Safe Staffing & Patient Safety Literature Review

Please note that the following document was created by the former Australian Council for Safety and Quality in Health Care. The former Council ceased its activities on 31 December 2005 and the Australian Commission for Safety and Quality in Health Care assumed responsibility for many of the former Council’s documents and initiatives. Therefore contact details for the former Council listed within the attached document are no longer valid. The Australian Commission on Safety and Quality in Health Care can be contacted through its website at http://www.safetyandquality.gov.au/ or by email [email protected] Note that the following document is copyright, details of which are provided on the next page.

mailto:[email protected]

The Australian Commission for Safety and Quality in Health Care was established in January 2006. It does not print, nor make available printed copies of, former Council publications. It does, however, encourage not for profit reproduction of former Council documents available on its website. Apart from not for profit reproduction, and any other use as permitted under the Copyright Act 1968, no part of former Council documents may be reproduced by any process without prior written permission from the Commonwealth available from the Department of Communications, Information Technology and the Arts. Requests and enquiries concerning reproduction and rights should be addressed to the Commonwealth Copyright Administration, Intellectual Copyright Branch, Department of Communications, Information Technology and the Arts, GPO Box 2154, Canberra ACT 2601 or posted at http://www.dcita.gov.au/cca

SAFE STAFFINGAND PATIENT SAFETYLITERATURE REVIEW

Final Report – 31 January 2003

Prepared by: The Australian Resource Centre for Hospital InnovationsFor: The Australian Council for Quality and Safety in Health Care

Consultant: Dr Maggie Haertsch, Director Systabytes P/LProject Manager: Charmaine Weeks, Executive Director, ARCHI

Email: [email protected]

2

© Commonwealth of Australia, 2003

DisclaimerNo representation or warranty, expressed or implied, is made as to the relevance, accuracy,completeness or fitness for purpose of this document in respect to any particular user’scircumstances. Any person who uses this document should satisfy themselves concerning itsapplication to, and where necessary, seek expert advice about this situation. The CommonwealthDepartment of Health and Ageing and the Australian Resource Centre for Hospital Innovationsand its contractors shall not be liable to any person or entity with respect to any liability, loss ordamage caused or alleged to have been caused directly or indirectly by this publication. Theviews expressed in this document do not necessarily represent the views of the CommonwealthDepartment of Health and Ageing, the Australian Resource Centre for Hospital Innovations or itscontractors.

AcknowledgementsThanks for the following people and organisations who have provided information and contactsfor this project:Naida C Hutton, Professor Christine Duffield, Karen Court, Professor Kathleen Fahy, BruceGreetham, Thomas Stoddart, Sue Behan, Carol Mirco, Cathie O’Neill, Robin Cassumbhoy, JennyHunt, Sally Lowe, Helen Myers, William Tarnow-Mordi, Siobhan Rowe, Eleanor Jackson Bowers,Dr Phillip Hoyle, Rhonda Topp, Michael Rodgers, Andrew Schultz, Patricia Burritt.

The assistance of Ms Lisa Fisher, ARCHI Information Manager is gratefullyacknowledged.

3

CONTENTS

1. Executive Summary 5

2. Overview of the project 6

3. Method 6

3.1. Search terms used for the health industry literature 6

3.2. Search terms used for industry not related to health 7

3.3. Inclusion criteria 7

3.4. Exclusion criteria 7

3.5. Process for culling the literature 8

3.6. Search strategy 8

3.7. Process for examining and summarising the findings 9

4. Results 12

4.1. Overview of the findings 12

4.2. Staff physical and mental health 15

4.2.1. Summary findings 15

4.2.2. Staff physical and mental health - health literature summary tables 17

4.2.3. Staff physical and mental health – non-health literature summary tables 20

4.2.4. Recommendations 23

4.3. Communication and feedback 24


4.3.2. Communication and feedback – health literature summary tables 26

4.3.3. Communication and feedback – non-health literature and summary tables 34


4

4.4. Hours of work, shift work, numbers of staff and fatigue 41


4.4.2. Hours of work, shift work, numbers of staff and fatigue - health literature summary tables 43

4.4.3. Hours of work, shift work, numbers of staff and fatigue- non-health literature summary tables 57


4.5. Competency, supervision and staff mix 70


4.5.2. Competency, supervision and staff mix – health literature summary tables 73

4.5.3. Competency, supervision and staff mix – non-health literature summary tables 104


5. Recommendations 114

6. Appendices 116

6.1. List of web sites used in the search 116

7. References 123

5

1. EXECUTIVE SUMMARY

The Australian Resource Centre for Hospital Innovations (ARCHI) has beencontracted by the Commonwealth Department of Health and Ageing on behalf ofthe Australian Council for Quality and Safety in Health Care to undertake acomprehensive review of published and unpublished literature on safe staffingand patient safety.

The literature review aims to identify staffing factors that are associated withpatient safety or the effectiveness of safety cultures within non-health industries.It also aims to identify the quality of evidence and gaps in research.

This comprehensive review includes 230 papers from both health (152 papers)and non-health literature (78 papers) that are case studies, literature reviews,research studies or guideline documents containing data. The papers weregrouped by the domain areas of:

• Staff physical and mental health• Communication and feedback• Hours of work, shift work, numbers of staff and fatigue• Competency, supervision and staffing mix.

The results indicate clear gaps in the health literature relating to staffing factorsand patient safety, and in particular, a dearth of well-designed interventionstudies.

There is a need to increase understanding of effective strategies that reducefatigue and fatigue related errors, the contributing factors that lead to poorcommunication in the workplace, physical and mental health of staff and theirability to provide safe patient care and systems that prevent rule-based and skill-based errors in health care delivery. The non-health literature reviewed hereoffers some insight for the development of these strategies and has beenincorporated into the recommendations for further research.

6

2. OVERVIEW OF THE PROJECT

The Australian Resource Centre for Hospital Innovations (ARCHI) has beencontracted by the Commonwealth Department of Health and Ageing on behalf ofthe Australian Council for Quality and Safety in Health Care to undertake acomprehensive review of published and unpublished literature on safe staffingand patient safety. The literature review aims to identify staffing factors that wereassociated with patient safety or the effectiveness of safety cultures within non-health industries. The review also aims to identify the quality of evidence andgaps in research.

Staffing factors include staff competency, communication, supervision, fatigue,rostering and working hours. Patient safety includes the variables that limit oraffect preventable adverse patient outcomes and errors.

3. METHOD

3.1 Search term used for the health industry literature

The search terms for health related literature were based on the standardmedical subject headings (MeSH) and include any or all of the following terms:

• Manpower, doctors, medical staff, nursing staff, health personnel, staffing,personnel staffing, scheduling, personnel administration, workload#

• Nursing staff, hospital/supply, distribution• Quality assurance• Hospital safety management• Patient safety, patient incidents, patient adverse health effects, clinical

risk, quality management, quality care, risk management, quality ofhealthcare, quality indicators, outcome assessment/methods,hospital/standards, accident prevention, safety management, adverseevents, adverse drug reactions

• Staff* rostering, staff fatigue, staff supervision, work place harassment,staff dissatisfaction, clinical governance, staff competency.

# The term ‘out-of-hours’ was not searchable in the majority of databases used for this review. Asa general rule, the term ‘of’ is not searchable in a database as it is considered a ‘stop’ word thatoccurs too frequently to make search results relevant. ‘Out-of-hours’ is found in the searchingrelated to workload, scheduling and working hours.

* The term “staff” will include doctors, nurses, midwives, managers, surgeons, anaesthetists,physicians, allied health professionals.

7

3.2 Search terms used for non-health industries

Search terms for other literature not related to the health industry will includeany or all of these terms:• Workers safety, customer/ consumer/ client/ passenger incidents, safety

risk, quality management, customer service• Staff* rostering, staff fatigue, staff supervision, work place harassment,

staff dissatisfaction, clinical governance or staff competency.

*The term “staff” will also include the specific title of the industry workers such as pilot, airlinesteward, bus driver, truck driver, police officer etc.

3.3 Inclusion criteriaAll literature to be included in the review had the following characteristics. Thepaper was:

• Published not earlier than 1992• Written in English• Related to the search terms• Based on either qualitative or quantitative data• A review, research paper or report, guideline, or case study that describes

the relationship between staffing variables and patient or customeroutcomes

3.4 Exclusion criteria

Papers are not included if they do not provide data that links at least one staffingvariable with at least one aspect of patient care outcomes or as in the case ofnon-health industry, at least one aspect of industry safety.

Literature that relates to occupational health and safety of staff working in thehealth industry is not included unless it contains data on adverse patient events.For example, violence directed at staff from patients is not included. However,papers describing violence from staff towards patients is included because theadverse event was focused on the patient. Papers that examine system typefailures and patient safety, equipment and environmental factors and patientsafety without a human factor or staffing variable included do not meet theinclusion criteria.

8

3.5 Process for culling the literature

The following questions were asked for each citation (where there wasinsufficient information in the citation, the paper was located):

3.6 Search strategy

Published health literatureExtensive searching has been undertaken in the following electronic databases:• ACP Journal Club (ACP)• Cochrane Controlled Trials Register (CCTR)• Cochrane Database of Systematic Reviews (COCH)• Database of Abstracts of Reviews of Effectiveness (DARE)• MEDLINE• Pre-MEDLINE• PsycINFO,• CINAHL• EMBASE• HealthStar• OSH-ROM

Unpublished health literatureThe following strategies were used to access unpublished health literature:• Posting a request on several pages of the ARCHI website

Is the paper related to health or another industry other than health?

For health related papers:

1. Does the paper explore staffing factors (such as competence, skill, stress, fatigue,rostering, workload, supervision) and this association with patient safety (such aspatient injury, harm, outcome, preventable adverse events)?

2. Is the paper a review containing data, a case study, a research study or guidelinedocument with supporting data?

For non-health related papers:

1. Does the paper examine the relationship between human factors and workperformance that can impact on safety?

2. Is the paper a review containing data, a case study, research study or guidelinedocument with supporting data?

Paper was included when the answer was “yes” for each point.

9

• A teleconference with the ARCHI state-based satellite officers to identify “grayliterature” (6 officers)

• Teleconferences of the Queensland, Victoria, Tasmania, South Australia andNew South Wales ARCHI Clinician’s Advisory Committees

• ARCHI change management e-mail discussion group (92 members)• ARCHI bed management e-mail discussion group (106 members)• Postings on ARCHI Net News (approximately 3200 subscribers)• Emails to National Rural Health Alliance, Health Canada, American Academy

of Family Physicians, Agency for Healthcare Research and Quality, AustralianIndustrial Relations Commission, National Occupational Health and SafetyCommission, Civil Aviation Safety Authority, NSW Safe Hours Group

• Individual discussions with leaders in the area of clinical governance andpatient safety including the Health Care Complaint's Commission, theAustralian Patient Safety Foundation and the Institute for ClinicalEffectiveness.

Non-health related literatureA search of electronic databases included the following:• ABI Inform• ANRO - Australian Agriculture & Natural Resources Online• CCOHS - Canadian Centre for Occupational Health & Safety• AGIS• WORKLIT• IREL - Industrial Relations, Australian Transport Index• ELIXER - Natural Resources• STREAMLINE - Natural Resources• ANSTI - Nuclear Science• BUILD - Building & Construction• AUSTROM• Business Australia• AGSM

3.7 Process for examining and summarising the findings

Endnote libraries and databases were established for tracking the searches andcataloguing of the "gray literature".

The papers in the review were classified using the hierarchy of study designslisted in table 1. This classification is helpful in summarising the study design inrelation to quality of evidence and was used only for research papers. A rating ofthe evidence was made based on the quality of evidence rating seen in table 2and applied to each paper. This rating is useful in determining the strength ofevidence and was applied to all of the types of papers used in the reviewincluding case studies.

10

The literature is categorised into summary tables and is divided into two maingroups, health related literature and the non-health related literature. Thedocuments are reported using the following headings:

• Author, year, country• Type of document (published or unpublished, study, guideline or case

study)• Rating using a level of 1 - 8 using the hierarchy of study designs

This was applied to research papers only and categorises the studydesign by ranking the design. Studies that minimise bias such asrandomized controlled trials are ranked more highly than designs that aresubject to bias and few controls such as descriptive studies and expertopinion.

• Rating the level of evidence 1 - 4.• Industry type (for non-health related literature)• Staffing variables examined and related patient outcomes (staffing

variables examined and factors contributing to a safety culture are usedfor the non-health related literature)

• Study design and summary description• Findings and comments

Four major domain areas relating to staffing variables have been identifiedacross both the health and non-health related documents. These are:

• Staff physical and mental healtho Health literature included: staff with infectious diseases and cross

infection to patients; and, mental health and appropriate clinicaldecision-making.

o Non-health literature included: staff anxiety and threats to safetypractices; and, impact of airline pilots with heart disease andpotential for an airline crash.

• Communication and feedbacko Health literature included: staff errors in communication leading to

adverse events; and, telephone orders leading to inappropriatetreatment.

o Non-health literature included: communication between staff at shifthand-over and mining disasters; and, staff behaviour in relation tosafety reporting and feedback.

• Hours of work, shift work, numbers of staff and fatigueo Health literature included: medical practice and fatigue and the

impact on patient injury; and, nurse to patient ratios and its effectson mortality and morbidity.

o Non-health literature included: fatigue prevention initiatives for airforce pilots; and, the impact of extended working days in an under-ground mine.

• Competency, supervision and staffing mixo Health literature included: hospital reviews and the staffing factors

that were associated with patient safety; and, child maltreatment byhospital staff.

11

o Non-health literature included: motor vehicle injury among dutyarmy officers; and, accidents among airport ground personnel.

Many of these staffing variables were inter-related however the variable that wasthe primary focus of the paper was used to determine the domain area.

TABLE 1 Hierarchy of research designs in decreasing level ofimportance1

Classification ofthe study

design

Type of research design

12345678

Randomised controlled trialsNon-randomised controlled trialsCohort studiesCase-control studiesComparisons between time and placeUncontrolled experimentsDescriptive studiesExpert opinions

TABLE 2 Quality of evidence ratings2

Levels Controlled trials Prevalence, risk factors andsensitivity studies

I Evidence obtained from a systematicreview of all randomised controlledtrials

Evidence from a systematic review ofall available population-based studies

2 Evidence obtained from at least oneproperly-designed randomisedcontrolled trial

Evidence obtained from a well-designed population-based studyrepresentative cohort study

3 – 1 Evidence obtained from well-designedcontrolled trials without randomisation

3 – 2 Evidence obtained from well-designedcohort or analytic studies, preferablyfrom more than one centre or researchgroup

3 – 3 Evidence obtained from multiple timeseries with or without the intervention

Evidence obtained from a well-designed case control study, cohortstudy or less well-designed population-based study

4 Opinions of respected authorities,based on clinical experience,descriptive studies, or reports of expertcommittees

Evidence obtained from a descriptivecase series, clinical experiences,respected authorities, or reports ofexpert committees

1 Report of the US Preventive Services Task Force. Guide to clinical preventive services.Baltimore, USA, Williams and Wilkins, 1996.2 Quality of Care and Health Outcomes Committee, National Health and Medical ResearchCouncil. Guidelines for the development and implementation of clinical practice guidelines.Canberra. AGPS, 1995.

12

4. RESULTS

4.1 Overview of the findings

A total of 20 people have provided information including reports, reference lists,web sites, papers and contact details for other people to contact.

Contact was made with aviation, transport, mining and forestry industries and atotal of 77 web sites have been reviewed for relevant documents.

A total of 10 755 citations were generated using the search terms under theMeSH terms and criteria described in the search strategy. Of these, 859 paperswere examined for suitability to be included in the literature review. The reviewincludes 230 papers that are a case study, a literature review, a research studyor a guideline document containing data. Almost all of the papers are publishedin a journal, as a report or book chapter, or are available on-line. Health relatedliterature accounts for 152 (66%) documents and non-health literature totals 78(34%) documents.

Table 3 provides a description of the domain areas by main grouping.

TABLE 3 Papers included in the review by domain area

Domain area Health relateddocumentsn=152 (%)

Non-health relateddocuments

n=78 (%)

Totaln=230

(%)Staff physical and mental health 7 (5) 7 (9) 14 (6)Communication and feedback 18 (12) 13 (17) 31 (13)Hours of work, shift work, numbers ofstaff and fatigue

36 (23) 31 (39) 67 (29)

Competency, supervision andstaffing mix

91 (60) 27 (35) 118 (52)

Table 4 provides a summary of the type of paper for each domain area. Thesenumbers include health and non-health literature.

13

TABLE 4 Summary of the types of papers in the review by domain area

Staff physical andmental health

N=14

Communicationand feedback

N=31

Hours of work,shift work,

numbers of staffand fatigue

N=67

Competency,supervision and

staffing mixN=119

TotalN=230

Health Non-Health

Health Non-Health

Health Non-Health

Health Non-HealthPublished papers

7 7 18 13 35 30 91 27 227Unpublished papers 0 0 0 0 1 1 1 0 3

LiteratureReview

0 0 1 0 1 2 6 0 10Guidelines 2 1 0 0 1 1 4 2 11CaseStudies

3 1 2 4 3 3 30 13 59

Type ofpaper

ResearchStudy

2 5 15 9 31 25 51 12 150

Hierarchy of study designs and quality of evidence

The quality of evidence available on any aspect of safe staffing and patient safetyis very limited. The broad search strategy was able to capture papers that maynot have been readily available through a search confined to the key electronichealth literature databases. The absence of quality research papers on this topicdemonstrates that well designed studies are needed. There are few researchstudies and of these a small selection are a randomised controlled trial(approximately two papers) and provide a limited contribution to understandingstaffing and patient outcomes as in both cases these were not the primary end-points in the study. There are no meta-analytical studies.

TABLE 5 Summary of the quality of evidence of the papers in the review

Staff physical andmental health

N=14

Communicationand feedback

N=31

Hours of work,shift work,

numbers of staffand fatigue

N=67

Competency,supervision and

staffing mixN=119

Health Non-Health

Health Non-Health

Health Non-Health

Health Non-Health

TotalN=231

1 0 0 0 0 0 0 0 0 02 0 0 1 0 3 0 4 0 83 0 3 3 5 14 13 21 4 63

Levels ofEvidence

4 7 4 14 8 19 18 66 23 1591 0 0 1 0 1 0 3 0 52 0 0 0 0 2 0 1 1 43 0 0 0 1 2 0 5 0 84 0 0 0 0 1 2 0 1 45 1 1 2 2 9 10 13 3 416 0 0 1 0 1 1 5 0 87 1 4 11 6 15 12 24 7 80

Hierarchyof StudyDesign

(Applicableto research

studiesonly)

8 Not included for studies using original data

14

The majority of the research studies are cross-sectional descriptive surveys.Very few use an intervention / experimental design with or without randomisation.There was one evidence based literature review included in the domain hours ofwork, shift work and number of staff. This review Making health care safer: Acritical analysis of patient safety practice3 included only level 1 – 3 evidenceusing the levels outlined in table 2. This review was not focused on safe staffingvariables, however, it contains sections on “nurses staffing and staffing models”and “fatigue, sleepiness and medical errors”.

The authors identify that the methodological issues in combining the studies areproblematic and there still remains conflicting outcomes with no clearunderstanding if certain nurse staffing models and number of staff improve orreduce patient safety. The same applied to the number of hours worked andfatigue experienced by medical practitioners and the effects on performance.

There were only five randomized controlled trials located in this review, one inthe area of communication and feedback, one in the area of hours of work, shiftwork, numbers of staff and fatigue and three in the area of competency,supervision and staffing mix.

3 Shojania KG et al (eds) Making health care safer: a critical analysis of patient safety practices.Evidence Report / Technology Assessment No. 43 (Prepared by the University of California atSan Francisco – Stanford Evidence-based Practice Centre) Agency for Healthcare Research andQuality, July 2001.

15

4.2 Staff physical and mental health

4.2.1 Staff physical and mental health – summary findings

There were seven health and seven non-health papers included in this reviewthat examined the relationship between staff and safety. The following tablesprovide a description of the types of papers.

Papers for staff physical and mental health by type of paper andclassification

Health Non-Health

TOTAL

Published papers 7 7 14Unpublished papers 0 0 0

Literature Review 0 0 0Guidelines 2 1 3Case Studies 3 1 4

Type of paper

Research Study 2 5 71 0 0 02 0 0 03 0 3 3

Levels of Evidence

4 7 4 111 0 0 02 0 0 03 0 0 04 0 0 05 1 1 26 0 0 07 1 4 5

Hierarchy of StudyDesign

(Applicable to researchstudies only)

8 0 0 0

Papers relating to health practitioners focused on physical stamina and effectivecardiac compressions during a simulated cardio-pulmonary resuscitationexercise [1], health professionals who have HIV infection and the potential toinfect patients [2] and the association of staff anxiety and patient incidents [3].The conclusion of these papers, whilst seen in the context of their designlimitations, identify the following themes that merit further examination:

• Self-perceptions of physical stamina may interfere with effectiveness ofcarrying out physically demanding procedures such as cardiaccompressions before being aware of the fact

• Staff that are more anxious may increase the likelihood of adverse events• Hospitals and health service organisations that employ practitioners who

have HIV or other infections and restrict their practice may face additionalissues relating to discrimination and human rights violations when theevidence is not clear that there is any increased risk to patient safety.

16

Papers relating to staff working in other non-health related industries provided aslightly increased number of studies although all but one are descriptive.Consistent with the health related study examining nurse stress and the increaselikelihood of incidents, a study examining staff worry and safety concerns on anoff-shore mining platform found that increase in job satisfaction, appropriateworkload and a range of other measures were associated with a reduction in riskperception and risk behaviour [4]. The non-health literature identified thefollowing themes that may be useful in examining further:

• The relationship between staff anxiety and workplace morale in improvingperformance and reducing adverse events

• Identification of the early signs of dangerous practice by considering staffmental health, physical health and the assessment of their fitness topractice.

4.2.2 Staff Physical and Mental Health – Health Literature Summary Tables

Author / Year /Country

Type ofdocument*

Staffing variables (SV) examined and patient outcomes(PO)

Study design / Summary description Comments / Findings

* P = published or U = unpublished. L = literature review, S = study containing data, G = guideline or CS = case study. Hierarchy of study design rating = Level 1 -4 (See table 2) COSD = classification of study design 1 - 8 (see table 1)**SS = sample size; RR = response rate; DS = data source

17

Poole et al /2002 / UK [5]

P, GLevel 4

SV: Staff fitness to practicePO: Abuse to paediatric patients, infection topatients, incompatibility with safe practice

This is a guideline documentthat describes standards ofhealth for clinical health careworkers, their suitability forcertain positions and the effectsof their health status on theirperformance and clinicaljudgment.Citations: 52.

The paper covers 15 areasrelating to fitness to practiceincluding: depression,substance misuse, anorexiaand bulimia, personalitydisorders, anxiety and visualimpairment.

Ochoa et al /1998 / Spain[1]

P, SLevel 4COSD 5

SV: Staff fatigue whilst doing chest compressionsduring CPR practicePO: Resuscitation

The clinicians were assessedby ineffective chestcompressions with on-set offatigue.SS: 38RR: n/aDS: Observed cardiaccompressions using doctorsand nurses on a manikin

Rescuer fatigue occurredbefore 60 chest compressions.After this point the chestcompressions were observedas being ineffective. Rescuersdid not recognize their fatigueuntil almost 2 minutes later.Efficacy of the chestcompressions did not differbetween clinical groups.

AIDS/TBCommittee ofthe SocietyforHealthcareEpidemiologyof America /1998 / USA[6]

P, GLevel 4

SV: HIV, hepatitis B or C infected health careprofessionsPO: Infection from provider

This guideline statement usesavailable evidence of providerto patient infection rates as abasis for recommendingadoption of these guidelines tolimit litigation and clarify patientrisk.Citations: 111

The transmission rate of HIV,hepatitis B or C is extremelylow. The guidelinesrecommend that practitionersexercise usual universalprecautions that infectedpractitioners have their privacyrespected, no specialmonitoring is required andinfection status does notconstitute a basis for barringthe provider from any patientcare activity including invasiveprocedures. There were 38instances of provider to patientHepatitis B transmission from1972 – 1994.



Type ofdocument*




18

Dugan et al /1996 / USA[3]

P, SLevel 4COSD 7

SV: The effect of stress on nursingPO: Improved quality of care

The article describes a studythat measured through aquestionnaire; staffingproblems, (turnover rates),nurse incidents, (absenteeism,back injuries, needle sticks)and patient incidents(medication errors, falls) over a3 month period and levels ofself-reported stress by nurses.SS:293RR:49%DS: Registered and LicensedPractical Nurses self-report inone hospital

Results showed that the higherthe reported levels of nursestress (Stress ContinuumScale) the greater likelihood ofpatient incidents (0.59correlation).

Reason /1995 / UK [7]

P, CSLevel 4

SV: Recognition, memory, attention and selectionfactorsPO: Morbidity and death

This is a case that describes 2adverse events to illustrate atheoretical model ofcategorizing human error. Onecase study describes a patienthaving repeated doses ofradiation leading to prematuredeath and the seconddescribes the dislodgment of aniridium source wire causing 90people to be irradiated.

The conceptual frameworkcategorizes human causes ofan adverse event as a violationor an error, active or latenthuman failure. Errors aredefined as mistakes orexecution errors. Case studiesare used to illustrate Reason’stheoretical model.

Shuster /1993 / USA[2]

P, CSLevel 4COSD 7

SV: The role of infected physicians and healthprofessionalsPO: Patient safety from infectious disease

This single case study looks atthe issues surrounding aphysician with acquiredimmune deficiency syndrome(AIDS) and the potential forphysician to patient infection.

Legal issues raised includeinformed consent anddisclosure of identifiable risk.The paper concludes thatmore reasonable approachesneed to be considered ratherthan continuing the guidelinesestablished in 1991 by theCentres for Disease Controlthat called for infected



Type ofdocument*




19

physicians to cease performinginvasive procedures or informtheir patients. It was decidedthat each state of the USA is toset its own rules in complianceto the Center for DiseaseControls guidelines.

Hudson /1992 / USA[8]

P, CSLevel 4COSD

SV: HIV Positive Health care workersPO: Decreasing patient exposure to infectiousdisease

This single case study looks atthe issues surrounding apharmacist with acquiredimmune deficiency syndrome(AIDS). The pharmacist wasbarred from preparing IVsolutions because of his HIVstatus.

Issues raised in this hospitalbased case study include legalissues, reaching compliancewith the Center for DiseaseControl (CDC), cost issues,fitness for duty, traininglogistics and exposuremanagement. Theimplementation of the CDCguidelines provided a legalconflict between the employerand employee on the basis ofdiscrimination.

4.2.3 Staff Physical and Mental Health - Non- Health Literature Summary Tables


Type ofdocument*


Study design / Summarydescription**

Comments / Findings


20

Joy / 1999 /UK [9]

P, GLevel 3 -3

AviationSV: Cardio-vascular disease in professional flightcrewSC: minimizing plane disasters

This paper synthesizes theconclusions of an internationalconference on aviationcardiology withrecommendations forcardiology assessment andappropriate standards for pilotlicensing.

Human factors are responsiblefor approximately 80% ofaircraft accidents. The papercites that of the 1000 fatalaccidents in a 40-year period,half occurred in private orcommercial aircraft, 4.7% weredue to medical causes ofwhich about 2.35% were dueto a cardiovascular incident.The implications of an agingpilot are discussed.

Rundmo et al/ 1998 /Norway [10]

P, SLevel 3-2COSD 5

Offshore miningSV: Job stress and physical working conditionsand commitment and involvement in safety workSC: Improvement of safety attitudes and accidentprevention

Comparative analysis twocross-sectional surveysundertaken in 1990 and in1994.SS: 9 companies, 12 offshoreplatforms, 912 employeesRR: 92%DS: Off-shore personnel

Improvements were found injob satisfaction, physicalworkload, safety andcontingency measures. Thesefactors affected risk perceptionand risk behaviour.

Edkins et al /1997 /Australia [11]

P, SLevel 3-2COSD 7

Railway transportSV: MoraleSC: Reduction of rail accidents

A retrospective analysis ofrailway accidents over a 3-yearperiod and cross-sectionalsurvey of train driversexamining the extent to whichRailway Problem Factorsinterfere with doing their jobSS: 112 accidents and 190 traindriversRR: n/a for accidents and 42%for train driversDS: Incident reports of trainaccidents and survey of traindrivers.

Sustained attention to railwaysignals was the majorcontributing human factoracross all types of railwayaccidents, particularlyinattentiveness to signals.Train drivers identified lowmorale was the most seriousproblem in safe conduct oftheir job.



Type ofdocument*



Comments / Findings


21

Raymond etal / 1995 /USA [12]

P, CSLevel 4

AviationSV: Emotional stressSC: Impact on pilot performance

Three case studies all resultingin fatal aircraft accidents.Explanations suggest that theimpact of emotional stressresulting from family problemsand other social stress, careerinstability, worry, aircraftaccidents and difficult flightschedules as a contributingfactor to pilot performance.

The paper suggests that theaviator "at risk" may exhibitwarning signs such asdefensiveness, arrogance,hostility, financialirresponsibility, excess inroutine habits, fatigue,deteriorating performance, orincreased risk taking.

Haugli et al /1994 /Norway [13]

P, SLevel 4COSD 7

AviationSV: Heath, sleep and mood perceptionsSC: Health promotion

This prospective questionnaireforms part of a major survey,which looks at the health ofcockpit and cabin crews.SS: 1240RR: 83%DS: Cock-pit and cabin crew.

Common problems reported bymore than 30% included dryskin, lower back pain, colds,fatigue, and sleepdisturbances. Pilots reportleast, while female cabinattendants register mostproblems. Long distance crewsreported more problems thanshort distance crews. Healthconcerns in cabin crews needto be considered along withfactors such as aircraftfacilities and equipment, jobdemands, duty schedules,work conditions, robustness ofpersonnel, and effectsassociated with length ofservice.

Sutherland /1993 /Belgium [14]

P, SLevel 4COSD 7

Offshore installationsSV: StressSC: Vulnerability to accidents

A stress audit examines thefactors relating to stress,accidents and personal life.SS: Random sample of 310males working on offshoredrilling and production rigsRR: 32%

Significant differences relatingto stress. Workers with jobinsecurity were more stressedthan those with job security.Workers had higher levels ofstress during the Wintermonths. Workers were more



Type ofdocument*



Comments / Findings


22

DS: Personnel working on off-shore mining platforms

stressed if they hadexperienced an accident andthe level of stress increased ifthe accident was severe.Workers with type Apersonality classifications weremore vulnerable to stress thanother workers.

Ore / 1993 /Australia [15]

P, SLevel 4COSD 7

Stevedoring / TransportSV: MoraleSC: Occupational accidents

A retrospective analysis ofaccidents by stevedoringemployees.SS: 124,279 accidents and losttime incidents.RR: n/aDS: Database of accidents andincidents from the Associationof Employers of WatersideLabour over a 20-year period.

A significant decrease in thefrequency of accidents wasevident with a significantincrease in the severity of theaccident. Most of the accidentsoccurred between 8am and 12noon on Mondays resultingfrom slipping and falling. Thepaper hypothesizes that therecould be a culture amongworkers that an accident orsporting injury may haveoccurred on the weekend sothe worker comes to work “tohave their accident” on theMonday. Morale improvesgenerally as the weekendapproaches.

23

4.2.4 Recommendations

The impact of long hours, shift work and stress are known to affect mood, mentalhealth and emotional wellbeing. This also has an impact on clinical decision-making, alertness, vigilance and effective communication.

There are clear gaps in the literature given that few papers in this review focuson staff health and patient safety. There needs to be further research thatprovides a better understanding of the link between staff physical and mentalhealth and patient safety particularly in the following areas:

1. Evidence-based guidelines for employing staff that provides a basis forassessing their fitness for practice

2. Understanding of the factors that affect staff mental health and effectiveclinical decision-making

3. Interventions that improve staff health which result in an increasedcapacity for improved performance particularly vigilance, resilience,alertness and effective communication.

24

4.3 Communication and feedback

4.3.1 Communication and feedback – summary findings

There were 18 health papers and 13 non-health papers included in this reviewthat examined the relationship between staff and safety. The following tableprovides a description of the types of literature within this domain area.

Health Non-Health

TOTAL



Type of paper

Research Study 15 9 241 0 0 02 1 0 13 3 5 8

Levels of Evidence

4 14 8 221 1 0 12 0 0 03 0 1 14 0 0 05 2 5 76 1 0 17 11 6 17



8 0 0 0

Papers that examined the relationship of communication and feedback betweenhealth practitioners and patient safety contained the following themes:

• Equipment and systems that alert, communicate or provide feedbackabout patients’ conditions have the potential for early intervention,avoidance of error and promote quality feedback processes.

• The impact of litigation has a major impact on doctors’ self-esteem,anxiety and mental state. Those who have experienced litigation oftenbelieve that inadequate communication between staff and the patient isthe major cause of accidents.

• Practitioners’ fear and low self-esteem inhibit open disclosure.• Patients have the least confidence in junior medical officers• Inexperienced staff combined with difficult communication increase the

potential for adverse events• Systems that have a high level of feedback and high level of programming

such as the use of policies and procedures, clinical guidelines and criticalpathways are perceived to be providing the best care.

• High level of staff coordination has an impact on reducing patientmorbidity in surgical units

• Complaints are seldom about clinical treatment alone but includedissatisfaction with personal treatment.

25

• The impact on patients when they are confused or have little knowledge ofthe adverse event may lead to increased anxiety if they are faced with asimilar situation.

The papers that examined communication and feedback and its relationshipbetween staff factors and the influence of a safety culture highlight the followingpoints:

• Improvements in communication within an organisation improve the safetywithin the organisation

• Sub-cultures within an organisation such as professional groups orparticular workgroups have an impact on the organisation’s teamwork andcommunication effectiveness.

• Social relationships on the shop floor are the strongest predictor of safetycompliance.

• In one study the predictors for work disability for men are monotonouswork, neuroticism and job dissatisfaction and for women these wereinterpersonal conflict at work and a combination of marital and workconflict.

• Role overload and negative perceptions of safety decreased safetybehaviour

• In an aviation study the frequency of crew changes was stressful. Regularmeetings, supervision and support had a positive influence on informationexchange

• Implementing goals and feedback processes to encourage safetycompliance is not enough where there is high absenteeism, difficultcommunication between management and staff.

• Improvements were identified in the reduction of incidents in oneorganisation when an occupational health and safety (OH&S) committeewas established, an organisational OH&S policy was developed and staffroles and responsibilities were clearly states and documented.

4.3.2 Communication and Feedback – Health Literature Summary TablesAuthor / Year /Country

Type ofdocument*



Comments / Findings

P = published or U = unpublished. L = literature review, S = study containing data, G = guideline or CS = case study.Hierarchy of study design rating = Level 1 -4 (See table 2) COSD = classification of study design 1 - 8 (see table 1)**SS = sample size; RR = response rate; DS = data source

26

O’Cathain etal / 2002 /UK [16]

P, SLevel 3-3COSD 5

SV: Acceptability of an emergency medicaldispatch (EMD) system to people who callambulance servicesPO: Improve patient care and satisfaction withambulance services

Prospective cross sectionalsurvey in which postalquestionnaires were sent to twosystematic random samples ofapproximately 500 namedcallers to one ambulanceservice before and 1 year afterthe introduction of EMD. TheEMD consisted of telephoneinformation for first aid andurgency of the incidence.SS: 493 (before) 466 (after)RR: 72% (before) 63% (after)DS: Callers to an ambulanceservice

Callers reported increases infirst aid information (7% to43%), general information(13% to 58%) with satisfactionon advice given rising from71% to 73%. Two main issueswere found with EMD, somecallers were advised to takeactions that were not neededand secondly, some callers feltthat ambulance crew did nottreat the situation as seriouslyas they would have liked.

Taylor et al /2002 /Australia [17]

P, SLevel 4COSD 7

SV: Emergency department (ED) complaintsPO: Decreasing patient complaints in the ED

This study is a retrospectiveanalysis of patient complaintsfrom 36 Victorian ED’s during a61 month time frame. Data wasobtained from the HealthComplaint Information Programwith the aim of analyzingcomplaints in order to identifyprocedures/practices thatrequire change.SS: 3418 complaintsRR: N/ADS: Patients in emergencydepartments

2419 ED patients complainedabout 3418 issues (15.4% ofall hospital complaints). 47.8%were made by telephone,34.4% were received by letter,63.1% were made by a personother than the patient. Highestcomplaint rates were receivedfrom patients who werefemale, born in non-Englishspeaking countries and werevery young or old. Remedialaction was taken in 3.2% (109)of cases and compensationpaid to 8 patients.

The BristolInquiry /2001 / UK[18]

P, CSLevel 4

SV: Communication amongst health professionalsand management / peer reviewPO: Deaths of infants undergoing cardiac surgery

This hospital review examinesthe factors that led to a patternof high mortality in infants agedless than 1 year undergoingcardiac surgery. Evidence wasobtained from 577 witnesses

There were 30 – 35 excessdeaths during 1991 – 1995 forchildren under the age of 1year. Mortality rates were atleast twice as high as those forthe UK. Major findings


Type ofdocument*



Comments / Findings


27

and reviews of over 1,800medical records. Activities forthe hospital were examined forthe period of 1984 – 1995.

included poor teamwork, nosystem to monitor outcomesand performance, a ‘club’culture of either being includedor being treated as an outsiderin the hospital management,confusion aroundresponsibilities for monitoringquality of care and norequirement for clinicians toensure their skills andknowledge are up-to-date.

Walton /2001 /Australia [19]

P, LLevel 4

SV: Communication with patients about adverseeventsPO: Reduction in litigation, improvedunderstanding of medical conditions

This systematic literaturereview examines documents onopen disclosure, informedconsent, ethics andcommunication.Papers Reviewed: 91

The review concludes thatearly disclosure by clinicians ofthe adverse event to thepatient minimizes litigation andimproves the patient'sunderstanding of the event.The review also identifies opendisclosure is inhibited byclinicians fear of litigation, lossof self-esteem and being seenas a failure by colleagues.

Greco et al/ 2001 / UK[20]

P, SLevel 4COSD 7

SV: Patient assessment of interpersonal skillsPO: Improved quality of health care

The prospective pilot studyused a questionnaire to providedoctors and nurses, with theperceptions held by patientperceptions of theirinterpersonal skills, and toevaluate the process in termsof its impact on professionaldevelopment and ongoingtraining.SS: 21 consultants, 10registrars, 10 senior nurses and1416 patients

Consultants had a mean scoreof 82%, juniors 79% andnurses 92% (the higher %score the better the perceivedcommunication skills).Consultants scored highest in`respect shown to patient’ and‘patients confidence in ability’.Junior doctors scored highestin ‘respect shown to patient’and ‘warmth of greeting’ andlowest in ‘patients confidencein ability’.


Type ofdocument*



Comments / Findings


28

RR: UnspecifiedDS: Questionnaire given topatients

Pichert et al /1999 / USA[21]

P, SLevel 4COSD 7

SV: Communication with patientsPO: Complaints about dissatisfaction with care

Review of hospital complaintsover a 7-year period.SS: 6,419 reports containing15,631 individual complaintsRR: N/ADS: Complaints recorded byhospital staff

The complaints were: negativeperceptions of care andtreatment (29%) related todiagnosis, problem withtreatment or discharge;communication (22%)classified as general failure orinability to listen; billing andpayment (20%); humanenessof staff (13%) related torudeness, blaming the patientand name calling; access tostaff (9%); cleanliness orsafety of the environment(7%).

Randolph etal / 1999 /USA [22]

P, SLevel 4COSD 6

SV: Accuracy of telephone orders in nursinghomesPO: Reduction in medical errors

A cohort of 4 physicians and 1nurse participated. The studyinvolved recording telephoneinstructions of the physiciansand then comparing the ordersnot less than 12 hours later.The clinicians then judged themedical accuracy of the orderto see if any medical errors hadbeen made.SS: 820 verbal orders in 10monthsRR: UnspecifiedDS: 100 patients from 7facilities

Types of orders madeincluded: medication (34%),lab tests (21%), diet (9%),nursing procedures (9%),equipment (4%), wound care(4%), monitoring (3%),physician procedures (3%),transfer (2%) and mobility(2%). The significant error ratewas found to be 6.1 per 1000.This rate is consistent withother studies.

Young et al /1998 / USA[23]

P, SLevel 3-2COSD 7

SV: Coordination and feedback within surgicalteamsPO: Reduction in morbidity and mortality

This cross-sectional studydescribes the views of surgicalnurses, attending surgeons and

Surgical units that had highfeedback (ie. caseconferences, hand-over


Type ofdocument*



Comments / Findings


29

attending anaesthetists onperceptions of quality of careand coordination approach aswell as patient morbidity andmortality data from 44university affiliated surgicalunits.SS (staff survey): 7,364RR: 73%SS (patient data): 60,000operations.DS: Survey of surgical staff,clinical data from patient chartsand mortality and morbidityfrom Department of VeteranAffairs.

reports) and high programming(ie use of policies andprocedures, clinical guidelinesand collaborative decision-making processes) had thebest perceived quality of carehowever there were nosignificant differences inpatient morbidity and mortality.Staff coordination is moreimportant for improvingmorbidity than mortality insurgical services.

Fiesta /1998/ USA [24]

P, CSLevel 4

SV: Liability for fallsPO: prevention of patient falls

This case report uses 4 casestudies to examine theproblems of patient falls andthe surrounding litigation.

For example, a 92 year oldpatient tripped over a salerepresentative’s suit case onthe way to the examinationroom and broke her hip. Thecourt ruled 40/60pharmaceutical company andthe hospital. The court ruledthat it was the employee’s dutyto perceive open and obvioushazards and to see that thepatient did not sustain injury.

Chen et al /1998 / HongKong [25]

P, SLevel 4COSD 7

SV: Incident reporting in acute pain managementPO: Preventing adverse patient events inpostoperative pain management

Prospective study looking atvoluntary incident reporting inpain management. Over a 12-month period incidents werereported in 1275 patients whoreceived pain relief treatments.SS: 53 incidentsRR: N/A

The most common incidentsinvolved delivery circuits,delivery pump and drugadministration. 81.4% of theincidents were thought to bepreventable. Human factorswere involved in 41.9% mostcommonly associated with


Type ofdocument*



Comments / Findings


30

DS: Hospital incident reportingsystem

inattention, inadequatecommunication andinexperience.

Bark et al /1997 / UK[26]

P, S,Level 4COSD 7

SV: Views of consultants and registrars providingfeedbackPO: Reducing adverse patient events andlitigation

A cross-sectional self-reportpostal survey to seek the viewsof consultants and seniorregistrars on ways of reducingpatient adverse events andlitigation.SS: 769RR: 76%DS: Survey of consultants andsenior registrars in acutehospitals

Of those respondents, 37%had experienced some form oflitigation. Anger, distress andfeeling personally attackedwere common responses.Suggestions for reducinglitigation included a need forchange at the clinical levelincluding supervision of juniordoctors, workload and trainingin communication skills.

Beckman etal / 1996 /Australia [27]

P, SLevel 4COSD 7

SV: Development and evaluation of an incidentreporting system in intensive care (IC)PO: Improving patient safety

The incident reporting (IR) formincluded both a narrativesection and a multiple choicesection to gain more detail. Theevaluation questionnaire wasdesigned to assess staffattitudes to the study,understanding and identify anyproblems.SS: 129 incidentsRR: 88%DS: 3 Intensive Care Units inNSW

More than 90% of respondentsshowed a good understand ofthe incident monitoring study.The IR form allocated fivemain incidents types, airway(20.5%), drugs (25%),procedures (21.2%),environment (22.7%) andmanagement (10.6%). The IRreport form was redesigned(simplified) for use in anongoing national study.

Kuperman etal / 1996 /USA [28]

P, SLevel 3-3COSD 5

SV: Alerting system for physiciansPO: Improved quality of inpatient care

This study developed andevaluated an early alertingsystem that notifies thephysician via a pager that analert is present and offerspotential therapies for theinpatient’s condition at the timehe views the alert information,over a six month period.

Physicians responded to 70%of alerts for which they werepaged. 82.5% were respondedto in less than 15 minutes.They said they would takeaction in 71.5% of cases and39.4% placed an order directlyfrom the alert screen. Furtherevaluation is needed to


Type ofdocument*



Comments / Findings


31

SS: 10,064 admissions whichgenerated 1945 alertsRR: UnspecifiedDS: Mumps database

determine if the systemimproves processes oroutcomes of care.

Crowther /1995 /Germany[29]

P, SLevel 4COSD 7

SV: Women's perceptions of the adequacy ofcommunication by staff about the stillbirth orneonatal death during a pervious pregnancyPO: Anxiety with current pregnancy

Pregnant women wereinterviewed who had previouslyexperienced a stillbirth orneonatal death.SS: 48RR: unspecifiedDS: Pregnant women

Women who had either poor orconfused informationsurrounding the deathaccounted for 25 of the 48interviewed. 29% weresatisfied with the information,one third were admitted tohospital because of anxietyabout the current pregnancy.

Anderson etal / 1994 /USA [30]

P, SLevel 2COSD 1

SV: Comparisons of adverse event reports bypatient and physicianPO: Reported adverse events

Two parallel randomised,double-blind multicentre clinicalcomparisons of quality of life inmen under treatment with anti-hypertensive agents werecollected using physicianreports, patient reports to asymptom distress checklist,physician interviews andlaboratory reports for adverseevent information.SS: 2,318RR: UnspecifiedDS: Physician interviews, labreports, responses from thepatient checklist, andresponses from the concurrentphysician checklist

Symptoms in the checklistwere generally more widelyreported by patients than byphysicians. Large disparities inthe frequency of symptomsreported by the physiciangroup (whose informationcame from consulting with thepatient) was on occasion fourtimes higher rates of reportingby patients using the 51-itemchecklist. The paper exploresissues around the reliability ofthe checklist, physiciansreporting patterns based ontheir knowledge of theimportance of reporting onsome and not other symptomsand the patients' ability torecall symptoms over time.This study indicates thatunder-reporting of adversedrug events is high.


Type ofdocument*



Comments / Findings


32

Bark et al /1994 / UK[31]

P, SLevel 4COSD 7

SV: Clinical complaintsPO: Improving quality of care

Cross sectional survey toestablish the reasons forclinical complaints as well ascomplainant’s feelings andmotivations.SS: 1007 complaintsRR: 49%DS: Complainants who hadwritten to 20 hospitals

Complaints were seldom aboutclinical treatment alone (11%)most (72%) included a clinicalcomponent and dissatisfactionwith personal treatment. Lackof detailed information andstaff attitudes were identifiedas important criticisms.Recommendations for stafftraining on communicating withdistressed and dissatisfiedpatients.

Vincent et al/ 1994 / UK[32]

P, SLevel 4COSD 7

SV: Impact of litigation (obstetricians andgynecologists O/G )PO: Prevention of injury

A questionnaire was sent to alldoctors in the North Thamesregion. Respondents wereasked to rate on five pointscales the main causes ofaccidents and the effect oflitigation on them personallyand professionally. Only theO/G’s results have beenreported.SS: 63RR: 84%DS: Senior O/G’s

It was found that 75% (47/63)of the O/G’s in the area hadbeen involved in some kind oflitigation. Of those 47, 53 hadhad an award made againstthem. Communication betweenpatients and staff was seen asthe main cause of accidents.Inadequate supervision ofjunior staff was also seen asimportant.

Hicks et al /1993 / UK[33]

P, SLevel 4COSD 7

SV: Hospice management of patients receivingchemotherapyPO: Improving patient quality of care

Retrospective analysis of datawas conducted using the notesof patient who received hospiceand chemotherapy careconcurrently.SS: 52 patients over 15months.RR: UnspecifiedDS: Hospital patient databaseas well as medical and nursingcase notes.

It was found that 24 out of 52referral forms had no mentionof chemotherapy, hospitalrecords or photocopied noteswere available for 14/52.These findings illustrate thepoor quality of communicationbetween professionals. Thelimited understanding ofpatient care was found tocause significant difficulties for


Type ofdocument*



Comments / Findings


33

patients. One third of patientsdied 1 week after ending theirchemotherapy.

4.3.3 Communication and Feedback – Non- Health Literature Summary TablesAuthor / Year /Country

Type ofdocument*



Comments / Findings


34

Nicol / 2001 /Australia [34]

P, CS,Level 4

Hazardous industries / oil and gas processingplantsSV: Improved inter-and intra companycommunication, recruitment of company staff toensure corporate memorySC: Avoidance of plant processing disasters

This paper examined thecircumstances around theLongford gas explosion where 2people died and several wereinjured. It recommendschanges to the plant,employment and system issuesfor its ongoing sustainability.

Strategies include improvedcommunication within thecompany and its contractors,identification of safety andengineering hazards,recruitment of youngerengineers to ensure corporatememory is continuedthroughout the lifetime of theorganisation, development of acontinuous hazardidentification system andavailability of engineering,operating and maintenanceskills at all times.

Cox et al /2000 / UK[35]

P, SLevel 4COSD 7

Offshore miningSV: Communication issuesSC: Improving the safety of the work place

The paper examines a two-partprocess of developing a surveyinstrument (a series of focusgroups and employee survey)to assess employeeperceptions of the safety culturein offshore work environments.SS (employee survey): 221RR: 63%DS: Questionnaires, focusgroups, behavioralobservations, situational auditsof onshore and offshorepersonnel in 3 separateorganizations

The survey results found thatcommunication issues,supportive environment andemployee involvement werethe three areas of greatestneed for improvement.Significant differences werefound between drilling teamswho rated lower evaluations oncommunication, supportiveenvironments and personalpriorities compared withproduction teams andmanagers. This may be due todrilling teams belonging toanother sub-culture within theorganisation with differentcommunication channels.


Type ofdocument*



Comments / Findings


35

McDonald etal / 2000 /Ireland [36]

P, SLevel 4COSD 7

AviationSV: CommunicationSC: Compliance to safety procedures

This study examined theattitudes about safety andcompliance with procedureswithin 4 organisations.SS: management: 33,technicians, crew managementand other support staff: 286RR: UnspecifiedDS: In-depth interviews(management) and a crosssectional survey (other staff)

The data from this study wasused to further develop amodel for safety managementsystems. It found markeddifferences betweenorganisations’ expressedcommitment to safety,standards, planning andorganisation, monitoring,feedback and change. Sub-groups within the organisationwere relatively homogenouswith key groups such astechnicians having a strongcommitment to safety.

Hofmann etal / 1998 /USA [37]

P, SLevel 4COSD 7

All IndustriesSV: Role of safety climate and communication inaccident interpretationSC: Safety intervention for workers

This cross sectional studysurveyed sampled 2 groups.Sample 1 – 2,566 outdoorworkers in a utility company.RR: 49.4%Sample 2 – 1318 workers fromdifferent divisions of the sameutility company.RR 49.5%Sample 3 - surveys were alsosent to 83 supervisors with aRR of 99%.

The results for both samplesindicated that contextualfactors such as climate safetyand communication onaccident interpretationssignificantly influencedaccident attributions.Organisational factors aboutcommunicating negativeevents such as industrialaccidents can positively affectstaff to create a learningenvironment and improvedsafety measures.

Simard et al /1997 /Canada [38]

P, SLevel 4COSD 7

ManufacturingSV: Cooperation in workgroups, compliance withsafety rules, supervisor’s controlSC: level of supervisor’s perception of risk tooccupational injury

This is a cross-sectional surveyexamining the determinants ofsafety compliance behaviour.SS: not clearly stated butrepresents 1062 workgroupsfrom a random sample of 97manufacturing plants.

Social relationship variables onthe shop floor are the strongestpredictor of safety compliance.These social relationships areinfluenced by managerialdecisions and actions.


Type ofdocument*



Comments / Findings


36

RR: UnspecifiedDS: Self-completedquestionnaires of managersand supervisors.

Rundmo et al/ 1998[10] /Norway

P, SLevel 4COSD 5

MiningSV: Off shore oil rig workers perception of safetyand risk of injurySC: Worry about platform movements

This cross-sectional surveyexamines the level of worry andconcern about safety.SS: 179RR: 100% on the first shift. RRwas difficult to calculate on the2 remaining shifts asemployees overlapped shiftsDS: Employees over 3 shifts

There is a relationshipbetween perception of beingunsafe and unsafe workpractices. Mental imagery andrationalistic approachesimproved worry and concernusing videos and direct contactwith an independent contractor(not company manager)explaining the event andpossible consequences.

Appelburg etal / 1996 /Finland [39]

P, SLevel 3-2COSD 3

Industry non-specificSV: Interpersonal conflict at workSC: work disability, martial conflicts, neuroticism,life dissatisfaction and general health status

This population based cross-sectional survey used anational sample of twins toidentify interpersonal conflict asa predictor of work disability.SS: 15,348RR: 84%DS: Existing twins register

Predictors for work disability:for men are monotonous work(RR1.61), neuroticism (RR1.96), life dissatisfaction (RR1.63) and stress of dailyactivities (RR 1.66); for womenare interpersonal conflict atwork (RR 1.56) andsimultaneous marital and workconflicts (RR 2.54).

Hoffman et al/ 1996 / USA[40]

P, SLevel 3-3COSD 7

Chemical processing plantSV: Group process, safety climate, intentions toapproach others in the work group andperceptions of role overloadSC: Frequency of reporting unsafe workpractices

This cross sectional surveyconducted in a chemicalprocessing plant representingthe 21 working groups at alllevels of operation.SS: 204RR: 92%DS: Anonymous questionnairegiven to all employees

Role overload (performance isaffected by inadequate time,training and resources) wassignificantly associated withunsafe behaviours. Individualinterventions within the groupprocess to mediate unsafebehaviours were notsignificant. The group processwas not significantlyassociated with accidents. The


Type ofdocument*



Comments / Findings


37

safety climate was significantlyassociated with between -team differences in unsafebehaviours. Workersperceptions of a safety climatewere significantly associatedwith unsafe behaviours.

Skogstad etal / 1995 /Norway [41]

P, SLevel 3-2COSD 7

AviationSV: Satisfaction with communication betweenflight crew and cabin crewSC: Effective safety management and jobsatisfaction

This cross sectional self-reportsurvey of cabin and flight crewwas completed by members ofSAS Norway.SS: 1240RR: 84%DS: Questionnaire, cabin andflight crews

Half of the sample wasdissatisfied with thecommunication andinformation exchange betweencabin and flight crews.Frequency of meetings,supervision and supportissues, and support schemesshowed a significantrelationship with informationexchange and cooperationbetween the two types ofcrews. The frequency ofchanging crews was identifiedas stressful for 86% of thesample.

Anonymous /1994 /Australia [42]

P, CS,Level 4

Food processingSV: Unsafe work practices and hazards in thework placeSC: Decrease in occupational injury

This case study outlines thestrategies used by a major foodmanufacturer in reducingoccupational injury in theworkplace, improving a cultureof safety through thedevelopment of an occupationalhealth and safety (OHS) policyincluding clear roles andresponsibilities and thedevelopment of an occupationalhealth and safety committee.

These strategies improvedcommunication between alllevels of the workforceparticularly the large number ofwomen from non-Englishspeaking backgrounds.Controlling hazards at theirsource by early riskassessment, periodicinspections, integration ofmanagement in OHS initiativesand training in OHS. Days lostas a result of workplace injures


Type ofdocument*



Comments / Findings


38

per 100 employees hasdecreased from 330 in 1991/92to 26 for 1993/94.

Cooper et al /1994 / UK[43]

P, SLevel 3-3COSD 5

ManufacturingSV: unsafe work practicesSC: occupational injury and accidents

This is an intervention studythat examines the effects ofgoal setting and feedback onreducing work-place accidentsand injury in a manufacturingplant employing 540 people.SS: 171RR: 32.6%DS: Production plant workers

The study used observers toassess workers’ safety againstpredefined checklists detailingthe safety criteria. The studyresults showed that theprovision of goals andfeedback is not enough toimprove workers safety in acomplex environment evidentby increases in injury rateswhen there is highabsenteeism and poorcommunication betweenmanagement and staff.

Pate-Cornell /1993 / USA[44]

P, CSLevel 4

MiningSV: Off shore oil rig workersSC: Death

This paper describes thelessons learned from theoffshore platform Piper Alphaaccident that killed 167 peoplein 1988.

Key issues relating to humanfactors were under trainedoperators in under experiencedpeople running operations,negative experiences of pastnear-misses and a culture ofignoring safety incidents andnot rectifying problems. Poorcommunication betweenworkers from one shift to thenext.

Clarke / 1993/ Hong Kong[45]

P, CSLevel 4

Transport - RailwaysSV: Safety management system (SMS)SC: Maintaining and improving safety standards

This paper outlines theimplementation of a SMS. TheSMS includes policy, safetytasks, safety responsibilitystatements, safety audit, workgroup audit, activity audit,safety communications, safetytraining, quality initiatives and

Fourteen safety tasks werenominated; information, safesystems, buildings, plant andsoftware systems, protectiveequipment, fire and security,human resources,communication on safety,contractors and visitors, design


Type ofdocument*



Comments / Findings


39

the role of the safetydepartment.

and project management,accident reporting andinvestigation, safetyinspections, safetyperformance monitoring,funding for safety and review,as crucial to the SMSimproving safety standards.

40


The literature identified that communication breakdown is, in many cases, acritical factor in disasters, preventable accidents and adverse events. Theliterature also indicates that high staff turn-over, inexperienced staff and personalfactors such as low self-esteem, irritability from tiredness and interpersonalconflict are contributing factors to ineffective and inappropriate communication.There still remain gaps in the literature particularly focusing on interventions thatpromote improved communication and the reduction of adverse events.Further research is needed using more robust methods which may providepromising strategies to improve incident reporting, the reduction of adverseevents and the improvement of staff morale. These include:

1. Automated or semi-automated alerting systems for staff to identifypotential risk associated with patient care

2. Organisational structures that improve social relationships amongcolleagues providing direct patient care

3. Feedback on performance and safety issues that increases teamwork andperformance.

41

4.4 Hours of work, shift work, number of staff andfatigue

4.4.1 Hours of work, shift work, number of staff and fatigue –summary findings

There were 36 health and 30 non-health papers included in this review thatexamined the relationship between staff working hours or fatigue and patientsafety. The following tables provide a description of the types of papers.

Health Non-Health

TOTAL



Type of paper

Research Study 31 22 531 0 0 02 3 0 33 14 10 24

Levels of Evidence

4 19 19 381 1 0 12 2 0 23 2 2 44 1 0 15 9 9 186 1 1 27 15 11 26



8 0 0 0

Papers that reported health professionals’ working hours, rostering and staffingmix and a relationship between patient safety contained the following centralthemes:

• Fatigue is a major concern for health professionals working long hours orrotating shifts. Fatigue affects performance particularly the speed at whichprocedures are undertaken, alertness and vigilance to identify problemsand affects mood which is likely to interfere with effective communication.Fatigue is commonly an issue reported for medical officers. Some papersexplored the introduction of maximum hours at work and workload. Drivingaccidents on the way home from work were also a risk for fatigued staff.

• Contributing factors for adverse events were found to be high workloaddue to the high acuity of patients, inappropriate staffing mix particularlywith regard to number of registered nurses and non-registered nurses.Papers identified a range of quantifiable patient outcomes in relation tonurse-patient ratios. Inadequate ratios in one paper [46] showed an affectwith increased patient re-admission rates at 30 days after discharge.Another paper [47] identified low registered nurse to patient ratios

42

increased the likelihood of patients acquiring a urinary tract infection orpneumonia, thrombosis or pulmonary compromise after surgery. A largestudy undertaken in the United States [48] reported that when theregistered nurse to patient ratio increased to 87.5% there is a reduction inmedication errors and patient complaints. Once the ratio exceeded thispoint these errors increased indicating that the assessment of anappropriate staff mix such as registered and non-registered nurses, is animportant consideration in reducing adverse events.

• Rostering of staff was also identified as a contributing factor to adverseevents. Enabling a longer sleep at night can be associated with improvedclinician performance, improved quality of care, fewer errors and fewmissed calls.

Fatigue was a major focus in the non-health related papers. The followingthemes were explored repeatedly in a number of papers:

• Condensed working hours into a longer shift and shorter working weekappeared to be of benefit to both the organisation and individual worker.The studies report an increase in productivity, improved job satisfactionand greater flexible for leisure time.

• In industries where working long hours are part of the job such as aviationpilots and truck driving, the effects of fatigue result in serious safetyissues. These effects include drowsiness, confusion, decreased alertnessand a greatly increased likelihood of making a serious potentially fatalerror.

• Shift workers are more likely to have motor vehicle accidents particularlywhen coming home from the shift. The most dangerous times for fatiguewith shift workers are when changing from day to night shift and from nightshift to day shift.

• The effects of fatigue are commonly reported to include physicalconditions such as headaches, back pain, nose congestion as well asmood changes including irritability.

• Some studies indicate that inexperienced operators are more likely tohave accidents as a result of fatigue compared to more experiencedoperators. Other fatigue related accidents may result from truncated andfragmented sleep patterns.

4.4.2 Hours of work, shift work, number of staff and fatigue - Health Literature Summary TablesAuthor / Year /Country

Type ofdocument*



Comments / Findings

P = published or U = unpublished. L = literature review, S = study containing data, G = guideline or CS = case study.Hierarchy of study design rating = Level 1 -4 (See table 2) # COSD = classification of study design 1 - 8 (see table 1)**SS = sample size; RR = response rate; DS = data source

43

JCAHO /2002 / USA[49]

P, GLevel 4

SV: Numbers of nursesPO: Patient length of stay, health outcomes

This guideline documentsynthesizes the literature andmakes recommendations toalleviate the nursing crisis inUSA hospitals.Citations: 132

Recommendations include:addressing the retention ofnursing staff; improvingcompensation and pay ratesfor nurses; reducing theburden of paperwork andadministrative duties; havezero-tolerance policies forabusive behaviours byphysicians and other healthcare professionals; and, setnurse / patient staffing ratiosbased staff competency andskill mix applicable to patientmix and acuity.

Goodman etal / 2002 /USA [50]

P, SLevel 3-2COSD 5

SV: Numbers of neonatologistsPO: Neonatal mortality

A retrospective analysis thatused national birth outcomedata, neonatal intensive carebeds and numbers ofneonatologists by hospital unitwas examined to determine therelationship between neonatalmortality and numbers ofneonatologists.SS: 3,892, 208 newborns over500gramsRR: N/ADS: Data was used from theAmerican Medical Association(AMA) and the AmericanOsteopathic Association as wellas surveys of beds to calculatesupply of neonatologists andneonatal intensive care beds.

Increases in the number ofneonatologists were notassociated with greaterreductions in neonatal death.


Type ofdocument*



Comments / Findings


44

Aiken et al /2002 / USA[46]

P, SLevel 3-2COSD 7

SV: Numbers of nursesPO: Mortality and failure to rescue (deathsfollowing complications) among surgical patients

A cross-sectional study ofnurses’ hours of work, jobsatisfaction and burnout thatcompares patient outcomedata.SS: Nurses - 10,184RR: 52%DS: Data from 168 hospitals

The study identifies asignificant relationshipbetween high patient-to-nurseratios and 30-day mortalityrates, nurse burnout and jobdissatisfaction. An increase ofa 7% risk of patient mortalitywas found for each additionalpatient added to the nurses'average workload highlightingthe role of nurses as anaround-the-clock surveillancesystem for the hospital todetect and intervene whenpatients' health deteriorates.

Kovner et al /2002 / USA[47]

P, SLevel 4COSD 5

SV: Staffing hoursPO: Venous thrombosis / pulmonary embolism,pulmonary compromise after surgery, urinary tractinfection, pneumonia

A cross-sectional analysis ofhospital outcome data using adata set of between 534 – 570hospitals over a 7-year period.Examination of the associationbetween registered nurse hoursand adverse patient outcomesfor 4 post-surgicalcomplications was undertaken.SS: 530 – 570 hospitals foreach of the years from 19901996 with 187 hospitals havingdata for all seven yearsRR: N/ADS: Annual HospitalAssociation (AHA) annualsurvey of hospitals, NationalInpatient Sample

An inverse relationship wasfound on 4 of the patientoutcomes with an increase innurse, physician and residentintern hours. Significance atp<0.5 level for pneumoniaonly.

Tucker et al /2002 / UK[51]

P, SLevel 3-1COSD 2

SV: Neonatal Intensive Care Units (NICU’s)PO: Improvement in neonatal care and hospitalmortality

Non-randomized controlled trialwhose primary outcomes werehospital mortality, mortality orcerebral damage, and

Data was available from 99%of infants. Mortality was raisedwith increasing workload in alltypes of NICU’s. Assessment


Type ofdocument*



Comments / Findings


45

nosocomial bacteraemia. Theaim of the study was to assesswhether patient volume, staffinglevels, and wok-load areassociated with risk-adjustedoutcomes, and with costs ofstaff wellbeing.SS: 13, 515 infantsRR: N/ADS: 54 randomly selectedNICU’s

of increased staffing levelscloser to those in adultintensive care may be moreappropriate.

Wolfe et al /2002 /Australia [52]

P, SLevel 3-3COSD 5

SV: Rural Australian hospital emergencydepartment (ED)PO: Reducing adverse patient events (ADE)

A before and after interventiondesign was used to determine ifmedical record screening andreview could detect and reduceADE’s in ED’s.SS: 20,050 patients attendedthe ED in the allocated timeperiod.RR: N/ADS: ED at Wimmera Basehospital 300km from Melbourne

An ADE was confirmed in1.24% of all attendances with32% being major and 68%being minor. Over two yearsthe number of ADE’s fell from84 to 12. Thus the studyconcludes that ADE’s in ED’scan efficiently be detected andtheir rate reduced usingretrospective medical recordscreening together with clinicalreview, analysis and action toprevent recurrences.

Davis et al /2002 /Australia [53]

U, SLevel 4COSD 7

SV: Implementation of a patient sittersPO: Prevention of patient self-harm

A quality improvement activitywas undertaken using patientsitters observing patients at riskof self-harm in an acutehospital setting.SS: 6 patientsRR: UnspecifiedDS: Staff surveys

Of the six patients in thisactivity, there were no falls, noabsconding or other self-harmincident while the sitter waspresent. Patients wereobserved to be less agitatedwith the sitter present.

Sovie et al /2001 / USA[54]

P, SLevel 3-2COSD 5

SV: Hospital restructuring and its impact onnursing staffPO: Quality of care

Prospective cross sectionalsurvey looking at the effect orrestructuring on delivery ofpatient care using selectedpatient outcomes (full rate,

The study found that increasedhours worked per patient perday were associated withlower fall rates, lower urinarytract infections and higher


Type ofdocument*



Comments / Findings


46

nosocomial pressure ulcer,urinary tract infections, andpatient satisfaction).DS: 29 university teachinghospitals (300 beds)SS: UnspecifiedRR: UnspecifiedDS: Department level datacollected following aDonabedian model from eachhospital

general levels of patientsatisfaction (from 79% to81%) as well as specificallywith pain management.

Weinburg etal / 2002 /USA [55]

P, SLevel 4COSD 4

SV: Reduced staff on weekendsPO: Impact on the rate of falls and medicationerrors

Prospective quality indicatordata was collected during a 6-month period in a 14 bed sub-acute unit. The majority of thepatients were admitted forrehabilitation following a bonefracture or stroke.SS: 31 patientsRR: N/ADS: All sub-acute unit residentsadmitted to the unit weremonitored for the presence ofthe specific outcome measures

Significant differences werefound between patient careduring weekdays compared toweekends. Essentialdocumentation in the nurses'notes was less on weekends(0.3% vs 3.8%), the falls ratewas increased on weekends(019% vs .77%) and nodifference was found inmedication errors. Nursingstaff levels was less on theweekends (3 vs 4).

Trafford /2001 / USA[56]

P, CSLevel 4

SV: Hospitalised patient use of private nurses orsittersPO: Quality of patient care

This single case study looks atthe effect of a private nurse orsitter on the effect of quality ofpatient care.

One patient recovering from aleg operation hired a sitter athis own expense, he also didthis when he underwent aheart operation. The patientsaid that to improve the qualityof care the only thing he coulddo was have somebody there24 hours a day. He felt hecould not get the service herequired from the hospital.


Type ofdocument*



Comments / Findings


47

Boodman /2001 / USA[57]

P, CSLevel 4

SV: Fatigue in hospital internsPO: Accident and injury prevention

This article presents 3 casestudies in order to look at theproblem of fatigue amongmedical interns.

For example a plastic surgeryresident had been workingmore than 50 hours withoutsleep when they started toperform colon surgery andbriefly nodded off with theinstruments in their hand.Problems of fatigue for internsmust be addressed byhospitals to improve quality ofcare.

AcuteHospitalPortfolio –Review ofNationalFindings /2001 / UK[58]

P, SLevel 4COSD 7

SV: Ward staffingPO: Increased quality of care

This study looks retrospectivelyat ward staffing and collectsinformation on patientcomplaints, pressure ulcers andpatient accidents.SS: Data from 3600 wards.RR: N/ADS: The database was createdby the Audit Commission

It was found that ward staffingis the largest single budgetitem, with large wards costingless per bed on average. Mostaccounting systems were notstandardized. Trying tostandardize the amount ofmoney on ward staffing is acomplex task because alltrusts have mixes of differenttypes of wards and patients.Although data on patientcomplaints and accidents iscollected it is not reported.

Hendrix et al/ 2001 / USA[59]

P, SLevel 3-3COSD 7

SV: Nurse staffing levels for optimal long termcarePO: Increased quality of patient care

This study is a cross section ofmultiple observations at a pointin time (1994) for all US nursinghomes. The study attempts todefine optimum staffing levelswhereby quality is optimized,personnel are conserved andpublic burden is minimized.SS: 12,000 nursing homesRR: UnspecifiedDS: Databases plus an onlinereporting system in nursing

Data revealed thatapproximately 60% of USnursing homes were usingfewer registered nurses thanand 100% were using fewernurses’ assistants than optimalmeasured using a nursesensitive patient outcome ofdecubitus ulcers.


Type ofdocument*



Comments / Findings


48

homesPatton et al /2001 / USA[60]

P, CSLevel 4

SV: Sleep deprivation among resident physiciansPO: Effects on adverse events

This paper used 2 case studies.One case study examines thelegal implications of a womanwho died whilst being treated inthe emergency department byinexperienced, unsupervisedand sleep-deprived residentmedical officers. The secondcase study examines the caseof a rail worker who was madeto work 27 hours and fell asleepwhile driving and collided into acar on his way home from work.The victims of the accidentsuccessfully sued the railworkers’ employers. The paperexamined the legal implicationsfor sleep-deprived doctors suchas direct hospital liability forfatigue-related residentmalpractice, vicarious liability ofteaching hospitals for torts offatigued residents, informedconsent and the duty todisclose skill and status risks.

The woman died and herfather attempted to use thiscase to reform legislationaround the safe working hoursfor doctors. As a resultlegislation was passed torestrict medical staff andpostgraduate trainees’ workhours to 12 consecutive hoursper on-duty assignment inhospitals with over 15,000unscheduled visits to theemergency service annually.Post-graduate trainees shallnot exceed 80 hours per weekover a 4-week period and shallnot be scheduled to work morethan 24 hours.

Needlemanet al / 2001 /USA [61]

P, SLevel 3-3COSD 7

SV: Numbers of registered nurses per patientPO: Outcomes Potentially Sensitive to Nursing(OPSN) which were urinary tract infections, skinpressure ulcers, pneumonia, DVT, length of stay,mortality, failure to rescue, upper GIT bleeding,complication of the CNS, sepsis, shock, woundinfection, pulmonary failure, metabolicderangement

Analysis of hospital patientdischarge data and financialreports or hospital staffingsurveys were undertaken toidentify the relationshipbetween level of registerednurse staffing level and OPSN.SS: A total of 3357 hospitals in11 states were used in variousstages of the analysiscontaining 12,261,737 records.

Total hours of inpatienthospital nursing time averaged11.4 hours / day. Registerednursing hours averaged 7.8hrs / day. An association wasfound between urinary tractinfections, pneumonia, lengthof stay, upper GIT bleedingand shock in medical patientsand only failure to rescue inmajor surgical patients. Higher


Type ofdocument*



Comments / Findings


49

RR: N/ADS: hospital database systems

RN staffing was associatedwith 3-12% reduction inOPSNs. Low to high overallnurse staffing variables (RN,aides and licensed practical /vocational nurses) wasassociated with 2-25%reduction in OPSN rates.

Moreno et al/ 2001 / TheNetherlands[62]

P, SLevel 3-2COSD 3

SV: Organ system failure and nursing workload atdischargePO: Mortality

This prospective cohort studyexamined whether post–intensive care unit (ICU)discharge mortality isassociated with the presence ofand severity of organdysfunction/failure just beforeICU discharge.SS: 4621 patients including2958 discharged alive

13.4% of patients died in ICU,32.8% died before hospitaldischarge. From the remaining4000 6% were dischargehome, 74% to the generalward and 9% to other ICUcare. Results showed that it isbetter to delay the discharge ofa patient with organdysfunction /failure from theICU unless adequatemonitoring and therapeuticresources are available.

Reed et al /2000 / UK[63]

P, SLevel 3-3COSD 6

SV: Inpatient care of mentally ill people in prisonPO: Improved quality of health care

Experimental study using semi-structured inspectionsconducted by a doctor and anurse to investigate thefacilities for inpatient care of thementally ill.SS: 348 bedsRR: N/ADS: 13 prisons with inpatientbeds

No doctor in charge ofinpatients had completed anyspecialist training with only24% of nursing staff havingmental health training, and32% non-nursing trainedhealth care officers. Mostpatients were unlocked foraround 3.5 hours per day withthe average length ofseclusion being around 50hours. Prison facilities wereoften poor, low staff numberswith insufficient training. In-mates spent too long secludedand without appropriate


Type ofdocument*



Comments / Findings


50

psychological interventioncompared with care expectedelsewhere.

Morris et al /2000 /Australia [64]

P, SLevel 4COSD 7

SV: Anesthesia and fatiguePO: Incident monitoring

A retrospective analysis wasdone using the AustralianIncident Monitoring Study(AIMS) database. Reports,which listed fatigue as acontributing factor, wereexamined.SS: 152 reports (2.7% of allreports, total reports = 5600)RR: N/ADS: AIMS database

Of the fatigue positive reportsother contributing factorsincluded haste (45%),inattention (37%), failure tocheck equipment (35%), faultof technique (29%), distraction(19%), pressure to proceed(9%), other stress (10%), Druglabel (10%) and otherequipment problems (10%).The papers suggest thatdefinitive prospective studiesbe targeted at these relatedinterventions.

Tarnow-Mordi et al /2000 / UK[65]

P, SLevel 2COSD 7

SV: Staff workload in an Intensive Care Unit (ICU)PO: Hospital mortality

This retrospective analysis wasbased on a prospective cohortstudy of all admissions to theICU in a hospital in Scotlandwith the objective of studyingmortality rates in relation to theworkload of hospital staff.SS: 1050 patientsRR: N/ADS: ICU staff

Variations in mortality can bepartly explained be excess ICUworkload. Patients exposed tohigh ICU workload are morelikely to die than thoseexposed to lower workload.The 3 measures of ICUworkload most stronglyassociated with mortality werepeak occupancy, averagenursing requirement peroccupied bed per shift, and theratio of occupied toappropriately staffed beds.

Schulmeister/ 1999 / USA[66]

P, SLevel 4COSD 7

SV: Nurses description of chemotherapymedication errorsPO: Reducing patient medication errors

Descriptive survey whichcontained 24 demographic andopen-ended questions. The aimof the study was to expand theknowledge on errors inchemotherapy medication.

Chemotherapy medicationerrors were reported to haveoccurred in the workplace of63% of the respondents and140 errors were described.Errors included under/over


Type ofdocument*



Comments / Findings


51

SS: 620 randomly selectedoncology nursing membersRR: 160 (26%)DS: Oncology nursing societymembers employed in directpatient care positions

dosing, schedule and timingerrors, omission of drugs,infusion rate errors andtherapy given to the wrongpatient. Trends in theworkplace identified by nursesas contributing to the error rateinclude: increased acuity ofpatients, inappropriatelyqualified nurses, reduction inthe number of registerednurses providing direct patientcare.

Peterson etal / 1999 /Australia [67]

P, SLevel 4COSD 7

SV: Pharmacists attitudes towards dispensingerrorsPO: Medication error reduction

A prospective cross sectionalsurvey of pharmacist’s attitudestowards dispensing errors wasundertaken in Tasmania.SS: 419 surveysRR: 49.9%DS: Registered pharmacists inTasmania

82% of pharmacists felt thatdispensing errors were on theincrease, 17% felt they weren’tand 3% were unsure.Pharmacists felt that 150prescription items per daywere a safe number. Factorscontributing to high error ratesincluded high prescriptionvolumes, fatigue, overworkand interruptions.

Cohen et al /1999 /Canada [68]

P, SLevel 4COSD 5

SV: Nursing workload associated adverse eventsin post-anesthesia care unit (PACU)PO: reduction in patient adverse events

Patients were observed andassigned points according tothe Project Research in Nursing(PRN) workload system.SS: 2031 patientsRR: N/ADS: The data from the PRNwere merged with the data fromthe hospital’s anesthesiadatabase.

Patients without any adverseevents had on average anumber of 29 points, going upthe scale to critical respiratoryevent at around 54 points. Itwas found that the rate ofpostoperative adverse eventsaffects the amount of nursingstaff resources needed in thePACU.

Blegen et al /1998 / USA[48]

P, SLevel 3-3COSD 5

SV: Numbers of nursing staffPO: Reduction of adverse patient outcomes(medication errors, patient falls, urinary and

Retrospective analysis ofmedical records and staffinglevels in a large university

Hospital units with high patientacuity had low levels ofmedication errors. When


Type ofdocument*



Comments / Findings


52

respiratory infections, skin breakdown, patientcomplaints and mortality)

teaching hospital over a 12-month period.SS: around 3000 beds acrossthe hospitals RR: N/ADS: Data from 39 units in 11hospitals

controlling for patient acuity aninverse relationship was foundbetween registered nurse (RN)hours and medication errors,decubiti and patientcomplaints. A directassociation was foundbetween total hours of carefrom all nursing personnel,complaints and mortality. Asthe RN proportion of care roseto 87.5% there was a lowerincidence of adverseoutcomes. However theseoutcome rates rose when thelevel of RN care increasedbeyond this level.

Hunt et al /1998 / UK[69]

P, SLevel 3-3COSD 7

SV: nurse to patient ratioPO: Mortality within 30 days of emergencyadmission and readmission within 30 days ofhospital discharge

Analysis of nursing full-timeequivalent data and hospitalbed occupancy (nurse / patientratio) was used to examine theassociation with patientoutcomes.SS: 49 hospital trustsRR: N/ADS: Data was obtained fromthe Information Services (ISD)Division of the CommonServices Agency for the NHS inScotland

The paper explores thepossibility of using routinelycollected data using multipleregression analysis. There wasno significant associationbetween nurse / patient ratioand mortality but an inverserelationship was foundbetween nurse / patient ratioand patient 30-dayreadmission rate.

Marck et al /1998 / USA[70]

P, SLevel 4COSD 7

SV: Staffing numbersPO: Safety concerns and complaints

A review of telephone calls in a10-month period to a registerednurse organisation providingtelephone advice and support.SS: 575 telephone callsRR: N/ADS: Telephone calls to the

Of the 575 calls, 30% (168)were related to safetyconcerns in Alberta statecovering concerns aboutpatient care, inadequatestaffing and inappropriate skillmix. There has been a five-fold


Type ofdocument*



Comments / Findings


53

AARN increase in calls relating tosafety concerns between thisreview and one previously (3years prior).

Kovner et al /1998 / USA[71]

P, SLevel 3-3COSD 5

SV: Nurse staffing levelsPO: Adverse patient events following surgery

Retrospective analysis ofhospital discharge data andnurse staffing levels.SS: 589 acute-care hospitals in10 states in the USA.RR: N/ADS: Hospital records from a

Full-time equivalent registerednurse to total hospital inpatientdays was shown to have asignificant inverse relationshipfor urinary tract infections aftermajor surgery (p< 0.001) andpneumonia after surgery (p<0.001). A less robust butsignificant inverse relationshipwas shown for thrombosisafter major surgery (p< 0.01)and pulmonary compromiseafter major surgery (p< 0.05).

Smith-Coggins et al/ 1997 / USA[72]

P, SLevel 2COSD 1

SV: Adaptation to night shift for emergencyphysiciansPO: Reduced adverse outcomes by improvedphysician performance

A double-blind, active placebo-controlled study conducted onattending physicians in aUniversity hospital emergencydepartment. The interventioninvolved a fatiguecountermeasure program usedfor commercial airline pilotsinvolving an educationcomponent, a regular workschedule and fatiguecountermeasure strategies.SS: 6 physiciansRR: Unspecified.DS: Emergency physicians

Overall, physicians' vigilancereaction times and times forintubation of the mannequinwere slower on night shift withno difference in theperformance on EGC analysis.Mood rating was morenegative on night shift. Theexperimental group used thefatigue countermeasure 85%of the time and did not affectphysician performance.

Kirkcaldy etal / 1997 /Germany[73]

P, SLevel 4COSD 7

SV: Safe working hours for medical officersPO: Accidents at work and on the road

This is a cross-sectional surveyof medical officers using apostal questionnaire thatexamines their accidentbehaviour and its relationship to

Doctors who worked morethan 48 hours in a week were5 times more likely toexperience a car accident andreported significantly higher


Type ofdocument*



Comments / Findings


54

job stress and working hours.SS: 2500RR: Unspecified, 8000questionnaires were mailed outand the first 2500 completedthat arrived back were used asdataDS: Practicing doctors

levels of job-related stress.

Brown et al /1996 /Canada [74]

P, SLevel 4COSD 5

SV: Number of ambulance crew sizePO: Differences in intervention and time on thescene

A review of data for advancelife support calls for ambulanceassistance for seizures andchest pain for two 1-monthperiods comparing theprocedures and time on thescene with 2 and 3 personambulance crews.SS: 126 casesRR: N/ADS: Two person teamemergency medical services(EMS) professionals and Threeperson team EMSprofessionals

No difference was found in thetotal number of calls and typesof procedures used. Significantdifferences were foundbetween the 2 team sizes withon-scene times. Two-personcrews were longer by anaverage of 8.4 minutes forseizures and 1.8 minutes forchest pain compared to 3-persons crews.

Booker et al /1995 / USA[75]

P, SLevel 4COSD 5

SV: Seasonal patterns of hospital medicationerrors in AlaskaPO: Prevention of medication errors

Length of daylight is evaluatedin relation to medication errorsin a medical center (140 beds)in the far north of Alaska. Thisdata was collectedprospectively over 5 years.SS: 262 errorsRR: NADS: All nursing staff

The data showed that 58% ofall medication errors occurredduring the first quarter of theyear. Medication errors were1.95 times more likely in Decthan Sept. Thus medicationerrors appear to follow apattern that is closelyassociated with the annualcycles of daylight anddarkness.

Edge et al /1994 / USA[76]

P, SLevel 4COSD 5

SV: Specialised compared to non-specialisedinter-hospital transport staffPO: Intensive care related adverse events

Data was collected over a 2and 3-year period looking atpatient outcomes and the

Significant differences werefound (p< 0.05) betweenadverse events in the


Type ofdocument*



Comments / Findings


55

experienced by paediatric patients association between staffingskill mix using 2 hospitals’ inter-transport service.SS: 141 patientsRR: unspecifiedDS: Patient transportedbetween two tertiary pediatricICU’s

specialised and non-specialised (staff who workedin ICU but not specificallydedicated to inter-hospitaltransportation) care team.Specialised teams have feweradverse events.

Haw et al /1994 / USA[77]

P, SLevel 4COSD 5

SV: Absenteeism of senior medical officers in apsychiatric unitPO: Suicide

Retrospective analysis ofregional suicide data over a 9-year period and examination ofpsychiatric hospital events.SS: 34 current patients whocommitted suicide.RR: N/ADS: Riverside HealthAuthorities Mortality Database(1987-1989) and from theOffice of Population andCensus Surveys (OPCS) for1981 to 1986.

A cluster of 14 suicides wasfound in a 12-month periodand a "temporal" associationcould be found between thetiming of these suicides andsignificant "life events" of thehospital in particular a periodof uncertainty about thehospital's future and higherrates of senior medical officerabsenteeism.

Laine et al /1993 / USA[78]

P,SLevel 2COSD 3

SV: Restricted working hoursPO: Quality of patient care (in hospital mortality,length of stay, medical complications, delays inordering tests and procedures, transfers to ICU’s,resuscitation attempts, discharge disposition)

Retrospective cohort study toexamine the impact on patientcare of a New York stateregulation that restrictedmedical house staff workinghours.SS: 281RR: 94%DS: Patients hospitalised onthe same general medicineteaching service of The NewYork Hospital

Results showed that restrictedmedical house staff workinghours were associated withdelay in test ordering (17% v2%) and in-hospitalcomplications (35% v 22%).No significant differences werefound in the more seriousoutcomes such as in-hospitalmortality, length of stay,discharge disposition ortransfers to ICU’s.

Barone et al/1993 / USA[79]

P, SLevel 4COSD 7

SV: Reduction in surgical staffPO: Quality of patient care

This prospective crosssectional study looked atwhether patients suffered anyloss to quality of care when 24

Of the 659 trauma patients, 86had undergone surgery within12 hours. Patient injuries werefound to be similar with no


Type ofdocument*



Comments / Findings


56

hour surgical staff were not in alevel 2 trauma center.SS: 659 trauma patientsRR: N/ADS: 1 level 2 trauma centercompared to 3 affiliated level 1trauma centers

differences found in mortalityrates. No unexpected adverseevent could be ascribed to thelack of 24-hour or staffing inthe level 2 trauma center.

Mann et al /1993 / USA[80]

P, SLevel 3-1COSD 2

SV: Effect of shift work and sleep deprivation onEmergency Departments (ED) radiology errorsPO: Improved patient care

The non-randomized controltrial compared the performanceof the usual (control) shift withthe performance of the ‘nightstalker ‘ (night shift rotation)shift.SS: 26,421 cases werereviewedRR: N/ADS: After hours ED

The ‘night stalker’ was found tobe averaging around 5.75hours sleep on night stalkerdays compared with theresident who was averagingaround 2.75 hours sleep with 2interruptions. The ‘nightstalker’ shift improved thepatient quality of care andappropriateness andtimeliness of patient care. Thenight stalker made fewer errorsand had fewer missed callsdaily.

Leung et al /1992 / USA[81]

P, LLevel 3-3

SV: Sleep deprivation experienced by doctorsPO: Inadequate assessment and slower responseto monitoring devises

This literature review examines14 studies assessing theeffects of sleep depriveddoctors.

The quality of the papers isvaried and makes it difficultmake clear conclusions.Generally the studies indicatethat sleep deprivation has aneffect on clinical performanceand deterioration in mood.

4.4.3 Hours of work, shift work, number of staff and fatigue – Non-Health Literature SummaryTablesAuthor / Year /Country

Type ofdocument*



Comments / Findings


57

Caldwell et al/ 2002 / USA[82]

P, SLevel 4COSD 7

AviationSV: FatigueSC: On-the-job alertness

A cross-sectional study of self-completed survey of militaryaviators and crew examiningthe impact of sleep andperception of performance.SS: 361RR: 95% estimateDS: Military aviation crew

The average hours of sleep is6 - 7 hours, at least 1 hour lessthan the recommended 8hours of sleep. Approximately45% of respondents havedosed off to sleep while onduty, 49% of pilots and 25% ofcrew have cancelled ordeclined a mission because offatigue, 81% of pilots and 90%of crew thought that fatiguewas the reason for increases inaviation accidents/incidentsand approximately 60% ofrespondents felt that theirsafety was compromised byfatigue or lack of adequaterest.

NationalOccupationalHealth andSafetyCommission /2002 /Australia [83]

P, LLevel 4

All types of industriesSV: Staff fatigueSC: Safety and performance implications

A review of the literaturedefines the scope and themagnitude of the problemsrelating to staff fatigue andworkplace safety.Papers Reviewed: 129

Industries that identify fatigueas a major problem areconstruction, transportation,mining and the health sector.This review outlines fatiguemanagement strategies whichincludes: performancemanagement, subjective ratingof fatigue, readiness-to-perform and fitness-for-dutytests, use of hardwaretechnologies and softwaretechnologies, regulatory andnon-regulatory approaches. Italso outlines therecommended hours of workstandards for industries that


Type ofdocument*



Comments / Findings


58

have such standards.AustralianCouncil ofTrade Unions/ 2002 /Australia [84]

P, SLevel 4COSD 7

13 different industriesSV: Long working hoursSC: Productivity and occupational injury

A qualitative analysis of 54semi-structured interviews withindividuals working in 13different industries.SS: 54 individuals and theirpartner (if they had one)RR: UnspecifiedDS: Interviews with workers

The analysis focused on theeffect of long working hours onthe quality of family and socialrelationships. The section onwork safety identified that inmost cases the participantexperienced a work relatedinjury as a result of fatiguewhich included falling asleep atthe wheel of a car or truck,taking longer to perform aprocedure, not being as alert tosafety signals and signs.

Lilley et al /2002 / NewZealand [85]

P, SLevel 3-3COSD 7

ForestrySV: hours of work, pay, fatigueSC: occupational injury

This cross-sectional study usinga self-administeredquestionnaire examined thepredictors of working conditionsand work place injury.SS: 367 forestry workersRR: 97%DS: Self-report from forestryworkers

The most significant predictorsof work place injury or near-miss injury events were fatiguedue to long working hours.

AustralianIndustrialRelationsCommission(AIRC) / 2002/ Australia[86]

P, GLevel 4

Construction, forestry, mining, hospitality, health,retail, public service, teachers, aviationSV: fatigueSC: occupational injury and accidents

This report is the full benchdecision of the AIRC containingthe arguments for and againstamending 14 awards onreasonable working hours,overtime and meal breaks.

The document presents legalarguments providing a range ofreferences to national andinternational studies comparinghours of work and standardsfor protecting workers rightsand workers safety. The ACTUclaims for a test case standardwere rejected with theexception of conditions relatingto working overtime.


Type ofdocument*



Comments / Findings


59

Rosenberg etal / 2001 /Israel [87]

P, SLevel 4COSD 7

Aviation/ MilitarySV: Military aviator fatigueSC: Fatigue prevention

This study used a voluntarysurvey questionnaire regarding14 current primary, secondaryand tertiary fatigue preventioninitiatives.SS: unspecifiedRR: 37%DS: Flight squadroncommanders in the Israeli airforce

The most popular primaryprevention (87%) dealt withreservist pilots. The mostpopular secondary measure(88%) was to utilize a stimulantdrug such as caffeine oramphetamines. Leading the listof tertiary prevention measures(77%) was that squadronsdebrief the incidence of aviatorfatigue.

Sussman etal / 2000 /USA [88]

P, CSLevel 4

Railway transportSV: Hours on the jobSC: Railway accidents

This case study has a focus on3 rail disasters where theengineer / train driver wasasleep while driving the train.

The paper explains the impactof sleep quality, the breaksrequired between shifts andthe issues related to legalrequirements for off-the-jobtime and incident under-reporting of fatigue relatedaccidents. A minimum of 12hours between shifts isrequired under USA federallaws.

Phillips /2000 /Australia [89]

P, SLevel 4COSD 7

SeafaringSV: Fatigue of staffSC: shipping and fishing boat accidents

A retrospective analysis ofIncident at Sea Reports.SS: 44RR: N/ADS: Accident investigator’sreports

Of the1 injury, 1 tank over-pressurization, 21 groundingsand 21 collisions, 38 incidentswere associated with watchkeeping and sleepiness.Truncated and fragmentedsleep patterns for watchkeepers and an unfavourablesleep environment on fishingboats contributed to accidents.


Type ofdocument*



Comments / Findings


60

Oron-Gilad etal / 2000 /Israel [90]

P, SLevel 4COSD 7

MilitarySV: Fatigue among military truck driversSC: Accident prevention

This prospective study usedfocus groups and a large-scalesurvey with the objective ofdetecting fatigue in militarytruck drivers.SS: 314 male truck driversRR: 93%DS: Military truck drivers’ self-report

The focus groups showed thatmost drivers had fallen asleepat the wheel at least once inthe past. On the survey 39% ofdrivers reported falling asleepat the wheel. Results showedthat quality of sleep was moreimportant than enforcing nightsleep and prohibiting nightdrives as part of fatiguemanagement. It was alsofound that mandatory servicedrivers (young, lessexperienced drivers, lowermilitary rank) fall asleep moreoften and to a greater extentthan other drivers. Thus resultsindicated that it is important toprovide drivers with more-in-vehicle, accessiblecountermeasures to alleviatefatigue.

Simons et al /2000 / TheNetherlands[91]

U, SLevel 3-3COSD 5

AviationSV: The effects of early starts on sleep, alertnessand vigilanceSC: Accident prevention

This prospective time seriesanalysis measured captainsflying short haul operationsduring two 4 week periods.Subjects were equipped with apalm top computer and anactigraph for objective andsubjective measurement ofsleep parameters, alertnessand performance on a vigilancedual task.SS: 6RR:N/A

It was found that pilotsreporting before 6am had asignificantly shorter total sleeptime, impaired sleep quality,and impaired performanceboth pre-flight and at top ofdescent.


Type ofdocument*



Comments / Findings


61

DS: Pilot self-reportWilliamson etal / 2000 /Australia [92]

P, S,Level 3-1COSD 4

TransportSV: Effects of alcohol and fatigueSC: Road crash risk

This case control studydeveloped measures todetermine the effects of fatiguecompared to alcoholconsumption on a series ofeight psychological tests onlong haul truck drivers andother workers in the transportindustry (controls). Participantsstayed awake for 28 hours inone occasion and on thesecond occasion drank alcoholuntil their blood level rose to0.1%.SS: 39RR: N/ADS: Employees volunteeredfrom both the transport industryand the army

Performance deficits due toalcohol were evident for alleight tests. Performance dueto sleep deprivation wasevident in most but not alltests. Sleep deprivation had noeffect on the Visual Search testor the Logical Reasoning test.Performance deficitsequivalent to 0.5% bloodalcohol concentration wereseen at 17-19 hours of sleepdeprivation.

Shanley et al/ 1999 / USA[93]

P, CSLevel 4

Chemical processingSV: Fatigue and shiftworkSC: Accident prevention

The case study describes asystem approach used by alarge chemical processingcompany that has a largeproportion of shiftworkers. Theorganisation implemented aseries of two types of educationprograms for workers andfamily and the other formanagers.

The company identified that atany one time 200 shiftworkershad control of over US$7billion of company assets fortwo-thirds of the plant'soperating hours. The companywas concerned about theimpact of shiftwork on staffhealth and life expectancy.Following the educationprogram employees reportedimproved sleep, improvedfamily life, less fatigue and lessgastrointestinal problems.


Type ofdocument*



Comments / Findings


62

Smith et al /1998 / UK[94]

P, LLevel 4

All types of industriesSV: 8 and 12-hour shiftsSC: safety in the workplace

The literature review examinesthe implications of effects thedifferent length of shift on arange of personal andperformance issues. The papercontains 85 citations.

The paper concludes that 12-hour shifts do not automaticallyinduce a significant decrementin safety.

Health andSafetyExecutive /1998 / UK[95]

P, SLevel 4COSD 7

ManufacturingSV: Size of organisationSC: Differences between large and smallorganisations and rates of injuries

A review of rates of fatal andnon-fatal work place injuriesreported under the Reporting ofInjuries, Diseases andDangerous OccurrencesRegulations 1995 (RIDDOR)between 1996/97 and 1997/98.SS: 86,761 employee injuriesthat can be attributable toorganisation size.RR: N/ADS: Injury statistics databases

Small workplaces (50 or lessemployees) have twice asmany fatal and amputationinjury in the workplace thanlarge organisations. The rate ofnon-fatal injury is smallest insmaller workplaces.

Gander et al /1998 / USA[96]

P, SLevel 3-3COSD 5

AviationSV: Fatigue experienced by pilotsSC: reduction of sleep loss on over-night cargooperations

This cohort of B-727 crewmembers was examined forphysical effects of fatigueexperienced by long flights overat least one time zone.SS: 34RR: N/ADS: Observations of simulatedflight

Sleep quality during the daywas not as good as at night.Total sleep time was 1.2 hoursshorter per 24-hour on dutydays. Crew members atesnacks and had increasedreports of headaches,congested nose and burningeyes during the long hauls.

Gander et al /1998 / NewZealand [97]

P, SLevel 4COSD 7

TransportSV: Fatigue management program for tankerdriversSC: Accident prevention

A prospective questionnairewas sent to all tanker driversworking for six Australasiancompanies addressingsleepiness levels and theprevalence of other risk factors,and possible effects of shiftcharacteristics.

Twenty-one percent of driversreported snoring every nightand one third of had a largeneck size > 42 cm. (Previousstudies have shown thataccident rates 42% higheramong drivers with large neckscompared with those with


Type ofdocument*



Comments / Findings


63

SS: 163 driversRR: 70%DS: Drivers’ self-report

small necks.) A drivereducation package wasdeveloped, managementeducation on driver fatigue,company medicalexaminations and rosteringguidelines were developed. Nosimple solution was found tofatigue management intrucking operations.

Gander et al /1998 / USA[98]

P, SLevel 3-3COSD 5

AviationSV: fatigue experienced by pilotsSC: avoid adverse events

This cohort of pilots use alogbook to record their sleep,heart rate, feelings of fatigue,hours on at work for multipleshort haul flights of 4.5 – 5.5hours in a duty day averaging10.6 hours over a 3-4 days ofwork. Comparisons were madebetween patterns duringworking periods and days-off.SS: 44RR: N/ADS: Pilot self-report

Sleep during a working weekwas on average an hourshorter. Pilots experiencedmore headaches, nosecongestion and back pain andhad greater difficulty in gettingto sleep. Their consumption ofcaffeine and snacks wasgreater while on duty.

Vega et al /1997 / USA[99]

P, SLevel 4COSD 5

PolicingSV: Hours of work in a weekSC: Quality of policing and productivity

A postal questionnaire wascompleted by police officerswho had worked a 40-hourweek over 5 days for 1 yearand in the subsequent yearworked 3 13-hour and 20-minute shifts a week for oneyear. Productivity data wasused from a police activitydatabase.SS: 34RR: 40.5%

An increase in productivity wasfound when police officersworked the compressed week.Increases were found in thenumber of cases handled andavailability to respond to callfor service. Ninety one percent of respondents werepositive about the compressworking week. Respondentsalso believed that there wasmore opportunity to gain


Type ofdocument*



Comments / Findings


64

DS: Police officers’ self-report additional employment,although the rate for this wasless during the period ofcompressed week than in the5-day working week period.There was more time to devoteto family members and leisureactivities.

Luna et al /1997 / USA[100]

P, SLevel 3-3COSD 5

AviationSV: Night shiftSC: Performance and aircraft safety

Cohort of air traffic controllerswas examined for sleep, mood,subjective assessment offatigue and general activitiesduring a 2.5 shift cycle periodcovering 2 weeks of duty shifts,post-shift, day-of-shift and dutylocation.SS: 43RR: UnspecifiedDS: Air traffic controllers’ self-report

There was significantly moresleep, fatigue, less vigour andconfusion, while on night-shiftcompared to day-shift.

Samel et al /1997 /Germany[101]

P, SLevel 3-3COSD 5

AviationSV: Hours of flight timesSC: Performance and aircraft safety

Cohort of flight crew wereexamined for sleep, task load,fatigue and stress measured byEEG, ECG, motor activity andsubjective ratings during andafter a 9-hour and 15-minunteflight, 13-hour and 30-minutelayover and 9-hour and 53-minute return flight.SS: 22 flight crewRR: N/ADS: Pilots’ self-report

Layover sleep was shortenedby 2 hours. 2 consecutive nightshifts had a total of 9.3 hourssleep loss. Increasing level offatigue was evident in-flightand was more pronounced onthe return flight home basemeasuring critical levels.Drowsiness, low state ofvigilance and alertness wereevident in both night flight butmore pronounced on the returnflight.


Type ofdocument*



Comments / Findings


65

Fell et al/1996 /Australia[102]

P, SLevel 3-2COSD 4

TransportSV: Driver fatigueSC: Minimising road accidents

This is a case control studyusing a telephone interview toexamine the prevalence andcharacteristics of driving fatiguein a sample who had a previousdriving incident and those whodid not (controls).SS: 301RR: UnspecifiedDS: Drivers’ self-report

Shift workers have a higherprevalence of driver-fatiguerelated accidents in the city.People traveling on work tripshave a high prevalence forfatigue related accidents incommuter times.

Emonson etal / 1995 /USA [103]

P, SLevel 4COSD 7

Aviation (Military)SV: FatigueSC: Effectiveness of performance, aircraft safety

A self-report questionnaire wascompleted by air force flightcrew deployed in operationDesert Shield / Desert Storm toassess the use ofamphetamines to overcomefatigue on-duty. Theseoperations involved flying east-ward for 15 hours across aminimum of 7 time zones,without auto-pilot andcompleting multiple air-to-airrefueling.SS: 464RR: 70%DS: Pilots’ self-report

Of the respondents, 65% usedamphetamines duringoperations. Most frequentindications for use wereaircrew fatigue and missiontype. 58 - 61% considered theiruse as beneficial or essentialto operations.Dextroamphetamine 5mg 4/24was used without side effects.

Smith et al /1995 / UK[104]

P, SLevel 3-3COSD 5

NuclearSV: Shiftwork effects on nuclear power workersSC: Staff safety

This prospective study used ahand held computer to recordlevels of alertness, cognitivetask performance and workloadratings every 2 hours overselected shifts.SS: 22 workersRR: 50% to 72% depending onthe task

Workload remained stableacross shifts. Night shift wasassociated with poor qualityand shorter sleeps withsignificantly lower levels ofalertness and poorerperformance on components ofthe performance measures. Nomajor difference by shift type


Type ofdocument*



Comments / Findings


66

DS: Shift workers’ self-report or time on shift interactioneffects were found.

Beilock /1995 / USA[105]

P, SLevel 4COSD 7

TransportSV: Hours of service and speed limit violationsamong tractor – trailer driversSC: Accident prevention

A prospective survey plusinterviews were conductedamong long distance truckdrivers on the way into or out ofthe Florida peninsula.SS: 498RR: 84%DS: Truck drivers’ self-report

Results showed thatdepending upon the averagespeed limit scenario, up to56% of drivers had violationsuspect schedules or violationinducing schedules. Solodrivers, refrigeration drivers,regular route drivers and thosewith longer trips were muchmore likely to have violationsuspect schedules. Theaverage driver drives 46 hoursper week and works a total of58 hours.

Williamson etal / 1995 /Australia[106]

P, SLevel 3-2COSD 5

All industriesSV: Time of shiftSC: Fatalities at work

An analysis of coronial reportswere examined over a 2-yearperiod. Accidents wereexamined by type of humanerror that related to the fatalityand the shift in which theaccident occurred.SS: 1020 casesRR: N/ADS: Coronial reports

Examination revealed that fatalaccidents more than doubledon night shift when controllingfor the proportion of workerson duty. Behavioral factorswere the most commoncauses, errors in automaticprocessing (skill-based error)were the most common andwere not specific to shift time,Rule-based errors were morecommon during the day andknowledge-based errors werethe most common in afternoonand night shifts.

Duchon /1994 / USA[107]

P,SLevel 3-1COSD 6

MiningSV: Extended WorkdaysSC: Decreasing worker fatigue

For this comparative study datawere collected in 2 phases. Thecontrol group was kept on theusual 8 hour shifts. In phase 1

Results indicated unanimousacceptance and improvedsleep quality with the 12 hourshifts. Fatigue sensitive


Type ofdocument*



Comments / Findings


67

experimental crew was on 8 hrbackward rotating. For phase 2,4x4 12hour shift schedule wasintroduced.SS: 31 experimental and 10control subjects were used(100% male)RR: N/ADS: Mine workers self-report

behavioral and physiologicalperformance measuresshowed either no change orimprovement with the 12 hourshifts.

French et al /1994 / USA[108]

P, SLevel 3-2COSD 5

Aviation (Military)SV: fatigueSC: Effectiveness of performance, aircraft safety

This cohort study examines theexperience of crew members inthree simulated B-1B bombermissions lasting 36 hours with33 – 35 hours rest betweenmissions.SS: 32RR: N/ADS: Flight crew self-report

Subjective fatigue, anger,confusion, depression andtension were significantlygreater in the first mission thenwith subsequent missions.Recovery from the missionsoccurred within 48 hours of thethird mission.

Parkes / 1993/ UK [109]

P, SLevel 4COSD 5

Offshore installationsSV: Shift workSC: Alertness

Offshore operators wereexamined for the effects of a 2-week offshore shift cycle ontheir cognitive performance.SS: 31RR: UnspecifiedDS: Offshore operators’ self-report

Alertness and cognitiveperformance were mostaffected adapting from nightshift to day shift and againadapting from day shift to nightshift. Disruptive effectsincluded changes in mood,sleep loss, low alertness andslowing of cognitiveperformance.

Salter et al /1993 / USA[110]

P, CSLevel 4

MilitarySV: Safe sleepingSC: Injuries and death

This paper describes 4incidents of army soldiersexperiencing an injury whilesleeping. These related tosleeping around a vehicle,inside or on a vehicle, in anapparently “safe” area and in

Soldiers were not seen by theircolleagues and experiencedeither injury or death by beingrun over or from moving partsor a tank. The paperrecommends the developmentof guidelines to assist army


Type ofdocument*



Comments / Findings


68

open areas. personnel when in the field.The paper suggests thatstandards be written on thelocation of where soldiersshould sleep. The averagenumber of hours on dutybefore a sleep-related accidentwas 16.8hours, indicating ahigh level of fatigue.

Phillips et al /1992 / UK[111]

P, SLevel 4COSD 7

Manufacturing (Paper products)SV: Shift workSC: Occupational injury

This paper examines thefactors that affect adaptation toshift work on a rotating rosterusing a self-reportquestionnaire.SS: 239RR: 48%DS: Factory workers in amanufacturing plant

Sixteen percent of the 239 inthe sample reported having aninjury since starting shift work.Significant predictors ofworkers’ ability to adapt to shiftwork included drowsiness thenext day, difficulty going tosleep, preference for meals atfixed times and ability to skip anight’s sleep.

Feyer et al /1992 /Australia[112]

P, SLevel 4COSD 7

TransportSV: Work practices and fatigue in long distancedriversSC: Accident prevention

This paper used a prospectivesurvey plus interviews to look atwork practices and fatigue inthe long distance road industry.SS: 960 long distance truckdrivers (658 – questionnaireand 302 – interviews)RR: UnspecifiedDS: Truck drivers’ self-report

Results showed that the mostcommon type of operationinvolved single drivers (89%),with most being employees ofmedium to large companies(76%), mean distance covered1259.8km with mean duration27 hours and 78% of driversdid some unloading. Mostdrivers (73.3%) reportedfeeling fatigued. Shorter tripswith more flexibility in timingand scheduling were related tolower levels of fatigue.

69


The papers in this review that focus on nurse staffing levels enhance ourunderstanding of nurse to patient ratios and its effect on patient outcomes. Giventhe large amount of resources that are required to provide patient care, thedevelopment of cost effective, efficient and safe models of care is of priority. Theliterature in both the health and non-health areas indicate that too few and toomany of the same type of staff can lead to poor patient outcomes. Increasing thisarea of research to include staffing models that are collaborative from a range ofprofessions, use a rostering system that is conducive to continuity of care yetenables adequate rest and sleep for clinicians can make a significant contributionto safe staffing and the promotion of patient safety.

There are some other promising areas for further research which are:1. Condensing the working week to enable longer shifts with more time away

from work2. Examining the impact of doctor to patient and allied health worker to

patient ratios and the impact on adverse events3. The development of a model that promotes continuity of care rather than

continuity of carer through the use of use of feedback and programming(such as the use of guidelines, policies and procedures)

4. The development of strategies that enable the inexperienced practitionerto work safely in the delivery of patient care such as specific education onmentoring, promotion of feedback and the use of checklists or protocols toclearly define roles and responsibilities

5. The promotion of effective messages on managing shiftwork, personal lifeand sleep.

70

4.5 Competence, supervision and staff mix

4.5.1 Competency, supervision and staff mix – summaryfindings

There were 91 health and 27 non-health papers included in this review thatexamined the characteristics of competent staff, staff supervision and staff mixand the relationship with patient safety. The following table provides a descriptionof the types of papers.

Health Non-Health

TOTAL



Type of paper

Research Study 52 12 641 0 0 02 4 0 43 21 4 25

Levels of Evidence

4 66 23 891 3 0 32 1 1 33 5 0 54 0 1 15 13 3 166 5 0 57 24 7 31



8 0 0 0

Papers that examined the characteristics of competent staff, staff supervisionand staff needs and its relationship on patient safety contributed a substantialvolume of information to this review. In relation to health, a large proportion ofpapers examined drug related adverse events, an area in the broader literaturealready well explored. A number of themes emerged relating primarily tocompetent staff and these are not only confined to staff knowledge and skills butalso to the systems and environment in which staff work. The following pointsemerged:

• Hospitals that have been reviewed because of high levels of patientcomplaints and adverse events were identified to have problems aroundcare coordination, poor management, poor documentation and a limited orabsent system for reporting adverse events.

• Patients have, in certain conditions, a considerably high risk of being re-admitted to hospital if discharged from hospital with a number of unstableconditions. One study [113] quantified the risk of death or readmission tohospital as being five times greater than the rest of the population if therewere two or more instabilities in the patients’ condition on discharge.

71

• Inexperienced or temporary staff, especially junior medical officers, posean increased risk to patient safety if not adequately supervised.Supervision is critically important as well as having clear processes forcommunication and role delineation.

• Patients who need an inter-hospital transfer may be placed at a higher riskof experiencing adverse events if the transport teams are not specialisedin inter-hospital transfers for the acutely ill.

• Delays in diagnosis may have a major impact on the longer term wellbeingof the patient in a number of clinical situations particularly acutemyocardial infarction.

• The development of communication processes for patients with lowliteracy levels or limited English skills may reduce adverse events.Diagrams, illustrations, symbols and signs may assist in helping patientsunderstand hazards and environmental dangers in a clinical area.

• Staff abuse of patients is clearly a serious issue and papers cited in thisreview identify the inability of other staff to intervene when aware of thesebehaviours.

Papers relating to staff working in non-health industries and related to workercompetency, supervision and worker needs provided similar content to the healthpapers however the commercial imperative was foremost in the analysis. Lostworker time was often reported as a company cost and papers would oftendescribe the responsibility staff had in protecting and maintaining companyassets. The following themes were identified:

• Inexperienced staff may contribute to accidents. Outsourcinginexperienced workers above employees because of potential cost savingmay increase risk of accident and injury.

• There is an economical imperative to identify and eliminate design faultsquickly which includes the implementation of rules and guidelines as wellas improving structures to ensure commercial viability.

• Poor judgment together with reduced visibility and ineffectivecommunication can lead to very serious, sometimes fatal accidents suchas rail and aviation disasters.

• A failure to perceive, comprehend and project situations where safety andaccident prevention is required may increase workers’ risk to creating anaccident.

The non-health industry literature contained a larger number of papers thatfocused on finding solutions to unsafe worker practices. These included:

• The implementation of an occupational health and safety program withinthe organisation. This program has the following characteristics ofleadership, organisational policy, hazard risk assessment and continuousimprovement. One company [114] greatly reduced there lost time injuryrates as a result of implementing such a program.

• The use of technology to prepare people for potentially hazardoussituations such as simulated mining experiences.

72

• The use of computer feedback on performance. This has been used in theaviation industry and was found to increase teamwork and performance.

• The use of video to improve visibility. This is a recommendation for largemining trucks where there was a risk to people on the ground of being runover.

• Improving the preparation of workers for their roles such as acomputerised site entry education program. The program was suggestedfor people working in confined spaces and would include emergency andrescue procedures.

4.5.2 Competency, supervision and staff mix – Health Literature Summary TablesAuthor / Year /Country

Type ofdocument*



Comments / Findings

* P = published or U = unpublished. L = literature review, S = study containing data, G = guideline or CS = case study.Hierarchy of study design rating = Level 1 -4 (See table 2) # COSD = classification of study design 1 - 8 (see table 1)**SS = sample size; RR = response rate; DS = data source

73

AustralianCouncil forSafety andQuality inHealth Care /2002 /Australia[115]

P, CS(series)Level 4

SV: Credentialing, role and responsibilities andquality monitoring processesPO: Serious adverse events and poor outcomesfor women and their families

This report outlines an inquiryinto health care delivery andhospital processes of a majortertiary hospital. The reviewinvolved information from over1600 patient files, 293 writtensubmissions, interviews from70 former hospital staff,consultant’s reports,comparisons between otherhospitals’ performance data,Case Studies: 106 transcriptsfrom current and formerhospital staff.

Findings included systemissues relating to sub-standardcare coordination, poormanagement of emergenciesand high-risk cases, lack ofsupervision of junior doctors,sub-standard documentationand non-existent systems forreporting adverse events.

Piotrowski etal / 2002 /USA [116]

P, GLevel 4

SV: CompetencyPO: Death and morbidity

This paper provides a set ofguidelines or standardsdeveloped as a result of fatalerrors in an intensive care unit.It uses examples of patientcase studies describing the useof specific check lists inminimizing errors.Citations: 18

The paper provides data oncompliance rates with thestandards by professionalgroup. The paper is not able todescribe the impact ofstandards and check lists onpatient outcomes due to dataand methodological limitations.

Halm et al /2002 / USA[113]

P, SLevel 3-2COSD 3

SV: Inappropriate hospital dischargePO: Improved quality of health care

The prospective observationalcohort study used informationon daily vital signs and clinicalstatus, to define and validate asimple, usable measure ofclinical stability on discharge forpatients with communityacquired pneumonia.SS: 680 patientsRR: N/ADS: patients with community-acquired pneumonia

Results showed that 19.1% ofpatients left the hospital withone or more instabilities.Instability on discharge wasassociated with higher riskadjusted rates of death orreadmission and failure toreturn to usual activities.Patients with 2 or moreinstabilities had a 5 five foldgreater risk-adjusted odds ofdeath or readmission.


Type ofdocument*



Comments / Findings


74

Paavilainenet al / 2002 /Finland [117]

P, SLevel 4COSD 7

SV: Identification of child maltreatment inhospitalsPO: Prevention of child patient mistreatment

This prospective crosssectional survey looked at hownurses and physicians ratedtheir ability to identify child(under 18) maltreatment(physical, sexual,psychological) by a parent ofcaregiver.SS: 513RR: 62%DS: Total population of staffcaring for children

Of the respondents 40%estimated that they had nevertreated a child maltreatmentcase and 2/3 believed that ifthey did they would be able toidentify the issues. However71% rated identification ofchild maltreatment as difficult.Staff believed that the mostdistinct signs to identifymaltreatment by werefractures, multiple bruises, andfrequent injuries.

Davidson etal / 2002 /UK [118]

P, SLevel 4COSD 7

SV: Management of ectopic pregnancyPO: Improved quality of healthcare for pregnantwomen

A prospective audit was carriedout on women diagnosed withectopic pregnancy, casenoteswere obtained as well as aquestionnaire sent to allspecialist registrars andconsultants within the unitasking about operative skill andconfidence in managing ectopicpregnancies.SS: 50 ectopic pregnantwomen and 36 physiciansRR: N/ADS: specialist registrars andconsultants

The ectopic pregnancy wasremoved laparoscopically in62% of cases with 80% beingdischarged on the firstpostoperative day. In aminority of cases elements ofsubstandard care were presentincluding failure to operatewhen ectopic pregnancy wasfound on ultrasound. Fewerthan 50% of physiciansreported competency inlaparoscopic management ofectopic pregnancy.

McDonnell etal / 2002 /USA [119]

P, SLevel 4COSD 7

SV: Hospital admissions from preventableadverse drug reactions (ADR’s)PO: Prevention of medication errors

A retrospective chart reviewover an 11 month period at auniversity hospital wasconducted to assess thepotential preventability ofADR’s directly related to apatients hospital admission.SS: 437 ADR’sRR: N/A

Of the 437 ADR’s, 158 werefound to be directly related tohospital admissions with97.4% due to drug exposure.62.3% of these wereconsidered preventable with24% considered lifethreatening. Most ADR’sresulted from inadequate


Type ofdocument*



Comments / Findings


75

DS: patient charts monitoring of therapy (67%) orinadequate dosing (51%).Patient non-compliance (33%)and drug interactions (26%)were also common causes.

Durie / 2002 /Australia[120]

P, SLevel 4COSD 7

SV: Incidents related to arterial canulationsPO: Prevention of medical errors

This report used data from theAustralian Incident MonitoringStudy (AIMS –ICU) database toidentify common problems andfactors associated with use andmaintenance of arterial lines.SS: 251 reports outlining 376accidentsRR: N/ADS: AIMS database

Of the incidents identified 15%described line insertionproblems, 66% line use andmaintenance problems and19% patient injuries. As aresult of the incident 49% ofpatients suffered no ill effect,28% minor effect and 15%major adverse effect. Thisstudy highlights the need touse meticulous insertiontechniques, line set up,securing, frequent lineassessment and the earlyremoval of lines that are nolonger necessary.

Dean et al /2002 / UK[121]

P, SLevel 4COSD 6

SV: Prescription errors in hospital inpatientsPO: Prevention on medical errors

A prospective study whichinterviewed pharmacists at aUK teaching hospital toestablish the prevalence ofmistakes made, causes oferrors and methods of errorreduction. Pharmacistsidentified prescribing errors andnotified the research team. Theteam invited the doctor to beinterviewed, complete aquestionnaire and the medicalrecord was examined.SS: 88 errors / 44 doctorsRR: Not specifiedDS: Doctors who worked at a

Skill based slips were mostfrequent (57%), rule basedmistakes (39%) and violations(4%). Seventy per cent of thetime doctors blamed beingbusy and 30% beinginterrupted.


Type ofdocument*



Comments / Findings


76

UK teaching hospitalPronovost etal / 2002 /USA [122]

P, CSLevel 4

SV: Safety in an intensive care unit (ICU)PO: Preventing patient adverse events

This single case study is usedas an example to outline thecomplex chain of medical andadministrative system failuresthat can result in an adverseevent.

After an uncomplicatedsurgical procedure, a 72 yr oldman developed hospitalacquired pneumonia and wasadmitted to ICU for oxygensupport, while in the ICU hedeveloped atrial fibrillation witha rapid ventricular response.Because the patient’s heartrate was 163 beats per minuteit was decide he would betreated with medication.Esmolol was administered andthe patient suffered cardiacarrest but was successfullyresuscitated. System failuresare identified.

Farbstein etal/ 2001/USA [123]

P, CSLevel 4

SV: Medication safety in hospitalsPO: Prevention of medical errors

This is a series of 6 casestudies used to identify 16 bestpractices to reduce adversedrug events across a networkof 6 hospitals.

Illustrates both long/short termstrategies in place such asreview of clinical order entry,computer systems, dispensingstations on wards, bedside barcoding and monthly meetingsto facilitate the safety process.

Kahn et al /2001 / USA[124]

P, SLevel 4COSD 5

SV: Ambulance driver competencyPO: Fatalities

This is retrospective analysiswith the objective of describingfatal ambulance crashcharacteristics.SS: 339 ambulance crashesinvolving fatalitiesRR: N/ADS: Road accident databasefrom 1987 – 1997

Most fatalities occurred whilstthe ambulance is driving in anemergency. The majority tofatalities occurred withpedestrians and not in theambulance. The most seriousand fatal injuries in theambulance were in the rear, toimproperly restrainedoccupants. Although the rateof ambulance officers with apoor driving record is similar to


Type ofdocument*



Comments / Findings


77

the driving population, thepaper concludes thatambulance officers shouldhave a higher level ofcompetence.

Shojanania(eds) et al /2001 / USA[125]

P, LLevel 3

SV: Technical skill, clinical decision-makingPO: Adverse events

This systematic literaturereview is a substantive volumeof work using a minimum oflevel 3 evidence for inclusion. Itsynthesizes evidence on 55topic areas

The study identifies a range ofclinical interventions thatimprove patient outcomeshowever, is limited inidentifying staffing variablesthat impact on patient safety.

Kinley et al /2001 / UK[126]

P, SLevel 2COSD 1

SV: Scope of nursing practicePO: Improved patient care

This prospective randomizedcontrol trial also includedqualitative assessment ofpatient care, staff perceptionsand economic evaluations.SS: 1907 patients (1874completed) who presented forassessment prior to generalanestheticRR: UnspecifiedDS: Four NHS hospitals in theUK

The study found that the use ofappropriately trained nurses(ATN’s) was acceptable topatients, and cost neutral inpre-operative assessment inelective general surgery.

Hebl et al /2001 / USA[127]

P,SLevel 4COSD 6

SV: Effect of weight on treatmentPO: Equality of patient care

Six cell randomized designwhere physicians evaluated apatient’s medical chart(average weight, overweight,obese) who presented with amigraine headache. Physiciansindicated how time was spentand which of the 41tests/procedures would beconducted.SS: 122 physicians in 1 of 3hospitals located in Houston,TexasRR: Unspecified

Physicians spent less time withheavier patients, prescribedmore tests, and viewed themmore negatively on 12 of 13indices. The study concludedthat physicians in Houston areplaying an important role in thelowering of the quality of healthcare that overweight andobese people receive.


Type ofdocument*



Comments / Findings


78

DS: Medical chartMorrison etal / 2001 /Australia[128]

P, SLevel 4COSD 7

SV: Nursing staff inexperiencePO: Eradication of adverse patient experiences inintensive care units (ICU’s)

Incidents related to nursingstaff inexperience (NSI) wereextracted from the AustralianIncident Monitoring Study inICU’s (AIMS – ICU) databaseand analysed using descriptivemethodology. The aim of thepaper was to identify effects ofincidents associated with NSIon quality of care and PO’s.SS: 735 reports covering 1472incidents were identified.RR: N/ADS: AIMS database

Of the incidents, 317 wererelated to airway ventilation,219 procedure lines andequipment, 234 patientenvironment, 234 unitmanagement. In 20% of thereports the result was a majorundesirable patient outcome.Therefore when rostering andemploying staff, NSI needs tobe considered.

Tang et al /2001 /Australia[129]

P, SLevel 4COSD 7

SV: Mental Health care for non-English speakersin SydneyPO: Quality of health care

A prospective survey wasconducted to identify the factorsaffecting the provision of qualityof care to people speaking alanguage other than English(LOTE) from a mental healthperspective in two area healthservices in Sydney.SS: 488 health practitionersRR: 56% (271 practitioners)DS: Randomly selected healthpractitioners from two AHS’s

It was found that only 39% ofrespondents were happy withthe quality of care. Shortage ofbilingual practitioners andinadequate organization ofservices were found to be thetwo factors affecting theprovision of the quality of care.

Emerson etal / 2001 /UK [130]

P, SLevel 3-2COSD 2

SV: Hospital pharmacistsPO: Reducing patient Adverse Events (AE’s)

This prospective observationalstudy examined the feasibilityof pharmacist led intensivemonitoring of adverse events(AE’s) associated with newlymarketed drugs.SS: 303 patients admitted tohospital who were prescribednewly marketed drugs duringtheir inpatient stay in 1998

Of the patients taking newdrugs there were 21 (7%)suspected ADR’s detected.The types of AE’s thatoccurred were broadly similarto those identified by generalpractice based prescriptionmonitoring. Howeverbiochemical changes didfeature more frequently than in


Type ofdocument*



Comments / Findings


79

RR: N/ADS: Inpatients who wereprescribed selected newlymarketed drugs

general practice.

Bouvier –Colle / 2001 /France [131]

P, SLevel 4COSD 7

SV: Severe obstetrical haemorrhagePO: Patient quality of care

This research used aretrospective questionnairesurvey design to determinewhat factors related to healthservices in France mightexplain substandard care ofsevere morbidity due toobstetric haemorrhage.SS: 165 women who had givenbirth in the previous yearRR: Not specifiedDS: Three administrativeregions in France

Of the 165 cases identified51% were vaginal, 19%operative vaginal and 30%caesarean. The leading causeof hemorrhage was uterineatony, with 68% of casesreceiving appropriate care,24% inadequate care and 14%mixed care. The main factorsassociated with sub-standardcare were lack of 24-hour onsite anesthetist and lowvolume of deliveries.

Sheikh et al /2001 / USA[132]

P, SLevel 4COSD 7

SV: Urban – Rural differences for Acutemyocardial infarction (AMI)PO: Quality of care

Data from medical records wasabstracted to see if the urban-rural differences in health careutilisation in Kansas exist in thequality of inpatient care forpatients with AMI.SS: 2521 medical recordsRR: N/ADS: patients (65+) who hadsurvived AMI from 12 urban, 31semi-rural, and 76 ruralhospitals

Patients with AMI received alower standard of health carein rural hospitals. This wasespecially true for use ofaspirin during hospital stay andat discharge, heparin, and IVnitroglycerin.

Sharma et al/ 2001 / India[133]

P, SLevel 4COSD 7

SV: TransfusionsPO: Prevention of medical errors

This paper used errors reportedby the blood bank staff and theresidents in charge of thepatient. The study time framewas over a period of 1 year andclassified based on the site ofoccurrence.SS: 123 errors

A total of 123 errors weredetected, of these 107 (86-99%) occurred outside theblood bank and 16 (13%)inside the blood bank. Errorsoutside the blood bank brokedown into bedside errors(99%) and in transit errors (6-


Type ofdocument*



Comments / Findings


80

RR: N/ADS: Blood bank staff

9%). Errors included: adifferent blood group from theoriginal sample, labeling errorssuch as overwriting thepatients name, mismatchbetween vial and centralregistration number andpatient’s name and requestform. Transit errors includedetachment of patient’sidentification label andincorrect labeling of bloodunits.

Callum et al /2001 / UK[134]

P, SLevel 4COSD 7

SV: Transfusion safetyPO: Improved mortality rates

A prospective audit oftransfusion related errors wasperformed to determine theability of a no-fault medicalevent reporting system fortransfusion medicine (MERS-TM) to identify the frequencyand patterns of errors.SS: 819 events over a 19month periodRR: N/ADS: Reported errors by medicalstaff

No serious patient outcomeoccurred in the study periodhowever 7.4% of bloodtransfusions were near-missesand were potentially lifethreatening or could have ledto permanent injury. Of mostconcern were 3 samplescollected from the wrongpatient, 13 mislabeledsamples, and 22 requests forblood from the wrong patient.The MERS-TM allowed therecognition, analysis of errors,determination of pattern errors,and monitoring for changes infrequency of errors aftercorrective was implemented.

Dunn et al /2001 / USA[135]

P, CSLevel 4

SV: Reduction in prescription errorsPO: Increasing patient safety

Three case studies are used toprovide examples of thehazards associated with thecommon Latin abbreviation of“qd” by showing copies ofactual medication orders. Other

An example from a case studywould be, ‘upon secondexamination of theprescription, both pharmacistsrealised the recommendeddosing interval of once daily


Type ofdocument*



Comments / Findings


81

dangerous abbreviations arealso discussed.

(“qd”) had been misinterpretedas 4 times a day (“qid”)’. Afterre-labeling the prescription vialthe patient was counseled.

White et al /2001 / UK[136]

P, SLevel 4COSD 7

SV: Adverse events following acupuncturePO: Patient safety

A prospective survey wasundertaken using intensiveevent monitoring to determinewhether the benefits ofacupuncture outweigh its risks.SS: 2178 adverse eventsoccurring in 31,822consultations.RR: Not clearly specifiedDS: 78 acupuncturists (alldoctors and physiotherapists)

The incidence of adverseevents was 684 per 10,000consultations. The mostcommon minor adverse eventsbeing bleeding, needling pain,and aggravation of symptoms(followed by a resolution ofsymptoms in 70% of cases).Avoidable adverse eventsincluded forgotten patients,needles left in patients,cellulitus and moxa burns.

Feldman etal / 2001 /USA [137]

P, CSLevel 4

SV: Abuse of pediatric patients by hospital staffPO: Improve patient safety

This study reports on aretrospective series of casesfrom one pediatric hospitalbetween 1982 and 1996, it alsoincludes a children’s hospitalsurvey conducted from 1990through 1995. The objective ofthis study was to describe achild protection team’s (CPT)efforts to develop andimplement a protocol forsystematic and management ofaccusations that hospital staffhave abused pediatric patients.Case Studies: 34 cases inhospital 1 and 27 hospitals inthe survey (RR:25%)

Thirty four cases of childabuse by staff were reviewed.Seventeen of physical abusecases included bruising,fractures, rough handling andverbal abuse and 18 sexualabuse complaints were madeincluding touching and sexualstatements. Complaints weresubstantiated in 23% of cases,with 1/3 of staff resigning orbeing fired. Results form thesurvey indicated that nearly60% knew of allegations ofabuse by staff members withonly 19% having internal policyon how to handle suchmatters.

Brilli et al /2001 / USA[138]

P, SLevel 4COSD 7

SV: Best practice in intensive carePO: Reducing medical error rates in intensivecare units

This paper outlines models ofbest practice in the intensivecare unit (ICU). It includes a

About twenty three per cent ofpatients were treated in an ICUutilizing full time intensivist


Type ofdocument*



Comments / Findings


82

non-systematic review of theliterature as well as data fromthe Committee on Manpowerfor the Pulmonary and CriticalCare societies (COMPACCS,1997).SS: COMPACCS – 5979 ICU’swith 72,500 beds with 77%occupancy rate – 53,000RR: NADS: 31 health care professionaland practitioners

model. In summary itconcludes that one model isnot unanimously supported asa model of best practice,however the best models allhave common elements suchas a multidisciplinaryapproach, physiciancomponent, nursingcomponent, pharmacycomponent and respiratorytherapy component.

Rex et al /2000 / USA[139]

P, SLevel 3-3COSD 5

SV: Temporary nursing staff, inexperienced staffor experienced staff in unfamiliar work area.Delays in communication, garbling ofcommunication or inability to verify accurateinformationPO: Serious adverse drug events

This study examines seriousadverse drug events in a majortertiary hospital in a 29-monthperiod. The paper describes theapplication of root causeanalysis to create change andimprove patient safety.SS: 23 ADE’sRR: N/ADS: Staff employed at thehospital

In the 29-month period ofreview, the hospital had 23serious adverse drug events.The main contributing factorsin the root cause analysis werefound to be 15 related to theuse of temporary,inexperienced or experiencedstaffing working in unfamiliarwork settings; and, 9 related tonight, weekend or change ofshift.

Kohn et al(eds) / 2000 /USA [140]

P, GLevel 4

SV: Improving staff performance and creation ofsystems to reduce human errorPO: Minimise the risk of adverse events in thehealth system

This extensive report reviewssystem errors in the healthservice delivery and providesrecommendations that aim tobuild safer delivery of healthcare.

The report contains 8 majorrecommendations in thefollowing areas: thedevelopment of a centre thathas as its focus, theimprovement of patient safetyand the development of aresearch agenda; nationalmandatory reporting system;development of voluntaryreporting programs; legalprotection of data and for peerreview; improve performance


Type ofdocument*



Comments / Findings


83

standards and expectations ofhealth care organisations;improve the performancestandards and expectations forhealth professionals; increaseattention on the pre and postmarketing of drugs; ensurethat patient safety is a clearaim within health servicedelivery; implementation ofmedication safety practices.

Glick et al /2000 / USA[141]

P, CSLevel 4

SV: Competency in diagnosing suspectedconversion disorder in an emergency departmentPO: Wrong diagnosis and inappropriate treatment

This paper describes 6 casestudies where patientspresenting to an emergencydepartment had significantneurological impairment andpain and were inappropriatelytreated. Cognitive andattitudinal factors by thephysician and caregivers areexamines.

The paper suggests 6 steps tominimize misdiagnosis whichincludes: the physician to beself-aware and reflectiverespecting difficult cases.

Wolff et al /2000 /Australia[142]

P, GLevel 4

SV: Leadership, competencyPO: Adverse events

This paper provides aframework for dealing withcomplex medical errors andorganizational factors to asuccessful clinical riskmanagement program.Citations: 18

The framework is theoreticaland is not supported with data.The framework is derived fromprevious work examiningmedical records for adverseevents.

Khunti / 2000/ UK [143]

P, SLevel 4COSD 5

SV: General Practitioner competencePO: Death

This descriptive studydescribes the use of autopsyreports in improving diagnosticskills for GPs.SS: 651 consecutive deathsover a four year periodRR: N/ADS: Six doctor inner citytraining practice

The study describes medicolegal reasons as the mostcommon for requesting anautopsy however there is valuein requesting clinical autopsiesto improve clinical care andreduce incorrect diagnoses orimprove the detection ofmedical conditions.


Type ofdocument*



Comments / Findings


84

Health andSafetyCommission/ 2000 / UK[144]

P, CSLevel 4

SV: Injuries in residential care homesPO: Injury prevention of employees and membersof the public

This report describes the maintypes of injuries sustained byemployees and staff acrossmany residential care settingsin the time period from 94/95 to98/99.Case studies: 3

6435 injuries were reported ofwhich 22 were fatal, 3743(58%) were fatal and 2670(41%) were over 3 day injuries.The key changes that wereimplemented includedsimplification of the RIDDORsystem in terms ofclassification of injuries,inclusion of suicide andtrespass on railways.

Hadfield /2000 / UK[145]

P, SLevel 4COSD 7

SV: Accountability in clinical supervision (CS)PO: Reduction in pediatric medical errors

This is a qualitative, exploratoryand descriptive study, usingsemi-structured interviews,which aims to gain anunderstanding from the ‘users’perspective of the impact of CSon pediatric nurses. Itreconstructs narratives in aninvestigative research format.SS unspecifiedRR unspecifiedDS female pediatric nurses

The paper describes CS, itsimpact on clinical practice andthe expected commitment toCS. It also examines theimportance of safety,impartiality, support, trust andrespect in CS and states thatthe CS relationship is valuablefor the development of goodpractice.

Espinosa etal / 2000 /USA [146]

P, CSLevel 4

SV: Reduction in number of clinically significanterrors on radiographs in the EmergencyDepartment (ED)PO: Improved quality of care

This single case study looks atlongitudinal data to review allclinically significant radiographerrors in the ED at monthlymeetings.

With the meetings in place andcase studies being used forteaching purposes the rate offalse negatives fell from 3% to1.2% and then to 0.3%. Thestudy concludes that systemsof radiograph interpretationsthat optimize physician skillscan significantly reduce errorrates.

Hourigan etal / 2000 /Australia[147]

P, SLevel 4COSD 7

SV: Decreasing patient treatment timePO: Quicker thrombolytic treatment for acutemyocardial infarction, mortality

A comparative observationalstudy using prospectivelycollected data. The interventionbeing that all patients with

Median door to needle timeswere less for patients treatedin the ED (37 minutes) than inthe CCU (80 minutes). Eighty


Type ofdocument*



Comments / Findings


85

acute myocardial infarctionreceive treatment by theemergency department (ED)instead of the usual coronarycare unit (CCU)SS: 89RR: N/ADS: Patients in a coronary careunit and emergency departmentof an Australian teachinghospital

three per cent of patients weretreated in less than 60minutes. Overall mortality wassimilar in the ED and the CCU.

Schenkel /2000 / USA[148]

P, LLevel 4

SV: Patient safety in Emergency Departments(ED)PO: Reduction in medical errors

This article reviews thedefinitions, detection andpresentation of error inmedicine and emergencymedicine (EM).Recommendations are madebased on the current literature.Citations: 127

Recommendations are madefor detection and reporting,teaching and prevention.

Thomas et al/ 2000 / USA[149]

P, SLevel 4COSD 7

SV: Adverse events in elderly patientsPO: Patient safety

This paper is a review of arandom sample of medicalrecords (2 stage process, stageI –nurses, stage 2- physicians)to detect adverse events. 2study investigators then judgedpreventability.SS: 15,000 hospitalisedpatients discharged in 1992RR: N/ADS: Hospital database

Results showed that theelderly (5.29%or 7419 adverseevents) had significantly moreadverse events than non-elderly patients (2.8% or 8901adverse events). Reviewersfelt that this was because ofthe clinical complexity of thecare rather than age baseddiscrimination.

Cote et al /2000 / USA[150]

P, SLevel 4COSD 7

SV: Medications used for sedation in pediatriccarePO: Prevention of adverse sedation events(ASE’s)

This paper used both a surveyof pediatric specialists as wellas data from case reports toperform a systematicinvestigation of medicationsassociated with adversesedation events in paediatric

Ninety-five incidents werefound to fulfill the incidentcriteria where the 4 reviewersagreed on causation. Sixtyincidents resulted in death orpermanent neurological injury.Adverse sedation events were


Type ofdocument*



Comments / Findings


86

patients.SS: 118 case reportsRR: N/ADS: Survey/ case reports

frequently associated with drugoverdoses and druginteractions, particularly where3 or more drugs were used.

Frey et al /2000 /Switzerland[151]

P, SLevel 4COSD 7

SV: Critical incident monitoring in a neonatalpediatric intensive care unitPO: Improve quality of care

This prospective surveyexamined the occurrence ofcritical incidents (CI’s) in orderto improve quality of care.SS: 467 admissions over a 1year periodRR: N/ADS: neonatal pediatric intensivecare unit (ICU)

CI’s were found to be verycommon in pediatric intensivecare. There were 211 IC’s,30% major, 25% moderate and45% minor. CI categories weremanagement/environment29%, drugs 29%, procedures18%, respiration 14%,equipment dysfunction 7% andnosocomial infections 3%.Most severe was CI’s involvingto drugs.

Oliver et al /2000 / UK[152]

P, LLevel 4

SV: Hospital fall prevention programsPO: Patient safety

This paper reviews theliterature with the objective ofanalysing hospital fallprevention programs todetermine whether there is anyeffect on fall rates, to review thequality of those programs andto provide direction for futureresearch.Papers Reviewed: 21 articles

Of the 21 only 10 containedsufficient data to allowcalculations of confidenceintervals. 3 papers wererandomised control trials, 7prospective studies withhistorical control. Theremaining 11 studiescontained data on fall ratesonly. To conclude individualcomponents of interventionsshowed no significant benefit.

Van denBemt / 2000 /TheNetherlands[153]

P, LLevel 4

SV: Drug related problems in hospitalised patientsPO: Improved quality of health care

This paper reviews theliterature related to drug relatedproblems such as medicationerrors (prescribing, dispensing,administration) and adversedrug events.Papers Reviewed: 115 articles

Of the 115 articles found 31were excluded. Of theremaining 84, 69 were originalinvestigations and were 15review articles. The paperlooks at definitions,frequencies, costs, causes andmethods for prevention of drugrelated problems.


Type ofdocument*



Comments / Findings


87

HCCC / 1999/ Australia[154]

P, CSLevel 4

SV: Surgical proceduresPO: Quality of surgical care

This report reviews 22 patientcases who experiencedadverse events after cataractsurgery at Dubbo Hospital.

It was found that 12 of the 19patients who underwentsurgery experienced significantpermanent damage to theircorneas and visualimpairment. Based on theevidence available it wasfound that the solution'eyestream' caused theadverse events. Thecommission maderecommendations in relation tosurgery at Dubbo Hospital.

HCCC / 1999/ Australia[155]

P, CSLevel 4

SV: Adverse events during surgeryPO: Quality of patient care

This report investigatesincidents in the operatingtheatre at Canterbury Hospital.SS: 24 patients were identifiedwho had a total of 28procedures with most of thepatients being exposed to thewrong solution

A number of problems wereidentified to the supply of asolution containing phenolduring Endoscopic RetrogradeCholangio PancreatographyERCP) procedures tooperating theatres. The reportprovides an outline of thesystems failures that occurredand recommendations toprevent further adverse eventsduring surgery.

Grant et al /1999 /Australia[156]

P, SLevel 3-3COSD 6

SV: Implementation of a Rapid Assessment Team(RAT)PO: Reduced waiting times

During a 3-month period a RATwas implemented along with acomputerized database torecord waiting times. Statisticsrecorded included medianwaiting time, length of stay and% of patients seen within theacceptable waiting time.SS: n = 5877RR: N/ADS: Doctor and triage nurse

The RAT reduced waitingtimes, 59% of patients wereseen within the accepted timelimit compared with 39% fromthe year before.

Pullen et al / P, S SV: Falls in hospitals This prospective study Results showed no increased


Type ofdocument*



Comments / Findings


88

1999 /European[157]

Level 3-3COSD 5

PO: Prevention of patient falls recorded all falls (time, place,risk factors, circumstances) in ageriatric hospital over a oneyear period.SS: 536 falls in a total patientpopulation of 1617RR: N/ADS: internal database ofgeriatric ward

frequency of falls were foundat any particular times,patients were not suffering fallswhile receiving therapy orbeing examined, but ratherwhile they were alone in theirrooms. 97.9% of fallshappened unobserved by staff.

Straand /1999 /Norway [158]

P, SLevel 4COSD 7

SV: General PracticePO: Improved quality of prescriptions for theelderly

This cross sectional, descriptivestudy aimed to describe thedrug prescribing in generalpractice for elderly patients.Patients’ age, sex, encounters,indications for prescriptions andthe occurrence of somepredefined inappropriate drugprescriptions were used.SS: patients = 16874,prescriptions = 16774RR: Not SpecifiedDS: General Practitioners

Prescriptions were issuedduring 2/3 of all contacts and63% of patients were female.On average, 99.4 prescriptionswere made per 100 GP-patientcontacts with 72.1% ofprescriptions were repeat.Prescriptions forcardiovascular diseasediagnosis made up 19% withinsomnia/anxiety/depressionmaking up 14%. The studyconcluded that inappropriatedrug prescriptions for theelderly are common in generalpractice.

Reid et al /1999 / USA[159]

P, CSLevel 4

SV: Technical competencePO: Misuse of peritoneal dialysis (PD) cyclers

This paper describes 2 casestudies involving the misuse ofequipment for patientsundergoing peritoneal dialysis.

The paper outlines thedesirable design andinstruction characteristics ofthe PD cycler machine.

La Duke /1999 / USA[160]

P, CSLevel 4

SV: Ambulance driver competencePO: Mortality and morbidity

A single case study thatdescribes issues around patientsafety when ambulance driversuse Emergency OperatingPrivileges which enables theambulance to use lights andsirens, to speed, go throughtraffic lights and disregard

This case study focuses onfactors relating to anambulance road accidentwhere a nurse was fatallyinjured. It recommendsimprovements in the design ofambulances to includeminimizing sharp corners and


Type ofdocument*



Comments / Findings


89

traffic laws. The paper cites anaverage of 385 ambulanceaccidents occurred each year(1993 -1999) resulting in 640injuries in the state of NewYork.

objects, securing objectsinside the vehicle, providingdriver training and providingsafety restraints for staff whiletreating patients.

Anonymous /1999 / UK[161]

P, CSLevel 4

SV: Staff corruptionPO: Patient safety from abuse and assault bothadult and child

This editorial is a summary of alarger report, the Fallon Inquiry.It looks at a single case studyto revise issues of patientabuse and assault by medicalstaff, physician incompetence,and the lack of security in largehospitals.

Highlighted is the need forcorrect security and hiringprocedures in all health caresettings.

Keyes et al /1999 / USA[162]

P, CSLevel 4

SV: Supervision of junior staffPO: Improvement in the quality of health care

This single case studyexamines the quality ofsupervision of junior staff andits relationship to quality ofhealth care.

The case study describes anelderly patient in the care of ajunior resident who becomescomatose overnight and dies 2weeks later as a resultextensive abdominal bleeding.The case outlines the need forappropriate supervision andlevels of responsibility withjunior medical residents.

Jelinek et al /1999 /Australia[163]

P, SLevel 3-3COSD 5

SV: Staffing and functional changes in anEmergency Department (ED)PO: Increased quality of patient care

This before and after studyused a range of variables tomeasure the changes in theED.SS: 35,000 per annumattendance with a 45%admission rate.RR: N/ADS: 573 bed in a teachinghospital in Western Australia

After making staffing andfunctional changes greatimprovements were foundclinically. Decreased waitingtimes, complaints rates,misdiagnosed fractures, timeto thrombolysis in acutemyocardial infarction andshortened length of stay forpatients with acute conditions.Data also indicatedimprovements in teaching andresearch.


Type ofdocument*



Comments / Findings


90

Leape et al /1999 / USA[164]

P, SLevel 2COSD 1

SV: Pharmacist participation on physician roundsPO: Prevention of medication errors

This randomised control trialused a before and aftercomparison including a controlgroup to measure the effect ofpharmacist participation onmedical rounds in the IntensiveCare Unit (ICU) on the rate ofpreventable adverse drugevents (ADE’s) caused byordering errors.SS: 75 patients (intervention)and 50 patients (control)RR: N/ADS: All admissions to the studyunit

With the pharmacist makingrounds with the ICU team,remaining in the ICU forconsultation in the morningand on call throughout the daythe rate of preventableordering ADE’s decreasedfrom 66% before interventionto 3.5% after the intervention,with the control groupremaining unchanged. ThusThe presence of thepharmacist substantiallylowered ADE’s caused byprescribing events.

Malpass et al/ 1999 /Australia[165]

P, SLevel 4COSD 7

SV: Adverse drug events (ADE’s)PO: Morbidity/ mortality

This retrospective medicalrecord review/ incidentmonitoring study used theAustralian Incident MonitoringSystem (AIMS) and the Qualityin Australian Health Care Study(QAHCS) databases to reviewpatient ADE’s.SS: AIMS – 6250 reports, 20%were ADE’s (1303) QAHCS – 2352 reports,17% were ADE’s (394)RR: N/ADS: AIMS database

Using data from two differentsources this study supports thefindings of previous studies ofADE’s concluding that errors inthe use of medication are asignificant contributor to injuryand illness.

Graber /1999 / USA[166]

P, SLevel 3-3COSD 5

SV: Expanding the goals of peer review to detectboth practitioner and system errorPO: Prevention of medical errors

A peer review committeeperformed all peer reviews for amedical service. Four areaswere discussed, quality of care,practitioner negligence, systemflaws as a contributing factorand elimination of system flaws.SS: 40 physicians

Comparison of results showedthat the intervention did notchange the identification ofpractitioner negligence (pre21% to post 25%) but that itsignificantly increased thedetection of system errors (pre11.5% and post 45.5%) and


Type ofdocument*



Comments / Findings


91

RR: N/ADS: Veterans AdministrationMedical Service

the number of improvementprojects per year (pre 1 andpost 12).

Bates et al /1999 / USA[167]

P, GLevel 4

SV: Frequency, consequences and prevention ofAdverse drug events (ADE’s)PO: Improved quality of health care

Descriptive data was gatheredto assess medication errorsand recommend guidelines forprevention of ADE’s.This paper uses also uses datafrom a previous study at theBrigham Women's' andMassachusetts GeneralHospital in Boston (SS: n= 530)to outline the importance ofmedical errors and makesrecommendations for strategiesto be used preventing ADE’s.Citations: 12

It was found that 1 in 100medication errors resulted inan ADE with dose errors beingthe most common (53%).Seven general preventionstrategies are looked atincluding standard processesfor prescribers to entermedication orders directly intocomputer systems, medicationbar coding and bettermonitoring and reportingsystems for ADE’s.

Fitzpatrick /1998 / USA[168]

U, SLevel 2COSD 3

SV: Strategies for chest painPO: Safe, early discharge for chest pain

Randomized prospective studyusing 2 cohorts of patientspresenting to the emergencydepartment (ED). Two studieswere run simultaneously. Study1 to validate the use of RSalgorithms and 2 to costeffectiveness and safety ofearly discharge.SS: Not specifiedRR: Not specifiedDS: all first time chest painpatients presenting to thehospital – 12/1 to 5/597

Early discharge was found tobe cost effective and butfurther research is needed onthe safety of the procedure.ED staff correctly predictedpatient risk 77% of the timeand only underestimated in 5%of cases. Twelve per cent wereinappropriate admissions and21% were dischargedinappropriately.

Taylor-Adams et al /1998 / UK[169]

P, CSLevel 4

SV - High workload, failure to identify clinical risk,competency, inadequate leadership, poor morale,inadequate supervision of junior doctors,inadequate communicationPO: Postpartum haemorrhage

A single case study thatdescribes a structured andsystematic approach to theinvestigation of adverse events.The case study focuses on awoman experiencing a major

This descriptive study is limitedto examination of one case.This paper provides a usefulchecklist (using 28 questions)described as PerformanceInfluencing Factors (PIFs). The


Type ofdocument*



Comments / Findings


92

postpartum haemorrhage andfactors surrounding her inreceiving treatment. Theframework involves theexamination of active failures,triggering factors and latentfailures.

PIF includes availability tocase notes, language/culturalproblems, agreement oninterpretation of test resultsand effectiveness ofcommunication between theclinical team.

Stearley et al/ 1998 / USA[170]

P, SLevel 3-1COSD 7

SV: Intra-hospital transportationPO: Decreasing mortality and morbidity

Monitoring and interventiondata were collected forinstances of patienttransportation between ahospital intensive care unit andits radiology suite. A speciallytrained team was put in placeand results compared withnational studies oncomplication rates associatedwith intra-hospitaltransportation of patients.SS: 237 instancesRR: N/ADS: Reports of instances duringtransportation

Patients moved by thespecially trained transportteam had a 15.5% overallcomplication rate (10.2%minor, 2.5% moderate and2.8% severe). The reportednational complication rates areas high as 75%.

Grilli et al /1998 / Italy[171]

P, LLevel 3 –3

SV: Medical specializationPO: Mortality at 3 and 5 years, optimal treatment,number and type of surgical interventions, pain

A systematic review of literaturethat attempts to identify medicalspecialization andspecialisation of hospital anddifferences in outcomes forcancer patients.Papers Reviewed: 56

The results are inconclusivedue to the methodologicalflaws in the literature.

Senior / 1998/ Australia[172]

P, SLevel 3-3COSD 5

SV: Initiation of medical care before review by theconsultantPO: Earlier treatment of patients experiencing amyocardial infarct

This is retrospective studywhich describes the use ofauditing medical records todetermine delays frompresentation to the emergencydepartment until receivingthrombolyic therapy.

The 5 strategies includingavailability of an ECGmachine, changing staffprotocols is thought to havereduced these delays fromgreater than 60 minutes to amean of 43 minutes.


Type ofdocument*



Comments / Findings


93

SS: 129RR: N/ADS: Hospital medical records

Bullard /1998 /Taiwan [173]

P, SLevel 4COSD 5

SV: Compliance with laws during inter-hospitaltransferPO: increase in patient safety during inter-hospitaltransfers

A prospective, cross sectional,observational study conductedat the Linkou Chang Gungmemorial hospital from otherhealth care centers,SS: 1056 patients transferred in1997RR: N/ADS: Reports of complicationsduring inter-hospital transfers

357 were critically ill, and only160 had received pre-transferstabilization. Whilst beingtransferred major omissionsincluded failure to intubate, nointravenous line, inadequate ivlines, lack of instruction. Mosttransfers were unaccompaniedby physicians or nurses withphone contact only 10% of thetime.

Wilson et al /1998 / UK[174]

P, SLevel 3-2COSD 3

SV: Medication errors in pediatric practicePO: Reduction in medication errors

A 2 year prospective cohortstudy was undertaken using anadverse incident reportingscheme, with the objective ofassessing incidence andconsequences of medicationerrors, highlight sources ofrecurrent error and institutechanges in practice to preventtheir recurrence.SS: 441 reported medicationerrors in the study periodRR: N/ADS: 682 patients were admittedfor 5315 inpatient days

The data showed that errorswere 7 times more likely tooccur in the intensive caresetting. Doctors accounted for72% of errors with errorsdoubling when new doctorsjoined the medication rounds.Most errors (68%) weredetected prior to drugadministration. During the 2nd

year of the scheme theincidence of all reported errorsfell, administration errors andserious errors fell but theprescription error rateremained constant. Toconclude errors occurredcommonly in this study butadverse consequences wererare.

Lattimer et al/ 1998 / UK[175]

P, SLevel 2COSD 1

SV: Nurse telephone consultationPO: Patient safety

A block randomised controlledtrial over 1 year to determinethe safety and effectiveness of

Nurses managed 49.8% of thecalls during the interventionwithout referral to a general


Type ofdocument*



Comments / Findings


94

nurse telephone consultation inout of hours primary care byinvestigating adverse eventsand the management of calls.SS: 14492 calls (7308 control,7184 intervention)RR: N/ADS: Internal database ofoutcomes

practitioner (GP). A 69%reduction in telephone advicefrom a GP and a 38%reduction in patientattendance, 23% reduction inhome visits were observedduring the intervention. Nursetelephone consultationprovided a 50% reduction inGP workload with no increasein adverse events.

Dale et al /1997 / USA[176]

P, SLevel 4COSD 6

SV: Wristband ErrorsPO: Increased accuracy and quality of patientcare

This study compared wristbanderrors for 204 small hospitalswith the objective of finding theaccuracy rates of patientidentification.SS: 451,436RR: unspecifiedDS: Hospital wristbands

Of the 451,436 wristbandsexamined 25,800 had errorsidentified (error rate foridentification 5.7%). Of theerrors reported 64.6% had nowristband, 12.4% had missinginformation, 12.1% hadmultiple wristbands withinconsistent information, 6.7%had illegible writing and 3.5%had erroneous information.

O’Hara et al /1997 /Australia[177]

P, SLevel 3-3COSD 5

SV: CompetencyPO: External injury to patients includingmisadventures, abnormal reactions to surgicaland medical procedures, drugs and biologicalsubstances causing an adverse effect.

This cross-sectional studydescribes the prevalence ofadverse events in using a state-wide database of all public andprivate acute-care hospitals.SS: 1248021 separationsRR: N/ADS: 135 public and 112 privatehospitals

Five percent (62,949) ofhospital separations had anadverse event recorded on themedical record. Surgicalcomplications were the mostfrequent adverse event (80%),adverse drug effects (19.3%)and misadventures (1.7%).The most commonmisadventure was anaccidental cut, puncture,perforation, haemorrhage(76%). This study providescomprehensive large scaleprevalence rates. Caution


Type ofdocument*



Comments / Findings


95

needs to be exercised on thevariations between public andprivate hospitals as theseriousness of the adverseevent is not measured andpatients from a private hospitalmay transfer to a publichospital for care.

No author /1997 / USA[178]

P, CSLevel 4

SV: Nursing competencePO: death / suicide

This single case studydescribes the incident of apsychiatrically ill patienttransferred from a psychiatricunit to an emergencydepartment with heartproblems. After staying in theemergency room for 48 hours,the patient left the area andwas found soon after leavingthe hospital in a dumpster. Hehad hanged himself.

The patient’s wife sued thehospital for failing to provideadequate care. The nurse wasblamed for not acting on thepatient leaving the hospital andfailing to call security.

Brown et al /1997 / USA[179]

P, CSLevel 4

SV: Patient controlled analgesia pumpsPO: Patient safety and error reduction

This case report outlines 9 casestudies in which patientcontrolled analgesia pumpswere responsible for anadverse patient event. Threemain areas are looked atnamely, shipping insert(packaging), drug concentrationmis-programming and devicepackaging.

For example one 42 year oldwomen was found markedlysedated and with respiratorydepression. It was found thatthe pump had been mis-programmed for a morphineconcentration of 1.0 mg/mlinstead of 5.0mg/ml, in whichthe patient received a 10-foldoverdose.

Anonymous /1997 / USA[180]

P, CS,Level 4

SV: Nurse knowledge of medicationPO: Prevention of adverse drug reactions

Discussed in this paper is asingle case study in which adrug was administered, thepatient was not monitoredcorrectly for an adverse effectwhich lead to the patient dying70 days later.

The result of the case wasruled as 70% nurse fault and30% physician fault. Thecourts decision was based onthe general principle that it is anurse’s responsibility in caringfor a patient to appreciate the


Type ofdocument*



Comments / Findings


96

potential for an adverse drugreaction and to perceive whenone is taking place.


P, CS,Level 4

SV: Staff responsibility for patient escapePO: Prevention of patient self harm

Discussed in this paper is asingle case study in which apatient was known to be anelopement and suicide risk, thepatient escaped from the unitand was hit by a truck whichkilled him.

The court ruled that when apsychiatric patient has beenassessed as potentiallysuicidal, suicide andelopement precautions mustbe undertaken at once(continuously monitored, ortransferred to a more securesetting).


P, CS,Level 4

SV: Standards for medical surgical nurses caringfor a psych patientPO: Accident and injury prevention

Discussed in this paper is asingle case study in which apsychiatrically ill patient wasput into acute care hospital. Hebecame paranoid anddelusional and jumped out a 3storey window suffering minorinjuries.

The nurses were found to benot negligent as they wereacute medical care nurses andhad no training in psychiatricnursing.

Anonymous /1996 / USA

P, CS,Level 4

SV: Staff conduct and reliabilityPO: prevention of sexual assault

This single case study outlinesthe incidence of sexual assaultin a small 131 bed hospital.

Victims of the assaults settledout of court. Although allpatients reported the assaultsat the time they occurred theirclaims were not takenseriously. So identified is theneed to listen to patients andrun employee backgroundchecks.

Coroner’sOfficeVictoria /1996 /Australia[183]

P, CSLevel 4

SV: Staffing level at nightPO: Reduce the risk of death from fire

This single case study outlinesthe coroners report describesfindings of the inquest into thedeaths of 9 people withintellectual disabilities who livedin a residential care facility. Thefire was thought to have startedwith 1 resident who, having a

A range of issues included theneed for the state governmentto have up-graded the firesystem in the facility, thestaffing levels of 2 staff onnight shift were identified asbeing deficient to adequatelydeal with a fire in a facility


Type ofdocument*



Comments / Findings


97

fascination for lighters, set fireto his bed withoutunderstanding theconsequences.

known to be high risk of fire,training of staff on evacuationand fire safety, identificationand early management of theresidents with fire-lightingbehaviours.

Wright / 1996/ USA [184]

P, CSLevel 4

SV: IlliteracyPO: Adverse events

This paper describes two casestudies. One where a patientignites his bed while on oxygensmoking a cigarette. He couldnot read the “no smoking” sign.The second case studyprovides an example of acasual cleaner notunderstanding the red wastebag labeled “radioactive waste”.

The paper concludes that theintroduction of signs, audioand video instructions are animportant aspect of riskmanagement for staff andpatients with low literacylevels.

Beckmann etal / 1996 /Australia[185]

P, SLevel 3-2COSD 6

SV: Australian Incident Monitoring Study inIntensive Care (AIMS-ICU)PO: Patient safety and quality of care

The AIMS-ICU is a prospectivecohort study set up to develop,introduce and evaluate ananonymous voluntary incidentreporting system for intensivecare.SS: 536 reports, whichidentified 610 incidentsRR; N/ADS: 7 ICU’s

Incident breakdown includedairway (20%), procedures(23%), patient environment(21%), and ICU management(9%). Incidents were mostfrequently detected byrechecking the patient or theequipment or by priorexperience. Incidentmonitoring was found to be asuccessful technique forimproving patient safety in theICU.

Short et al /1996 / HongKong [186]

P, SLevel 3-2COSD 5

SV: Incident reporting in anesthesiaPO: Improved quality of care

The role of an anestheticreporting incident program inimproving anesthetic safetywas studied using aprospective reportingprocedure.SS: incidents reported at 3hospitals was 1000 over a 4

Results showed that 69% ofincidents were considered tobe preventable with humanerror contributing to 76% of allincidents, and violation ofpractice contributing to 30%.The program was effective inits ability to detect latent errors


Type ofdocument*



Comments / Findings


98

year periodRR: N/ADS: Incident reporting programin a Hong Kong hospital

in the anesthesia system, andwhen these were corrected itwas found that incidents didnot recur.

Spittal et al/1995 / UK[187]

P, SLevel 4COSD 7

SV: Critical incidents attributable to anesthesiaPO: Reduction in medical errors associated withanesthesia

The study is a prospectiveanalysis of adverse anestheticevents and their consequencesin a general hospital with 300surgical beds over one year.SS: 338 critical incidentsRR: N/ADS: record of adverse eventsby anesthesiologists

Critical incidents werecategorized into airway (29%)circulation (28%) patient (5%)equipment (31%) andpharmacology (7%). Incidentfrequency was 6.68% or 1 in15. Morbidity was 0.53%. Thestudy strongly recommendsthe monitoring of standardsand procedures in anesthetics.


P, SLevel 3-2COSD 3

SV: System analysis of adverse drug events(ADE’s)PO: Prevention of patient ADE’s

This study used a systemanalysis of events from aprospective cohort study toidentify and evaluate thesystems failures that underlieerrors causing ADE’s andpotential ADE’s.SS: 334 errorsRR: N/ADS: All admissions to 11medical and surgical units in 2tertiary hospitals over a 6month period

Results showed 334 errorsthat were detected as thecauses of 264 preventableADE’s. Sixteen major systemsfailures were identified as theunderlying causes of theerrors. Most common errorswere, dissemination of drugknowledge (29%), inadequatepatient information (18%) withseven systems failuresaccounting for 78% of theerrors.

Tammelleo /1995 / USA[189]

P, CSLevel 4

SV: Negligence on admission for nursesPO: Prevention of patient injuries and accidents

This single case study providesan example of a patient beingadmitted to hospital and thenfalling out of bed due totwisting. He sued the hospitalfor failure to exercisereasonable and ordinary care.The hospital denied theallegations saying that he hadnot filed an experts affidavit in

The court ruled that the patientwas required to attach theexperts affidavit to his claim.


Type ofdocument*



Comments / Findings


99

support of his claim.Tammelleo /1995 / USA[190]

P, CSLevel 4

SV: Supervision of nurse aidesPO: Improved quality of care

This single case studydescribes a situation in which anurse manager failed to reportmisconduct of a nurses’ aid.

The court ruled that the nursemanager was required toreport the behaviour and thenurse manager was dismissedfrom her position.


P, CSLevel 4

SV: Sub-standard quality of carePO: Improved quality of health care

This single case studydescribes a situation in which aman was allegedly given sub-standard medical care, whichled to the symptoms becomingworse. He had only one witnessto support his allegations.

The witness was unable totestify.


P, CSLevel 4

SV: Timeliness in nurses’ notification of patientssymptomsPO: Improved quality of health care

This single case studydescribes a situation in which acase of abdominal pain was notrecognized as being caused bypregnancy.

The patient brought a wrongfuldeath claim against thephysician. The court found thephysician negligent.

Evans et al /1994 / USA[191]

P, SLevel 3-3COSD 5

SV: Hospitalised patientsPO: Preventing adverse drug events

Prospective intervention study,which aimed to look at theeffects of a computerisedsurveillance system in order toidentify methods to reduce thenumber of type b (allergic oridiosyncratic reactions) ADE’sin, hospitalised patients.SS: 79 719 hospitalisedpatients during a 44 monthperiod.RR; N/ADS: Computerised reportingsystem of ADE’s

Results showed that in the firststudy period 56 ADE’s wereidentified during 120,213patient days. In the secondstudy period 8 occurred in113,237 patient days, 18during 107,868 patient days.Thus the study concluded thatprospective surveillance ofcomputer based medicalrecords for known drugallergies and appropriate drugadministration rates canreduce the number of type bADE’s.

Barry et al /1994 / UK[192]

P, SLevel 4COSD 7

SV: Adverse events (AE’s) during inter-hospitaltransfer of the critically illPO: Improved quality of care

This observational study lookedat children transferred over asix month period to determinewhat complications children

Results showed that 75% ofthe children had adverseclinical events during transfer.In 13 cases the event was life


Type ofdocument*



Comments / Findings


100

have during inter-hospitaltransfer for intensive care andhow often these complicationsoccur.SS: 56 childrenRR: N/ADS: AE reports

threatening. Children whosubsequently died were morelikely to have had complicatedtransfers than who survived.

Cohen et al /1994 / USA[193]

P, CSLevel 4

SV: Dangerous abbreviationsPO: Prevention of medication errors

Discussed in this paper are twocase studies describingpotential adverse eventsrelating to drug administration.

A nurse misread ‘IVR’ to be‘IVP’ and administered a drugincorrectly. The abbreviation‘IVR’ was not usually used atthis hospital as it was thoughtto be too confusing. Thesecond case study discusses asituation in which a first yearmedical resident ordered aninjection for a child that wastoo potent. The pharmacistthought it was unusual andquestioned the doctor and theadverse event was averted.


P, SLevel 3-3COSD 7

SV: Technical competencePO: Adverse events

A medical record review of1133 patients with adverseevents was examined by aphysician to determine if thesewere avoidable and rankedthem into categories ofseriousness, and type ofmistake in performance orthought.SS: 1133RR; N/ADS: Medical records

The four most common typesof medical error were error indiagnosis, technical error, drugerror and inappropriate care.Over 75% were classified aspreventable. The mostcommon errors were drugrelated (19.6%), woundinfections (13.6%), technicalcomplication relating tosurgery (12.9%).

Singleton etal / 1993 /Australia[195]

P, SLevel 4COSD 7

SV: Physical injuries and environmental safety inanesthesiaPO: Reduced incidence of injury

This case report study uses theAustralian Incident MonitoringStudy (AIMS) to look at injuriesto both staff and patients.

Of the first 2000 incidentsreported to AIMS, 56 involvedphysical hazards or injuries topatients and staff. Theses


Type ofdocument*



Comments / Findings


101

SS: 2000RR; N/ADS: AIMS database

were categorised as oraltrauma (17), operating table(10), dermal or epithelial injury(6), electrical hazards (6),patient transport (5),monitoring hazards (4),needlestick (4) andmiscellaneous (4).

Van Der Waltet al / 1993 /Australia[196]

P, SLevel 4COSD 7

SV: Recovery room incidentsPO: Prevention of accidents and injuries

This paper uses the first 2000incidents reported on theAustralian Incident MonitoringStudy (AIMS).SS: 2000RR; N/ADS: AIMS database

Of the first 2000 incidents, 6%occurred in the recovery room,over 2/3 involved respiratorysystem, 19% werecardiovascular, 3% involvedthe nervous system 9% weremiscellaneous in nature.


P, CSLevel 4

SV: Nursing competencePO: Abduction of a baby

This single case studydescribes an incident of a nurserecognizing a woman’srepeated attempts to hold ababy in a maternity ward.

The hospital had a clear policyon name badges for staff andthis was communicated in ahospital protocol for patientsand visitors. This alerted avisitor to whom the allegedabductor requested wanting tosee the baby. A nurserecognized inappropriatebehaviour and alerted securitywho then intervened.

Barboni et al /1993 / Italy[198]

P, SLevel 3-3COSD 5

SV: Effects of a quality improvement project inemergency treatment of bronchospastic attacks(BA) in the Emergency department (ED)PO: Quality of patient care

This study was conducted in 3phases. Phase 1 was aretrospective analysis of 20cases. Phase 2 researchersassessed the treatmentprovided to patientsexperiencing BA and measuredpatient length of stay andoutcomes such as a clinicalcondition on discharge. Phase3 assessed the same

Results showed that after theimplementation of theguidelines there was a 56%improvement in physicianbehaviour. An improvementwas also shown in theoutcome of patients treated forBA in the ED.


Type ofdocument*



Comments / Findings


102

parameters as phase 2 but afterthe guidelines wereimplementedSS: Phase 1 – 22 cases, Phase2 – 33 forms, Phase 3 – 59formsRR: N/ADS: ED, General Hospital ofItaly

Bates et al /1993 / USA[199]

P, SLevel 3COSD 3

SV: Incidence and preventability of adverse drugevents (ADE’s)PO: decrease in ADE’s

Prospective cohort study in anurban tertiary care hospital,which evaluated the incidenceand preventability of adverseevents and determined the yieldof several strategies foridentifying them.SS: 2976 patients across sevenunits over a 37 day periodRR: N/ADS: 7 units in an urban tertiarycare hospital

Results showed that 73 ADE’swere found, 27 were ADE’s, 34potential ADE’s and 12problem orders. Physicianswere responsible for 72% ofthe incidents with theremainder divided evenlybetween nursing, pharmacy,and clerical. Of the 27 ADE’s,5 were life threatening, 9 wereserious, 13 were serious. 56%were judged as preventable.

McKee et al/1992 / UK[200]

P, LLevel 4

SV: Junior Medical DoctorsPO: Increased quality of patient care

The study uses a literaturereview and interviews withjunior doctors to look at effect ofthe current system of hospitalmedical staffing (junior doctors)on quality of care.SS: 62 interviews wereconducted but number ofarticles reviewed was not statedRR: Not specifiedDS: Junior doctors from fournational health service hospitalsin South East London.

The paper suggests that thecurrent use of junior medicaldoctors decreases the qualityof patient care throughmistakes associated withinadequate supervision andtiredness. It concludes that thesystem needs to change inorder to provide supportiveworking environments forjunior doctors.

Bulau / 1992/ USA [201]

P, CSLevel 4

SV: Failure to treat a technology dependentinfant in a home healthcare settingPO: Accident prevention

This single case study presentsthe history of a situation thatwas mismanaged and resulted

During the discovery phase ofthe lawsuit, several significantissues were identified


Type ofdocument*



Comments / Findings


103

in a very large payout for thepatient who was severelyimpaired neurologically as aresult of poor crisis intervention.It highlights individualresponsibility in providing safehigh care quality in the home.

regarding the standard of carefor providing skilled nursingservices, staff qualifications,and nursing staff supervision.Standard of care, lack of casemanagement, lack of a plan ofcare, lack of an admissionnursing assessment, lack ofphysician orders, unsafe clientenvironment and lack ofemergency care were all foundto contribute to themismanagement of the case.

4.5.3 Competency, supervision and staff mix – Non- Health Literature Summary TablesAuthor / Year /Country

Type ofdocument*



Comments / Findings


104

Fowler / 2002/ UK [202]

P, SLevel 4COSD 7

TransportSV: Workplace control measuresSC: Reduce risk

This study used a mail-outquestionnaire to companiesfrom each identified industry.The main objectives being togain information aboutworkplace transport safety,including the number ofvehicles on site, safe worksystems and use of protectiveequipment.SS: 2000RR: Not specifiedDS: questionnaire sent tocompanies

All companies reported havingimplemented some measuresto control workplace safetywith 52% already havingconducted a workplace riskassessment.

Anonymous /2002 / France[203]

P, CSLevel 4

AviationSV: Causes of the 2000 Concorde crashSC: Accident and death prevention

This single case study looks atthe causes of the 2000Concorde crash at Charles deGaulle airport that killed 109passengers and 4 people onthe ground.

The report states that a metalstrip that fell off an earlierplane burst the Concorde’styres. The explosion sentrubber debris toward the fueltank and started a fuel leakand fire that caused the planeto crash. All Concorde aircrafthave since been fitted with fueltank liners of bullet proofKevlar, a reinforcedundercarriage and strongertyres.

Anonymous /2001 / Brazil[204]

P, CSLevel 4

Mining – OffshoreSV: Oil industry accidentsSC: Accident and injury prevention/Environmental protection

The single case study looks atthe sinking of the world’slargest oil platform (40 storeyshigh, costing $350 million). Italso describes other accidentsin the industry since 1980.

Two people were killed in theincident with another 9 missingpresumed dead. The accidentis the worst since 1984 were36 people were killed in anexplosion. There wereconcerns about the use of out-sourcing to inexperiencedworkers at the cost of savingmoney.


Type ofdocument*



Comments / Findings


105

Struttman etal / 2001 /USA [205]

P, CSLevel 4

LoggingSV: Fatal injuries in LoggingSC: Prevention of fatal injuries

This case report looks at 7 casestudies, which outline the fatalinjuries caused by logs rollingoff trucks. The deaths wereidentifies by the KentuckyFatality Assessment andControl Evaluation Project.

From 1994 to 1998, 7 incidentswere reported in which aworker was killed by a logrolling off a truck, accountingfor 15% of deaths related tologging. Strategies identified toprevent such injuries occurringinclude limiting load heights ontrucks, installing unloadingcages at sawmills andprohibiting overloaded truckson public roadways.

Swift / 2001 /USA [114]

P, CSLevel 4

ManufacturingSV: Safety managementSC: Prevention of lost time injury

This single case study looks atJohn Deere and his companies'commitment to OH&Spractices. Inherent in hisstrategies is the managementof preventable accidents. Heestablished his OH&S programin 1975.

Since 1977 John DeereDomestic Facilities havereceived more than 440National Safety CouncilAwards, 32 in 2000. Altogetherit has reduced its lost timeinjury by 93%, 18% in 2001.For example, in the Davenportworks facility there were 8.2million hours of production timewithout a lost time injury. Theprogram has the managementof safety as its focus startingwith leadership, hazard riskassessment, through to healthand safety programs andmanaging change. Thecompany goal is to now makehis employees safe at home aswell as on the job.

Menser /2001 /Australia[206]

P, CSLevel 4

MiningSV: Managing safety in miningSC: Prevention of lost time injury

This single case study looks atthe CSIRO's mining division.The study describes severalprojects undertaken forimproved safety in the

One of these projects is thedevelopment of the 3D virtualmine as a predictive tool, itallows complex geological,geophysical and geotechnical


Type ofdocument*



Comments / Findings


106

workplace. data to be displayed in 3D.This information can then betransferred quickly to decision-makers. The research team iscurrently working on theintegration of time as the 4th

dimension.Bell et al /2000 / USA[207]

P, SLevel 3-3COSD 7

Military: ArmySV: Risk taking behavioursSC: Prevention of motor vehicle injuries

This descriptive study trackedpersonnel who had completedthe risk appraisals for 6 years.They evaluated speeding, seatbelt use, drinking patterns anddemographics.SS: 99,981RR: N/ADS: Army personnel whocompleted Health RiskAppraisal surveys in 1992

A total of 429 soldiers werehospitalized for motor vehicleinjury. Associated factorsincluded heavy drinking,drinking and driving, speeding,low seat belt use, younger age,minority race/ethnicity, andenlisted rank. Neither smokingnor gender was associated.The study suggests thatprograms targeting thesebehaviors that meet the needsof young and minority soldiersare needed.

Belke / 2000 /USA [208]

P, SLevel 4COSD 7

Chemical ProcessingSV: Accident risk from hazardous chemicals andfacilitiesSC: Prevention of chemical accidents

The study draws together15,000 facility reports into anew regulatory program calledthe Risk Management Program(RMP). The study aims toincrease information chemicalhazards and impact publicfacility risk.SS: 14828 facilities containing20210 chemical processesRR: N/ADS: RMP info database

Information includes thefacilities’ accident history,accident prevention programsin place and the potentialconsequences of hypotheticalchemical release. Accidentrates for each type of chemicalare outlined.

Bailey et al /2000 / USA[209]

P, SLevel 3-2COSD 5

AviationSV: Performance feedback for air trafficcontrollersSC: Improving performance

The study used a computerplayback of performance undersimulator conditions as atraining tool in building effective

It was found that observing acomputer playback enhancesperspective of how theirperformance is affected and


Type ofdocument*



Comments / Findings


107

air traffic control teams.SS: 240 adults participated(53% female).RR: Not specifiedDS: paid participants

affects others. This perspectiveenhanced teamwork andincreased aircraft density thatreached destination within thetime limit.

Anonymous /2000 /Australia[210]

P, CSLevel 4

Transport - RailwaysSV: System failures and human factors in theGlenrock train crashSC: Prevention of accidents

This single case studyexamines the events andcauses of the train collision atGlenrock, Sydney. It also looksat risk management proceduresand safety improvements toprevent further accidents.

Causes and factors included;signal failure, limited visibility,delay of the other train, humanjudgment of the automaticsignals and communicationsbetween drivers and the signalbox.

Laitinen et al/ 1999 /Finland [211]

P, SLevel 4COSD 7

ConstructionSV: Accidents on building/construction sitesSC: Decreasing accident rates

Using observational methodsthis study monitoredconstruction sites andcompared results with accidentfigures form the same sites.Observed safety aspectsincluded working habits,scaffolding, ladders, machinesand equipment, protectionagainst falling, lighting andelectrical and order andtidiness.SS: 164 accidents, 1204 manyearsRR: N/ADS: 305 construction sites

A significant correlation wasfound between the observedsafety index and the accidentrate of the site groups. Thesites with the lowest observedsafety index had an average ofa 3 times higher accident ratethan the sites with the highestsafety index.

Harrell / 1999/ USA [212]

P, SLevel 4COSD 5

Swimming pool lifeguardsSV: competency of life guardsSC: minimize adverse events of children inswimming pools

This observational studyexamines the scanningbehaviour of lifeguards at publicswimming poolsSS: 4 lifeguardsRR: N/ADS: three public swimmingpools.

The study concludes that asthe child-adult ratio in the poolincreases the lifeguard scansare less frequent. The numberof scans increased when thechild-adult ratio decreased.When the total number ofchildren increased lifeguardvigilance decreased primarily


Type ofdocument*



Comments / Findings


108

due to rule violations of childand the issuing of reprimands.

Baron et al /1998 / USA[213]

P, CSLevel 4

ConstructionSV: Roadway ‘flagger’ safetySC: Accident and injury prevention

The 3 case studies describeinjuries incurred by flaggers(fatal and non-fatal) by vehiclesboth on the work site and publicvehicles. The paper suggestsreasons for accidents such asnoisy working environment e.g.becoming accustomed to noisymachinery.

The incidents are described aslargely preventable with amore defensive flagger trainingprogram and more specificsafety equipment such as theuse of rear view video camerason large trucks.

Anonymous /1998 /Australia[214]

P, CSLevel 4

Utilities/ElectricitySV: Workplace safetySC: Reduction of lost time injuries and generalaccident prevention

This single case studydescribes how Northpower hasinvolved all levels of staff inmaking safety a part ofeveryday work including theirrisk management strategies.Northpower employs 1200 staff.

The paper looks at the successof Northpower safety and riskmanagement program, findingan 85% reduction in time lostinjuries throughimplementation of strategiessuch as daily work site riskand hazard controlassessment and monthly depotwork practices meetings.

Wuerz et al /1997 / USA[215]

P, SLevel 3 -2COSD 2

AviationSV: Instrument proficiency in helicopterEmergency Medical SituationsSC: Decreasing accidents in EMS situations

Controlled experimental studyusing a full motion-visualhelicopter simulator. Instrumentproficient pilots (13) were testedagainst non-instrumentproficient pilots (15). Thehelicopter crash rate is 3.1crashes per 100,000.SS: 28 pilotsRR: N/ADS: EMS pilots with commerciallicense (does not require acurrent instrument license) andprevious simulator experience

Instrument proficient pilotsmanaged more safely ininstrument meteorologicconditions (IMC) or badweather. The instrumentproficient pilots (15%) lostcontrol less often than the non-proficient pilots (67%).

Isaac / 1997 / P, CS Construction This case study examines the The analysis uses the models


Type ofdocument*



Comments / Findings


109

New Zealand[216]

Level 4 SV: Competency with designSC: Stability of construction

Cave Creek incident where 14people died as a result of aviewing platform collapse.

of Reason to illustrate theactive and latent failuresassociated with this disaster.Failures included poorconstruction of the platform,inadequate skill and support,concerns of the platform notinvestigated, lack of staff, andpressure of work foremployees.

Sanchez et al/ 1997 / Spain[217]

P, GLevel 4

MiningSV: Development of emergency plans (EP)SC: Accident injury and prevention

This paper describes the firstapproach to developing acomplete EP. It looks at thelaws and regulationsconcerning the plan, riskanalysis of the operations,human and material resourcesavailable and emergencymanagement.Citations: 5

The EP was developed withspecific characteristics for anunderground coal mine and tothose situations considered inhigh-risk analysis, for examplea fire in a belt conveyor.However the EP is flexible andcan be modified to be suitablefor other coalmines.

Jones et al1996 / USA[218]

P, SLevel 4COSD 7

AviationSV: Situation awareness (SA) errorsSC: Decrease in SA errors

This study describes SA errorsin a flight environment using theAviation System ReportingSystem (ASRS) database.Errors were classified into 3categories.SS: 143 incidents (111 – flightcrew, 32 air traffic controllers)RR: N/ADS: ASRS database

The results showed thatCategory 1 – failure toperceive the information(76.3)% was the largest area,then Category 2 – failure tocomprehend the situation(20.3%) and Category 3 –failure to project the situationinto the future (3.4%). Otherproblems included vigilance,automation problems and poormental models.


Type ofdocument*



Comments / Findings


110

Cavenagh /1996 /Australia[219]

P, CSLevel 4

MiningSV: Safety plans implemented by BHPSC: Prevention of deaths

This single case study outlinesa mining explosion at BHP'sMoura Coal Mine(underground), which killed 11people. After a secondexplosion a decision was madeto seal the mine and the bodieswere never recovered.

A six-point plan is beingimplemented following theaccident. This plan involves;developing a risk managementplan including an emergencyresponse plan, developing afull time high level audit team,technical training for personnelengaged in hazardousactivites, raising the profile ofsafety, changing the culture ofBHP to value safety andfunding additional research insafety.

Pettit et al /1996 / USA[220]

P, SLevel 4COSD 7

All Industries/Confined SpacesSV: Deaths in confined spacesSC: Prevention of deaths

A cross sectional study whichuses the National TraumaticOccupational Facilities 1980 to1989 (NTOF) database toidentify the number of deathsacross industries that occur inconfined spaces.SS: a. 585 fatal accidentsclaiming 670 victims. b. 572 trench cave-ins.RR: N/ADS: NTOF database

Fatalities were highest inmanufacturing (152),agriculture (128), construction(90), transportation (77) andmining/oil/gas (63). Other dataidentify atmospheric conditionsand loose materials. Trenchcave ins caused a further 606deaths with the constructioninjury accounting for 77%. Thestudy recommends specificguidelines and site entryprograms (eg entry andrescue procedures), to reducethe number of deaths.

Ashton / 1994/ Canada[221]

P, CSLevel 4

ManufacturingSV: Enhancing safety cultureSC: Promoting OH&S

This report uses 2 case studiesto outline the effects humanfactors and safety attitudeshave on OH&S.

For example when Dupontrestructured in the 1980’sproductivity, profits andaccident rates (1984, 0 losttime accidents, 1986, 5, 19988) increased. Dupontresponded by implementingOH&S strategies and by 1991


Type ofdocument*



Comments / Findings


111

lost time accidents were backdown to 0.

Gully et al /1995 / USA[222]

P, SLevel 3-2COSD 5

Traffic accidentsSV: Accident type and frequencySC: Decreasing the number of preventableaccidents

This time series analysis tested47 police officers over a 4 yearperiod using a driverperformance measure (DPM).The purpose of the study wasto investigate the relationshipbetween driver behavior andaccident involvement.SS: 47 police officersRR: Not specifiedDS: moderate sized policedepartment

Results indicated that scoreson the DPM predictedinvolvement in preventableaccidents but notunpreventable accidents.

Shappell /1995 / USA[223]

P, SLevel 4COSD 7

Military/ NavalSV: Flight deck injuriesSC: Decreasing flight deck injuries

This study provides a review ofinjuries sustained by staffworking on naval flight decksbetween Jan 1977 and Dec1991.SS: 918 flight deck injurieswere reported.RR: N/ADS: US Naval Safety CenterDatabase

Injuries included 43 fatalities, 5permanent total disabilities, 42permanent partial disabilitiesand 828 major disabilities.Injuries included fractures,traumatic amputations,lacerations, dislocations,contusions, concussions,burns, crushing injuries,sprains and strains with 90%being attributable to humanerror.

Ribak et al /1995 / Israel[224]

P, SLevel 4COSD 7

AviationSV: Accidents among airport ground personnelSC: Accident prevention

This study examines injury-producing accidents amongground personnel. All accidentswere reported from 1998 to1992.SS: 2000 ground workers if theinjury resulted in more than 3days or more of absence.RR: N/ADS: Accident reports

The most common accidentswere found to be slip, trips andfalls (40.3%), lifting andcarrying accidents (20,4%) andmachinery accidents (18.7%).Transport accidents (12.8%)and chemical exposures(7.5%) were less common.


Type ofdocument*



Comments / Findings


112

Rundmo /1994 /Norway [225]

P, SLevel 4COSD 7

Mining/Offshore PetroleumSV: Organizational factors and safety measuresSC: Attitudes towards safety culture

A self-administered surveyamong offshore petroleumpersonnel was conducted. Thesurvey used 5 companies, 8installations. The studyobjectives were to determinethe association betweenorganisational factors andsafety and contingencymeasures.SS: 915RR: 92%DS: Employees of an offshorepetroleum mine

Results showed that employeeperception of greatermanagement commitment,social support, and subjectiveevaluations of priorities ofsafety versus production goalswere all important predictorvariables for employeesatisfaction with safety andcontingency measures.

Jackson /1994 /Australia[226]

P, CSLevel 4

ManufacturingSV: Good design of manufacturing equipmentSC: Accident prevention

This single case study looks atthe design of equipment and itsrelationship to safety. It isstated that as people areprimarily unreliable it is gooddesign that will help to preventaccidents.

As part of new standards insafety the aims will be toeliminate hazards very early inthe design process. Themodern approach to safety iswhere both the designer andthe employer will share theresponsibility for safetywhereas at the moment theprimary responsibility restswith the employer.

Young / 1992/ USA [227]

P, GLevel 4

MediaSV: Safety ManagementSC: Accident prevention

This study describes theprocess of putting into placenew safety procedures. 8 x 2hour sessions of training weregiven as well as an 11 membersteering committee set up. 100accidents since start up wereprofiled.

The accident rate of the milldropped from 9.09 reportableevents in 1991 to 1.99 to in1992. The report outlines theimportance of safetyprocedures that suit theworking environment.

113


Staff performance is not only dependent on them having the knowledge and skillsto do their job, their performance is also influenced by a number of other factors.These include the environment in which they work, their mental and physicalhealth, their relationship with the people around them and their ability tocommunicate as well as a range of system related issues. The work done in thenon-health industries provides some potentially useful research considerations.The use of technological aides to reduce errors is currently in use in the area ofdrug prescribing. This could be expanded to the area of clinical decision-making.

Of the health related papers the following strategies for reducing patient adverseevents may warrant further consideration. These are:

1. The use of autopsy reports to review the detection and accuracy ofdiagnosis.

2. The effectiveness of multidisciplinary case meetings and review of testresults to improve reliability and treatment diagnosis.

3. The implementation of specialist transport teams for inter-hospitaltransfers of acutely ill patients.

4. The implementation of organisational structures and resources to improvethe speed between the patient’s diagnosis and receiving their treatment.

114

5. RECOMMENDATIONS

Staff physical and mental health

The impact of long hours, shift work and stress are known to affect mood, mentalhealth and emotional wellbeing. This also has an impact on clinical decision-making, alertness, vigilance and effective communication.

There are clear gaps in the literature given that few papers in this review focuson staff health and patient safety. There needs to be further research thatprovides a better understanding of the link between staff physical and mentalhealth and patient safety particularly in the following areas:

1. Evidence-based guidelines for employing staff that provides a basis forassessing their fitness for practice

2. Understanding of the factors that affect staff mental health and effectiveclinical decision-making

3. Interventions that improve staff health which result in an increasedcapacity for improved performance particularly vigilance, resilience,alertness and effective communication.

Communication and feedback

The literature identified that communication breakdown is, in many cases, acritical factor in disasters, preventable accidents and adverse events. Theliterature indicates that high staff turn-over, inexperienced staff and personalfactors such as low self-esteem, irritability from tiredness and interpersonalconflict are contributing factors to ineffective and inappropriate communication.

There still remain gaps in the literature particularly focusing on interventions thatpromote improved communication and the reduction of adverse events as theprimary end-point. Further research is needed using more robust methods whichmay provide promising strategies to improve incident reporting, the reduction ofadverse events and the improvement of staff morale. These include:

1. Automated or semi-automated alerting systems for staff to identifypotential risk associated with patient care

2. Organisational structures that improve social relationships amongcolleagues providing direct patient care

3. Feedback on performance and safety issues that increases teamwork andperformance.

Hours of work, shift work, number of staff and fatigue

The papers in this review that focus on nurse staffing levels enhance ourunderstanding of nurse to patient ratios and its effect on patient outcomes. Giventhe large amount of resources that are required to provide patient care, thedevelopment of cost effective, efficient and safe models of care is of priority. Theliterature in both the health and non-health areas indicate that too few and too

115

many of the same type of staff can lead to poor patient outcomes. Increasing thisarea of research to include staffing models that are collaborative from a range ofprofessions, use a rostering system that is conducive to continuity of care yetenables adequate rest and sleep for clinicians can make a significant contributionto safe staffing and the promotion of patient safety.

There are some other promising areas for further research which are:1. Condensing the working week to enable longer shifts with more time away

from work2. Examining the impact of doctor to patient and allied health worker to patient

ratios and the impact on adverse events3. The development of a model that promotes continuity of care rather than

continuity of carer through the use of use of feedback and programming (suchas the use of guidelines, policies and procedures)

4. The development of strategies that enable the inexperienced practitioner towork safely in the delivery of patient care such as specific education onmentoring, promotion of feedback and the use of checklists or protocols toclearly define roles and responsibilities

5. The promotion of effective messages on managing shiftwork, personal lifeand sleep.

Competency, supervision and staff needs

Staff performance is not only dependent on them having the knowledge and skillsto do their job, their performance is also influenced by a number of other factors.These include the environment in which they work, their mental and physicalhealth, their relationship with the people around them and their ability tocommunicate as well as a range of system related issues. The work done in thenon-health industries provides some potentially useful research considerations.The use of technological aides to reduce errors is currently in use in the area ofdrug prescribing. This could be expanded to the area of clinical decision-making.

Of the health related papers the following strategies for reducing patient adverseevents may warrant further consideration. These are:

1. The use of autopsy reports to review the detection and accuracy ofdiagnosis.

2. The effectiveness of multidisciplinary case meetings and review of testresults to improve reliability and treatment diagnosis.

3. The implementation of specialist transport teams for inter-hospitaltransfers of acutely ill patients.

4. The implementation of organisational structures and resources to improvethe speed between the patient’s diagnosis and receiving their treatment.

116

6. APPENDICES

6.1 Websites Searched

Government Websites

Australian Governmenthttp://www.fed.gov.au/KSP/

Commonwealth Department of Health and Ageinghttp://www.health.gov.au/

NSW Department of Healthhttp://www.health.nsw.gov.au/

QLD Department of Healthhttp://www.health.qld.gov.au/

NT Department of Health & Community Serviceshttp://www.nt.gov.au/health/

SA Department of Human Serviceshttp://www.health.sa.gov.au/

VIC Department of Human Serviceshttp://www.dhs.vic.gov.au/

Health Department of WAhttp://www.health.wa.gov.au/

TAS Department of Health and Human Serviceshttp://www.dhhs.tas.gov.au/

ACT HealthFirsthttp://www.healthfirst.net.au/entry.ser

Australian Health WebsitesAustralian Centre for Evidence Based Clinical Practicehttp://www.acebcp.org.au/

Australian College of Health Service Executiveshttp://www.achse.org.au/

117

Australian Council for Quality and Safety in Healthcarehttp://www.safetyandquality.org/

Australian Council of Healthcare Standardshttp://www.achs.org.au/

Australian Healthcare Associationhttp://www.aushealthcare.com.au/

Australian Health Policy Institute (University of Sydney)http://www.usyd.edu.au/chs/ahpi/

Australian Institute of Health & Welfarehttp://www.aihw.gov.au/

Australian Institute of Primary Care (La Trobe University)http://www.latrobe.edu.au/aipc/

Australian Medical Councilhttp://www.amc.org.au/

Australian Medical Associationhttp://www.ama.com.au/

Australian Medical Workforce Advisoryhttp://amwac.health.nsw.gov.au/

Australian Patient Safety Foundationhttp://www.apsf.net.au/

Australian Safety and Efficacy Register of New Interventional Procedures -Surgicalhttp://www.racs.edu.au/open/asernip-s.htm

Centre for Health Economics Research & Evaluation (University of Sydney)http://www.chere.usyd.edu.au/

Centre for Health Program Evaluationhttp://chpe.buseco.monash.edu.au/

Clinical Information Access Program (CIAP)http://www.ciap.health.nsw.gov.au/

Clinical Risk Management (VIC Health)http://clinicalrisk.health.vic.gov.au/

Clinicians Health Channelhttp://www.clinicians.vic.gov.au/

118

Health and Community Services Complaints Commission (NT)http://www.nt.gov.au/omb_hcscc/HCSCC/welcomehh.htm

Health Leaders Networkhttp://www.hln.com.au/

Health Rights Commission (QLD)http://www.hrc.qld.gov.au/

Health Workforcehttp://www.health.gov.au/workforce/

HealthInsitehttp://www.healthinsite.gov.au/

Joanna Briggs Institutehttp://www.joannabriggs.edu.au/

National Health & Medical Research Councilhttp://www.health.gov.au/nhmrc/

National Health Priorities & Qualityhttp://www.health.gov.au/pq/

National Health Priority Action Councilhttp://www.nhpac.gov.au/

National Institute of Clinical Studieshttp://www.nicsl.com.au/

NSW Healthcare Complaints Commissionhttp://www.hccc.nsw.gov.au/

NSW Rural Doctors Networkhttp://www.nswrdn.com.au/

NT Remote Health Workforce Agencyhttp://www.ntrhwa.org.au/

Office of Health Review (WA)http://www.healthreview.wa.gov.au/

Office of the Health Services Commissioner (VIC)http://www.health.vic.gov.au/hsc/index.htm

Office of Safety & Quality in Health Care (WA)http://www.health.wa.gov.au/safetyandquality/faq/index.cfm

119

Patient Safetyhttp://www.clinicalrisk.com/

Primary Healthcare Research & Information Servicehttp://www.phcris.org.au/

Private Health Insurance Ombudsmanhttp://www.phio.org.au/home.php

Public Health SAhttp://www.health.sa.gov.au/pehs/Default.htm

Quality Branch (NSW Health)http://www.health.nsw.gov.au/quality/

Quality Improvement Council (La Trobe University)http://www.latrobe.edu.au/qic/

QLD Health Council on Safety and Quality in Healthcarehttp://www.health.qld.gov.au/quality/qc.htm

QLD Centre for Public Healthhttp://www.sph.uq.edu.au/qcph/

Royal Australasian College of Medical Administratorshttp://www.racma.org.au/

Royal Australian College of General Practicehttp://www.racgp.org.au/

Royal Australian College of Physicianshttp://www.racp.edu.au/

Royal College of Nursing, Australiahttp://www.rcna.org.au/

Tasmanian Health Complaints Commissionerhttp://www.justice.tas.gov.au/health_complaints/home.html

WA Centre for Remote and Rural Medicinehttp://www.wacrrm.uwa.edu.au/wacrrm.nsf/docs/49S7YE?opendocument

120

Other Australian WebsitesACT Workcoverhttp://www.workcover.act.gov.au/

Australian Council of Trade Unions (ACTU)http://www.actu.asn.au/

Australian Industrial Relations Commissionhttp://www.airc.gov.au/

Department of Employment and Workplace Relationshttp://www.dewrsb.gov.au/

Risk Management (Standards Australia)http://www.riskmanagement.com.au/

University of South Australia- Work & Stress Research Grouphttp://www.unisa.edu.au/workstress/

- Group for Research on Employment and Workplace Changehttp://business.unisa.edu.au/research/grewc/

- Sleep Centre Researchhttp://www.unisa.edu.au/sleep/research/main.htm

Workcover Corporation (SA)http://www.workcover.com/

Workcover New South Waleshttp://www.workcover.nsw.gov.au/

Workcover Queenslandhttp://www.workcoverqld.com.au/index.htm

Workcover Western Australiahttp://www.workcover.wa.gov.au/

Workplace Health & Safety (QLD Government)http://www.whs.qld.gov.au/index.htm

Workplace Standards Tasmaniahttp://www.wst.tas.gov.au/node/WST.htm

121

International WebsitesAgency for Healthcare Research and Quality (US)http://www.ahcpr.gov/

American College of Healthcare Executives (US)http://www.ache.org/

Bureau of Health Professionals (US)http://bhpr.hrsa.gov/

Centers for Disease Control and Prevention (US)- Division of Healthcare Quality Promotionhttp://www.cdc.gov/ncidod/hip/default.htm

Department of Health (UK)www.doh.gov.uk

Department of Work and Pensions (UK)http://www.dwp.gov.uk/

Food and Drug Administration (US)http://www.fda.gov/

- MedWatch – The FDA Safety Information and Adverse Event ReportingProgram (US)http://www.fda.gov/medwatch/getforms.htm

Healthcaresafetyinfo.com (US)http://209.213.127.236/index.cfm

Health Canadahttp://www.hc-sc.gc.ca/

Health Workforce Advisory Committee (NZ)http://www.hwac.govt.nz/

Institute of Medicine (US)http://www.iom.edu/iom/iomhome.nsf?OpenDatabase

Institute for Healthcare Improvements (US)http://www.ihi.org/

Institute for Safe Medication Practices (US)http://www.ismp.org/

122

International Reform Monitor: Social Policy, Labour Market Policy, IndustrialRelationshttp://www.reformmonitor.org/

National Coordinating Council for Medication Error Reporting and Prevention(US)http://www.nccmerp.org/

National Patient Safety Agency (UK)http://www.npsa.org.uk/

National Task Force on Violence Against Social Care Staff (UK)http://www.doh.gov.uk/violencetaskforce/index.htm

Quality Interagency Coordinating Task Force (US)http://www.quic.gov/

Society of Office-Based Anesthesia (US)http://www.soba.org/

U.S. Department of Labor Occupational Health & Safety Administration (US)http://www.osha.gov/

World Health Organisation (WHO)http://www.who.int/

Veterans Affairs – National Center for Patient Safety (US)http://www.patientsafety.gov/

123

7. REFERENCES

1. Ochoa, F.J., et al., The effect of rescuer fatigue on the quality of chestcompressions. Resuscitation., 1998. 37(3): p. 149-52.

2. Shuster, E., A surgeon with acquired immunodeficiency syndrome: athreat to patient safety? The case of William H. Behringer. AmericanJournal of Medicine., 1993. 94(1): p. 93-9.

3. Dugan, J., et al., Stressful nurses: the effect on patient outcomes. Journalof Nursing Care Quality., 1996. 10(3): p. 46-58.

4. Rundmo, T., H. Hestad, and P. Ulleberg, Organisational factors, safetyattitudes and workload among offshore oil personnel. Safety Science,1998. 29(2): p. 75-87.

5. Poole, C.J., et al., Guidance on standards of health for clinical health careworkers. Occupational Medicine (Oxford)., 2002. 52(1): p. 17-24.

6. AIDS/TB Committee of the Society for Healthcare Epidemiology ofAmerica and Anonymous, Management of healthcare workers infectedwith hepatitis B virus, hepatitis C virus, human immunodeficiency virus, orother bloodborne pathogens. Infection Control & Hospital Epidemiology.,1997. 18(5): p. 349-63.

7. Reason, J., Understanding adverse events: human factors. Quality inHealth Care., 1995. 4(2): p. 80-9.

8. Hudson, T., P. Eubanks, and K. Lumsdon, HIV-positive health careworkers pose legal, safety challenges for hospitals. Hospitals, 1992.66(18): p. 24-30.

9. Joy, M., Introduction and summary of principal conclusions of the SecondEuropean Workshop in Aviation Cardiology. European Heart JournalSupplements., 1999. 1(Suppl D): p. D1-12.

10. Rundmo, T. and L. Sjoberg, Risk perception by offshore oil personnelduring bad weather conditions. Risk Analysis, 1998. 18(1): p. 111-118.

11. Edkins, G.D. and C.M. Pollock, The influence of sustained attention onrailway accidents. Accident Analysis & Prevention., 1997. 29(4): p. 533-9.

12. Raymond, M.W. and R. Moser, Aviators at risk. Aviation, Space, andEnvironmental Medicine, 1995. 66(1): p. 35-39.

13. Haugli, L., A. Skogstad, and O.H. Hellesoy, Health, sleep, and moodperceptions reported by airline crews flying short and long hauls. Aviation,Space, and Environmental Medicine, 1994. 65(1): p. 27-34.

14. Sutherland, V.J., The use of a stress audit. Leadership & OrganizationDevelopment Journal, 1993. 14(1): p. 22-28.

15. Ore, T.A.D.D.P. and T.C.F.D.A.P.O.B.M.S.M.V. Economics, Accidentrates in the Australian stevedoring industry. J Occup Health Saf Aust NZ,1993. 9(1): p. 31-42.

124

16. O'Cathain, A., J. Turner, and J.P. Nicholl, The acceptability of anemergency medical dispatch system to people who call 999 to request anambulance. Emergency Medicine Journal, 2002. 19(2): p. 160-163.

17. Taylor, D.M., R. Wolfe, and P.A. Cameron, Complaints from emergencydepartment patients largely result from treatment and communicationproblems. Emergency Medicine (Fremantle, W.A.)., 2002. 14(1): p. 43-9.

18. Learning the lessons: the report of the public inquiry into children's heartsurgery at the Bristol Royal Infirmary 1984-1995.. 2001.

19. Walton, M., Open disclosure to patients or families after an adverse event:a literature review.. 2001, National Open Disclosure Consortium. p. 55.

20. Greco, M., et al., Patient assessment of interpersonal skills: a clinicalgovernance activity for hospital doctors and nurses. Journal of ClinicalExcellence, 2001. 3(3): p. 117-124.

21. Pichert, J.W., et al., Identifying medical center units with disproportionateshares of patient complaints. Joint Commission Journal on QualityImprovement, 1999. 25(6): p. 288-99.

22. Randolph, J.F., et al., A study of the accuracy of telephone orders innursing homes in southern California. Annals of Long Term Care, 1999.7(9): p. 334-8.

23. Young, G.J., et al., Patterns of coordination and clinical outcomes: a studyof surgical services. Health Services Research., 1998. 33(5 Pt 1): p. 1211-36.

24. Fiesta, J., Liability for falls. Nursing Management (Springhouse), 1998.29(3): p. 24-6.

25. Chen, P.P., et al., Incident reporting in acute pain management.Anaesthesia, 1998. 53(8): p. 730-5.

26. Bark, P., et al., Impact of litigation on senior clinicians: implications for riskmanagement. Quality in Health Care., 1997. 6(1): p. 7-13.

27. Beckmann, U., et al., The Australian Incident Monitoring Study in IntensiveCare: AIMS-ICU. The development and evaluation of an incident reportingsystem in intensive care. Anaesthesia and Intensive Care, 1996. 24(3): p.314-9.

28. Kuperman, G.J., et al., Detecting alerts, notifying the physician, andoffering action items: a comprehensive alerting system. Proc AMIA AnnuFall Symp, 1996: p. 704-8.

29. Crowther, M.E., Communication following a stillbirth or neonatal death:room for improvement. British Journal of Obstetrics & Gynaecology., 1995.102(12): p. 952-6.

30. Anderson, R. and M.A. Testa, Symptom distress checklists as acomponent of quality of life measurement: comparing prompted reports bypatient and physician with concurrent adverse event reports via thephysician. Drug Information Journal, 1994. 28(1): p. 89-114.

31. Bark, P., et al., Clinical complaints: a means of improving quality of care.Quality in Health Care., 1994. 3(3): p. 123-32.

32. Vincent, C., et al., The impact of litigation on obstetricians andgynaecologists. Journal of Obstetrics & Gynaecology, 1994. 14(6): p. 381-387.

125

33. Hicks, F. and G. Corcoran, Hospice management of patients receivingcytotoxic chemotherapy: problems and opportunities. British Journal ofCancer, 1993. 68(6): p. 1205-1209.

34. Nicol, J., Have Australian's major hazard facilities learnt for the Longforddisaster? an evaluation of then impact of the 1998 ESSO Longfordexplosion on major hazard facilities in 2001.. 2001, The Institute ofEngineers, Australia: Canberra. p. 55.

35. Cox, S.J. and A.J.T. Cheyne, Assessing safety culture in offshoreenvironments. Safety Science, 2000. 34(1-3): p. 111-129.

36. McDonald, N., et al., Safety management systems and safety culture inaircraft maintenance organisations. Safety Science, 2000. 34(1-3): p. 151-176.

37. Hofmann, D.A. and A. Stetzer, The role of safety climate andcommunication in accident interpretation: implications for learning fromnegative events. Academy of Management Journal, 1998. 41(6): p. 644-657.

38. Simard, M. and A. Marchand, Workgroups' propensity to comply withsafety rules: the influence of micro-macro organisational factors.Ergonomics, 1997. 40(2): p. 172-188.

39. Appelberg, K., et al., Interpersonal conflict as a predictor of work disability:a follow-up study of 15,348 Finnish employees. Journal of PsychosomaticResearch, 1996. 40(2): p. 157-167.

40. Hofmann, D.A., A cross-level investigation of factors influencing unsafebehaviors and accidents. Personnel Psychology, 1996. 49(2): p. 307.

41. Skogstad, A., A. Dyregrov, and O.H. Hellesoy, Cockpit-cabin crewinteraction: satisfaction with communication and information exchange.Aviation Space & Environmental Medicine, 1995. 66(9): p. 841-848.

42. Best practice case study: Herbert Adams Bakeries. J Occup Health SafAust NZ, 1994. 10(3): p. 275-8.

43. Cooper, M.D., et al., Reducing accidents using goal setting and feedback:a field study. Journal of Occupational & Organizational Psychology, 1994.67(3): p. 219-240.

44. Pate Cornell, M.E., Learning from the Piper Alpha accident: a postmortemanalysis of technical and organisational factors. Risk Analysis, 1993.13(2): p. 215-232.

45. Clarke, R.J. and T.G. Adams. The safety imperative - managing safety. inConference on Railway Engineering: "Contracting Railways Safety,Standards and the Surroundings". 1993. Newcastle, Australia.

46. Aiken, L.H., et al., Hospital nurse staffing and patient mortality, nurseburnout, and job dissatisfaction. Journal of the American MedicalAssociation, 2002. 288(16): p. 1987-1993.

47. Kovner, C., et al., Nurse staffing and postsurgical adverse events: ananalysis of administrative data from a sample of U.S. Hospitals, 1990-1996. Health Services Research, 2002. 37(3): p. 611-629.

48. Blegen, M.A. and T.E. Vaughn, A Multisite study of nurse staffing andpatient occurrences. Nursing Economics., 1998. 16(4): p. 196-203.

49. Joint Commission on Accreditation of Healthcare Organizations, Healthcare at the crossroads: strategies for addressing the evolving nursing

126

crisis.. 2001, Joint Commission on Accreditation of HealthcareOrganizations. p. 43.

50. Goodman, D.C., et al., The Relationship between the availability ofneonatal intensive care and neonatal mortality. New England Journal ofMedicine., 2002. 346(20): p. 1538-1544.

51. Tucker, J., et al., Patient volume, staffing, and workload in relation to risk-adjusted outcomes in a random stratified sample of UK neonatal intensivecare units: a prospective evaluation. Lancet, 2002. 359(9301): p. 99-107.

52. Wolfe, A.M. and J. Bourke, Detecting and reducing adverse events in anAustralian rural base hospital emergency department using medical recordscreening and review. Emergency Medicine Journal, 2002. 19(1): p. 35-40.

53. Davis, J. and W. Rowell, Safety innovations in practice: patient sitters trial.Final report.. 2002, Royal Hobart Hospital: Hobart. p. 6.

54. Sovie, M.D. and A.F. Jawad, Hospital restructuring and its impact onoutcomes: nursing staff regulations are premature. Journal of NursingAdministration., 2001. 31(12): p. 588-600.

55. Weinberg, A.D., et al., Quality care indicators and staffing levels in anursing facility subacute unit. Journal of the American Medical DirectorsAssociation, 2002. 3(1): p. 1-4.

56. Trafford, A., Private 'sitters' fulfill patients' needs where nursing staff can't:because hospital personnel is so short-handed, many in need of medicalattention are hiring personal assistants., in Los Angeles Times. 2001: LosAngeles. p. S8.

57. Boodman, S.G., Waking up to the problem of fatigue among medicalinterns., in Los Angeles Times. 2001. p. S1, S7.

58. Commission, A., Ward staffing. Acute Hospital Portfolio: review of nationalfindings.. 2001, Audit Commission: London. p. 19.

59. Hendrix, T.J. and S.E. Foreman, Optimal long-term care nurse-staffinglevels. Nursing Economics, 2001. 19(4): p. 164-175.

60. Patton, D.V., D.R. Landers, and I.T. Agarwal, Legal considerations ofsleep deprivation among resident physicians. Journal of Health Law.,2001. 34(3): p. 377-417.

61. Needleman, J., et al., Nurse staffing and patient outcomes in hospitals..2001, Health Resources Services Administration: Boston.

62. Moreno, R., et al., Mortality after discharge from intensive care: the impactof organ system failure and nursing workload use at discharge. IntensiveCare Medicine., 2001. 27(6): p. 999-1004.

63. Reed, J.L. and M. Lyne, Inpatient care of mentally ill people in prison:results of a year's programme of semistructured inspections. BritishMedical Journal, 2000. 320(7241): p. 1031-4.

64. Morris, G.P. and R.W. Morris, Anaesthesia and fatigue: an analysis of thefirst 10 years of the Australian Incident Monitoring Study 1987-1997.Anaesthesia and intensive care, 2000. 28(3): p. 300-4.

65. Tarnow-Mordi, W.O., et al., Hospital mortality in relation to staff workload:a 4-year study in an adult intensive-care unit. Lancet, 2000. 356(9225): p.185-189.

127

66. Schulmeister, L., Chemotherapy medication errors: descriptions, severity,and contributing factors. Oncology Nursing Forum., 1999. 26(6): p. 1033-42.

67. Peterson, G.M., M.S. Wu, and J.K. Bergin, Pharmacist's attitudes towardsdispensing errors: their causes and prevention. Journal of ClinicalPharmacy and Therapeutics, 1999. 24(1): p. 57-71.

68. Cohen, M.M., et al., Nursing workload associated with adverse events inthe postanesthesia care unit. Anesthesiology., 1999. 91(6): p. 1882-90.

69. Hunt, J. and S. Hagen, Nurse to patient ratios and patient outcomes.Nursing Times, 1998. 94(45): p. 63-66.

70. Marck, P., D. Allen, and D. Phillipchuk, AARN practice consultants findRNs still reporting high rate of safety concerns. AARN News Letter., 1998.54(3): p. 10-1.

71. Kovner, C. and P.J. Gergen, Nurse staffing levels and adverse eventsfollowing surgery in U.S. hospitals. Image the Journal of NursingScholarship, 1998. 30(4): p. 315-21.

72. Smith-Coggins, R., et al., Rotating shiftwork schedules: can we enhancephysician adaptation to night shifts? Academic Emergency Medicine,1997. 4(10): p. 951-61.

73. Kirkcaldy, B.D., R. Trimpop, and C.L. Cooper, Working hours, job stress,work satisfaction, and accident rates among medical practitioners andallied personnel. International Journal of Stress Management, 1997. 4(2):p. 79-87.

74. Brown, L., et al., Does ambulance crew size affect on-scene time ornumber of prehospital interventions? Prehospital & Disaster Medicine,1996. 11(3): p. 214-7; discussion 217-8.

75. Booker, J.M. and C. Roseman, A Seasonal pattern of hospital medicationerrors in Alaska. Psychiatry Research, 1995. 57(3): p. 251-257.

76. Edge, W.E., et al., Reduction of morbidity in interhospital transport byspecialized pediatric staff. Critical Care Medicine, 1994. 22(7): p. 1186-91.

77. Haw, C.M., A cluster of suicides at a London psychiatric unit. Suicide &Life-Threatening Behavior., 1994. 24(3): p. 256-66.

78. Laine, C., et al., The impact of a regulation restricting medical house staffworking hours on the quality of patient care. Journal of the AmericanMedical Association, 1993. 269(3): p. 374-378.

79. Barone, J.E., et al., Is 24-hour operating room staff absolutely necessaryfor level II trauma center designation? Journal of Trauma-Injury Infection &Critical Care., 1993. 34(6): p. 878-82; discussion 882-3.

80. Mann, F.A. and P.L. Danz, The night stalker effect: quality improvementswith a dedicated night-call rotation. Investigative Radiology, 1993. 28(1):p. 92-6.

81. Leung, L. and C.E. Becker, Sleep deprivation and house staffperformance. Update 1984-1991. Journal of Occupational Medicine.,1992. 34(12): p. 1153-60.

82. Caldwell, J.A. and S.R. Gilreath, A survey of aircrew fatigue in a sample ofU.S. Army aviation personnel. Aviation Space & Environmental Medicine,2002. 73(5): p. 472-480.

128

83. National Occupational Health and Safety Council, Draft watching brief onfatigue.. 2002, National Occupational Health and Safety Council. p. 35.

84. Australian Council of Trade Unions, Fifty families.. 2002, ACTU.85. Lilley, R., et al., A survey of forest workers in New Zealand: do hours of

work, rest, and recovery play a role in accidents and injury? Journal ofSafety Research, 2002. 33(1): p. 53-71.

86. Working hours case.. 2002, Australian Industrial Relations Commission.87. Rosenberg, E. and Y. Caine, Survey of Israeli Air Force line commander

support for fatigue prevention initiatives. Aviation Space & EnvironmentalMedicine, 2001. 72(4): p. 352-356.

88. Sussman, D. and M. Coplen, Fatigue and alertness in the United Statesrailroad industry part I: the nature of the problem. Transportation ResearchPart F: Traffic Psychology & Behaviour, 2000. 3(4): p. 211-220.

89. Phillips, R., Sleep, watchkeeping and accidents: a content analysis ofincident at sea reports. Transportation Research Part F: TrafficPsychology & Behaviour, 2000. 3(4): p. 229-240.

90. Oron Gilad, T. and D. Shinar, Driver fatigue among military truck drivers.Transportation Research, Part F - Traffic Psychology and Behaviour, 3F,4, 2000-12, pp 195-209., 2000.

91. Simons, M. and P.J.L. Valk. Early starts: effects on sleep, alertness, andvigilance. 2000.

92. Williamson, A., et al., Development of measures of fatigue: using analcohol comparison to validate the effects of fatigue on performance.Report, CR, 2000. 189(07): p. 0-642.

93. Shanley, A., Assuring alertness on the job: from the control room to theboardroom, sleep deprivation affects everyone. Here's how one companyis battling fatigue. Chemical Engineering, 1999. 106(5): p. 147.

94. Smith, L., et al., Work shift duration: a review comparing eight hour and 12hour shift systems. Occupational and Environmental Medicine, 1998.55(4): p. 217-29.

95. Executive, H.S., Key fact sheet on rates of injury within small and largemanufacturing workplaces.. 1998, Health & Safety Executive

Prepared by Government Statistical Service,: Merseyside. p. 17.96. Gander, P.H., et al., Flight crew fatigue V: long-haul air transport

operations. Aviation Space & Environmental Medicine, 1998. 69(9, Sect2): p. B37-B48.

97. Gander, P., et al. An integrated fatigue management programme fortanker drivers. in International Conference on Fatigue and Transportation,3rd. 1998. Fremantle, Western Australia: Murdoch University. Institute forResearch in Safety and Transport, Murdoch, Western Australia, Australia.

98. Gander, P.H., et al., Flight crew fatigue II: short-haul fixed-wing airtransport operations. Aviation Space & Environmental Medicine, 1998.69(9, Sect 2): p. B8-B15.

99. Vega, A. and M.J. Gilbert, Longer days, shorter weeks: compressed workweeks in policing. Public Personnel Management, 1997. 26(3): p. 391-402.

129

100. Luna, T.D., J. French, and J.L. Mitcha, A study of USAF air trafficcontroller shiftwork: sleep, fatigue, activity, and mood analyses. AviationSpace & Environmental Medicine, 1997. 68(1): p. 18-23.

101. Samel, A., et al., Two-crew operations: stress and fatigue during long-haulnight flights. Aviation Space and Environmental Medicine, 1997. 68(8): p.679-87.

102. Fell, D. and B. Black, Driver fatigue in the city. 1996: p. 5471-7.103. Emonson, D.L. and R.D. Vanderbeek, The use of amphetamines in U.S.

Air Force tactical operations during Desert Shield and Storm. AviationSpace & Environmental Medicine, 1995. 66(3): p. 260-263.

104. Smith, L., P. Totterdell, and S. Folkard, Shiftwork effects in nuclear powerworkers: a field study using portable computers. Work & Stress, 1995. 9(2-3): p. 235-244.

105. Beilock, R., Schedule-induced hours-of-service and speed limit violationsamong tractor-trailer drivers. Accident; analysis and prevention, 1995.27(1): p. 33-42.

106. Williamson, A.M. and A.-M. Feyer, Causes of accidents and the time ofday. Work & Stress, 1995. 9(2-3): p. 158-164.

107. Duchon, J.C., C.M. Keran, and T.J. Smith, Extended workdays in anunderground mine: a work performance analysis. Human Factors, 1994.36(2): p. 258-268.

108. French, J., et al., Crew fatigue during simulated, long duration B-1BBomber missions. Aviation, Space, and Environmental Medicine, 1994.65(5): p. A1-A6, Section II (Supplement).

109. Parkes, K.R., Health, and E. Safety, Human factors, shift work andalertness in the offshore oil industry. HMSO Books, P.O. Box, 1993.276(94): p. 00.

110. Salter, C.A., J.S. Crowley, and L.W. Allan, Sleep-related injuries in theU.S. Army, 1984-1991. Military Medicine, 1993. 158(12): p. 782-5.

111. Phillips, J.A. and K.C. Brown, Industrial workers on a rotating shift pattern:adaptation and injury status. Aaohn Journal, 1992. 40(10): p. 468-76.

112. Feyer, A.M. and A.M. Williamson, Work practices and fatigue in the longdistance road transport industry in Australia. Human Factors Society,1992. 2: p. 975-9.

113. Halm, E.A., et al., Instability on hospital discharge and the risk of adverseoutcomes in patients with pneumonia. Archives of Internal Medicine.,2002. 162(11): p. 1278-84.

114. Swift, B., Worker safety: John Deere proves it's possible. RiskManagement, 2001. 48(12): p. 10.

115. Australian Council for Safety and Quality in Health Care, Lessons from theinquiry into obstetrics and gynaecological services at King EdwardMemorial Hospital 1990 - 2000.. 2002, Australian Council for Safety andQuality in Health Care: Canberra. p. 64.

116. Piotrowski, M.M. and D.B. Hinshaw, The safety checklist program:creating a culture of safety in intensive care units. Joint CommissionJournal on Quality Improvement, 2002. 28(6): p. 306-15.

130

117. Paavilainen, E., et al., Identification of child maltreatment while caring forthem in a university hospital. International Journal of Nursing Studies.,2002. 39(3): p. 287-94.

118. Davidson, E.J. and S.C. Nicholson, Management of ectopic pregnancy ina Scottish teaching hospital: Implications for training. Journal of Obstetrics& Gynaecology, 2002. 22(2): p. 197-200.

119. McDonnell, P.J., et al., Hospital admissions resulting from preventableadverse drug reactions. Annals of Pharmacotherapy, 2002. 36(9): p. 1331-1336.

120. Durie, M., U. Beckmann, and D.M. Gillies, Incidents relating to arterialcannulation as identified in 7,525 reports submitted to the Australianincident monitoring study (AIMS-ICU). Anaesthesia & Intensive Care.,2002. 30(1): p. 60-5.

121. Dean, B., et al., Causes of prescribing errors in hospital inpatients: aprospective study. Lancet, 2002. 359(9315): p. 1373-8.

122. Pronovost, P., et al., Building safety into ICU care. Journal of Critical Care,2002. 17(2): p. 78-85.

123. Farbstein, K. and J. Clough, Improving medication safety across amultihospital system. Joint Commission Journal on Quality Improvement.,2001. 27(3): p. 123-37.

124. Kahn, C.A., R.G. Pirrallo, and E.M. Kuhn, Characteristics of fatalambulance crashes in the United States: an 11-year retrospectiveanalysis. Prehospital Emergency Care., 2001. 5(3): p. 261-9.

125. Shojania, K.G., et al., Making health care safer: a critical analysis ofpatient safety practices.. 2001, Agency for Healthcare Research andQuality (AHRQ): San Francisco.

126. Kinley, H., et al., Extended scope of nursing practice: a multicentrerandomised controlled trial of appropriately trained nurses and pre-registration house officers in pre-operative assessment in elective generalsurgery. Health Technology Assessment, 2001. 5(20): p. 1-87.

127. Hebl, M.R. and J. Xu, Weighing the care: physicians' reactions to the sizeof a patient. International Journal of Obesity & Related MetabolicDisorders., 2001. 25(8): p. 1246-52.

128. Morrison, A.L., et al., The effects of nursing staff inexperience (NSI) on theoccurrence of adverse patient experiences in ICUs. Australian CriticalCare., 2001. 14(3): p. 116-21.

129. Tang, K.C., et al., Factors affecting quality of mental health care to peoplespeaking a language other than English in Sydney. Journal of MentalHealth, 2001. 10(1): p. 87-97.

130. Emerson, A., et al., Prospective cohort study of adverse events monitoredby hospital pharmacists. Hospital Adverse Event Monitoring Study(HAEMS) Group. Pharmacoepidemiology & Drug Safety., 2001. 10(2): p.95-103.

131. Bouvier-Colle, M.H., et al., Evaluation of the quality of care for severeobstetrical haemorrhage in three French regions. British Journal ofObstetrics & Gynaecology, 2001. 108(9): p. 898-903.

131

132. Sheikh, K. and C. Bullock, Urban-rural differences in the quality of care forMedicare patients with acute myocardial infarction. Archives of InternalMedicine, 2001. 161(5): p. 737-743.

133. Sharma, R.R., S. Kumar, and S.K. Agnihotri, Sources of preventableerrors related to transfusion. Vox Sanguinis., 2001. 81(1): p. 37-41.

134. Callum, J.L., et al., Reporting of near-miss events for transfusionmedicine: improving transfusion safety. Transfusion., 2001. 41(10): p.1204-11.

135. Dunn, E.B. and J.J. Wolfe, Let go of Latin! Veterinary & HumanToxicology., 2001. 43(4): p. 235-6.

136. White, A., et al., Survey of adverse events following acupuncture (SAFA):a prospective study of 32,000 consultations. Acupuncture in Medicine,2001. 19(2): p. 84-92.

137. Feldman, K.W., C. Mason, and R.P. Shugerman, Accusations that hospitalstaff have abused pediatric patients. Child Abuse & Neglect., 2001.25(12): p. 1555-69.

138. Brilli, R.J., et al., Critical care delivery in the intensive care unit: definingclinical roles and the best practice model. Critical Care Medicine, 2001.29(10): p. 2007-2019.

139. Rex, J.H., et al., Systematic root cause analysis of adverse drug events ina tertiary referral hospital. Joint Commission Journal on QualityImprovement, 2000. 26(10): p. 563-75.

140. Kohn, L.T., J.M. Corrigan, and M.S.E. Donaldson, To err is human:building a safer health system.. 2000, Institute of Medicine: Washington,D.C. p. xxiv, 287p.

141. Glick, T.H., T.P. Workman, and S.V. Gaufberg, Suspected conversiondisorder: foreseeable risks and avoidable errors. Academic EmergencyMedicine., 2000. 7(11): p. 1272-7.

142. Wolff, A.M. and J. Bourke, Reducing medical errors: a practical guide.Medical Journal of Australia., 2000. 173(5): p. 247-51.

143. Khunti, K., Referral for autopsies: analysis of 651 consecutive deaths inone general practice. Postgraduate medical journal, 2000. 76(897): p.415-6.

144. National Statistics, Key fact sheet on injuries within residential care homesreported to local authorities 1994/95 to 1998/99.. 2000, Health & SafetyCommission: Merseyside. p. 10.

145. Hadfield, D., Clinical supervision: users' perspectives. Journal of childhealth care, 2000. 4(1): p. 30-4.

146. Espinosa, J.A. and T.W. Nolan, Reducing errors made by emergencyphysicians in interpreting radiographs: longitudinal study. British MedicalJournal, 2000. 320(7237): p. 737-40.

147. Hourigan, C.T., et al., Changing the site of delivery of thrombolytictreatment for acute myocardial infarction from the coronary care unit to theemergency department greatly reduces door to needle time. Heart, 2000.84(2): p. 157-163.

148. Schenkel, S., Promoting patient safety and preventing medical error inemergency departments. Academic Emergency Medicine, 2000. 7(11): p.1204-22.

132

149. Thomas, E.J. and T.A. Brennan, Incidence and types of preventableadverse events in elderly patients: population based review of medicalrecords. British Medical Journal, 2000. 320(7237): p. 741-4.

150. Cote, C.J., et al., Adverse sedation events in pediatrics: analysis ofmedications used for sedation. Pediatrics, 2000. 106(4 I): p. 633-644.

151. Frey, B., et al., Comprehensive critical incident monitoring in a neonatal-pediatric intensive care unit: experience with the system approach.Intensive Care Medicine, 2000. 26(1): p. 69-74.

152. Oliver, D., A. Hopper, and P. Seed, Do hospital fall prevention programswork? a systematic review. Journal of the American Geriatrics Society,2000. 48(12): p. 1679-89.

153. Van den Bemt, P., et al., Drug-related problems in hospitalised patients.Drug Safety, 2000. 22(4): p. 321-333.

154. Health Care Complaints Commission, Report on an investigation intoadverse outcomes following cataract surgery at Dubbo Base Hospital on 8February 1999.. 1999, Health Care Complaints Commission: Dubbo. p.27.

155. Health Care Complaints Commission, Report on an investigation ofincidents in the Operating Theatre at Canterbury Hospital 8 February - 7June 1999.. 1999, Health Care Complaints Commission: Dubbo. p. 38.

156. Grant, S., D. Spain, and D. Green, Rapid assessment team reduceswaiting time. Emergency Med, 1999. 11(2): p. 72-7.

157. Pullen, R., C. Heikhaus, and I. Fusgen, Falls in hospital: prospective studyof time, place and circumstances. European Journal of Geriatrics, 1999.1(2): p. 119-123.

158. Straand, J. and K.S. Rokstad, Elderly patients in general practice:diagnoses, drugs and inappropriate prescriptions. A report from the Moreand Romsdal Prescription Study. Family Practice, 1999. 16(4): p. 380-388.

159. Reid, M.H. and D. Sawyer, The human factors implications of peritonealdialysis: cycler overfill incident reports. International Journal of TraumaNursing, 1999. 5(2): p. 68-71.

160. LaDuke, S., Risky rides. Make ambulance transports safer for yournurses. Nursing Management (Springhouse)., 1999. 30(9): p. 29-31.

161. Anonymous, The horrors of Ashworth. Lancet., 1999. 353(9149): p. 251.162. Keyes, C. and J. Hammond, Supervision of junior staff. International

Journal for Quality in Health Care., 1999. 11(4): p. 357-8.163. Jelinek, G.A., et al., Re-engineering an Australian emergency department:

can we measure success? Journal of Quality in Clinical Practice, 1999.19(3): p. 133-8.

164. Leape, L.l., et al., Pharmacist participation on physician rounds andadverse drug events in the intensive care unit. [erratum appears in JAMA2000 Mar 8;283(10):1293]. Journal of the American Medical Association,1999. 282(3): p. 267-70.

165. Malpass, A., S. Helps, and W. Runciman, An analysis of Australianadverse drug events. Journal of quality in clinical practice, 1999. 19(1).

166. Graber, M.L., Physician participation in quality management. Expandingthe goals of peer review to detect both practitioner and system error. JointCommission Journal on Quality Improvement, 1999. 25(8): p. 396-407.

133

167. Bates, D., Frequency, consequences and prevention of adverse drugevents. Journal of quality in clinical practice, 1999. 19(1): p. 13-18.

168. Fitzpatrick, M. Evolving strategies in the triage of chest pain. in Past,Present and Future of Acute Hospital Admissions. Australian ResourceCentre for Hospital Innovations - First Tool Kit Seminar. 1998. SydneyAirport Hilton, Sydney: Australian Resource Centre for HospitalInnovations.

169. Taylor-Adams, S., C. Vincent, and N. Stanhope, Applying human factorsmethods to the investigation and analysis of clinical adverse events.Safety Science, 1999. 31(2): p. 143-159.

170. Stearley, H.E., Patients' outcomes: intrahospital transportation andmonitoring of critically ill patients by a specially trained ICU nursing staff.American Journal of Critical Care, 1998. 7(4): p. 282-7.

171. Grilli, R., et al., Do specialists do it better? the impact of specialization onthe processes and outcomes of care for cancer patients. Database ofAbstracts of Reviews of Effectiveness, 2002. 3: p. 3.

172. Senior, J. and N. Patel, Reducing thrombolytic therapy time delays in theemergency department. Journal of Quality in Clinical Practice., 1998.18(2): p. 99-107.

173. Bullard, M.J., et al., Compliance with the law and appropriate medicalstandards during interhospital transfers. Journal of the Formosan MedicalAssociation, 1998. 97(11): p. 770-776.

174. Wilson, D.G., et al., Medication errors in paediatric practice: insights froma continuous quality improvement approach. European Journal ofPediatrics., 1998. 157(9): p. 769-74.

175. Lattimer, V., et al., Safety and effectiveness of nurse telephoneconsultation in out of hours primary care: randomised controlled trial. TheSouth Wiltshire Out of Hours Project (SWOOP) Group. British MedicalJournal, 1998. 317(7165): p. 1054-9.

176. Dale, J.C. and S.W. Renner, Wristband errors in small hospitals: a Collegeof American Pathologists' Q-Probes study of quality issues in patientidentification. Laboratory Medicine, 1997. 28(3): p. 203-7.

177. O'Hara, D.A. and N.J. Carson, Reporting of adverse events in hospitals inVictoria, 1994-1995. Medical Journal of Australia, 1997. 166(9): p. 460-3.

178. Patient commits suicide; hospital, nursing agencies settle for $700,000.Hospital Security and Safety Management, 1997. 18(6): p. 8-10.

179. Brown, S.L., et al., Human error and patient-controlled analgesia pumps.Journal of Intravenous Nursing, 1997. 20(6): p. 311-6.

180. Anonymous, Adverse drug reaction: nurse must know possible sideeffects, monitor patient closely and take action, court rules. Legal EagleEye Newsletter for the Nursing Profession, 1997. 5(8): p. 5.

181. Anonymous, Elopement risk: psych facility's staff faulted for patient'sescape, suicide. Legal Eagle Eye Newsletter for the Nursing Profession,1997. 5(3): p. 7.

182. Anonymous, Suicidal patients: court looks at standards for medicalsurgical nurses caring for a psych patient. Legal Eagle Eye Newsletter forthe Nursing Profession, 1997. 5(6): p. 3.

134

183. Victoria-State-Coroner's-Office, Inquest findings, comments andrecommendations into fire and nine deaths at Kew Residential Services on8 April 1996.. 1997.

184. Wright, V., Illiteracy in patients and HCWs means safety concerns.Healthcare Hazardous Materials Management, 1996. 9(4): p. 8.

185. Beckmann, U., et al., The Australian Incident Monitoring Study in IntensiveCare: AIMS-ICU. An analysis of the first year of reporting. Anaesthesiaand Intensive Care, 1996. 24(3): p. 320-9.

186. Short, T.G., et al., Improvements in anaesthetic care resulting from acritical incident reporting programme. Anaesthesia, 1996. 51(7): p. 615-21.

187. Spittal, M.J., G.P. Findlay, and I. Spencer, A prospective analysis ofcritical incidents attributable to anaesthesia. International Journal forQuality in Health Care., 1995. 7(4): p. 363-71.

188. Leape, L.L., et al., Systems analysis of adverse drug events. ADEPrevention Study Group. Journal of the American Medical Association,1995. 274(1): p. 35-43.

189. Tammelleo, A.D., GA: side rails left down, pt. falls from bed: "ordinarynegligence" or "malpractice"? PA: charge nurse fails to report aide:nursing board's fine and reprimand upheld. Regan Report on Nursing Law,1995. 36(1): p. 3.

190. Tammelleo, A.D., MO: failure to recognize complications: admissibility ofnursing staff meeting minutes. Regan Report on Nursing Law, 1995.36(7).

191. Evans, R.S., et al., Preventing adverse drug events in hospitalizedpatients. Annals of Pharmacotherapy, 1994. 28(4): p. 523-7.

192. Barry, P.W. and C. Ralston, Adverse events occurring during interhospitaltransfer of the critically ill. Archives of Disease in Childhood, 1994. 71(1):p. 8-11.

193. Cohen, M.R., Dangerous abbreviation leads to potassium chlorideadministration IV push. Hospital Pharmacy, 1994. 29(12).

194. Leape, L.L., et al., Preventing medical injury. Quality Review Bulletin.,1993. 19(5): p. 144-9.

195. Singleton, R.J., et al., The Australian Incident Monitoring Study. Physicalinjuries and environmental safety in anaesthesia: an analysis of 2000incident reports. Anaesthesia and Intensive Care, 1993. 21(5): p. 659-63.

196. Van der Walt, J.H., et al., The Australian Incident Monitoring Study.Recovery room incidents in the first 2000 incident reports. Anaesthesia &Intensive Care., 1993. 21(5): p. 650-2.

197. Possible infant abduction foiled by 'best protected' maternity unit. HospitalSecurity and Safety Management, 1993. 14(7): p. 1-3.

198. Barboni, E., et al., Emergency treatment of bronchospastic attacks in anemergency medicine department: effects of a quality improvement project.Quality Assurance in Health Care., 1993. 5(2): p. 123-6.

199. Bates, D.W., L.L. Leape, and S. Petrycki, Incidence and preventability ofadverse drug events in hospitalized adults. Journal of General InternalMedicine, 1993. 8(6): p. 289-294.

135

200. McKee, M. and N. Black, Does the current use of junior doctors in theUnited Kingdom affect the quality of medical care? Social Science &Medicine, 1992. 34(5): p. 549-558.

201. Bulau, J.M., Alleged failure to treat a technology-dependent infant in thehome health care setting: a risk management case study. Journal of HomeHealth Care Practice, 1992. 5(1): p. 47-54.

202. Fowler, C., Review of workplace control measures to reduce risks arisingfrom the movement of vehicles: Phase 1 &Phase 2.. 2002, TransportResearch Laboratory: Berkshire. p. Phase 1 - 91p; Phase 2 - 133p.

203. Anonymous, Concord crash report. Engineers Australia, 2002. 74(2): p.23.

204. Anonymous, World's biggest offshore oil platform sinks. EngineersAustralia, 2001. 73(4): p. 32-33.

205. Struttmann, T.W. and A.L. Scheerer, Fatal injuries caused by logs rollingoff trucks: Kentucky 1994-1998. American Journal of Industrial Medicine.,2001. 39(2): p. 203-8.

206. Menser, N., Increasing mine safety through automation. EngineersAustralia: Civil Edition, 2001. 73(1): p. 50-51.

207. Bell, N.S., et al., Self-reported risk-taking behaviors and hospitalization formotor vehicle injury among active duty Army personnel. American Journalof Preventive Medicine, 2000. 18(3 SUPPL): p. 85-95.

208. Belke, J.C., Chemical accident risks in U.S. industry - a preliminaryanalysis of accident risk data from U.S. hazardous chemical facilities..2000, United States Environmental Protection Agency, ChemicalEmergency Preparedness and Prevention Office: Washington, D.C. p. 41.

209. Bailey, L.L. and R.C. Thompson, The effects of performance feedback onair traffic control team coordination: a simulation study, in FAA Office ofAviation Medicine Reports. 2000, ACTU. p. 10.

210. Anonymous, Causes of rail disaster. Engineers Australia, 2000. 72(7): p.16-17.

211. Laitinen, H., M. Marjamaki, and K. Paivarinta, The validity of the TR safetyobservation method on building construction. Accident Analysis &Prevention., 1999. 31(5): p. 463-72.

212. Harrell, W.A., Lifeguards' vigilance: effects of child-adult ratio and lifeguardpositioning on scanning by lifeguards. Psychological Reports, 1999. 84(1):p. 193-197.

213. Baron, J., T.L. Strome, and L.H. Francescutti, The constructionflagperson: a target for injury. Occupational Medicine (Oxford)., 1998.48(3): p. 199-202.

214. Anonymous, Northpower gets everybody involved in making safety its toppriority. Engineers Australia, 1998. 70: p. 46-47.

215. Wuerz, R.C. and R. O'Neal, Role of pilot instrument proficiency in thesafety of helicopter emergency medical services. Academic EmergencyMedicine., 1997. 4(10): p. 972-5.

216. Isaac, A., The Cave Creek incident: a REASONed explanation. AustralasJ Disaster Trauma Stud, 1997. 3(13p).

136

217. Sanchez, J.O., R.R. Aunon, and A.O. Chacon. Emergency planning forunderground coal mining. in 27th International Conference of Safety inMines Research Institutes. 1997.

218. Jones, D.G. and M.R. Endsley, Sources of situation awareness errors inaviation. Aviation, Space, and Environmental Medicine, 1996. 67(6): p.507-512.

219. Cavenagh, C., New safety plan for BHP. Engineers Australia, 1996. 68(3):p. 56.

220. Pettit, T.A., Worker deaths in confined spaces. Professional Safety, 1996.41(11): p. 22.

221. Ashton, D., Human factors/safety attitudes : developing a safety culture.Safety Management, 1994. 10(6): p. 38-39.

222. Gully, S.M., D.J. Whitney, and F.E. Vanosdall, Prediction of police officers'traffic accident involvement using behavioral observations. AccidentAnalysis & Prevention, 1995. 27(3): p. 355-362.

223. Shappell, S.A., Naval flight deck injuries: a review of naval safety centerdata, 1977-91. Aviation Space & Environmental Medicine, 1995. 66(6): p.590-595.

224. Ribak, J., B. Cline, and P. Froom, Common accidents among airportground personnel. Aviation Space & Environmental Medicine, 1995.66(12): p. 1188-1190.

225. Rundmo, T., Associations between organizational factors and safety andcontingency measures on offshore petroleum platforms. ScandinavianJournal of Work, Environment & Health., 1994. 20(2): p. 122-7.

226. Jackson, B., Good design will prevent accidents. Engineers Australia,1994. 66(3): p. 38.

227. Young, J., Newsprint south dramatically reduces mill accident rate. Pulp &Paper, 1992. 66(13): p. 92.

Safe Staffing & Patient Safety Literature Review

Documents

Transcript of Safe Staffing & Patient Safety Literature Review