertilization rate or embryoquality.

Do not perform this step. 1

1.As per manufacturer’s instructions and established optimized methodology.2.While this factor is non-contributory to the current issue, it does, however, represent a risk of possible aerosol contamination of the laboratoryif a tube breaks while being centrifuged. In accordance with good laboratory practice the centrifuge should be upgraded to one with sealed buckets.In this case, this could be achieved by replacing only the rotor.


the pellet in 1 ml of fresh medium. Spin again at 2000 rpm for 15minutes and again aspirate and discard the supernatant. Repeatthe washing cycle. Resuspend the final sperm pellet in a small vol-ume of fresh medium and overlay with 0.6 ml fresh medium thenallow to swim up for 30–60 minutes at 37 ◦C in a CO2 incubator.Recover the upper 2/3 of the overlay and assess.

Constructing the RCA The process map for sperm preparation isshown in Figure 7.5. It clearly illustrates the need to delve deeperuntil no further subordinate level processes remain because, inreality, the sperm preparation process actually comprises threeseparate processes. Only by having dissected the process to thislevel can the following comprehensive troubleshooting exercisebe undertaken. The Contributory Factors and their classificationsare listed in Table 7.5. Benchmark criteria are taken from therecommended method for using PureSperm density gradients asper the manufacturer’s package insert and previously publishedinformation.

Conclusions of the RCA While the method might not seem tobe too different from that used by many other labs, there arenumerous factors within the sperm preparation protocol thatcan contribute to reduced yield in quantitative and/or qualitativeterms. In many cases the individual variation might not cause a

Figure 7.5 A process map for sperm preparation in IDEF0 format. The first panel shows theobligatory parent process map which is then broken down into sub-processesdisplayed in multiple, hierarchically linked process maps. Each map should haveno more than 6 steps (although 9 could be used without breaking the formalnumbering convention). The first map (second panel of Figure) shows the basictasks involved in the process (steps 1 to 6), with step 2 being shown in greaterdetail in the third panel. The annotation “A2” outside the bottom right-handcorner of step 2 in the 6-step chart of process A-0 (its “node number”) showsthat the box has been detailed as a “child” diagram (i.e. the third panel of theFigure). In diagram A2, steps 1 and 2 are identified as being displayed in greaterdetail in child diagrams A21 and A22 (the fourth and fifth panels of the Figure). Forclarity, information concerning the controls and mechanisms have been omitted.


marked degradation in the outcome, but taken together there willbe substantial detrimental impact. The final conclusion is that thesperm preparation method should be replaced by one based on themanufacturer’s instructions and established optimized method-ology. Furthermore, because this change is not going to have anunknown or uncertain outcome on the process, there is no needto perform any validation studies of the “new” method.

Conclusions

The tools described in this Chapter are fundamental to all qualityand risk management activities. Familiarity with their principles, andwith their use, makes quality and risk management far less dauntingprospects. Indeed, with these tools available to you, there is no needto be worried about any of the “scary” concepts or procedures thatadministrators or business managers, quality managers and risk man-agers bandy about, such as “Troubleshooting” (see Chapter 8), “RiskManagement” (see Chapter 9) or “Benchmarking” (see Chapter 10).As scientists we already know and understand all the fundamental prin-ciples that underpin these tools, and most of them are no more thanformalized applications of scientific method.

Once tools like FMEA or RCA have been explained, many scientists,exclaim “but that’s just common sense!” – and so it is. The real problemlies in the number of people working in IVF (and not just in the lab)who aren’t aware of these tools, or can’t use them – or, far worse, whodon’t see the value in their use.

8

What’s gone wrong? Troubleshooting

There are several different conceptual ways of looking at problems (seeTable 8.1). While much of this book is about being proactive, no systemwill be perfect and sometimes you will need to deal with a problemthat has occurred or an issue that is affecting the lab, and you willhave to be reactive. This Chapter is about what to do when things havegone wrong, including dealing with problems and troubleshootingthem. Learning how to deal with these subjects is of interest to IVF labpeople. αlpha, the international society of scientists in reproductivemedicine (www.alphascientists.com), held an internet conference onthis subject in 1998 (Elder and Elliott, 1998), and the αlpha workshopat the 11th World IVF Congress held in Sydney in May 1999, structuredas a foundation workshop in reproductive biology, concluded with asession by Jacques Cohen on the practical application of this knowledgein the ART laboratory, with particular reference to troubleshooting.

Having to be reactive

Although we all believe (hope?) that we’re doing everything right, thatour success rates will be high (and remain high), and that things willcontinue to run smoothly, we all know that from time-to-time therewill be problems. Sometimes problems are caused by factors outsideour control, but sometimes they arise because we have not paid atten-tion to detail, or have not bothered keeping up-to-date on some lessinteresting aspect of the field, or because someone else (e.g. a supplier)has changed something and either not told us or we did not recognize

135


Table 8.1 Conceptual approaches to perceiving and dealingwith problems

Action Nature of problem Response type Outcome or effect

Remedial Existent Reactive Alleviate the symptom(s)Corrective Existent Reactive Prevent recurrencePrevention Non-existent Proactive Prevent occurrence

the importance of the change at the time. Regardless of the origin ofthe problem, sometimes we have to troubleshoot a part of our system.Of course, the more proactive we are the less likely this event will be,the less serious it is likely to be, and hopefully the easier the problemwill be to solve. But sometimes we just have to be reactive. In terms ofprocess analysis, Root Cause Analysis (RCA) is the conceptual basis oftroubleshooting.

Troubleshooting

A generic illustration of the troubleshooting process is provided inFigure 8.1. From this flow diagram it is clear that the application ofscientific method is fundamental to the process and, indeed, is essentialfor effective troubleshooting.

Hopefully, many readers will ask “But isn’t this all just commonsense?” – and the answer would have to be “Yes” – but it is notnecessarily so to someone who has not had the benefit of scientifictraining. For someone who has successfully learned scientific methodthis structured approach to problem-solving should have becomeintuitive – and a good scientist will essentially go through all the stepsof an RCA automatically, if subconsciously, whenever they are con-fronted by a problem. The need to formalize the procedure is to ensurethat less well-trained or less-experienced people can apply the sametechnique. It also provides the framework whereby the performance ofthe process is documented – an essential part of all laboratory work,and crucial within the context of accreditation.

137 What’s gone wrong? Troubleshooting

Figure 8.1 An illustration of the troubleshooting process, demonstrating its fundamentalequivalence to Root Cause Analysis (RCA). Redrawn from Mortimer (1999).

An effective troubleshooter can be likened to a scientific detective.Using just the same rigorous objective approach as Sherlock Holmes,one must examine every detail and decide whether it is relevant to themystery at hand (“contributory” in RCA terminology). Recognizingsome of the more obscure factors might require specialist knowledge –from a Dr. Watson on your staff – and perhaps failure to recognizesuch a factor might be a great hindrance to reaching the goal of solving


the mystery. This is why it is so important when teaching someone amethod that you take the time to explain “why” things are done theway they are (and perhaps why they are not done a certain way): justdescribing the “how” will not produce a truly competent scientist.

Identifying and measuring extrinsic factors that affect a processrequires a thorough understanding of the basic science of the pro-cess. Not just the biology, but also the chemistry, the physics and eventhe engineering that affect the process, as and when required. Withoutsuch comprehensive knowledge, your ability to minimize or eliminatethe problems will be greatly compromised.

Troubleshooting a process

An example of needing to look at many systems together when trou-bleshooting is illustrated in an investigation of low fertilization rates.It had been shown in numerous studies that a low glucose environ-ment was beneficial for embryo development from the zygote to 8-cellstage, and so many laboratories used this type of medium in place of theglucose-containing medium that they had previously used for all of thegamete handling, fertilization and early embryo development steps. Insome of these laboratories, the fertilization rate by IVF (rather thanICSI) was not as good as hoped, but the cleavage rate of the embryoswas very good. A side-effect of this outcome was that some peopletended to see IVF as a more “risky” approach than ICSI, in terms of“guaranteeing” fertilization, so there was a shift towards a higher pro-portion of ICSI than IVF cycles. Whether this was a good or a bad thingis highly debatable – but since it isn’t really the focus of this example,we’ll leave it there.

The reason for the lower-than-expected fertilization rates with IVF,and the relatively higher fertilization rates with ICSI, was often ratio-nalized as being patient-related; other laboratories considered thatit was probably medium-related, but felt that the improvement inembryo development at Day 3 was worth the lower number of embryosobtained. However, since the principles of Quality require providing


“more,” the apparently compromised fertilization rate was seen as areason for troubleshooting, to determine whether improvements couldbe made in the number of zygotes obtained from an IVF cycle.

The steps taken in this exercise were the same as those outlined inthe Troubleshooting flowchart (Figure 8.1):

There seems to be a problemAn apparently lower-than-expected IVF fertilization rate.

Collect data• Fertilization rates, and the incidence of complete failure of fertiliza-

tion:• for IVF cycles (retrospectively); and• for contemporaneous IVF and ICSI cycles.

• Embryo quality and cleavage rates (retrospectively).

Analyze dataReview of the fertilization rates and embryo quality.

Define the problem(s)(a) The new medium results in better embryo development (not a

problem); but(b) the number of zygotes is reduced for IVF cycles.

Design experiments(a) Develop a hypothesis:

• If the number of zygotes is lower than expected, then is it becausegamete function is impaired?• Since the number of zygotes from ICSI is not affected, the

oocytes are probably not affected.• Therefore, maybe sperm function is affected:• we know that spermatozoa metabolize glucose (reviewed by

Ford and Rees, 1990);


• we know that glucose is necessary for mouse sperm capacitationand hyperactivation (prerequisites for zona penetration) (Fraserand Quinn, 1981); but

• it had been reported that IVF could be achieved successfullyin medium without glucose and that the proportion of motilespermatozoa was not affected (Quinn, 1995).

• Therefore – could it be that sperm function is affected, ratherthan the proportion of motile spermatozoa? In other words, wasthe quality, rather than the quantity, of sperm movement beingaffected?

(b) Test the hypothesis: the procedure followed has been presented indetail elsewhere (Mortimer, 2002). However, briefly:• Semen samples that met the lab’s criteria for IVF were used

(n = 11).• The samples were washed as for IVF, using PureSperm gradients.• The pellets were harvested and resuspended to 0.5 ml, then

divided into two portions.• 4 ml of medium ± 2.8 mM glucose were added and the sperm

suspensions centrifuged.• Pellets were resuspended to 0.5 ml in medium ± 2.8 mM glucose

(according to treatment group), incubated for 60 minutes in a6% CO2 atmosphere.

• Sperm motility was assessed using Computer-Aided SpermAnalysis (CASA).

Collect dataThe results showed that while the proportion of motile spermatozoawas unchanged in the incubation period by the presence or absenceof glucose, the proportion of hyperactivated spermatozoa was signifi-cantly reduced in medium that did not contain glucose.

Prioritize problemsThere was only one problem being investigated, that of gamete (sperm)function, so no prioritization was necessary.


Resolve problemsThe problem was with the fertilization rate in low-glucose medium.Since it was apparently solved by the addition of glucose to the samemedium, a two-stage medium was created for all clinical IVF cycles.The fertilization medium was the low-glucose medium, supplementedwith 2.8 mM glucose, and the cleavage medium was the originallow-glucose medium.

Monitor outcomeThis had to be done retrospectively:

• The number of IVF cycles with complete failure of fertilization wasreduced significantly, and the fertilization rates achieved over all IVFcycles increased significantly.

• The embryo cleavage rate was not affected by the addition of glucoseto the fertilization medium.

• But Day 3 embryo morphology was significantly improved.

These outcomes led to the next question: Could it be that the incubationof gametes in glucose-containing medium confers a developmental advan-tage? But since the observation is a positive side-effect of the solution,we’re now past troubleshooting and into research.

Troubleshooting an incident

Troubleshooting can also be used as an opportunity for re-educationand training – take the story of the CO2 incubator (fortunately it wasa “holding” incubator, so no gametes or embryos were affected).

There seems to be a problemDuring a busy day, the temperature of a water-jacketed incubatordropped to 35 ◦C and stayed there, causing the incubator to alarm,setting off the external dialler.


Collect dataAll of the embryologists were questioned about the general perfor-mance of the incubator. A trainee volunteered the information that,unlike the other incubators, this one had been fast to lose, and slow toregain, its temperature during the day, but every morning the temper-ature checks showed that it was at 36.8 ◦C – so “obviously” there wasno problem.

Analyze dataThese observations indicated that there was a problem with the heatingor temperature maintenance function of the incubator.

Define the problem(s)The question to be answered then was “What keeps the temperatureof an incubator stable?”

Design experimentsSince an incubator’s temperature relies upon the heating current andthe heat sink (the water jacket) for temperature maintenance, the per-formance of each of these was checked.

Collect data• There were no obvious problems with the electricity supply to the

incubator, and it was connected to an uninterruptible power supply,so power surges were considered unlikely.

• Further investigation of the electrical circuits within the incubatorwas considered to be warranted only after all other avenues wereexhausted.

• The next area of investigation was the water jacket. Enquiries revealedthat the incubator’s water jacket had not been topped up as part ofthe lab’s routine maintenance.


Prioritize problemsBecause the water jacket was probably at a low level, this was consideredto be more likely to be the source of the problem than the electricalcircuit.

Resolve problemsThe water jacket was topped up with salt-enriched reverse-osmosiswater. The incubator regained its temperature almost immediately,and the temperature remained stable throughout the day.

Monitor outcomeNothing was returned to that incubator until the incubator’s perfor-mance was confirmed to have been stable for a week. Each day, theincubator’s door was opened and closed several times, and no problemwith regaining temperature was observed.

As a result of this incident:

• the trainee was instructed that all unusual observations must bewritten down and reported;

• all of the lab staff received an in-service refresher on the care andmaintenance of water-jacketed incubators; and

• the Lab Director updated the training manuals, the SOPs andthe troubleshooting manual so that similar incidents might beprevented.

Basically, all that is required for effective troubleshooting is the abilityto go through all the aspects of the perceived problem in an organizedway. However, there is an inescapable need for a thorough knowledgeof the basic science that underlies the field, as well as a moderate under-standing of practical considerations that affect how we do things in theIVF Lab. This latter need encompasses a broad range of areas of prac-tical knowledge including, for example, various aspects of engineeringas they relate to equipment design and operation, Clearly, being a goodIVF scientist, and especially an IVF Lab Director, requires a great deal


more of one than just being a skilled embryologist – and this is the basisof the standard accreditation requirement for being a “learning organi-zation.” Education is a lifelong process and as you become more seniorthere is a concomitant need for broader knowledge that includes manyareas that might previously have seemed extraneous or superfluous tobeing a good embryologist.

9

Risk management: being proactive

Risk management is all about being proactive. Risk analysis is under-taken to identify where things might go wrong. This does require someexperience and, indeed, the wider your experience the more likely youare to be able to recognize issues as problems or to identify poten-tial problems. The general principles of risk management are pre-sented quite lucidly in the Australia/New Zealand Standard AS/NZS4360:1999 (Standards Australia, 1999) and will be discussed later in thisChapter.

“Why bother with that? It’s never happened here!”

How often have you identified a potential problem, only to be told “Ohthat’s never been a problem here,” or “We’ve never had a problem withthat,” or “Why waste our time, that’s just so unlikely”? Of course, thetruth is that this head-in-the-sand mentality is exactly why some of theworst problems in IVF labs have arisen. We have personally experiencedsituations where an identified risk was pooh-poohed by the MedicalDirector, General Manager or equivalent, only to have just that problemoccur a few weeks later – although professional confidentiality clearlyprecludes quoting specific examples! The dreaded “It’s never happenedhere . . .” should probably be considered a warning bell that a proper riskassessment should be undertaken forthwith. After all, Captain EdwardJohn Smith hadn’t hit any icebergs before the maiden voyage of theTitanic, either!

145


Can we eliminate risk?

Of course, we would all like to believe that we can eliminate all riskfrom our laboratories, but in reality we must consider things in practicalterms and recognize that different risks are amenable to different levelsof resolution.

Risk elimination

There are a few risks that we can eliminate from our IVF labs, e.g.banning the wearing of perfume and aftershave to reduce the level ofvolatile organic compounds (VOCs).

Risk avoidance

In risk management terms, risk avoidance is used to describe aninformed decision not to become involved in activities that lead tothe possibility of the risk being realized. Prohibiting smoking in anarea where there are flammable, volatile solvents is a good exampleof risk avoidance. An extreme example of risk avoidance in the “realworld” is shown in Figure 9.1.

Risk reduction or risk minimization

These terms are essentially synonymous and are used in risk manage-ment to describe the application of appropriate techniques to reducethe likelihood of an adverse event, its consequences, or both. A realworld example of this type of action is shown in Figure 9.2. In theIVF lab, the management of the cryobank is a good area to see riskminimization strategies: not only should the cryotanks’ usage of liquidnitrogen be monitored on a regular basis (e.g. their level at weekly top-up for tanks not fitted with an autofill system) but they must also beconnected to low-level alarms in case of tank failure, especially outsidenormal working hours. Further dimensions of risk minimization in

147 Risk management: being proactive

Figure 9.1 A rather extreme example of risk avoidance. The sign was photographed by DMin a suburban park in Cape Town, South Africa.

this example are the maintenance of a spare cryotank, partially filledwith liquid nitrogen, ready to receive specimens from a suspect cry-otank, or the “split storage” of each patient’s specimens between twoseparate cryotanks. Risk reduction/minimization is discussed in somedetail later in this Chapter.

Risk transfer

This is a risk management concept that describes the shifting of theburden of the risk to another party. Probably (hopefully!) the mostcommon example of risk transfer in an IVF lab is insurance.

Risk acceptance or risk retention

This risk management term is used to describe an informed decisionto accept the consequences and likelihood of a particular risk (e.g. thepossibility of a meteor strike, discussed in Chapter 4).


Figure 9.2 A common example of risk minimization.

How do we manage risk?

An interesting illustration of how risk management typically usesseveral strategies to achieve a comprehensive solution is that of fire(Figure 9.3).

We have fire insurance so that if our premises are damaged or lostdue to fire we will be recompensed and be able to repair or rebuild. Butwhile such an event would be disastrous for us professionally, as wellas personally, it could be even worse for the patients. That is why wealso have fire drills to evacuate the premises in case of a serious fire, fire


Figure 9.3 An illustration of the multi-faceted approach usually taken towards fire; embrac-ing various risk avoidance, risk prevention, risk minimization and risk transfertactics.

extinguishers to put out any fires that might occur before they can domuch damage, and fire alarms to alert us that a fire has started. Evenso, we could suffer appreciable financial loss during repairs, or loss ofprofessional confidence that leads to a drop-off in referrals, a fall inworkload, and hence a down-sizing of the program with all that thisentails. Therefore, it is wise that we also have fire prevention to try andavoid a fire occurring in the first place.


This overall scheme might seem somewhat redundant, but it is reallyan example of strategic planning – we should look at its componentsin reverse order:

• Fire prevention as an expression of risk avoidance: we avoid doingthings or creating situations that increase the risk of fire, e.g. refrain-ing from smoking in areas where there are solvents, storing solventsin proper flammables cabinets, ensuring that electrical equipment isproperly wired.

• Fire extinguishers to reduce the amount of damage that a fire can do.• Fire alarms, to make sure that we get as much warning as possible of

a fire, so as to use a fire extinguisher sooner, or have more time toevacuate staff, patients and data backups.

• Fire drills to make evacuations more efficient.• Fire insurance as the last resort to allow us to rebuild and/or recom-

pense as effectively as possible.

This sort of multi-layered strategy should be applied throughout theIVF Center, including the laboratory, or, indeed, any workplace.

Risk reduction

Rather like the Quality Cycle (see Chapter 3), risk reduction is aniterative process that never reaches perfection. While we can reach anasymptotic state where all controllable risks have been avoided, elimi-nated, or controlled to the best of our abilities, the exclusion of all risk –“perfection” in this sense – is not a realistic goal. In the real world,constraints will be applied to this process, for example:

• cost-benefit scenarios, usually determined at the corporate level;• physical limitations, e.g. available space for backup facilities such as

spare incubators or cryogenic storage tanks; and• shortage of trained staff to ensure necessary “slack” in the system

(see Chapter 4).


Taking responsibility

As professionals, we all must accept responsibility for our actions. Thissubject was considered in the discussion on identifying “high risk”IVF labs in Chapter 4, but it is such an important aspect that it will bereiterated here.

When a team member does not take enough care to ensure that theyhave performed – and completed – all the tasks that were assigned tothem it can be either intentional or unintentional. In either case it isunprofessional behavior, but at least if it is unintentional it is only anexpression of inadequate or incomplete training, a problem for whichthere should be an easy remedy. The intentional omission of parts ofone’s job can only continue without jeopardizing outcomes if there areothers who take the time (and/or are prepared to make the time) toensure that the whole process is completed: in effect they are the “safetynet.” All professionals must be prepared to work without a safety net;if someone cannot do this then (s)he should not be working in an IVFlab.

The benefits of risk management

The following benefits will ensue from an effective risk managementprogram:

• Better knowledge of the process, thereby helping to eliminate theneed for re-work.

• More efficient processes and systems, leading to savings in time andresources, and hence money.

• Reduced stress levels for the staff.• Reduced risk of error.• Higher quality of service and hence customer satisfaction.• Reduced legal implications, and hence reduced potential for liability.• Lower risk rating can lead to reduced insurance premiums.• Documented organizational history.


Basically, poorly-designed processes or ones vulnerable to failure aretargeted for replacement by better-designed processes; existing pro-cesses also benefit by virtue of introducing improvements beforeadverse events occur. Properly conducted and documented FMEAsrecord the evolution of the laboratory’s processes. This should preventpast mistakes from being repeated, e.g. by well-intentioned but inex-perienced staff, and will enable new employees to learn the laboratory’ssystems and their operational characteristics more rapidly.

Developing a risk management program

Like an effective Quality Management System, a successful risk man-agement program must be integrated into all levels of the organization.Clearly, therefore, a risk management program must include the wholeIVF Center – restricting it just to the laboratory is short-sighted andwill severely limit its total benefits to the Center. After establishing theextent (scope) and purpose (goals) of the risk management program,an individual or team should be designated and assigned responsibilityfor its development and implementation.

Seven general principles create a framework within which effectiverisk management can be accomplished.

1. Global perspective Consider the IVF Center’s development withinthe context of the “industry” – competing Centers (getting results,referrals, etc), government regulations, licensing, accreditation, etc.Recognize the potential value of opportunity but also the potentialimpact of adverse effects.

2. Forward-looking view Always look to the future. Identify uncer-tainties and anticipate potential adverse outcomes.

3. Open communication Encourage and facilitate formal, informal,and impromptu communication and the free flow of information atand between all areas and levels of the Center. Always value the indi-vidual voice: anyone can contribute unique knowledge and insightthat can help identify or manage risk.


4. Integrated management Make risk management an integral part ofthe Center’s activities.

5. Continuous process Like the quality cycle, risk management is anever-ending process.

6. Sharedvision Build a vision of quality of service and care throughoutthe Center based on common purpose and shared ownership; focuson results.

7. Teamwork Everyone in the Center must work cooperatively towardscommon goals. No-one knows everything, so knowledge, skills andtalents should be shared among all members of staff.

Anyone who has experience of a formal laboratory accreditation pro-cess will recognize the above seven principles as core elements of suchschemes. For an IVF Center the ultimate goal of risk management isto reduce the likelihood of having to expend resources on dealing withcatastrophic incidents – resources which will probably not be readilyavailable when needed. This is why proactive risk management is soimportant, and why familiarity with the fundamental concepts andtools used in risk management should be an essential part of the train-ing for all scientists working in IVF labs.

A widely-accepted standard for developing a risk management pro-gram is AS/NZS 4360:1999 (Standards Australia, 1999). An outline ofthe steps involved in this process is shown in Table 9.1 and illustratedin Figure 9.4.

The Risk Register is the centralized archive of all documenta-tion pertaining to an organization’s management of its identifiedrisks.

Illustrative examples of risk management in the IVF lab

The following examples have not been presented in the strict format offormal risk analyses, but rather as explanations of the issues and howeach IVF lab should go about evaluating and managing the risks foritself.


Table 9.1 Major elements of a risk management plan as described in theproposed revised version of AS/NZS 4360 (Standards Australia:www.standards.com.au)

Clause Plan element Explanation

2.3.1 General Establish the purpose of the plan and embed the riskmanagement process into all the organization’s processes (thisshould be the primary objective of the plan).

2.3.2 Ensure support of seniormanagement

Without this any risk management is foredoomed.

2.3.3 Develop the risk managementpolicy

Defining and documenting the organization’s policy, includingmanagement’s commitment to it, is essential.

2.3.4 Communicate the policy Create an infrastructure to ensure that managing risk becomesembedded throughout the organization’s processes and intoits culture.

2.3.5 Establish accountability andauthority

Create the framework for delegated personnel to build andimplement the plan under the aegis of senior management.

2.3.6 Customize the risk managementprocess

Interpret and apply the Standard to create the organization’s riskmanagement processes, including specifying measures of itsperformance and criteria for judging its success.

2.3.7 Resourcing Identifying and allocating adequate resources is vital for thesuccess of the plan.

2.3.8 Organizational level riskmanagement plan

An integrated plan for managing risk at all levels of theorganization must be developed and implemented, includingits incorporation into all the organization’s processes andsystems.

2.3.9 Manage risks at the area, projectand team levels

Plans for each subordinate area within the organization must bedeveloped and implemented; these plans must be consistentwith, and integrated into, the organizational level plan.

2.3.10 Monitor and review Risks are not static so risk management must be a dynamicprocess.

Off-site collection of sperm samples

Even though it is better for any number of reasons that the semensample to be washed for use in IVF or IUI is produced on-site at theclinic, it often happens that the man produces it elsewhere and it isbrought to the clinic by the woman. While in virtually all cases itcould be correctly assumed by the clinic staff that the semen is fromthe woman’s partner, this is not always so. There have been incidentsin which a woman has brought someone else’s semen to the clinic,


Figure 9.4 A diagrammatic representation of the risk management process as per AS/NZS4360 (Standards Australia, 1999).

without her partner’s knowledge. To avoid the litigation and adversepublicity that might eventuate from such cases, the IVF Lab needs tohave a procedure in place in which the woman warrants that the semenis that of her partner. Similarly, if the man brings the sample to the Lab,he should attest formally that it is his. An example of the type of formone could use for this purpose is given in Figure 9.5.

Disposal of frozen embryos

Frozen embryos are held for a couple’s possible future attempts atpregnancy. It is standard that the embryos are considered to belongto the couple, and that any decision as to their disposition must beagreed to by both people. However, how the couple notifies the IVF


THE XYZ FERTILITY CENTRE

If sample delivered to the Lab by self: LABORATORY USE ONLY

Signed : _______________________

Lab Accession No.

Date of sample:

Time received by Lab:

XFC Chart No.

A0___–___________

__ / _______ / 200__

___ : ____hrs

_________________

Signed : _______________________

FRM-XFCLab012-20040324 Off-site release

If sample delivered to the Lab by partner or family member:

(Name: LAST, First)

(Name: LAST, First)

(Name: LAST, First)

(Date)

(Date)

I, ___________________________ confirm that

the semen sample that I have delivered to The XYZ

Fertility Centre today ______________ is my own

semen, produced off-site.

I, ___________________________ confirm that the semen sample that I have delivered to The XYZ Fertility

Centre today ______________ is the semen of my husband / partner,

which was produced off-site.

___________________________,

RELEASE FOR SEMEN SAMPLES PRODUCED OFF–SITE

Figure 9.5 An example an off-site sperm sample form that can be used to manage risk inthe IVF lab.

Center of their wishes, particularly when it is to donate the embryos, orto discard them, can have important ramifications. For example, oneCenter might consider that if one partner calls the clinic and says thatthe couple wishes to have their embryos discarded, this is good enoughto order the lab to dispose of the frozen embryos – not consideringthat this might not be the wish of both partners and thereby exposingitself to the risk of litigation, as well as causing significant distressto the unwilling partner. Another Center might follow the protocolthat following the phone call, information regarding the options fordisposal of the embryos (for informed consent) is sent to the couple’slast known address, along with a “request to dispose of frozen embryos”form that requires the couple’s wishes to be expressed explicitly, andthat has to be signed by both partners with each signature witnessed bya third party. Alternatively, it might require that both partners attend acounseling session at the Center before signing the form. This Centermight also allow a “cooling off” period after the receipt of the completedform, in case the couple change their mind.


By considering the risks involved in handling a request to dispose offrozen embryos, the second IVF Center has done all it can to ensure thatboth partners’ opinions are heard. Apart from reducing the exposureof the Center to litigation, it is also a more respectful approach thatadheres to the principles of Quality.

Parallel processing of sperm samples

Often in a busy lab, more than one sperm sample has to be preparedat a time. This has enormous potential for disaster (i.e. mixing upsamples), and a system must be developed which addresses this. Apartfrom the witnessing which should take place whenever a specimen ismoved from one container to another (see “Tools vs. Solutions” inChapter 5), there should be policies and procedures in place whichprevent one person from actually manipulating more than one sampleat once. A good solution to this potential problem is to have separatetest tube racks to isolate the materials for each case (Figure 9.6).

Labeling OPU tubes

It is a standard accreditation requirement, and good laboratory prac-tice, that all containers holding a patient’s tissues and fluids be labeled(ideally with two unique identifiers; see Figure 5.12). An example ofwhere this may not be followed in some IVF labs is in the labelingof the OPU tubes. If the IVF Center is one in which there is alwaysa 30–60 minute gap between oocyte retrievals, then the managementmight not consider this risk to be very great. However, if the Center isa very busy one, with many oocyte retrievals in a day, and very shortlag times between cases, then the management might see the risk ofmis-identification of oocytes as significant, and insist upon a labelingprocedure.

Therefore, the decision as to whether non-labeling of OPU tubesconstitutes a risk (i.e. mixing tubes between patients) is one whichmust be addressed by each IVF Center individually. However, labelingeach container would seem to be a prudent risk minimization strategy.


Figure 9.6 A system to minimize the risk of cross-contamination between sperm specimensbeing processed in parallel. All the tubes, etc, for a given specimen are placed ina single rack so that even if two specimens are being centrifuged together, theyare each harvested and resuspended separately. Note that the semen collectionjar is labeled with the man’s name and the andrology lab reference number(A04-0682) as well as the date and time of collection. Each tube and pipetteare labeled with the name plus both the andrology lab number and the oocyteretrieval case number (R04-0179) as identifiers, along with the date and thepurpose of each item. There is no label on the rack because it is the identity ofeach of the individual items that must be verified at each stage of the process; ifthere was a large label on the rack there might be a tendency to read only thatwhen going to work on a sample.

Temperature control during oocyte retrieval

Because of the meiotic spindle, the oocyte is exquisitely temperature-sensitive. Cooling causes depolymerization of the spindle (Pickeringet al., 1990; Almeida and Bolton, 1995; Wang et al., 2001), releasing thechromosomes into the ooplasm. Although the spindle repolymerizes


upon warming, there is considered to be a significant risk that one ormore chromosomes might not become reattached to the spindle andhence a state of aneuploidy would be created at the second meioticdivision, which occurs in response to the spermatozoon activating theoocyte at fertilization. It has been hypothesized that repeated cyclesof warming and cooling during oocyte retrieval might increase therisk of aneuploidy in human embryos, and hence the degree to whichan IVF lab’s systems protect the oocyte from such stresses might beat least a partial explanation of the differences in embryo aneuploidyrates reported by various centers. Anecdotally, in the multicentre studyof treatment-related chromosome abnormalities in human embryospublished by Munne et al. (1997), the center with the lowest observedrate of aneuploidy was the only one using an IVF Chamber work station(see Chapter 11).

The following example is presented as an excellent illustration thatnot only do perceived problems represent opportunities for improve-ment, but that if a known negative factor is recognized in a system thenthe effort is better spent on minimizing it (ideally, eliminating it) thanproving whether it is actually a significant factor in this particular cir-cumstance. It can be considered as the tempering of scientific methodwith common sense – there is no point proving that a known deleteriousfactor (in this case cooling oocytes, risking spindle depolymerization)is actually affecting your results, just eliminate it and thereby excludeit from both current and future concerns!

In November 2002 we were invited to participate in a culture systemsworkshop at Peter Kemeter’s Institute for Reproductive Medicine andthe Psychosomatics of Sterility (IRMPS) in Vienna. While attendingthe World IVF Congress in Sydney in May 1999, Peter, one of thepioneers of IVF in Austria, saw the culture system we had developedand implemented at Sydney IVF in late 1997 (see D. Mortimer et al.,2002). He was so impressed by the systems and the improvementsin implantation and pregnancy rates that he went home to Viennaand introduced the Cook culture media and the MINC incubator.These changes resulted in a 47% increase in pregnancy rates, as well as


Figure 9.7 Photographs illustrating various aspects of the oocyte retrieval process at theInstitute for Reproductive Medicine and the Psychosomatics of Sterility (Vienna,Austria). See Table 9.2 for detailed explanation.


significant improvements in fertilization and cleavage rates (Kemeterand Lietz, 2002). During the week we were guests in his Institute wewere asked to review their systems to identify any possible factors thatmight still be contributing to sub-optimal outcomes. Peter Kemeter’sgraciousness and openness were most impressive and he has generouslyconsented for us to reproduce our findings, which were presented inthe last session of the November 2002 Workshop, here. The factorsidentified in the IRMPS system for oocyte retrieval are summarized inTable 9.2, with the accompanying Figure 9.7 illustrating some of theobservations.

A reduced prevalence of early spontaneous pregnancy loss from 33%to 20% was observed in the period following the technical changes tothe OPU procedure although, due to the relatively small numbers ofcases, the difference was not significant. Other changes, e.g. in stim-ulation protocol, would also confound analysis of any direct effect ofthe technical changes.

Packaging systems for cryobanking gametes and embryos

Of great current concern to those working in human gamete andembryo cryobanking are issues arising from concerns over the riskof contamination either by other specimens in the same cryotank orby contaminated liquid nitrogen. Even though such an occurrence hasnever been reported for sperm or embryos, and the risk is generallyaccepted to be vanishingly small, it does represent a finite risk andall reasonable measures should be taken to reduce the chance of itsoccurring. A further dimension to resolving this issue is the abilityto achieve the “correct” cooling and warming curves during freezingand thawing within the physical constraints of the various packagingdevices, especially the vexed and persistent argument of straws versuscryovials. These matters were recently considered within a generalframework of risk analysis and management, taking into account theavailable evidence and perceptions of “best practice” from both the


Table 9.2 List of factors that might have affected temperature and pHmaintenance during an oocyte retrieval procedure. Based on observations madeat the Institute for Reproductive Medicine and the Psychosomatics of Sterility(Vienna, Austria, November 2002). Information provided with the permission ofUniv. Doz. Dr. Peter Kemeter

Contributory factor Classification/Explanation Solution

1 Empty OPU tubes notkept warm

Contributory: Although placedon a warm plate underneaththe sterile drape, the emptyFalcon 2001 tubes will not bekept at 37 ◦C (see white arrowin Fig. 9.7A).

A Cook tube warmer had already beenordered.

2 Flush buffer not keptwarm

Not contributory: the Falcon2001 tubes were held in awarm block on a warm plate,beneath the drape, so thetemperature will bereasonably well maintainedduring the time course of theoocyte retrieval procedure.(see white asterisk inFig. 9.7A)

No change required.

3 Follicular fluid notkept warm duringaspiration

Contributory: The retrieval tubehangs free in the air duringfollicle aspiration (seeFig. 9.7B).

A Cook tube warmer had already beenordered.

4 Cooling effect oflaminar flow cabinet

Contributory: The verticallaminar flow cabinet was keptswitched on during the eggsearch procedure, hencecreating a cooling draught.

The laminar flow cabinet was switched offat the commencement of the egg searchprocedure.

5 Wash buffer not keptwarm during eggsearch procedure

Contributory: Although thedish was kept in the smallincubator inside the workstation (see Fig. 9.7C) untilthe first oocyte was found,after that it was left on a 37 ◦Cwarm plate adjacent to themicroscope (see Fig. 9.7D,white arrow) for 10 minutesor more.

A heating stage that could be calibrated toachieve the desired 37 ◦C temperatureinside the dish, rather than that beingthe temperature of the warm surface,was recommended.


Table 9.2 (cont.)

Contributory factor Classification/Explanation Solution

6 Wash medium not keptunder CO2 duringegg search procedure

Contributory: As per factor 5,the wash mediuma would nothave maintained its pH andoocytes already recoveredfrom the follicular aspirateswould have been exposed toan alkaline medium.

A properly-formulated HEPES-bufferedwash medium was to be used to hold alloocytes until the completion of the eggsearch procedure, at which time all theoocytes were washed through a dish ofCO2-equilibrated culture medium(held until that time in the smallincubator in the work station) beforetransferring into the IVF culturedishes.

7 Cooling ofcumulus-oocytecomplexes duringrecovery fromfollicular aspirates

Contributory:� A very large Petri dish (15 cm

diameter) was used for theaspirated material, takinglonger to search than thetypical 60 or 100 mm dishes;and

� When the first oocyte wasfound it was aspirated into thePasteur pipette (Fig. 9.7D,black arrow) and held there,exposed to the cooling effectof the air in the laminar flowcabinet, until all oocytes hadbeen found and recovered,taking a minute or more.

Using a smaller Petri dish to search theaspirates was recommended, alongwith transferring each oocyte to thewash dish as soon as it was recovered.The wash dish was to be kept on therecalibrated warm stage (see factor 5,above) until that time.

a The wash medium being used was a standard-buffered culture medium (25 mM bicarbonate) with7.5 mM HEPES added. This amount of HEPES would not have held the pH while under an air atmospheredue to the “stress” of the 25 mM bicarbonate buffer being left without a CO2-enriched atmosphere. A normalHEPES-buffered medium has 20 mM HEPES with the bicarbonate (which is an essential cofactor) reducedto 10 mM (see D. Mortimer et al., 2002).

medical and legal perspectives (Mortimer, 2004b; using technicaland biological information reviewed in Mortimer, 2004a). Table 9.3illustrates a formal risk analysis, in the format of an FMEA, per-formed on the four most common packaging devices used for cryopre-serving human gametes and embryos and leads to the clear conclusion

Table 9.3 An example of using FMEA to evaluate the safety of cryostorage packaging devices for gametes and embryos.R = the risk rating or consequence of the risk happening, L = the likelihood of the risk occurring and C = the calculated“criticality” of the risk (C = R × L). *Denotes conditions assuming a device is used correctly according to the manufacturer’sinstructions, and †identifies situations where it is assumed that best practice laboratory procedure has been followed. Fornumeric superscripts see notes, below. Reproduced from Mortimer (2004b) with permission.

Packaging types

IMV 0.25-ml IMV 0.50-ml CBS high-Consequence Cryovial straw straw security straw

Risk Description of failure R L C L C L C L C

Contamination of the outsidethe device during filling

Will carry contamination into thecryogenic storage vessel

4 7 28 8† 32 8† 32 0∗ 0

Microbial transmissionthrough the device wall

Risk of outward contaminationof cryogenic storage vessel orinward contamination ofspecimen

4 1 4 1 4 1 4 0∗ 0

Fragility of the device at –196◦C

Risk of breakage during handlingwhile in storage (e.g. audits)

7 0 0 3†1 21 2†1 14 0∗ 0

Secondary containmentneeded for safe use of deviceunder “extreme” storageconditions (i.e. −196 ◦C)

Ability to provide reasonableexpectation of hermeticintegrity of the specimen

6 9∗ 54 4† 24 4† 24 0∗ 0

Adverse practical sequelae ofthe secondary containmentsystem

Handling difficulties in attachingdevices to canes

5 5∗ 25 0 0 0 0 0∗ 0

Cooling curve of the specimendoes not follow theprogrammed rate closely

Proper cooling rate is notexperienced by the specimen,or rate is variable throughoutthe specimen

7 8 56 1 7 2 14 2 14

Warming rate of specimendoes not follow ambienttemperature closely duringthawing

Proper (rapid) warming ratecannot be achieved duringthawing

6 6† 36 1† 6 2† 12 2† 12

Risk of inadvertent warmingduring handling ofcryobanked device

Risk of ice recrystallization due tospecimen warming above theglass transition temperature ofwater(c. –132 ◦C)

7 2 14 6†2 42 4†2 28 4†2 28

Explosion hazard whenthawing specimen

Explosive over-pressure due toevaporation of liquid nitrogentrapped inside the device

5 4† 20 2† 10 2† 10 0∗ 0

ID information can be lost orsmudged duringcryostorage

Integrity of identifyinginformation of each unit stored

8 1 8 2† 16 1† 8 0∗† 0

Total criticality scores 245 1623 1463 543

1.These risk likelihood ratings reflect the typical practice in many human IVF cryobanks of storing straws in narrow visotubes attached to canes, rather than accordingto the manufacturer’s instructions to store straws in visotubes that are kept inside goblets in canisters (which would merit reducing these ratings by 1 rank). Whenattached to canes there is the possibility of (inadvertent) attempted flexion of the straws during their removal from the visotubes.2. These risk likelihood ratings reflect the typical practice in many human IVF cryobanks of handling straws in isolation, rather than inside visotubes where thesurrounding liquid nitrogen would guarantee their remaining at −196 ◦C. If, as per the manufacturer’s recommendations, straws were only handled in visotubes (exceptwhen removing for thawing) these risk likelihood ratings could be reduced to values of 1.3. If the correct practices described in Notes 1 and 2 were followed, these total criticality scores would be reduced to 113, 118 and 33 respectively.


that the High Security Straws from CryoBio System (Paris, France)represent the current best practice from the perspective of notonly the technical achievement of cryopreservation but also bio-containment.

Protecting IVF laboratory staff from unfair litigation

As risk management moves from the manufacturing industry to health-care there is a far greater – or at least more immediately obvious –impact of risk on customers’ wellbeing. Although the underlying issueand driving force for risk management is patient safety, the increasingfocus on financial risk, not just medical risk, has substantial implica-tions for healthcare workers – and hence IVF Centers.

A leader in the medical application of RCA stresses that “Errorsmust be accepted as evidence of systems flaws, not character flaws”(Leape, 1997). But even if the professions involved all agree thatadverse events are the result of happenstance, multiple human errorsthat combine in a particular configuration by chance (Reason, 1994;Bogner, 1994), a situation with extremely negative emotional impactcould arise if the unfortunate, and unintentional, victims obtained anRCA report describing the “causes” of their problem – and litigationcould ensue. Because personal liability is a far greater risk in health-care than in industry, issues of personal fear are correspondingly moreprominent.

In addition, some organizations operate under a managementculture of fear, which further compounds employees’ worries andincreases the likelihood of errors being concealed, cover-ups organizedand staff dishonesty in general. No Center should instruct its staff toconceal errors, or to lie to patients under threat of dismissal or otherpunishment. Such behavior is unacceptable. In a decent, supportivework environment, where risk management and Quality flourish, thefollowing principles must guide all human resource activities (and bedocumented in all employees’ contracts):


(a) no-one will be punished for making a mistake;(b) mistakes are always seen as opportunities for improvement;(c) if a mistake occurs it must be reported to the supervisor immedi-

ately and dealt with expeditiously;(d) anyone who lies about a mistake or attempts to cover one up should

be dismissed; and(e) all levels of management must adhere to and apply these same

principles.

Protection of employees from personal liability is essential if anymodern-thinking IVF Center is to function without its staff livingin permanent dread of litigation – a concern of everyone, not justclinical embryologists. To this end Centers must have fully-detailedstandard operating procedures (SOPs) documented and in place, com-plemented by comprehensive quality management and risk manage-ment programs. Then, so long as staff work within this framework, theCenter should have a legal (not just moral) obligation to indemnifyits employees against personal responsibility for any adverse event thatmight occur, since it will have been a fault of the system. Naturally,in cases of mischief, dishonesty or blatant malpractice the Center’sobligation would be nullified. Incompetence should also be the Cen-ter’s responsibility since staff selection and training are the employer’sresponsibilities, and staff who cannot perform any of their duties prop-erly should not be allowed to continue performing them. Systemsfor continuous employee appraisal, proficiency evaluation and com-petence testing should be implemented by the employer, with bothparticipation and satisfactory performance being mandatory.

Conclusion

Although the great majority of IVF Labs (actually, IVF Centers) prob-ably do not have a formal risk management program in place, we hopethat the material presented and discussed in this Chapter will have


provided sufficient insight to allow everyone not only to recognize theneed for such activity but also to make a start on developing risk man-agement processes for themselves. Beyond the jargon and formalizedprocesses of risk management lies another example of an area that isessentially common sense and which is readily amenable to the appli-cation of scientific method.

10

How are we doing? Benchmarking

Benchmarking, basically, is the proof of what is possible. In a traditionalbusiness setting, benchmarking is the continuous process of measuringone’s products or services against one’s strongest competitors or thoserenowned as world-leaders in the field. In its practical application forIVF Centers, benchmarking can be viewed at three levels:

• Internal benchmarking: comparisons between Centers within agroup or network.

• Competitive benchmarking: comparisons against the direct compe-tition.

• Functional or generic benchmarking: comparisons against the “best-in-the-world” Centers.

For an IVF lab, benchmarking can be seen as verifying that the lab-oratory outcomes and the Center’s clinical success rates are main-tained, monitoring the implementation or amendment of processes toimprove outcomes to match those of competing Centers, and evalu-ating the development of better processes or technology to meet, orexceed, the performance of other Centers. Benchmarking is the bestway to avoid complacency.

Like systems analysis and process control (see Chapters 5 and 6),benchmarking requires the use of Indicators, things that are mea-sured to determine how we are doing. However, because benchmark-ing requires us to compare Indicators between IVF Labs or Centers,it requires greater care in ensuring that these Indicators are calculatedthe same: we must not only compare apples with apples, but theymust be the same sort of apples, e.g. Granny Smiths, and be ripe. Such

169


considerations are covered later in this chapter after we have establishedwhat sort of performance Indicators we might want to use.

Sentinel indicators and adverse events

In addition to performance Indicators readers will often find referencesin quality management texts to “sentinel indicators.” Sentinel indica-tors are somewhat different from performance Indicators, although theterm is used with various meanings by different authorities. At its mostextreme, a sentinel indicator is an unexpected occurrence involvingdeath or serious physical or psychological injury, and is often qualifiedwith the phrase “or the risk thereof.” However, in other areas of qualitymanagement a sentinel indicator can be anything that can be measuredor quantified to monitor a process.

Because of the nature of IVF we need both types of measure-ment. Most of the high profile adverse events are generally rare,although every occurrence is critical and anything less than com-plete correction is less than adequate, i.e. would be seen as a systemfailure. However, many such accidents are probably due to “the insid-ious concatenation of often relatively banal factors, hardly significantin themselves, but devastating in their combination” (Reason, 1994) –making them more-or-less unpredictable. So, if nothing else, effec-tive risk management will help eliminate the occurrence of criticaladverse events that do not fall into this extreme classification. Butsome “sentinel events” might be tolerable with certain prevalence andit is the goal of risk management (essentially, quality improvement)to progressively reduce the frequency of such events, while accept-ing that some cannot be eliminated (which cannot be defined asfailure).

Therefore we need to use careful terminology to describe the differ-ent types of event. Critical adverse events can be termed just that, whilemeasurements of laboratory processes (or any other operational pro-cesses within the IVF Center) will not – hopefully – constitute criticaladverse events; they are merely assessments that form the quantitative

171 How are we doing? Benchmarking

basis for quality control and quality assurance. Indeed, their existenceis fundamental to the ability to monitor the results of any correctiveaction. To this end, such events can be termed “Indicators” and con-fusion avoided.

How to choose indicators

As already discussed, Indicators are crucial to the development, andmaintenance, of a quality system. In accordance with the maxim “Youcan’t control what you can’t measure,” the Indicators that are usedshould reflect the areas which, when controlled, will bring the mostmeasurable benefit to the IVF program. So, for example, even thoughpregnancy rate is an important indicator of the overall program per-formance, it is not necessarily the most useful one in terms of benefitto the program’s efficiency, finances and operation.

It is quite likely that the most useful Indicators will differ from oneIVF program to the next, since they will depend upon the areas beingtargeted for improvement. This is why accreditation authorities do notprovide a list of “mandatory” Indicators – it is expected that the areas ofneed will have been identified as part of each Center’s ongoing QualityImprovement process.

From the patients’ perspective, apart from cost and location, aCenter’s pregnancy rate and implantation rate are the most likely Indi-cators that will be sought. It is important, therefore, that everyone in theprogram understands that these rates are Indicators of the Center’s per-formance as a whole, and not simply of the laboratory’s performance.It is also important to realize that different IVF programs may have(indeed, will likely have) different definitions for their Indicators –therefore, it is essential that the definitions used for each Indicator areknown before any attempt at benchmarking is made. Further discus-sion on the use of reference populations for reporting results as well assome general considerations of definitions of success rates and honestyin reporting pregnancy rates follows the next section.


Operational and performance indicators

The following lists are only intended to be illustrative, to provide exam-ples of the types of Indicators which can be followed, under generalsub-headings. Each IVF Lab or Center should establish their own set ofIndicators that meet their specific needs from an operational perspec-tive, in addition to others that are needed for external comparisons.

Program Performance Indicators

• Pregnancy rates (should be broken down by female patient’s age andprocedure):� biochemical (positive β-hCG)� clinical (e.g. fetal sac or fetal heart at 7-week ultrasound)� ongoing (e.g. fetal heart at 7-week ultrasound or pregnancy)

� Implantation rates (should be broken down by female patient’s ageand procedure):� overall, e.g.

total number of fetal sacs seen at 7-week ultrasound

total number of embryos transferred to all patients in that age group and procedure type

� incidence of multiple implantation (e.g. proportion of pregnancieswith >1 fetal sac at 7-week ultrasound)

Laboratory performance indicators

• Oocyte grade and/or maturity (note: this is not actually an Indicatorof the laboratory’s performance, but it does provide a description ofthe “source material”)

• IVF fertilization rate (the proportion of inseminated oocytes whichhave 2PN the day after insemination)

• ICSI fertilization rate (the proportion of injected oocytes which have2PN the day after injection)

• Poor or failed fertilization rate (e.g. the proportion of cycles in which<25% of inseminated oocytes were fertilized)


• ICSI damage rate (the proportion of injected oocytes which degen-erate during, or immediately after, the injection procedure)

• Zygote grade• Cleavage rate (e.g. the proportion of zygotes which cleave to become

embryos)• Embryo development rate (e.g. the proportion of cleaved embryos

which are at the 4-cell stage 2 days after insemination; the proportionof cleaved embryos which are at the 8-cell stage 3 days after insemi-nation; and/or the proportion of embryos which are at the blastocyststage 5 days after insemination)

• Embryo fragmentation rate (e.g. the proportion of Day 3 embryoswith <5% fragmentation)

• Embryo score or “grade” (e.g. the proportion of Day 3 embryos withthe highest score)

• Embryo utilization rate (the proportion of cleaved embryos whichwere transferred or cryopreserved)

• Embryo cryosurvival rate

Efficiency

• Number of tests handled by individual operators• Time lag between receipt of an enquiry and the response• Proportion of patient records which are complete• Number of telephone calls answered by a person, rather than by

voicemail• Average delay between completion of a test and the publication of

results

Best practice

• Incident reports• Treatment complications• Infection and Accident Reports• Number of comments received per month (both positive and

negative)


Laboratory operations

• Number of each type of procedure performed each week or month• Equipment malfunction reports• Equipment performance (e.g. amount of liquid nitrogen required to

top up a storage dewar; amount of CO2 or pre-mixed gas used byeach incubator; incubator temperatures)

• Rate of utilization of consumables (e.g. plasticware)

Financial

• Comparison of the service fees with those of other Centers• Comparison of the cost of performing a procedure and the revenue

it generates• Accounts Payable and Accounts Receivable balances• Number of patient referrals per month

Reference populations

Often there is a definite need to be able to compare Indicators (usu-ally some measures of clinical outcome such as pregnancy rates or livebirth rates) across many IVF Centers, especially for such purposes asa national registry. In these situations, each Center’s results are notjust reported en masse, usually with stratification according to femalepatient age bands, but also in terms of some sort of “reference popula-tion” that is designed to be consistent between all the Centers to allowdirect comparison between the Centers.

Obviously the more tightly the reference population is defined,the smaller it will be for any clinic. Hence for smaller IVF Cen-ters the dataset from which their reference population Indicatorsare calculated might be very small, and hence the actual resultsbeing used for comparison will have different degrees of statisti-cal robustness. In this situation, confidence intervals are attachedto the reference population Indicators so that their uncertainty of


measurement can be stated and the Indicators compared meaning-fully between all Centers (e.g. as in the UK Human Fertilisation andEmbryology Authority’s annual Patients’ Guide to IVF Clinics; seewww.hfea.gov.uk/HFEAPublications/PatientsGuides).

A properly defined reference population must not only consider thedemographics and aetiology of the patients receiving treatment, butalso technical aspects of the treatment modalities employed – especiallythe number of embryos replaced. A major weakness of the so-called“HFEA league table” in the UK is that some Centers have a higherproportion of “elective” 3 embryo transfers than others, and hencemight “compensate” for what are actually lower success rates comparedto other Centers where all patients received only two fresh embryos. Forthis reason, many experts concerned with comparability of reportingof Clinical Indicators insist upon analyses based on the implantationrate per embryo transferred. The expansion of elective single embryotransfer programs and the advent of obligatory single embryo transfers(e.g. Belgium) will provide some very interesting insights over the nextfew years.

Honesty in reporting results

As was intimated in the preceding section, clinical results and indica-tors can be manipulated quite extensively by altering the compositionof the dataset. While the majority of IVF Labs are probably honest,there are certainly many IVF Centers, especially in highly competitivecommercial environments, who are less scrupulous in their definitionof Indicators. Perhaps fans of the BBC television series Yes, Ministerand Yes, Prime Minister might consider this to be one of HumphreyAppleby’s “irregular verbs”? I employ properly defined reference popu-lations to calculate my Indicators, your reference populations are subjectto some bias, and he cheats on his success rates.

This is not the forum for an extensive debate and analysis of honestyin reporting, but as scientists we should endeavour to define Indicatorsthat will restrict opportunities for introducing bias or cheating, and


that will facilitate communication and comparison between Centers.This will benefit not only the patients but us as professionals.

Illustrations of benchmarking

Internal benchmarking

This form of benchmarking considers comparisons between Centerswithin a group or network. In Australia, the UK and USA there areIVF organizations who provide services at multiple sites, and for orga-nizations such as these it is clearly important to be able to ensure thatcomparable performance and outcomes are achieved regardless of thelocation. However, the comparability of Indicators between such siteswill still depend on the achievement of operational standardizationbetween the sites – a daunting task to say the least!

Competitive benchmarking

In the world of IVF, comparisons against the direct competition are aneveryday occurrence. In the UK, the HFEA regularly publishes whatare effectively league tables of IVF Centers – which patients then usein deciding which private unit to attend. Even when a national registrypublishes center-specific but anonymous success rates, each Centerknows who they are, and the one with the highest rates will often“break cover” to try and gain commercial advantage.

Functional (generic) benchmarking

In a way, this is what Centers are doing when they make decisions aboutwhat stimulation drugs or protocol to use, or which culture technol-ogy or products to use. Unfortunately, it is usually a completely invalidbenchmarking exercise as there are very many more variables withinthe IVF process that need to be controlled than just the one under con-sideration. However, if an IVF Center were to look to a world-leading


program and attempt to replicate its operational systems as well as itstechnology, then comparisons against such “best-in-the-world” Cen-ters can be useful. Certainly when we undertake consultancy workin this area, we are employing generic benchmarking since the tech-nology that we have been involved in developing (e.g. D. Mortimeret al., 2002) has been successfully implemented – in conjunction witheffective Quality Management Systems – in many locations around theworld with very comparable performance indicators.

Recently, Cook IVF, whose Culture System products are founded onincubator design concepts and culture medium formulations devel-oped while DM was Scientific Director at Sydney IVF, has recentlyestablished a network of Global Reference Clinics to collate perfor-mance indicators on each new batch of culture medium rapidly as itis disseminated into the market. In this way Cook IVF will have rapidfeedback on the batch’s performance, including rapid warning of anysubstandard performance, and the Reference Clinics will be able tocompare their results with those of other participating Clinics. CookIVF was also the first (and, to the best of our knowledge is still theonly) culture medium manufacturer to provide “expected results” –i.e. benchmarks – for the performance of its products when used asan integrated culture system according to specifications (Cook IVF,1999).

11

Specifying systems

All the basic concepts and principles for the tasks that will be con-sidered in this Chapter have been covered in the preceding chapters.The purpose of this Chapter is to provide particular examples of thoseconcepts in practice and to place the process of specifying the systemsthat will be used in the IVF lab in the “real world.”

Regardless of whether we are choosing or specifying a techni-cal procedure or a piece of equipment, we must consider it withinthe context of a process. General principles for specifying a systeminclude:

• What does it need to do?This defines the technical specifications of the procedure or pieceof equipment, as well as required or acceptable tolerances in itsperformance.

• How well must it do it?This relates both to the performance as well as the reliability of theprocedure or piece of equipment.

• Will it last?For equipment, what is the mean time before failure (MTBF), whatis the offered (or available) warranty and the availability, quality andcost of after-sales service and repairs.

• Will help be available?What support is available, either from the manufacturer of a pieceof equipment or a reagent (e.g. culture medium), or from the origi-nators of a particular technical procedure or method?

178

179 Specifying systems

Selecting methods, devices, equipment, etc.

There are some simple rules to follow when selecting a method, pieceof equipment or a device.

1. Evidence-based considerations(a) Has the method, device or instrument been used by other IVF

Labs?(b) Are there established performance Indicators for the method,

device or instrument?(c) How do the established performance Indicators for the method,

device or instrument compare to your internal benchmarks?2. Regulatory approvals

(a) Does the method, device or instrument have necessary reg-ulatory approval(s) from the authorities that govern youroperations?

(b) Does the method, device or instrument have any other regu-latory approval(s) from other authorities, as indication of itsquality and fitness-for-use?

Possible regulatory approvals include those from authori-ties such as the US Food and Drugs Administration, e.g. a510(k) pre-market clearance, or CE marking within the Euro-pean Union.

3. Follow the manufacturer’s instructions(a) If the manufacturer has provided instructions on how to use

the method (e.g. reagents or kits), device or instrument, ensurethat everyone is aware of them and follows them.

(b) If the manufacturer has provided instructions on how to storereagents, a kit, or a device, ensure that everyone is aware of themand follows them.

(c) If you have modified any of the manufacturer’s instructions thenensure that your SOP describes the change(s) in detail. Maintainrecords of the development and validation studies that wereperformed to establish such changes in your lab.


4. Avoid known problems(a) If there are any known problems with the use of a method,

reagent, kit, device or instrument, ensure that everyone is awareof them and that your SOP includes all necessary details toensure that they will not be repeated in your lab.

There are a number of issues, both general and specific, that havebeen established as significant sources of decreased outcomes in IVFlab procedures, and it is the responsibility of the Lab Director to beaware of what is in the literature and to either deal with the issuedirectly or, where it might be outside his/her jurisdiction or directcontrol, to bring any possible factors that might create adverse out-comes for the IVF Center to the attention of the Center’s MedicalDirector. Failure to do so is not only unprofessional and unscientific,but could leave the Lab Director open to accusations of incompetenceand perhaps legal liability. However, if a Lab Director were to bringsuch issues to the attention of those in charge of an IVF Center andthey decided, for whatever reason, not to accept – or to just ignore – theLab Director’s advice then those “executive managers” have assumedall responsibility for that decision, as well as any and all possible futureliability.

While we do not recommend the routine use of such self-protectionactions, there are some IVF Centers where the lab has little or no say insuch decisions but, nonetheless, is often blamed for anything that goeswrong. A written memorandum outlining the lab’s scientific concerns,including references to the evidence upon which they are based, cansave great acrimony – and possible future liability.

Can semen analysis be standardized?

It has long been recognized that assessments of sperm concentra-tion, motility and morphology can be subject to wide variations dueto technical error. However, methods whereby this intra- and inter-observer variability can be reduced have been known for quite some


time (e.g. Mortimer et al., 1986, 1989; Mortimer, 1994; World HealthOrganization, 1999). The basic requirements are simple: (1) use robustmethods; and (2) train the staff in the correct performance of the meth-ods. The basic semen analysis courses run by the Andrology SpecialInterest Group of ESHRE, the European Society of Human Reproduc-tion and Embryology, have provided eloquent evidence of the validityof this approach (Bjorndahl et al., 2002). Therefore the only reasonthat sperm assessments are not performed more reliably in many IVFCenters can only be that the medical and scientific direction of thoseCenters just don’t care about sperm. This, of course, makes the opinionthat sperm counts, etc, don’t have any value a self-fulfilling prophesy –how could observations with error components of up to 50% be takenseriously or used intelligently?

Reactive oxygen species and sperm preparation methods

It has been known since the late 1980s that the centrifugal pelletting ofunselected ejaculated human spermatozoa can result in the generationof reactive oxygen species (ROS) within the pellet. These ROS candamage the spermatozoa to such an extent as to impair, or even destroy,their fertilizing ability (Aitken and Clarkson, 1988). A 1991 editorialraised an important risk management question: if a couple had anunsuccessful IVF cycle in which such iatrogenic sperm dysfunctionmay have occurred, and then had a successful IVF cycle in which amore appropriate sperm preparation technique was used, might theyhave grounds for legal action against the person(s) responsible for theearlier treatment attempt (Mortimer, 1991)? Subsequent research hasconfirmed the possible severity of this problem (Mortimer, 2000), andthe issue, therefore, remains a valid concern.

Tomcat catheters

Although these catheters are sold as veterinary products, they were usedextensively in the early days of assisted reproduction and are still used


Table 11.1 Results of studies comparing the Tomcat and othercatheters for embryo transfer

StudyPregnancy rate by catheter type

Tomcata TDTb Frydmanb Wallacec K-Softd Cook SIVFd

A 28% 16%B 47.0% 14.7%C 9.2–19.4% 32.3% 19.2%D 20.5% 27.1%E 20.5% 29.6%F 28% 52%

Sources:A: Gonen et al. Hum. Reprod., 6 (1991): 1092–1094.B: Meriano et al. Fertil. Steril., 74 (2000): 678–682.C: Wisanto et al. Fertil. Steril., 52 (1989): 79–84.D: van Weering et al. Hum. Reprod., 17 (2002): 666–670.E: McDonald and Norman Hum. Reprod., 17 (2002): 1502–1506.F: Mortimer et al. Fertil. Steril., 76, 3S (2002): S17–S18, Abstract O-045.Manufacturers:a Kendall, Mansfield, MA, USA.b Laboratoire CCD, Paris, France.c Smiths Medical, Hythe, Kent, UK.d Cook IVF, Eight Mile Plains, Queensland, Australia.

today in some IVF Centers for intra-uterine insemination and evenembryo transfer. In many jurisdictions their use for clinical applicationsis forbidden by the regulatory authorities, but not everywhere. Theirattraction is their very low cost and reports that they perform as well as“proper” embryo transfer catheters (see Studies A and B in Table 11.1).However, not only have more recent randomized trials demonstratedthat Tomcat catheters give worse clinical results (Studies E and F inTable 11.1) but both these trials were cancelled at interim data analysisdue to the unacceptable prejudice the Tomcat gave to those patients. Avery recent report has described endometrial lesions caused by morerigid catheters, including the Tomcat (Marconi et al., 2003). Therefore,from an analysis of efficacy and safety, as well as considerations ofregulatory approval, there would seem to be no place for the use ofTomcat catheters in a responsible IVF Lab.


Cryo buffers: the move from PBS to HEPES

In the mid-1990s when Sydney IVF was eliminating patient serum fromall its culture media, the serum component in the embryo freezing andthawing protocols was replaced by a solution of 45 mg/ml of HSA innormal saline (this being the albumin content of serum). However,in one lab the freezing solutions were based on a modified version ofphosphate-buffered saline (PBS) that contained phenol red and it wasnoted that, when straws were being seeded, the medium column hadturned from the normal pink colour to bright yellow. Clearly therewas a problem with pH buffering during cooling in the absence ofserum, and this was confirmed by pH measurements. Advice from Dr.John Critser led to the adoption of a TL-HEPES medium as the basalmedium for embryo freezing and thawing in August 1996. Not only didthe HEPES buffering allow for proper pH control during cooling, butthe implantation rate per thawed embryo transferred went from 6.5%to 16.2% (Cullinan et al., 1998). Moreover, even when embryos thathad been frozen in PBS-based solutions were thawed in TL-HEPESthere was a significant improvement in implantation rate to 14.3%,indicating that the major damage was probably being done duringthawing and washing to remove the cryoprotectant. Subsequently, aHEPES-buffered version of the cleavage medium (D. Mortimer et al.,2002) was developed and has been used since that time (Cook IVF,1999).

Research on the temperature stability of phosphate buffers hasrevealed that they are highly unstable at lower temperatures and there-fore unsuitable for freezing and thawing media. Early success withembryo cryopreservation using PBS-based solutions was probablyaided by the additional buffering capacity of the serum component,but with the replacement of serum by albumin it seems that embryofreezing and thawing solutions should not be based on PBS.

The mechanism for the adverse effect of the low extracellular pHremains unproven, but it has been shown that thawed hamster embryosare unable to regulate their intracellular pH for several hours, until


proper homeostatic mechanisms are restored, and that this may be atleast a partial explanation for their impaired oxidative metabolism anddecreased developmental competence (Lane et al., 2000).

Practical examples

The following sections provide worked examples to illustrate the prin-ciples of specifying systems described at the beginning of this Chapter.It is not the purpose of this book to provide specific recommendationsof one technology over another, nor to recommend any one particu-lar make or model of a particular piece of equipment. Rather, it is ourgoal to encourage readers to consider all relevant factors in making suchdecisions instead of simply continuing to do what they or their mentorshave done before. Nonetheless, we have provided some comment onour preferences in light of the factors discussed. Although the followingexamples are not structured as formal FMEAs, those general princi-ples clearly apply and interested readers should use the informationpresented here to construct FMEA tables (as described in Chapter 7)within the context of their own labs.

What sort of embryology work station is best?

There is a wide range of possible configurations of microscope workstation where one can perform the various technical proceduresinvolved in routine IVF. (This analysis does not relate equally to a workstation where embryos are processed for freezing and thawing sincethere are different requirements for temperature and pH control.)At the simplest level, the common modern alternatives include thefollowing permutations:

Style of cabinet• Vertical laminar flow (VLF) cabinet• Neonatal isolette-style “IVF Chamber”• Class II cabinet


A horizontal laminar flow (HLF) cabinet has not been considered as itprovides no protection for the operator.

Clearly, on the open bench is unacceptable since it provides noprotection from contamination for either the oocytes/embryos or theoperator.

Warmed stagesEveryone is aware that some sort of warm stage must be providedfor microscopes where oocytes and embryos are being examined. Theoocyte in particular is extremely sensitive to alterations in temperature:cooling causes the spindle to depolymerize, risking aneuploidy of theresulting embryo if not all chromosomes reattach to the spindle whenit repolymerizes as the oocyte re-warms to 37 ◦C (Pickering et al., 1990;Almeida and Bolton, 1995; Wang et al., 2001). In addition, tempera-ture shifts can affect trans-membrane transport and many intracellu-lar metabolic processes. Consequently, human oocytes and embryosmust be held as close as possible to a stable 37 ◦C. Furthermore, asignificant, but poorly recognized, confounding aspect of temperaturecontrol during the microscopic observation of oocytes and embryosin dishes is that the design of all traditional disposable plastic dishesdoes not allow the base of the dish to come into direct contact withthe microscope stage, so there is always an air gap (see Figure 11.2).Because air is a poor conductor of heat, this air gap greatly reduces theefficacy of heated stages, allowing the medium in dishes to cool belowthe temperature at which the heated surface is set.

Gassed enclosuresWhen working with bicarbonate-buffered culture media, a CO2-enriched atmosphere is essential in order to maintain the pH of themedium. A study by Blake et al. (1999) revealed that not only dobicarbonate-buffered media take a long time to reach equilibrium, butthey out-gas very much more quickly than previously assumed. Forexample, a Petri dish containing 5 ml of culture medium will out-gasafter removal from a CO2 incubator so that the pH has shifted above


Figure 11.1 Diagram illustrating the existence of an air gap between traditional culture dishesand incubator shelves or warm stages and the lack of such an air gap in the CookMINC mini-incubator. Because air is a poor conductor of heat, the temperatureof the medium in dishes in the MINC incubator is re-established much morequickly.

7.45 within 2 minutes – and after replacing the dish in the CO2 incu-bator it will take 15 minutes to re-equilibrate the pH (see Figure 11.4).These differences are due to the relative magnitudes of the differentialCO2 contents between the equilibrated medium and air and betweenthe incubator’s atmosphere and the partially out-gassed medium. Thisstudy is discussed further in the later section on the use and need forculture oil.

In addition to standard commercial laminar flow and Class II cab-inets there are also specialized work station cabinets that are mar-keted for IVF applications, including the “IVF Chamber” from HDScientific (Wetherill Park, NSW, Australia; www.hdscientific.com.au)or the Hoffman Chamber from MidAtlantic Diagnostics (Marlton,NJ, USA; www.midatlanticdiagnostics.com), and the widely-used “K-Systems” cabinets which are either VLF or Class II cabinets fitted withwarmed surfaces and gassing systems (K Systems, Birkerød, Denmark;www.k-systems.dk).


The pros and cons of the principal alternative styles of work stationhave been summarized in Table 11.2. Support for the contention thatthe IVF Chamber style work station will provide better quality embryosand clinical outcomes remains limited (Mortimer et al., 2001b), butseveral of the most successful IVF Centers have chosen to use suchwork stations.

Fortunately, the EU Tissues and Cells Directive (European Union,2004) did not include the strict requirements for air quality standardsfor tissue establishments that were contained in the draft legislation.Those proposed standards were the same as those contained in theUK Code of Practice for Tissue Banks (Department of Health, 2001),which said that critical work areas where tissue is manipulated openlyshould have Grade A (i.e. Class 100 000, or ISO Class 5) air and oper-ate under a Grade B background (which is equivalent to ISO Class7 under operational conditions, but the same as Grade A “at rest”).These air quality requirements were based, seemingly rather arbitrarily,on those applicable in the pharmaceutical manufacturing industry –which deals with chemicals, not living cells and tissues. Under suchregulations the only EU-compliant IVF work station would be onebased on a Class II cabinet and, in addition, would have to be usedin a clean room environment! Considering the technical realities ofperforming IVF under such environmental and physical conditions(especially the perceived sub-optimal temperature and gas controlconditions presently achievable in such work stations), the “primedirective” of IVF – optimizing the clinical potential of gametes andembryos – would have been compromised. Indeed, the general expertopinion was that the enforcement of such standards would have beenmore likely to have had a deleterious impact on clinical outcomesrather than be of any benefit to patients. Hopefully, the technicalannex(es) to the EU Tissues and Cells Directive will pay heed to thetechnical realities of working with gametes and embryos, and the cru-cial importance of maximizing their potential, rather than attempt toenforce arbitrary – and extremely expensive – air quality standards

Table 11.2 Summary of the pros and cons concerning the principal alternative configurations for IVF work stations inrelation to their ability to control extrinsic factors that can affect outcome

Type of workstationExtrinsic factor/variable

RankTemperature CO2/pH Air Evaporation/osmolarity

Vertical laminarflow cabinet

Configuration Heated stage onmicroscope

Sometimes a gas “funnel”is put over the disheswhile they are not on themicroscope, there is noprotection while on themicroscope.

Open to room airthroughout unless gasfunnel installed, in whichcase only exposed duringobservation.

Gas flowing into funnelor box is humidified bybubbling through a waterbottle.

#3

Effectiveness Moderate. Still have airgap between bottom ofdish and the heatedsurface.

Up to several minutes ofexposure to room airduring observation, alsoduring holding if no gasfunnel.

Exposed to VOCs andother contaminants in theroom air at least duringobservation.

Moderate effectivenesswhile under gas funnel orinverted box, noprotection duringobservation.

Ranking 3 = 2 = 2 = 2

K-systems cabinet

Configuration Heated stage built intothe cabinet workingsurface.

Gas funnel usuallyinstalled for when dishesare not on themicroscope, noprotection while on themicroscope.

Exposed to room air atleast throughout theobservation period.

Gas flowing into funnel ishumidified by bubblingthrough a water bottle.

#2

Effectiveness Moderate. Still have anair gap between bottomof dish and the heatedsurface.

Up to several minutes ofexposure to room airduring observation.

Exposed to VOCs andother contaminants in theroom air at least duringobservation.

Moderate effectiveness whileunder gas funnel, noprotection duringobservation.

Ranking 2 = 2 = 2 = 2

Configuration Entire chamber warmedto 37 ◦C

Entire chamber has aCO2-enrichedatmosphere.

Air inlet has 0.22 �mfilter, an in-linecarbon-filter can also beinstalled.

The atmosphere inside thechamber circulatescontinually over ahumidification pan.

#1IVF chamber Effectiveness All dishes inside thechamber are held close to37 ◦C whether on themicroscope stage or not.

All dishes are held undera CO2-enrichedatmosphere.

Exposure to room airminimized bysemi-closed system

Humidity is maintainedhigh (c. 80%, i.e.non-condensing)

Ranking 1 1 1 1


whose achievement is technically unrealistic within the constraints ofperforming clinical IVF to the maximum benefit of the patient.

Choosing a CO2 incubator

There are numerous factors that need to be considered when select-ing a CO2 incubator that go beyond the simple need for creating aCO2-enriched environment for culturing gametes and embryos. Someaspects are technical, while others are more practical (see Table 11.3for a description of these factors; also Figure 11.2).

Given the technical superiority of solid-state incubators using pre-mixed gas, we have been using these units exclusively since the late1990s. Indeed, the project to develop the MINC 1000 incubator (CookIVF, Eight Mile Plains, Qld, Australia) was initiated by us – although weshould state that because all development work was funded by CookIVF we receive no financial rewards for their sales. The beneficial resultsof the value of these incubators (D. Mortimer et al., 2002) have beenconfirmed by other users in terms of both physical performance andclinical results (e.g. Cooke et al., 2002; S. T. Mortimer et al., 2002; alsoFigure 11.1). One particular advantage of the MINC type of “solid-state” incubator is their much lower power requirement compared totraditional CO2 incubators, making it possible to run them for manyhours on battery-based uninterruptible power supply (UPS) units evenif emergency maintained mains power is not available.

Choosing a culture medium

Given our current understanding of the changing requirements ofmammalian embryos during their early development, the use of “stagespecific” or “sequential” media is clearly obligatory for the produc-tion of the best quality embryos (Bavister, 1995, 1999). The concept ofusing a series of culture media optimized for fertilization and differentstages of embryonic development is not new, having been proposed in


Figure 11.2 Graph showing the effects of incubators and culture media on implantation rate.Period A was using Forma incubators with a single-stage culture medium, whilein Period B patients were allocated into either Forma (data points shown ascircles) or Cook MINC (data points shown as squares) incubators, still using thesame single stage medium. Period C shows the further increase in implantationrate when sequential culture media were used in conjunction with the MINCincubators (the initial spike was due to January and February being quiet monthsin which pregnancy rate was affected disproportionately by small numbers ofembryo transfers). Data generously provided by Simon Cooke, John Tyler andGeoff Driscoll (IVF Australia – western Sydney).

the mid-/late-1980s (e.g. Mortimer, 1986; Leese, 1990). Similarly, theformulation of culture media for mammalian gametes and embryosbased on the composition of oviduct fluid is a well-established concept,dating back to the early-1970s (e.g. Tervit et al., 1972; Menezo, 1976;Quinn et al., 1985; Mortimer, 1986).

When choosing a culture medium to use in an IVF lab the followingcriteria are pertinent to making the best decision, not just from a


Table 11.3 Factors to be considered in choosing a CO2 incubator

General

consideration Factor Comments

Practical Capacity � How much of the chamber will be available for use? E.g. do you onlyplace dishes at the front of the shelves? Do you use modular incubatorsinside the main incubator chamber?

� How many cases will the incubator need to handle? Consider thenumber of dishes per case as well as the resultant frequency of dooropenings (a question of stability of the incubator’s conditions).

Practical Physical size � Are there any space constraints in the lab?� Is there any problem with limited access for delivery of the incubator to

your premises and getting it into the lab?Practical Ergonomics � Will all your staff be equally able to access safely all shelves within the

incubator? Besides the obvious issue of shorter embryologists notbeing able to each the topmost shelves without using a step (whichitself creates significant risk of injury and loss of embryos), willembryologists have to bend or kneel down to access the lower shelvesof an incubator that is sitting on the floor?

� Is the mains power switch liable to accidental operation? E.g. anincubator at floor level with the switch in the lower region of a sidecontrol panel could easily be knocked by someone using, for example,a floor cleaner.

Technical Temperaturecontrol

� What are the tolerances of the incubator’s operational performancearound the “set point”?

� How quickly does the incubator recover its temperature after a dooropening? E.g. is a water jacket required for adequate performance?

� How quickly do the contents of dishes placed in the incubator reachthe desired temperature after being placed in the incubator (e.g. Cookeet al., 2002; also Fig. 11.2)?

Technical CO2 control � Should you use an infra red (IR) or thermal conductivity (TC) basedcontroller? Whereas historically TC controllers were cheaper and morereliable than IR units, improvements in technology have largelyeliminated this consideration. However, the requirement for thetemperature and humidity to be re-established inside the incubatorchamber before a TC controller can commence re-establishing the CO2

level makes them far less stable than units employing IR controllers(see Fig. 11.3).

Technical O2 control � Do you want to use a low oxygen atmosphere for embryo culture? It iswell-established that a reduced pO2 improves mammalian embryoculture in vitro (e.g. Tervit et al., 1972; Bavister, 1995) and there isgrowing evidence that this also produces better quality humanembryos, especially if employing extended culture to the blastocyststage (Catt and Henman, 2000).

(cont.)


Table 11.3 (cont.)

General

consideration Factor Comments

� If using a low O2 system, should you use a tri-gas incubator orpre-mixed gas?

� If using a tri-gas incubator, what will you use as the source of nitrogen(e.g. compressed nitrogen, liquid nitrogen vapor, etc)?

Technical Air quality � The quality of the air inside incubators has been raised as a possiblesource of concern (Cohen et al., 1997; Mayer et al., 1999), thereforesteps should be taken to minimize the possible adverse impact of thisfactor on embryo culture.

� If using traditional incubators then the quality of the air in the lab thatwill enter the incubators must be considered and appropriate stepstaken to remove detrimental components such as VOCs.

� If using an incubator or internal vessel that uses pre-mixed gas (e.g.modular incubator units or the Cook IVF MINC incubator), then thatgas needs to be filtered to remove particulates and VOCs beforeentering the incubation chamber. This is an easier process to ensurethan a system that allows room air to contribute to the cultureatmosphere.

Technical Stability ofthe internalenvironment

� If the door will be opened within the recovery time for the incubatorthen should you buy a model with a split inner door?

� For temperature stability do you need to have a model with a waterjacket? While this was, historically, considered to be very important toensure even heating of the chamber and for continued stability in caseof power failure, this is far less important today due to improvementsin incubator design and the expected availability of some form ofemergency maintained power.

� How much power does the incubator draw? When considering theprovision of emergency maintained power this is a very importantconsideration, especially if the power is to be provided via abattery-based system.

scientific perspective, but considering important practical matters thatare vital to the IVF lab’s ability to provide a quality service.

Availability and stockThe product not only needs to be available in your geographic areabut ideally there should be a local agent or distributor who will holdsome reserve stock in case of problems. While most culture medium


Figure 11.3 Graphs showing the effect of opening incubators doors on the CO2 level and howits rate of recovery is affected by the type of CO2 controller. The left-hand panelshows the results for a Forma Model 3360 incubator using a thermal conductivitytype of CO2 controller (data from the Forma Scientific website) while the right-hand panel shows the results for a Galaxy Model 170 incubator using an infra-red CO2 controller (data courtesy of RS Biotech, Irvine, Ayrshire, Scotland). Bothincubators were fitted with a single inner door. Note the discrepancy betweenthe actual CO2 level and what was shown on the incubator display.

companies require standing orders for regular supply, there must be abackup plan in case of a lost or damaged shipment – an IVF lab cannotbe left without media under such circumstances.

Delivery and cold chainThe more stops along the way between the manufacturer and the IVFlab, the more difficult it is to guarantee the integrity of what is called the“cold chain,” i.e. the knowledge that the media have been kept withintheir prescribed storage temperature range throughout the entire inter-vening period. If the product is manufactured in one country andthen shipped to another via one or more intermediate places (e.g. thecourier company’s hubs of operation) then there must be a guaran-tee of how long it will take for this process to be completed, whichmust be within the ability of the packaging to maintain correct storage


temperatures. Then, the agent or distributor must guarantee (and pro-vide documented confirmation) that they maintained the proper stor-age conditions during their preparation of individual orders from theirbulk delivery. Finally, the local delivery must be accomplished in a sim-ilarly expeditious manner to ensure that the proper storage conditionswere maintained. Some companies ship temperature “tell-tales” oreven loggers with their shipments, and it is up to the IVF Lab Directorto decide how much documentation and confirmatory evidence (s)herequires to establish that the media were not compromised en route.

CostObviously the cost is an important factor in selecting products for usein IVF, perhaps especially in countries where the fees paid for IVF areless than in the more developed markets. For example, in India the costof an IVF cycle is far less than in, say, the USA – yet Indian labs haveto pay substantially higher prices for their culture media than US labsdue to higher freight costs plus customs duties and other taxes, as wellas the costs of the local agents/distributors.

Suitability for useObviously IVF labs are going to choose culture media that have beenestablished in the literature as suitable for human IVF and embryoculture. However, what is considered “best practice” changes over timeas our knowledge increases, and the use of systems other than sequentialmedia from reputable companies must be viewed with suspicion. Veryfew IVF labs now prepare their own culture media, especially as theregulatory environment tightens and requirements for manufactureof these media according to GMP standards (“Good ManufacturingPractice”) become more widespread. There are also different regulatoryrequirements in different parts of the world (e.g. FDA 510(k) pre-market clearance in the USA, CE marking as a device in the EuropeanUnion), and it is the responsibility of the IVF Lab Director to ensurethat only properly registered media are used.


Quality controlA further dimension here is the suitability of culture media for humanIVF and embryo culture use as defined by appropriate bioassays.All human IVF media manufacturers perform mouse embryo assay(“MEA”) testing on their products and typically certify that they “pass”this test. But what does this mean? Here we must consider not onlywhether the MEA is “foolproof” in detecting all embryotoxic contam-inants, but also just how reliable the MEA actually is. There have beensituations in recent times where media that have passed the MEA withflying colors were later found to contain contaminants that, while notaffecting the development of mouse embryos to the blastocyst stage,were certainly highly toxic to human oocytes or embryos in vitro. As fora comprehensive discussion on the reliability of the MEA, that could bethe subject of another book on its own! However, the simple principlesof quality management that govern the selection of products or servicesaccording to ISO 15189:2003 (see Chapter 6) require that the culturemedium supplier not only provides documentary evidence that theirproduct(s) “passed” whatever QC system they used (in this case theMEA), but that there is a statement of the Uncertainty of those mea-surements. From basic statistical sampling theory, the fewer embryosused to performed the MEA the less robust will be the typical “pass”level of >80% blastocyst development.

EfficacyWhile a particular culture medium product has been used by otherIVF labs with a level of success acceptable to them, will its performancemeet your expectations or requirements? Variations in other aspects ofthe culture system will likely change a product’s ultimate performance,and hence careful comparisons of systems should be undertaken beforedeciding on a particular product or accepting a report of its efficacy.Cook IVF was the first IVF medium company to provide benchmarkexpectations of performance for its culture media products (Cook IVF,1999), within the context of their complete culture system, based onthe experience of the center that developed the products (Sydney IVF,


under the scientific direction of one of the authors, DM). To the best ofour knowledge, no other IVF medium manufacturer has yet followedthis useful approach.

When/why do you need to use culture oil?

Why does an IVF lab use culture oil? What functions is it consideredto perform – and what advantages is it perceived to confer or possiblerisks might its use incur? We have always been of the opinion that oilis a nuisance in the IVF lab and have restricted its use to only thosesituations where its benefits are proven. The logic for that positionis considered briefly below – not in an attempt to “convert” otherembryologists but simply as the basic information upon which properinformed decisions can be made within the context of an FMEA, ratherthan a historical or “gut” opinion.

Prevents evaporation: Certainly a layer of oil over aqueous culturemedium will greatly reduce evaporative loss of water from themedium. But how much of a problem is this? Obviously whenworking with microdroplet culture an oil overlay is essential tomaintain the integrity of the droplets and prevent excessive evap-orative loss. But what is the risk of evaporative loss to the extentthat it causes an unacceptable shift in medium osmolarity forlarger volumes of culture medium? With incubators whose inter-nal atmosphere is properly humidified there is no serious risk ofsignificant evaporative loss when using “open” culture in suchvessels as Nunc plates or organ culture dishes (typical mediumvolumes = 0.5–0.8 ml).

Temperature stability: When questioned, many embryologistsexpress the opinion that the use of an oil overlay will help main-tain the temperature of the culture medium in the dish. Whenasked for the basis for that opinion, few are able to provide anysource for the information, leading us to conclude that it is onlydogma. Indeed, a recent study has shown that the presence of an


oil overlay not only does not protect medium from cooling, butthat the reverse effect is actually true (Cooke et al., 2002).

Gassing and pH: Measuring pH in culture medium under normalconditions of use (i.e. an equilibrated system with a CO2-enrichedatmosphere) is extremely difficult, even using micro pH probes –which are notoriously difficult to use and keep calibrated. Whendeveloping the original “M91” series of culture media that werecommercialized by Cook IVF as the Sydney IVF series of media (D.Mortimer et al., 2002) we did undertake some studies to verifythat the pH of media inside the MINC incubators was stablewithin the desired range when the chamber was supplied withpre-mixed gas (6% CO2/5% O2/balance N2). Having establishedthat the media performed properly under correctly controlledconditions we concluded that there was no need to monitor thepH of media inside the incubators since it was determined only bythe formulation of the medium, the temperature and the pCO2

of the gas phase – all of which can be controlled independently.Clearly this is a simpler, more robust approach to quality controlthan attempting to make routine measurements of pH knowingthat the proper measuring system is difficult to calibrate and veryprone to fluctuation.

But what about outside the incubator, for example during fer-tilization check or embryo assessments when the culture dish isexposed to room air? Or when culture dishes are placed in a tra-ditional CO2 incubator where the CO2 level has been reducedby the opening of the door? Our solution to the first prob-lem is to use IVF work stations with a controlled atmosphere,e.g. the IVF Chamber (see above), but what happens when thecontrol system is less robust? At the meeting of αlpha held inCopenhagen in 1999, Debbie Blake presented some studies on thedynamics of out-gassing of equilibrated culture medium eitheras drops under oil or as larger volumes in Petri dishes in a tra-ditional CO2 incubator (Blake et al., 1999). Some of her resultsare shown in Figure 11.4, which clearly illustrates the followingconclusions:


Figure 11.4 Graph showing the equilibration of either 50 �l drops of medium under oil(dotted line) or 5 ml of medium (broken line) in 60 mm diameter Falcon 3004dishes. Dishes were then taken out of the incubator (a Heraeus Cytoperm fittedwith a 6-section inner door and running at 6.5% CO2) and placed under an airatmosphere for 3 minutes before being replaced in the incubator. In both casesthe pH of the medium had exceeded 7.45 within 2 minutes of exposure to air,and re-equilibration took about 15 minutes for the 5 ml of medium in a dish and35 minutes for the 50 �l droplets. (Blake et al., 1999; data generously providedby Debbie Blake).

1. It takes longer to equilibrate 50 �l droplets of medium underoil than 5 ml of medium in an “open” dish.

2. In both cases the pH had exceeded the desired range (i.e. was>7.45) within 2 minutes of exposure to room air.

3. Re-equilibration after replacing the “out-gassed” dishes in theCO2 incubator took about 15 minutes for 5 ml of medium,but about 35 minutes for the 50 �l droplets under oil.

In summary, it appears that not only does oil not slow out-gassing(i.e. it does not help protect against pH shifts), it actually hinders


the re-equilibration of medium pH after exposure to room air.While further studies would be warranted to investigate this issuein more detail, in the meantime there would seem to be no benefitof using oil in terms of maintaining pH in culture media.

Creating a quality IVF lab

Creating a quality IVF lab is neither a simple nor short task. Cer-tainly it is easier to create such a lab when one is starting afreshwith a new facility, but most IVF labs don’t have that opportunity,they have to rebuild and reorganize what they have – usually whilehaving to maintain services with already limited human resources.Nonetheless, it is a daunting task that many Lab Directors will haveto face anywhere that accreditation is pursued, but especially withinthe European Union as the Tissues and Cells Directive is implemented(http://europa.eu.int/eur-lex/en/).

There is no single “right” way of doing this, no cookbook recipethat can be followed to achieve this goal. Each IVF lab will have todevelop its own strategic plan in response to its unique combination ofcircumstances. However, the general principles described in the earlierChapters of this book will provide the background and framework fordoing this, and the techniques and approaches we have described willprovide a basic toolkit. But an open mind combined with comprehen-sive multidisciplinary knowledge are vital pre-requisites for success.

In the final chapter of this book we will try and create a “road map” toreach this goal. But first we need to consider some aspects of managingthe IVF lab’s most precious resource: its people.

12

Human resourcesFinding (and keeping) the right staff

“Teamwork” is a huge buzzword in modern business, with the abilityto create and/or assemble a winning team considered to be one ofthe hallmarks of leadership. For a team to function well, there mustbe mutual trust, respect and cooperation. While each member of astrong, successful team has the knowledge, skills and confidence to be a“star” in their own right, they also understand that this talent is sharedby all the members of the team – and they each have the generosity ofspirit to allow everyone to shine. It is precisely because each person in awinning team is a “star” that they are sought after by competitors whoare hoping to create their own winning team. It is then incumbent uponthe manager of a winning team to ensure that the effort and success ofeveryone in the team is recognized and rewarded – otherwise the teammight be lost.

It is the same for the IVF Center, and for the IVF Lab, since a strong,functioning team is probably the greatest key to success. Recruitmentand retention of good embryologists is a challenge. However, it is achallenge which must be met, because if you don’t respect and lookafter your people, you have a fundamental flaw in your approach toQuality. This is also a fundamental failing for accreditation.

In this chapter, we will discuss the types of people you need to lookfor when developing your team, and some of the strategies for holdingonto the team once it is established and successful.

Who makes a good embryologist?

Of course, to have a strong team, you need to have good people.But how do you know who to hire? It is quite possible that good

201


embryologists are born, not made. We have found over the years thatsome of the best embryologists were hired as trainees on the basisof their personality traits, rather than their formal qualifications. Ofcourse, an embryologist must also have a good background in repro-ductive biology – a degree in biological sciences is a reasonable require-ment – but this should be the first step in the selection of staff, not theonly one.

In essence, a good clinical embryologist will have most or all of thesetraits:

• natural leader• well-developed sense of responsibility and accountability• able to work independently• self-starter• works well in a team• enjoys a challenge• perfectionist• strongly empathic• goal-oriented• energetic• honest• intelligent• creative

It is also important to ensure that each person hired has the abilityto meet the requirements of their job description. All applicants fora position must see the job description for that position as part ofthe interview process, and all of the requirements should be discussedfully.

Training: the importance of teaching “why,” not just “how”

After you have found the right people for your team, they need to betrained, or retrained. There must be a formal orientation and training

203 Human resources: finding (and keeping) the right staff

program, with a comprehensive review process to ensure that trainingis adequate. However, because you are trying to build a team composedof creative individuals, there is always a risk that even trained peoplewill think of a “creative” way to do a task, ignoring the SOP. As we havealready discussed in Chapter 6, this is a very dangerous tendency, andone which will adversely affect the IVF Center’s Quality ManagementSystem.

The best way to stop this potential drift away from the SOP is toensure that each person in the team understands why a particularprocedure is done in a particular way. This is illustrated in the storyof a young woman who moved away from home and had to cookon her own for the first time. Her flatmate noticed that whenevershe was going to roast a piece of meat, she always cut off a cornerof the raw meat, and asked her why she did that. The answer was“That’s how my mother always did it, and she’s a really good cook.”Eventually, this habit drove the flatmate crazy, so the girl called hermother and asked why she cut off a corner of the meat before sheroasted it. The answer was “I only have a small roasting pan, so ifthe piece is too big, the fat would drip into the oven, so I alwayscut off any part that doesn’t fit into the pan to keep the oven clean.”Clearly, the “why” asked a bit earlier would have helped the householdbudget!

In the lab, the best approach is to give the “why” as part of the“how.” In other words, the rationale for a particular procedure shouldalways be explained at the same time the procedure is being taught.This is helpful in ensuring that the SOP is respected, and it is also agood way to illustrate how all the processes in the lab are inter-related,and how they are each related to physiology. This should be sufficientto emphasize the importance of not deviating from the SOP, since, asa living document, it is the result of many years of experience. Fur-thermore, understanding the procedural and physiological bases fora procedure are necessary in troubleshooting, in developing improve-ments in methods, and in formulating research questions – each ofwhich are skills required in a good clinical embryologist.


Why other people should be trying to steal your staff (and why theywill be unsuccessful)

If you have managed to assemble and/or create a winning team, then,by definition, they will each have the knowledge, skills, experience andconfidence to be a star in their own right. It is a great compliment toyou as a team leader or manager if your stars are being actively recruitedby other clinics. However, since your staff are your greatest asset, youdon’t want to make a present of them to your competitors. The wayyou get people to stay with you is to:

(a) ensure that they understand their value to the team; and(b) perhaps more importantly, ensure that their contributions are rec-

ognized, illustrating that the team values each of its members.

Frustration as a result of inadequate recognition is a significant problemin a team environment like IVF, and it is one of the major reasons thatpeople will move themselves and their family across a city, a countryor the world just to work in another IVF Center.

Different types of reward

The size of a person’s salary is a concrete demonstration of their value,and is a reward for many years of effort. It seems logical, then, thatif a person is receiving a high salary, they will be satisfied enoughto stay working for you. However, the higher a person’s salary, theless motivating money becomes, and some people are motivated byfactors unrelated to money. If their needs are not satisfied, or at leastrecognized, these people will find somewhere more amenable to work.

To keep employees’ interest and motivation (and therefore to main-tain a healthy and productive lab), there must be a consideration of:

• internal motivation – personal belief in, and commitment to, thework; and

• external motivation – rewards, recognition and personal growth.


Internal motivation is achieved when employees have a strong per-sonal belief in the company’s goals and processes. In the case of anIVF Center, the contribution the lab team makes to people’s lives canbe a very strong motivational factor. Ensuring that the lab staff havecontact with patients, perhaps as cursorily as introducing themselvesto couples before procedures, is a way of ensuring that the impor-tance of their role is continually reinforced. Self-confidence is anothermajor factor for internal motivation. When a person feels that theyare performing an important role in the best possible manner, theirself-satisfaction is itself a reward. In the same way, if someone does nothave internal motivation, then they can never become a good clinicalembryologist.

External motivation comes from rewards given in formal recogni-tion of a job well done. The most obvious reward is salary, but it canalso be:

• more vacation days• flexible work hours to allow for family commitments• more office space• an enhanced health plan• increased contributions to a retirement plan• personal health programs, e.g. quit smoking classes, or gym mem-

bership• professional development, e.g. in-house education, support for for-

mal qualifications• development of research projects• time for formal presentation of research results (i.e. writing papers)• support to attend conferences (and associated travel opportunities)

Because each person’s needs are different, there should be a policysupporting two-way communication about the rewards program. Toensure staff morale is not damaged by rewards being offered indiscrim-inately, they should be given only as formal recognition of a job welldone. Poor performance should never be rewarded, and there must alsobe a policy for effective discipline and punishment. It is also important


that all staff be treated fairly, with rewards and/or discipline relatedstrictly to work performance.

It should be understood that external motivation is not going tomake someone with poor, or no, internal motivation into a goodworker. Its value is in helping to ensure that a good worker maintainstheir internal motivation, and so will be less likely to look somewhereelse for recognition and satisfaction.

Development of a career path

In addition to the rewards discussed above, creative, energetic, intelli-gent, goal-oriented people (like successful clinical embryologists) alsoneed to have long term goals, such as a career path.

In almost every organization, the normal structure of the organi-zational chart is that of a pyramid (Figure 12.1). However, in an IVFlab there are – or one hopes to have – more senior embryologiststhan junior ones, yet there is only one Lab Manager or Lab Director(Figure 12.2). So how can a Lab Director provide a framework for pro-fessional development within the necessarily restrictive career structureof clinical embryology? One tactic that has worked well is to identifyareas of specialized interest or responsibility within the laboratory thatcan be awarded to those senior embryologists who are motivated toseek them (Figure 12.3). Rewards can then be provided after they havedemonstrated achievement in the extra role.

Delegation

Delegation, if done well, can be another very useful tool in developingand motivating employees. From the manager’s point of view, it freesup time needed for other projects. From the delegate’s point of view, itis a tangible expression of the manager’s trust and confidence in theirabilities. It is also an opportunity for the delegate to demonstrate talentswhich may not be needed in their day-to-day work (e.g. graphic designskills in making a poster). However, if the wrong person is delegated


BOSS

MANAGER

SENIOR SENIOR

STAFF STAFF STAFFSTAFFSTAFF STAFF

TRAINEE

Figure 12.1 An organization chart of a generic company hierarchy.

to do a task, or if the ground rules are not laid out in advance, it canbe a disaster. For delegation to be successful:

• the right delegate must be selected – this is the same as when hiringsomeone, i.e. their skills must match the job description;

• the reason and goals for the task must be explained clearly;• there must be a strict deadline given for the presentation of the

completed project; and• once the task is delegated, there must be no interference from the

manager – the trust given in the delegation of the task must berespected.

Another benefit of judicious delegation is that it gives the managerthe opportunity to observe the delegate’s approach to completing theproject, which is a very useful way to test someone’s ability in a new,


LABDIRECTOR

EMBRYOLOGY LAB MANAGER

SENIOREMBRYOLOGIST

SENIOREMBRYOLOGIST

SENIOREMBRYOLOGIST

SENIOREMBRYOLOGIST

SENIOREMBRYOLOGIST

EMBRYOLOGIST EMBRYOLOGIST

TRAINEEEMBRYOLOGIST

SENIOREMBRYOLOGIST

Figure 12.2 An organization chart showing the ideal situation for an IVF laboratory.

expanded role, before taking the step of promoting them to it. Thisis crucial to ensure not only that the person could be happy with anew role, before proposing it to them, but also that they are capableof fulfilling the role. In this way, one can prevent the promotion ofsomeone into a role which is too difficult for them – thereby avoidingthe “Peter Principle” (after Laurence Johnston Peter [1919–1990]), i.e.“employees within an organization will advance to their highest level ofcompetence and then be promoted to and remain at a level at which theyare incompetent” (The American Heritage® Dictionary of the EnglishLanguage: Fourth Edition, 2000).

Other considerations

Apart from these principles of reward and recognition, there are someother factors which should be taken into consideration.

• It is critical that enough lab staff are employed to ensure adequatecoverage on the busiest days, and sufficient time off for all staff. Itis unreasonable, and a short-term solution at best, to rely on people


LABDIRECTOR

EMBRYOLOGY LAB MANAGER

CRYOBANKMANAGER

OVERSEASCOORDINATOR

TRAININGCOORDINATOR

PGD/PGSCOORDINATOR

R & DCOORDINATOR

EMBRYOLOGIST EMBRYOLOGIST

TRAINEEEMBRYOLOGIST

QUALITYMANAGER

Figure 12.3 An organization chart for an ”ideal” IVF laboratory showing how the assignmentof ”special responsibilities” can help create a career structure for the seniorembryologists.

being good-natured enough to sacrifice their personal time for theclinic; it will result in resentment and burnout.

• There needs to be good communication to ensure that any loomingproblems within the team are dealt with promptly and effectively.

• Space and time are important. In an environment of “knowledgeworkers,” there is a need for creative thought, and this is achievedmost effectively when there is sufficient time and workspace. It is notreasonable to expect someone to generate good ideas if they are notable to find the time to think, and if they are not given the appropriateinputs – such as access to journals – and a supportive environment.

Finally, it should be remembered that no matter how great the team, andhow supportive the work environment, people will leave. Hopefully,their decision will be due to personal circumstances rather than todissatisfaction. Therefore, as a Quality initiative, it is important thatexit interviews are carried out, to determine why the person is leaving,in case areas for improvement can be identified.

13

The well-run lab

We hope that the preceding chapters will have provided sufficient back-ground and introduction to the tools and techniques used in QualityManagement and Risk Management to allow any IVF Lab Director toembark upon the road towards creating the best lab in the world. This isnot a facetious remark because everyone has access to the same proto-cols, equipment, techniques, plasticware, culture media, etc, as anyoneelse – so why shouldn’t any lab, anywhere, have the same opportunityto be as good as any other?

But why would we want to expend what is, unarguably, a hugeamount of effort, on changing the nice comfortable lab that we’vebeen running for n years into one that will require us to spend a notinconsiderable amount of time monitoring and dealing with all theQC/QA issues, document control, etc? To our minds, the explanationcan be summed up as:

(a) being professional: the need always to do one’s best and adhere tothe principle of primum non nocere; and

(b) the advantages: better results, less risks, higher morale and confi-dence.

For those working in the private sector, the commercial advantage ofimproved success rates must then also be factored into the equation.

What does it take?

To develop a quality lab that achieves the highest success ratesand minimizes its risks requires a broad spectrum of resources, a

210

211 The well-run lab

shortfall in any one of which can cause the whole endeavour tofail.

1. For any organization to be able to change, there is an absolute needfor “slack” (DeMarco, 2001). Insufficient slack will compromise theavailability of vital human resources and the stress on the morale ofcritical personnel will destroy their commitment to the process ofchange.

2. Everyone in the organization must be committed and involved.Involvement and support from the highest to the lowest membersof the organization are vital.

3. Adequate resources: time and money. This might require additionalpersonnel, some permanent, but many can be employed on a tem-porary basis or as consultants.

4. The ability to manage change must exist within the organiza-tion (Heller and Hindle, 2003). All key players must understandhuman nature – our innate fear of change and the inertia that thiscreates.

5. Attitudes must be corrected where necessary: any culture of blamemust be eliminated, mistakes must be seen as opportunities forimprovement, not events that require scapegoats and punishment.Basically, any element of a “toxic workplace” must be eradicated(Coombs, 2001).

6. A vision of how things should be run to create a positive environ-ment for one’s staff as well as focussing on what your “customers”want.

In regard to point 6, remember the discussions on the definition ofQuality in Chapter 3 and the change in perspective from being a“product-out” company to a “market-in” company. The organizationchart for an IVF Center shown in Figure 13.1 might serve the organi-zation very well in terms of establishing its hierarchy and the lines ofauthority and responsibility, but the alternate view of the same organi-zation shown in Figure 13.2 eloquently illustrates the change in focuswhen a “market-in” organization provides its services.


Chief ExecutiveOfficer (CEO)

Medical Director Nursing Director Scientific DirectorGeneral ManagerQuality Manager

Medical StaffOfficeManager

FinanceManager

BOARD OF DIRECTORS

Day SurgeryManager

AndrologyLab Manager

EmbryologyLab Manager

GeneticsLab Manager

Senior PatientCoordinator

CounselorsAccountsReceivable Secretarial

Reception

Day SurgeryNurses

LaboratoryAndrologists

(Senior/Junior+ Trainees)

Embryologists(Senior/Junior

+ Trainees)

Geneticists:- FISH / PGD

-cytogenetics(+ Trainees)

PatientCoordinators

Ultrasono-graphers

AccountsPayable

Accounts clerks Phlebotomists Technicians TechniciansNursesRecoveryNurses

EXECUTIVE MANAGEMENTCOMMITTEE

EthicsCommittee

Figure 13.1 An organization chart for an IVF clinic reflecting the traditional (“product-out”)view of its hierarchy, defining lines of authority and responsibility.

Chief ExecutiveOfficer (CEO)

Medical Director Nursing Director Scientific DirectorGeneral ManagerQuality Manager

MedicalStaff

BOARD OF DIRECTORS

LaboratoryStaffCounselorsSecretarial Reception Patient

CoordinatorsUltrasono -graphers

AccountsStaff

Phlebot -omists

NursingStaff

EXECUTIVE MANAGEMENTCOMMITTEE

EthicsCommittee

P A T I E N T S

Figure 13.2 An alternate view of the organization chart for an IVF clinic that is focussed onbecoming a “market-in” company, more responsive to the needs of its patients.


ESTABLISHTHE GOALS

EDUCATEEVERYONE

DESIGNTHE PROCESS

BUILDTHE TEAMS

IMPLEMENTTHE

ACTION PLAN

CREATE THEACTION PLAN

MONITORTHE CHANGES

REVISE IFNECESSARY

DEFINETHE AREAS

FOR CHANGE

Figure 13.3 A road map of the journey to achieving excellence in the IVF Lab.

Achieving excellence in the IVF lab also requires the scientists whowork there to think and act as scientists. As has been seen numeroustimes throughout this book, scientific method is at the foundationof many of the concepts and approaches used in Quality Manage-ment and Risk Management, and so we as scientists should see theseareas as logical extensions of our scientific work and their achievementas being based on common sense. A valuable resource that we havefound not only highly enjoyable to read, but a wonderful expression ofwhat it takes to be a scientist, is the book by Jack Cohen and GrahamMedley entitled Stop Working and Start Thinking: A Guide to Becom-ing a Scientist (Cohen and Medley, 2000) – everyone working in a labanywhere should read their “little red book.”


How do we get there?

There are no short-cuts to achieving excellence, but a bird’s-eye per-spective of the “road map” is shown in Figure 13.3, the general conceptsof which should be pretty familiar by now.

1. Establish the goals and get “buy-in” from everyone in the organi-zation. This buy-in must be reinforced along the way, any setbacksor unresolved problems will create doubt in some of your people’sminds, and once someone has lost faith in the process it’s very muchharder to get them back on board.

2. Educate everyone in the tools and techniques that will be used:(a) systems analysis and process mapping (including the flowchart-

ing tools that will be used);(b) benchmarking;(c) audits;(d) how to perform FMEAs and RCAs; and(e) scientific method and troubleshooting.

3. Design the process of what you will be doing:(a) appoint a Quality Manager;(b) ensure that management recognizes that the time spent by

everyone working in their Teams is counted as work time(including overtime, if required), and not personal time;

(c) identify the areas for change and prioritize them (you can’tchange everything at once);

(d) define the specific goals; and then(e) set a realistic timetable.

4. Build the teams who will tackle each of the areas:(a) identify the Team Leaders;(b) recruit the Team Members; and(c) have each Team establish their scope of activities.


5. Define the areas for change:(a) each Team reviews their area and identifies the initial targets for

improvement;(b) each Team then prioritizes their areas for change in terms of

“bang for the buck” and risk minimization; and then(c) each Team sets their goals, constructs the tasks that will lead

to those goals, and identifies how each change will be mon-itored for effective implementation and efficacy (i.e. selectIndicators) – in other words, they build their Action Plan.

6. Implement the changes contained in the Action Plan(s).

7. Monitor the changes using the defined Indicators.

8. Review the outcomes – and then revise or repeat as necessary untilthe goal is achieved.

9. Continue the cycle because no system ever achieves perfection. Thisis the foundation of TQM and the underlying principle of all accred-itation schemes.

In all of this the role of the IVF Lab Director is not to run the wholeTQM program for the lab. While the Lab Director might well leadone or more of the lab Teams, (s)he will also be involved in Teamsworking in other areas of the Center (since TQM cannot be under-taken by the lab in isolation). Not only will the Lab Director not havethe time to run everything, but in attempting to do so they will bemissing one of the most important points of the process – that of cre-ating an environment in which all members of the lab staff not onlybuy-in to the process (which requires that they are directly involved)but also one in which there are opportunities for career and personaldevelopment.

In any well-run IVF Center the Lab Director must be a central,key figure in the organization. The lab cannot be “marginalized” and


seen as just a “back room” where “the techs do their thing” – IVFdoes not exist without the lab and its people. Indeed, at the αlphaFoundation Workshop held during the 1999 World IVF Congress inSydney, Bob Edwards opened the day with an uplifting perspective onthe role of the Scientist in ART, in which he made the point that theIVF laboratory must be run by the scientist, not the clinician, whilestill emphasizing the importance of the team approach and mutualrespect. At the conclusion of the talk he stated that scientists should bemore able to understand business and finance (the better to run IVFprograms) and suggested that αlpha or national clinical embryologistgroups should consider providing business training as part of theireducational remit. Throughout this presentation he emphasized thatit is the scientist, not the clinician, who does science. Yet again, wemarvel at the prescience of Bob Edwards, whose understanding of thefield of scientific endeavour that he created is being vindicated by theincreasing importance of the central role played by the IVF lab inaccreditation and regulatory schemes.

Is it all worth it?

Both of us have been intimately involved in accreditation systems and intaking private IVF Centers through the accreditation process. We can,therefore, put our hands on our hearts and say, without reservation,that “yes, it is all worth it.” Yes, it does take a lot of time and effort,but these investments will be returned many-fold in the future as thelab continues to operate “in control” and with the occurrence of crisesgreatly reduced. A true cost-benefit analysis will reveal that “qualityalways pays for itself” – in the long run.

As with everything in life, there is no guarantee of success orreward, but the benefits of embarking on the “quality journey” areirrefutable. Besides the obvious professional endpoints of improvedresults, increased efficiency and general job satisfaction, there are


financial advantages as well. However, it is important to recognize thatsavings might not be obvious in operational costs – since the technicalprocess often ends up costing more when we use properly-certifieddevices and other products, higher quality products, etc – but therewill be larger-scale savings by eliminating inefficiency and the need for“re-work.” There are also several highly beneficial corollaries to havinga well-run IVF Lab.

• Keeping stress levels down for yourself and your staff by virtueof well-designed and efficient systems that are unlikely to gowrong.

• Being able to sleep easy at night, without having to worry about whatwent wrong today, or what might go wrong tomorrow.

• That warm glow that comes from “being the best.”• The comfort of knowing that you are unlikely to suffer the embar-

rassing catastrophes that befall some other labs.• The confidence that next time you’re asked “What’s wrong in the

lab?” you’ll be able to give the prompt response “Nothing, everythingis running within its control limits.” This is not about “CYA,” but itcertainly does give you asbestos underpants!

Implementing Quality Management and Risk Management in IVFCenters should be based on carrots, and not sticks, but regulatoryauthorities will always have the last word: the EU Tissues and CellsDirective requires a formal Quality Management System, and so everyIVF Center in the EU has to create one. However, hopefully we haveconvinced you that professional self-respect should be sufficient moti-vation, and that prophylaxis is far better than cure.

References and recommended reading

Aitken, R. J. and Clarkson, J. S. (1988) Significance of reactive oxygen species

and antioxidants in defining the efficacy of sperm preparation techniques.

Journal of Andrology, 9: 367–376.

Almeida, P. A. and Bolton, V. N. (1995) The effect of temperature fluctuations on

the cytoskeletal organisation and chromosomal constitution of the human

oocyte. Zygote, 3: 357–365.

Alper, M. M., Brinsden, P. R., Fischer, R., and Wikland, M. (2002) Is your IVF

programme good? Human Reproduction, 17: 8–10.

Bavister, B. D. (1995) Culture of preimplantation embryos: facts and artifacts.

Human Reproduction Update, 1: 91–148.

(1999) Stage-specific culture media and reactions of embryos to them. In

Jansen, R. and Mortimer, D. (eds.), Towards Reproductive Certainty: Fertility

and Genetics Beyond 1999. Carnforth, UK, Parthenon Publishing, pp. 367–

377.

Bjorndahl, L., Barratt, C. L. R., Fraser, L. R., Kvist, U., and Mortimer, D. (2002)

ESHRE basic semen analysis courses 1995–1999: immediate beneficial effects

of standardized training. Human Reproduction, 17: 1929–1305.

Blake, D. A., Forsberg, A. S., Hillensjo, T., and Wikland, M. (1999) The practicali-

ties of sequential blastocyst culture. Presented at ART, Science and Fiction, the

Second International Alpha Congress, Copenhagen, Denmark, September

1999. [email protected].

Bogner, M. S. (ed.) (1994) Human Error in Medicine. Hillsdale, USA, Lawrence

Erlbaum Associates Inc.

Brison, D. R., Hooper, M., Critchlow, J. D., et al. (2004) Reducing risk in the IVF

laboratory: implementation of a double-witnessing system. Clinical Risk, 10:

176–180.

Catt, J. W. and Henman, M. (2000) Toxic effects of oxygen on human embryo

development. Human Reproduction, 15, Suppl. 2: 199–206.

218

219 References and recommended reading

Cohen, J. and Medley, G. (2000) Stop Working and Start Thinking: A Guide to

Becoming a Scientist. Cheltenham, UK, Stanley Thornes.

Cohen, J., Gilligan, A., Esposito, W., Schimmel, T., and Dale, B. (1997) Ambient

air and its potential effects on conception in vitro. Human Reproduction, 12:

1742–1749.

Cook IVF (1999) Cook Culture Systems. Instructions for Use. Eight Mile Plains,

Australia, Cook IVF.

Cook, R. R. (1999) Assessment of Uncertainties in Measurement for Calibration

and Testing Laboratories. Sydney, Australia, National Association of Testing

Authorities.

Cooke, S., Tyler, J. P., and Driscoll, G. (2002) Objective assessments of tem-

perature maintenance using in vitro culture techniques. Journal of Assisted

Reproduction and Genetics, 19: 368–375.

Coombs, A. (2001) The Living Workplace: Soul, Spirit and Success in the 21st

Century. Toronto, Canada, HarperCollins.

Cullinan, R. T., Catt, J. W., Fussell, S., Henman, M., and Mortimer, D. (1998)

Improved implantation rates of cryopreserved human embryos thawed

in a phosphate-free medium. 14th Annual Meeting of the European

Society of Human Reproduction and Embryology, Goteborg, Sweden,

June 1998. Abstract O-117. Human Reproduction, 13 (Abstract Book 1):

59–60.

Dale, B. and McQuater, R. (1998) Managing Business Improvement and Quality:

Implementing Key Tools and Techniques. Oxford, UK, Blackwell.

Damelio, R. (1996) The Basics of Process Mapping. Portland, OR, Productivity

Inc.

DeMarco, T. (2001) Slack: Getting Past Burnout, Busywork, and the Myth of Total

Efficiency. New York, NY, Broadway Books.

Department of Health (2001) A Code of Practice for Tissue Banks Providing Tissues

of Human Origin for Therapeutic Purposes. London, UK, Department of

Health. (http://www.doh.gov.uk/humantissuebanking/tissuebank.pdf).

Edwards, R. and Steptoe, P. (1980) A Matter of Life. London, UK, Hutchinson &

Co.

Elder, K. and Elliott, T. (1998) Problem Solving and Troubleshooting in IVF. Perth,

Australia, Ladybrook.

European Union (2004) Directive 2004/23/EC on setting standard of quality

and safety for the donation, procurement, testing, processing, preservation,


storage and distribution of human tissues and cells. Official Journal of the

European Union, L102: 48–58. (http://www.europa.eu.int/eur-lex/en).

Ford, W. C. L. and Rees, J. M. (1990) The bioenergetics of mammalian sperm

motility. In Gagnon, C. (ed.), Controls of Sperm Motility: Biological and

Clinical Aspects. Boca Raton, FL, CRC Press.

Fraser, L. R. and Quinn, P. J. (1981) A glycolytic product is obligatory for ini-

tiation of the sperm acrosome reaction and whiplash motility required for

fertilization in the mouse. Journal of Reproduction and Fertility, 61: 25–35.

Grout, J. R. (2004) Process mapping. http://campbell.berry.edu/faculty/jgrout/

processmapping/.

Harbottle, S. (2003) Are you working too hard? Annual Meeting of the Association

of Clinical Embryologists, Glasgow, UK.

Heller, R. and Hindle, T. (2003) Essential Manager’s Manual. New York, NY, DK

Publishing.

Hutchison, D. (1994) Total Quality Management in the Clinical Laboratory.

Milwaukee, WI, ASQC Quality Press.

IEEE Standards Association (1998) IEEE Standard for Function Modeling

Language: Syntax and Semantics for IDEF0. IEEE 1320.1-1998 (Replaces FIPS

PUB 183). Institute of Electrical and Electronics Engineers Standards Asso-

ciation, Piscataway, NJ, USA. http://standards.ieee.org/.

ISO (1993) Geometrical Product Specifications (GPS): Inspection by Measurement

of Workpieces and Measuring Equipment. Part 2: Guide to the Estimation of

Uncertainty in GPS Measurement, in Calibration of Measuring Equipment and

in Product Verification. ISO/TS 14253–2:1999. International Organization

for Standardization, Geneva, Switzerland.

(2000) Quality Management Systems: Requirements. International stan-

dard ISO 9001. International Organization for Standardization, Geneva,

Switzerland.

(2003) Medical Laboratories: Particular Requirements for Quality and Com-

petence. International Standard ISO 15189. International Organization for

Standardization, Geneva, Switzerland.

Jacka, J. M. and Keller, P. (2001) Business Process Mapping: Improving Customer

Satisfaction. New York, NY, John Wiley and Sons.

Kemeter, P. and Lietz, K. (2002) IVF/ICSI with 5% O2 in a Mini-Incubator Com-

pared with 21% O2 in a Conventional Incubator. Eight Mile Plains, Australia,

William A. Cook Australia.


Kennedy, C. R. (2004) Risk management in assisted reproduction. Clinical Risk,

10: 169–175.

Kuist, U. and Bjorndahl, L. (2002) Manual on basic semen analysis. ESHRE Mono-

graphs, Oxford, UK, Oxford University Press.

Lane, M., Lyons, M. A., and Bavister, B. D. (2000) Cryopreservation reduces

the ability of hamster 2-cell embryos to regulate intracellular pH. Human

Reproduction, 15: 389–394.

Leape, L. L. (1997) A systems analysis approach to medical error. Journal of

Evaluation in Clinical Practice, 3: 213–222.

Leese, H. J. (1990) The environment of the preimplantation embryo. In Edwards,

R. G. (ed.), Establishing a Successful Human Pregnancy, Serono Symposia

Publications Volume 66, New York, NY, Raven Press, pp. 143–154.

Marconi, G., Vilela, M., Bello, J., et al. (2003) Endometrial lesions caused by

catheters used for embryo transfers: a preliminary report. Fertility and

Sterility, 80: 363–367.

Mayer, J. F., Nehchiri, F., Weedon, V. M., et al. (1999) Prospective randomized

crossover analysis of the impact of an IVF incubator air filtration system

(Coda, GenX) on clinical pregnancy rates. Fertility and Sterility, Suppl. 1:

S42–S43.

Mayer, J. F., Jones, E. L., Dowling-Lacey, D., et al. (2003) Total quality improve-

ment in the IVF laboratory: choosing indicators of quality. Reproductive

Biomedicine Online, 7: 695–699.

Menezo, Y. (1976) Milieu synthetique pour la survie et la maturation des gametes

et pour la culture de l’oeuf feconde. Comptes rendus de l’Academie des Sciences

de Paris, Serie D, 282: 1967–1970.

Mortimer, D. (1986) Elaboration of a new culture medium for physiological

studies on human sperm motility and capacitation. Human Reproduction,

1: 247–250.

(1991) Sperm preparation techniques and iatrogenic failures of in-vitro fertil-

ization. Human Reproduction, 6: 173–176.

(1994) Practical Laboratory Andrology. New York, NY, Oxford University

Press.

(1999) Quality management in the IVF laboratory. In Jansen, R. and Mortimer,

D. (eds.), Towards Reproductive Certainty: Fertility and Genetics Beyond 1999.

Carnforth, UK, Parthenon Publishing, pp. 421–426.

(2000) Sperm preparation methods. Journal of Andrology, 21: 357–366.


(2004a) Current and future concepts and practices in human sperm cryobank-

ing. Reproductive Biomedicine Online, 9: 134–151.

(2004b) Setting up risk management systems in IVF laboratories. Clinical Risk,

10: 128–137.

Mortimer, D. and Mortimer, S. T. (1992) Methods of sperm preparation for

assisted reproduction. Annals of the Academy of Medicine of Singapore, 21:

517–524.

Mortimer, D., Demetriou, A., Cullinan, R., Henman, M., and Jansen, R. P. S.

(1995) A novel approach to IVF laboratory quality control. IXth World

Congress on In Vitro Fertilization and Assisted Reproduction, Vienna,

Austria, April 1995, Abstract OC-160. Journal of Assisted Reproduction and

Genetics, 12(3) Suppl. 66S.

Mortimer, D., Henman, M. J., and Jansen, R. P. S. (2002) Development of an

Improved Embryo Culture System for Clinical Human IVF. Eight Mile Plains,

Australia, William A. Cook Australia.

Mortimer, D., Shu, M. A., and Tan, R. (1986) Standardization and quality control

of sperm concentration and sperm motility counts in semen analysis. Human


Mortimer, D., Shu, M. A., Tan, R., and Mortimer, S. T. (1989) A technical note

on diluting semen for the haemocytometric determination of sperm con-

centration. Human Reproduction, 4: 166–168.

Mortimer, S. T. (2002) Practical application of computer-aided sperm analysis

(CASA). In van der Horst, G., Franken, D., Bornman, R., de Jager, T.,

and Dyer, S. (eds.), Proceedings of the 9th International Symposium on

Spermatology, Cape Town, South Africa. Bologna, Monduzzi Editore S.p.A.,

pp. 233–238.

Mortimer, S., Fluker, M., and Yuzpe, A. (2001a) Keeping the lab in control: use

of laboratory performance measures. 47th Annual Meeting of the Canadian

Fertility and Andrology Society, Whistler, BC, Canada, Abstract FP3.

(2001b) Effect of culture conditions on embryo cleavage rates and morphol-

ogy. 47th Annual Meeting of the Canadian Fertility and Andrology Society,

Whistler, BC, Canada, Abstract FP1.

(2002) Relative effects of culture conditions on fertilization rate and embryo

quality. 48th Annual Meeting of the Canadian Fertility and Andrology Soci-

ety, Manoir Richelieu, Quebec, Canada, Abstract FP23.

Motulsky, H. (1995) Intuitive Biostatistics. Oxford, UK, Oxford University Press.


Munne, S., Magli, C., Adler, A., et al. (1997) Treatment-related chromosome

abnormalities in human embryos. Human Reproduction, 12: 780–784.

Murray, C. T. and Mortimer, D. (1999) Monitoring laboratory performance mea-

sures for quality assurance in IVF. 11th World Congress on In Vitro Fertiliza-

tion and Human Reproductive Genetics, Sydney, Australia. Abstract O-058,

p. 90.

National Institute of Standards and Technology (1993) Integration Definition

for Function Modelling (IDEF-0). Draft Federal Information Processing

Standards Publication 183. Available from http://www.idef.com/.

NCCLS (2002) Clinical Laboratory Technical Procedure Manuals. Approved

Guideline: Fourth Edition, document GP2-A4. NCCLS, Wayne, PA, NCCLS.

www.nccls.org.

Pickering, S. J., Braude, P. R., Johnson, M. H., Cant, A., and Currie, J. (1990)

Transient cooling to room temperature can cause irreversible disruption of

the meiotic spindle in the human oocyte. Fertility and Sterility, 54: 102–108.

Quinn, P. (1995) Enhanced results in mouse and human embryo culture using

a modified human tubal fluid medium lacking glucose and phosphate ions.

Journal of Assisted Reproduction and Genetics, 12: 97–105.

Quinn, P., Kerin, J. F., and Warnes, G. M. (1985) Improved pregnancy rate in

human in vitro fertilization with the use of a medium based on the compo-

sition of human tubal fluid. Fertility and Sterility, 44: 493–498.

Reason, J. T. (1994) Foreword. In Bogner, M. S. (ed.), Human Error in Medicine.

Hillsdale, NJ, Lawrence Erlbaum Associates.

Sharpe, V. A. and Faden, A. I. (1998) Medical Harm: Historical, Conceptual,

and Ethical Dimensions of Iatrogenic Illness. Cambridge, UK, Cambridge

University Press.

Standards Australia (1999) Risk Management. AS/NZS 4360:1999. Standards

Australia, Sydney, Australia. http://www.standards.com.au.

Tervit, H. R., Whittingham, D. G., and Rowson, L. E. A. (1972) Successful culture

in vitro of sheep and cattle ova. Journal of Reproduction and Fertility, 30:

493–497.

US Department of Health and Human Services (USDHHS) (1992) Clinical Lab-

oratory Improvement Amendments of 1988: Final Rules and Notice. 42 CFR

Part 493. The Federal Register, 57: 7188–7288.

Wang, W. H., Meng, L., Hackett, R. J., Odenbourg, R., and Keefe, D. L.

(2001) Limited recovery of meiotic spindles in living human oocytes after


cooling–rewarming observed using polarized light microscopy. Human


World Health Organization (1999) WHO Laboratory Manual for the Examina-

tion of Human Semen and Sperm–Cervical Mucus Interaction, 4th Edition.

Cambridge, UK, Cambridge University Press.

Wikland, M. and Sjoblom, C. (2000) The application of quality systems in ART

programs. Molecular and Cellular Endocrinology, 166: 3–7.

Index

Page numbers in italics refer to tables and figures.

accreditation 12

defined 8, 12

generic process 18, 19

national schemes 13–14

ongoing commitment to quality 22–23

own processes 23

process 18, 19

external assessment 10, 12–13, 21

recommendations 21–22

self-assessment 12–13, 18–21

standards

features 12

operational areas 20–21

see also ISO standards

Accu-Beads 107

accuracy 111–112

ACHS 13

adverse events 51, 170–171

Alpha 135

andrology EQA schemes 117

anthrax bioterrorism 52

AS/NZS 4360 46, 145, 154, 155

audit 119

Australia, accreditation schemes 13

Australian Council on Healthcare

Standards (ACHS) 13

bar codes 82, 81–83

benchmarking 169–177

adverse events 51, 170–171

competitive 176

functional (generic) 176–177

honesty in 175–176

indicators

best practice 173

choice of 171

efficiency 173

financial 174

laboratory operations 174

laboratory performance 104–105,

172–173

programme performance 172

sentinel 170

internal 176

reference populations 174–175

best practice indicators 173

bioterrorism 52

calibrators 107

Canadian Council on Health Service

Accreditation (CCHSA) 13

Canadian Fertility and Andrology Society

(CFAS) 13

CAP 14, 91

CCHSA 13

certification 9

CFAS 13

Clinical Pathology Accreditation (UK)

Ltd. (CPA) 14

CO2 incubators

choice of 185, 186, 190, 191,

193–199

225

226 Index

CO2 incubators (cont.)

CO2 recovery after opening 194

troubleshooting example 141–144

College of American Pathologists (CAP)

14, 91

consequence for risks 123

control charts 101, 101–105, 103, 104

control limits 102

control mean 102

warning limits 102

see also process control, charts

controlled documents see document

control systems

Cook IVF 177, 196

MINC incubator see MINC incubator

Coombs, Ann 42

counting errors 109, 110

CPA 14

credentialing 9

criticality scores 121

cryobank management, risk

minimization 146–147

cryopreservation buffers 183–184

cryostorage packaging 164, 161–166

culture media, choice of 190–193

availability 193–194

cost 195

delivery and cold chain 194–195

efficacy 196–197

quality control 196

suitability 195

culture oil, factors influencing use

197

gassing and pH 194, 198–200

prevention of evaporation 197

temperature stability 197–198

customer expectations 33

Deming (PDCA) cycle 124, 126

Deming, W. Edwards 25

Department of Health (2001) 187

detection-based mentality 42, 42

disaster planning 52

documentation, poor 51

document control systems

forms 114

naming of electronic files 114–115

requirements 113, 112–114

SOP review and reissue 113, 114

documents, classifying 116, 116

double-witnessing 50, 79–81

duty of care 32

earthquakes 52

education of staff 36–37, 76–77,

202–203

efficiency indicators 173

electrical power failure 50

electronic files, naming conventions

114–115

embryo disposal 155–157

embryologists, recruitment of 201–202

embryology work stations 184

air quality requirements 187–190

cabinet styles 184–185, 186

gassed enclosures 185, 186

pros and cons summarized 188

warmed stages 185, 186

employee issues see human resources

employment regulations 11

EPCoT Systems 83

EQA (external quality assurance) 117

equipment

maintenance and monitoring 50

selection 179–180

errors

random 110

sampling 109, 110

systematic 110

system flows 166

ethics 32–33

European Union (2004) 17, 187

European Union Tissues and Cells

Directive 17–18, 187, 217

227 Index

external assessment 21

external audit 119

external quality assurance (EQA) 117

Failure Modes and Effects Analysis

(FMEA) 52–53

cryostorage packaging example 164

diagrammatic overview 120

flow chart 121

steps 122, 123, 120–124

uses 120

FertAid 117

fertilization rates, troubleshooting

example 138–141

financial indicators 174

FIPS PUB 183 67–68

fire, risk management example 149,

148–150

flow charts 63, 64

FMEA see Failure Modes and Effects

Analysis

forms 114

frozen embryo disposal 155–157

governance of organizations,

accreditation of 20

“hand-offs” or “hand-overs” 80

hazardous materials, regulations for use

11

honesty in reporting results 175–176

human error 125

human resources 201–209

accreditation 20

basic principles 166

education 36–37, 76–77, 202–203

embryologists 201–202

leadership 35–36

motivation, external 205

motivation, internal 206

protecting staff from litigation 166

retention of staff 204

career path 206, 207, 208, 209

delegation 206–208

rewards 204–206

other factors 208–209

salaries 204

staffing issues in risk analysis 48–49

staff involvement and commitment

37–38

staff motivation 204–205

teamwork 38–39, 201

training 36–37, 47, 76–77, 202–203

ICSI, top–down process map 65, 66

IDEF0 63, 68, 67–68, 132–133

implantation rates 172

incident reports 51

indicators used in benchmarking see

benchmarking: indicators

information management accreditation

20

information technology accreditation 20

inspection 9, 118–119

internal audit 119

ISO standards

development principles 14–15

ISO 9000 family 15–16

for laboratories 16–17

IVF centers, structure and organization 3,

4, 2–4, 5, 207, 208, 209, 212

IVF Chamber 186, 188

JCAHO 14

Joint Commission on Accreditation of

Healthcare Organizations

(JCAHO) (USA) 14

Juran, Joseph M. 25

Kemeter, Peter 159–161, 160, 162–163

K-Systems cabinets 186, 188

labeling in IVF laboratories 78, 77–79, 82,

157

228 Index

laboratories

“high risk” 48–52

ISO standards 16–17

“out of control” 6

well-run see well-run laboratories

laboratory operations indicators 174

laboratory performance indicators (LPIs)

104–105, 172–173

laboratory services, accreditation of 21

labor relations 11

laminar flow cabinets, vertical 188

legal aspects 33

licensing process

defined 8–9

IVF 10–11

likelihood of risks 122

litigation, protection of staff from

166–167

“market-in” company 212

measurement, uncertainty of 109,

107–109

MedCalc software 102

meteor strikes 52

methods

design 86–87, 88, 106, 179

implementation and validation 106

calibrators 107

need for accuracy 111–112

reference materials 107

statistical issues 108, 110, 109–111

uncertainty of measurement 109,

107–109

selection 179

microscopy, temperature control 185, 186

MINC 1000 incubator 185, 186, 190, 193

mouse embryo assay (MEA), reliability

196

NATA 13

National Association of Testing

Authorities (NATA) (Australia) 13

New Zealand, IVF accreditation 13

oil see culture oil, factors influencing

use

oocyte retrieval 160, 158–161, 162

organization chart

generic company hierachy 207

ideal IVF lab

defined roles 209

generic 208

IVF Center 3

large 3

“market-in” 212

“product-out” 212

small 4

IVF lab, large 5

Pareto methodology 123

PDCA 124, 126

Percoll 46

Peter principle 208

Plan–Do–Check–Act (PDCA) cycle 124,

126

power failure 50

pregnancy rates 172

problems

conceptual approaches to 136

dealing with 135

reactive approach 135–136

see also troubleshooting

process analysis 74, 75–76

process control 74–75, 101

charts

generic process 101, 101–102

laboratory performance monitoring

example 104, 104

recalculation of control limits 102,

103, 103–104

scenarios requiring action 102–103

software 102

processes

as components of systems 55, 54–55

defined 54

development for individual centers 23

generic 55

229 Index

process management software 83

process mapping 55–72

benefits 68–69, 84–85

“dinner party” illustration 56–57

intrinsic and extrinsic factors 56

IVF laboratory systems 58, 59, 60, 61,

62, 80

map creation

“bottom–up” approach 70

as the process is performed 69

recommended procedure 70–72

“top–down” approach 69–70

principle 55–56

tools 59

approaches compared 63

flow charts 63, 64

IDEF0 63, 68, 67–68

swim lanes 65, 67

top–down maps 64, 65, 66

“product–out” company 212

professional responsibility 49, 81, 151

quality

in action, example 27–29

assurance (QA) 26, 29

defined 26

external 117

concepts of 24–25

control (QC) 26, 29

defined 26

cycle 26, 29, 30

defined 26

defined 24–25

improvement (QI)

continual 29–31

defined 26

management

defined 25–26

history 25

tools 118

see also specific management tools;

Total Quality Management

(TQM)

management systems (QMS)

defined 27

manager 27

manual 27

in medicine 25–26

objective 27

ongoing commitment to 22–23

planning 27

policies 27

systems 27

terminology 26–27

total 31

radio frequency identification devices

(RFIDs) 82, 83

random errors 110

RCA see Root Cause Analysis (RCA)

reactive oxygen species in sperm

preparation 181

reference materials 107

reference populations 174–175

registration 8, 9

regulation of IVF 10–12, 33

regulations, defined 8–10

Reproductive Technology Accreditation

Committee (RTAC) (Australia)

13

resistance

active 41

passive 41

to change 40–41, 41

resource issues 49–50

responsibility see professional

responsibility

RFIDs 82, 83

risk 45

acceptance/retention of risk 147

avoidance of risk 146

consequence rating 123

elimination of risk 146

likelihood rating 122

reduction/minimization of risk

146–147, 150

230 Index

risk analysis of IVF laboratories 48

organizational issues 50–51

resource issues 49–50

risk management issues 51–52

staffing issues 48–49

risk compensation behavior 81

risk management 145–168

basics 46

benefits 46–47, 151–152

consequences of non-implementation

47

disaster planning 52

examples 153

cryostorage packaging 164,

161–166

fire, multi-faceted approach 149,

148–150

frozen embryo disposal 155–157

labeling of OPU tubes 157

off-site sperm collection 154–155,

156

Percoll use 46

sperm processing in parallel 157,

158

temperature control during oocyte

retrieval 160, 158–161, 162

importance for IVF centers 45

professional responsibility 151

program development 152–153, 154,

155

protection of staff from litigation

166–167

standard for 46

summarized 46

tools 52–53, 118

see also specific tools

transfer of risk 147

risk matrices 123

risk registers 153, 155

risk reduction 150

Root Cause Analysis (RCA) 52–53,

124–125, 127

flow chart 126

implications of 166

sperm preparation example 129, 133,

128–134

performing 127, 124–128

see also troubleshooting

RTAC 13

sample identification/verification

bar codes 82, 81–83

double-witnessing 50, 79–81

labeling 78, 77–79, 157

radio frequency identification devices

82, 83

system failure 79, 80

sampling errors 109, 110

self-assessment in accreditation 12–13,

18–21

semen analysis 180–181

sentinel indicators 170

service contracts 90, 87–90, 91

Shewhart (PDCA) cycle 124, 126

silos 73, 72–74

slack 49, 211

SmartDraw 64

SOPs see standard operating procedures

(SOPs)

sperm collection off-site 154–155,

156

sperm preparation

example SOPs 95, 96

in parallel 157, 158

reactive oxygen species in 181

troubleshooting using RCA 129, 133,

128–134

staff issues see human resources

standard operating procedures (SOPs)

90–100

considerations when writing

75–76

deviation from 50–51, 203

electronic manuals 91

231 Index

examples of good and poor versions

95, 96, 94–100

guidelines/requirements 90–91,

92

inadequate 95

review and reissue 113, 114

summary protocols 93

value of 93–94

see also top–down process maps

standards

accreditation see accreditation:

standards

defined 9–10

IDEF0 67–68

risk management 46, 154, 155

SOP requirements 90–91, 92

see also ISO standards

surveys for accreditation

process 10, 13, 21

recommendations 21–22

swim lane process maps 63, 65, 67

systematic errors 110

systems

defined 54

importance of understanding 84

systems analysis 55, 54–55

see also process mapping

systems specification 178–200

general principles 178

practical examples 184–200

CO2 incubator, choice of 185, 186,

190, 191, 193–199

cryopreservation buffers 183

culture media, choice of 190–197

work station, choice of 188, 184–190

culture oil, use of 197–200

selection of methods or equipment

179–180


temperature control, oocyte retrieval 160,

158–161, 162

third-party services 90, 87–90, 88–89,

91

Tomcat catheters 51, 182, 181–182

top–down process maps 63, 64, 65, 66

Total Quality Management (TQM) 31

defined 27

diagrammatic overview 32, 31–32

implementation 34–40

education and training 36–37,

76–77, 202–203

employee involvement and

commitment 37–38

importance of organization 34

leadership 35–36

measurement and feedback 39–40

requirements summarized 34–35


tools and techniques 37

in IVF

customer expectations 33

duty of care 32

ethics 32–33

legal obligations 33

liability 33–34

medical and scientific standards 32,

31–32

responsibility 32

ongoing process 40

origin of concept 25

reasons for failure 40

resistance to change 40–41

“toxic workplace” 42–44

terminology 26–27

“toxic workplace” 42–44, 211

training of staff 36–37, 49, 76–77,

202–203

troubleshooting 135–144

being reactive 135

CO2 incubator incident 141–144

conceptual approaches to problems

136

fertilization rates example 138–141

232 Index

troubleshooting (cont.)

flow diagram of process 137

sperm preparation example 129, 133,

128–134

see also Root Cause Analysis (RCA)

UK, accreditation schemes 14

uncertainty of measurement 109,

107–109

USA, accreditation schemes 14

user requirement specifications (URS)

86–87, 88

vertical laminar flow cabinets 188

warmed microscope stages 185, 186

well-run laboratories 210–217

benefits 216–217

Lab Director’s role 215

organization charts 212

requirements 210–213

road map to establishment 213,

214–215

scientists’ roles 215

WHMIS 11

Workplace Hazardous Materials

Information System (WHMIS) 11

work stations see embryology work

stations

Quality and Risk Management in the IVF Laboratory

Documents

Transcript of Quality and Risk Management in the IVF Laboratory