6th Annual Evidence-Based Practice & Research Conference

Preventing Alert Fatigue through use of Predictive Analytics Managed by a Virtual Quality Team

August 23,2019NorthBay Healthcare Green Valley Administration BuildingFairfield , CA

Session Presenters

Laura S. Maples MSN RN CCRN-e Quality Nurse Consultant – Clinical Quality Programs and Analytics – Northern California Quality

Danette Williams BSN RN Quality Nurse Consultant – Clinical Quality Programs and Analytics – Northern California Quality

Elizabeth Scruth RN PhD MPH CNS CCRN CCNS FCCM FCNS CPHQ, Director of Tele Critical Care – Clinical Quality Programs and Analytics – Northern California Quality

DISCLOSURES

Presenters have no conflicts of interest to report

Photos, except where noted are shown with permission and not in violation of copyright

Learning Objectives

4 | Copyright © 2017 Kaiser Foundation Health Plan, Inc.

1. Describe predictive analytics in the context of early warning systems

2. Describe components of an early warning system

3. Describe the use of a Virtual Quality Team to mitigate alert fatigue and ensure data capture for measuring outcomes

INTRODUCTION

5 | Copyright © 2017 Kaiser Foundation Health Plan, Inc.

Evolving healthcare technologies impose upon nurses the ever-growing challenge of Alert Fatigue

While caring for their patients, they are continuously exposed to:

•Hypersensitive Bed alarms•Acutely exaggerated IV pump alarms•Fatal alarms for a disconnected pulse oximetry•Talking Hill-Rom beds•Every 5-15mins vital signs•Neuro-checks q 2hrs•Hourly rounds•Ventilator alarms•BiPAP/CPAP alarms•--- and now …Electronic Health Record Chart Alarms

EHR Chart Alarms!!!

Electronic Health Record Chart Alarms!!!

...like Modified Early Warning Scores (MEWS)

Image:rapid.leeds.sc.uk

Key Goals of the AAM/Virtual Quality Team Program

FIRST-

• Promotion of early recognition of physiological deterioration

• Predicting today who will be in ICU tomorrow

SECOND-

• Reduction of alert fatigue for bedside clinicians

• Support the facility level standardized workflows

Key Goals of the AAM/Virtual Quality Team Program

Background/Significance

• Early detection of deterioration in non ICU patients is challenging

• Failure to recognize and rescue deterioration are preventable quality and patient safety concerns

• Increased morbidity and mortality in these patients1-5

PREDICTIVE ANALYTICS

Predictive Analytics allows us to predict deterioration and address patient needs proactively

PREDICTIVE ANALYTICS

• Predictive analytics began as an early warning system strategy

• Use of big data allow algorithm derived predictions

• Many predictive analytic programs exist-Advance Alert Monitor (AAM)-Modified Early Warning Score (MEWS)

PREDICTIVE ANALYTICS

The predictive analytics AAM program was developed by Kaiser Permanente Division of Research

Implemented by Quality Division

COMPONENTS OF AAM: Advance Alert Monitor

• 99 parameters

• 12 hour lead time

• Reactive to proactive approach

THE AAM/VIRTUAL PROGRAM

• Assists clinicians with clinical pathway decisions for rescue therapies

• Integrates supportive care for patient-centered social work

• Provides palliative care preferences

The Virtual Quality Team

• Masters prepared nurses

• Critical care backgrounds

• Support the facility level standardized workflows

THE AAM/VIRTUAL QUALITY TEAM PROGRAM

Quality Nurse Consultant:• Receives alert

• Reviews alert

• Calls Rapid Response RN at the hospital –gives SBAR

• RRT RN assesses patient/notifies MD

PROCESS IMPROVEMENT

• Expand an existing virtual e-Hospital platform• Spread AAM from pilot Alpha and Beta sites

• Develop a remote surveillance team of Quality Nurse Consultants

+ +

AAM IMPLEMENTATION

A multi-disciplinary leadership team of:• Quality Nurses/

• Rapid Response Nurses(RRT)

• Hospital-Based Specialists (HBS)

• Palliative Care

• Social Services

Aligned with:• Hospital Leadership

• Regional Leadership

AAM/VIRTUAL QUALITY TEAM SPREAD

PROGRAM DEVELOPMENT AND IMPLEMENTATION

The Virtual Quality Team ensured

• Standardized continuous monitoring process

• Workflow refinement

Data Data Data

The AAM Program:

• Supports robust data analytics

• Provides performance improvement data on a SharePoint website

• Allows facilities to view their weekly and monthly data

PROGRAM INDICATORS

AUTOMATED EARLY WARNING SYSTEMS CAN MITIGATE FAILURE TO RECOGNIZE, COMMUNICATE, OR ACT ON SIGNS OF

EARLY CLINICAL DETERIORATION

A PATIENT’S DECLINE IS GENERALLY PRECEDED BY CHANGES IN A PATIENT’S

RESPIRATORY RATE, HEART RATE, OXYGENATION AND MANY OTHER CLINICAL TRIGGERS APPROXIMATELY 6 TO 24 HOURS PRIOR TO CLINICAL DETERIORATION (6–10)

EARLIER RECOGNITION AND ACTION CAN REDUCE THE ASSOCIATED

ADVERSE EVENTS SUCH AS UNPLANNED ADMISSIONS TO THE ICU

AND UNEXPECTED DEATHS (11)

How does AAM differ from MEWS?

The AAM/Virtual Quality Team Program differs from MEWS in three ways:• Technology

• Process

• Multidisciplinary Collaboration

Technology:

MEWS scores only: • heart rate • blood pressure • respiratory rate • temperature • mentation

www.caregiverology.com/mews-score.html

The AAM program’s algorithm has over 99 parameters including:

--LAPS- laboratory-based acute physiology score

--COPS- comorbidity point score

AAM is more robust and each parameter is weighted according to risk

COPS2

COmorbidity Point Score (vs 2) (COPS2) – a dimensionless number based on all diagnoses a patient has received in the preceding 12 months.

The higher the COPS2, the greater the comorbidity burden

LAPS2

Laboratory-based Acute Physiology Score (vs 2) (LAPS2) – a dimensionless number based on physiologic data from the preceding 24 or 72 hours (vital signs, neurological status, pulse oximetry, and 16 laboratory tests)

The higher the LAPS2, the more unstable the patient

AAMScore

AAM Score – discrete probability estimate of the likelihood of deterioration in the next 12 hours

Based on age, sex, COPS2, LAPS2, care directive, individual vital signs, vital signs trends, individual laboratory tests, and how long a patient has been in the hospital

AAM and The Algorithms Feeding Into It

Process:

AAM/Virtual Quality Team Program

The AAM program captures data

inputted to the EMR every hour

and sends alerts to the AAM virtual quality team nurse

Modified Early Warning Score

MEWS operates as an alarm

within the chart for the primary

nurse to review, adding to other

environmental and chart alarms

AAM/Virtual Quality Team Program

The AAM program captures data

inputted to the EMR every hour and

sends alerts to the AAM virtual quality team nurse

Modified Early Warning Score

MEWS operates as an alarm

within the chart for the primary

nurse to review, adding to other

environmental and chart alarms

and alert fatigue

Multidisciplinary Collaboration

The AAM Quality Nurse Consultant activates multidisciplinary collaboration between

• Rapid Response Nurse

• Social Services/Palliative care services

• HBS- whose orders generate action from1. Laboratory2. Radiology3. Respiratory4. any other department

Translating Predictive Analytics into Highly Reliable Care

Identify highest risk inpatients to prevent deterioration and reduce mortality and LOS

Respect patient goal of care

Shift safety culture from reactive to proactive

Quality Virtual Nurse performs Expert review of AAM Alert

Maximize benefits of EMR and data science

Standard Work for alert response by RN &Physician to assess and evaluate the patient, communicate robust plan and expedite orders

Integration of palliative care and Social services teams for symptom management and life-care planning

On-site Education, coaching, implementation, support to train staff & leaders

Notifies RRT RN Reduces Alarm Fatigue

Integration of AAM model into KP HealthConnect to effectively scale & maintain technology

EVALUATION

• In many of our hospitals, there is inconsistency between when the clinical deterioration occurs and when it is detected

• It was also noted that there is variation in clinician response

The AAM Program ensures:• Standardized actions and responses• Reduced alarm fatigue• Improved interdisciplinary communication

CONCLUSIONS/OUTCOMES

• AAM is a technology for proactive detection.

• We have developed a standardized spread process alongside a highly skilled virtual quality team.

• Team of experts , using common language and agreements on how we work together• Reduction in LOS and mortality

• A difference in differences study was done and the results with currently in review for publication

• We are now working with other KP Regions to assess for inter-operational transferability

• The AAM Program is a first step toward a vision where predictive analytics and remote monitoring ensure patients remain safe from harm

• Ensures their goals are incorporated into treatment decisions before adverse outcomes occur

LESSONS LEARNED

• The overall goal of implementing AAM in 21 KPNC hospitals was achieved through leadership alignment.

• Regional and Hospital leadership teams jointly prioritize, promote, and implement resources.

• A system-wide approach of leadership and multidisciplinary collaboration made this innovation possible

Questions

References:1. Gerber DR, Schorr C, Ahmed I, Dellinger RP, Parrillo J. Location of patients before transfer to a tertiary care

intensive care unit: impact on outcome. J Crit Care. 2009 Mar;24(1):108–13.

2. Escobar GJ, Greene JD, Gardner MN, Marelich GP, Quick B, Kipnis P. Intra-hospital transfers to a higher level of

care: contribution to total hospital and intensive care unit (ICU) mortality and length of stay (LOS). J Hosp Med.

2011 Feb;6(2):74–80.

3. Bapoje SR, Gaudiani JL, Narayanan V, Albert RK. Unplanned transfers to a medical intensive care unit: causes and

relationship to preventable errors in care. J Hosp Med. 2011 Feb;6(2):68–72.

4. Liu V, Kipnis P, Rizk NW, Escobar GJ. Adverse outcomes associated with delayed intensive care unit transfers in

an integrated healthcare system. J Hosp Med. 2012 Mar;7(3):224–30.

5. Delgado MK, Liu V, Pines JM, Kipnis P, Gardner MN, Escobar GJ. Risk factors for unplanned transfer to intensive

care within 24 hours of admission from the emergency department in an integrated healthcare system. J Hosp

Med. 2013 Jan;8(1):13–9.

6. Ludikhuize J, Smorenburg SM, de Rooij SE, de Jonge E. Identification of deteriorating patients on general

wards; measurement of vital parameters and potential effectiveness of the Modified Early Warning Score. J

Crit Care. 2012;27(4):424.e7-13.

7. Prytherch DR, Smith GB, Schmidt PE, Featherstone PI. ViEWS--Towards a national early warning score

for detecting adult inpatient deterioration. Resuscitation. 2010;81(8):932–7.

8. Mapp ID, Davis LL, Krowchuk H. Prevention of Unplanned Intensive Care Unit Admissions and Hospital

Mortality by Early Warning Systems. Dimens Crit Care Nurs. 2013;32(6):300–9.

9. Boniatti MM, Azzolini N, Viana M V., Ribeiro BSP, Coelho RS, Castilho RK, et al. Delayed medical

emergency team calls and associated outcomes. Crit Care Med. 2014;42(1):26–30.

10. Smith GB, Prytherch DR, Meredith P, Schmidt PE, Featherstone PI. The ability of the National Early

Warning Score (NEWS) to discriminate patients at risk of early cardiac arrest, unanticipated intensive care

unit admission, and death. Resuscitation. 2013 Apr;84(4):465–70.

11. Mitchell IA, McKay H, Van Leuvan C, Berry R, McCutcheon C, Avard B, et al. A prospective controlled trial

of the effect of a multi-faceted intervention on early recognition and intervention in deteriorating hospital

patients. Resuscitation. 2010;81(6):658–66.