The Lessons of Bristol (Models for the Future)

The Lessons of Bristol(Models for the Future)

Associate Professor Stephen Bolsin

Department of Perioperative Medicine

The Geelong Hospital

Barwon Health

The Lessons of Bristol

• Other examples• Manitoba Paediatric Cardiac Surgery Inquest (Canada)• Dr Harold Shipman Inquiry (UK)• Dr Michael Swango (USA)• Bristol Royal Infirmary Inquiry (UK)• 40 deaths in private clinic 2000 (France)• Dr Reimers charged with manslaughter 2001 (Aus)

Safety in Health Care

Stimulus to change

• 1995 ‘Quality in Australian healthcare study’

• 1999 ‘To err is human’ Institute of Medicine, USA

• 44-98K patients die through error in US hospitals

• 3-6K patients die through error in Australian care

• Bristol Inquiry 2001 “could be happening now in NHS”

• Public demand for change

Safety in Health CareQuality in Australian Health Care Study 1992

Wilson RMcL, Runciman WB, Gibberd RW, et al. The Quality in Australian Health Care Study. Med J Aust 1995; 163: 458-471.

• 16.6% of admissions had iatrogenic injury• 14,000 admissions to 28 hospitals (NSW & SA)• Extrapolate <230,00 preventable adverse events• Extrapolate <14,000 preventable deaths

Safety in Health CareCauses of Adverse Events (50% preventable)

34.6% complication/failure of technical performance

15.8% decision/action failure

11.8% failure to arrange procedure/investigate/consult

10.9% lack of care/attentionAn analysis of the causes of adverse events from theQuality in Australian Health Care Study

Ross McL Wilson, Bernadette T Harrison, Robert W Gibberd and John D Hamilton

MJA 1999; 170: 411-415


Limited adverse occurrence screening:

using medical record review to reduce

hospital adverse patient events

Alan M Wolff

MJA 1996; 164: 458


Detecting and reducing hospital adverse events: outcomes of the Wimmera clinical risk management program

Alan M Wolff, Jo Bourke, Ian A Campbell and David W Leembruggen

MJA 2001; 174: 621-625

Safety in Health Care Wolff M A. MJA 1996; 164: 458

General patient outcome criteria used for screening

of medical records – Death.

– Return to operating theatre within 7 days.

– Transfer from general ward to intensive care unit.

– Unplanned readmission within 28 days of discharge.

– Cardiac arrest.

– Transfer to another acute-care facility.

– Length of stay greater than 35 days (reduced to 21 days in 1993-1994).

– Theatre booking cancelled.

Safety in Health CareWolff M A. MJA 1996; 164: 458

15,912 patients screened

1,465 (9.21%) screened positive for criteria

155 (0.97%) screened positive for AO

110 major 45 minor

88 (56.8%) cases minor or not preventable

67 recommendations to patient care committee

66 recommendations for changes in policy

Changes in policy clinical & administrative


Wolff M A. MJA 1996; 164: 458• Reasonable rate of detection on screening by clerks • Reasonable rate of confirmation by clinicians• LAOS will detect circa 50% of adverse events• Requires 10% review of medical records• Fast & accurate• Costs <0.1% of total hospital budget • Reduction in adverse events by >50% in 3 years


Wolff AM et al. MJA 2001; 174: 621-625

49,834 inpatients screened

20,050 EMD patients screened

Inpatient record review

EMD record review

Clinical incident reporting

GP reporting


Inpatient adverse events down from 1.35% - 0.74%

(Reduction from 69-49 events in 8 years p<0.001)

EMD adverse events decreased from 3.26% - 0.48%

(Reduction from 84-12 events in 8 quarters p<0.001)

“Adverse events can be detected...”

“...and their frequency reduced using…detection methods and clinical improvement strategies...

Safety in Health CareCauses of Adverse Events (50% preventable)

34.6% complication/failure of technical performance

15.8% decision/action failure

11.8% failure to arrange procedure/investigate/consult

10.9% lack of care/attentionAn analysis of the causes of adverse events from theQuality in Australian Health Care Study

Ross McL Wilson, Bernadette T Harrison, Robert W Gibberd and John D Hamilton

MJA 1999; 170: 411-415

ANZCA Personal Professional Monitoring Project

• Pietroni 1993 Ann RCS;75:200-2.

• de Leval et al. 1994 J Thorac Cardiovasc Surg;107:914.

• Kestin 1995 BJA;75:805-9.

• Ellis 1995 BJA;75:673-4.

• Day & Bolsin 1998 Short Practice of Anaesthesia.

• Bolsin 2000 Int J Qual Health Care;12:367-369.

• Bolsin & Colson Int J Qual Health Care;12:433-8.

• Bolsin 2001 Aust Health Review;24:1-4.


Graph 1

-2

0

2

4

6

8

10

12

0 20 40 60 80 100 120 140 160 180 200

Attempt Number

CumulativeFailure

PROBABLY ACCEPTABLE

DEFINITELY ACCEPTABLE

PROBABLY UNACCEPTABLE

DEFINITELY UNACCEPTABLE


Graph 2

-15

-10

-5

0

5

10

15

0 20 40 60 80 100 120 140 160 180 200

Attempt Number

Cusum


Graph 3

-2

0

2

4

6

8

10

12

14

16

0 20 40 60 80 100 120 140 160 180 200

Attempt Number

Cumulative Failure

PROBABLY ACCEPTABLE

DEFINITELY ACCEPTABLE

PROBABLY UNACCEPTABLE

DEFINITELY UNACCEPTABLE


Graph 4

-15

-10

-5

0

5

10

15

0 20 40 60 80 100 120 140 160 180 200

Attempt Number

Cusum


• Recruits – 1st year Anaesthetic Registrars– Supervisors of Training

• Collects Electronically– Log Book data on procedures– Data on procedural performance– Data on adverse incidents


• Sponsors– ANZCA (Australian & New Zealand College of

Anaesthetists)– United Medical Protection– PALM Corporation of Australasia– VMIA– Sync International


Procedures monitored• IV line insertion• IA line insertion• CVP line insertion• Epidural insertion• Spinal anaesthetic• Brachial Plexus block• Other blocks


• PALM III handheld computers

• Personal log book

• Customised synchronising programmes

• Electronic data collection

• Electronic data retrieval

• Secure electronic data transmission

• Remote Analysis

• Secure return of analysed data


• Data collection in <1 minute

• Cultural change has been achieved

• Enthusiasm is palpable

• Other Specialities are interested

• Other Professions are interested

• Registrars are Specialists of the Future

• Data from any procedures collectable









Palm

Desktop

SecureServer

Anaesth.College

APSF


Phase 1 Project

• Six 1st or 2nd year registrars recruited

• 3 Centres in Australia & New Zealand

• 4-7 month data collection

• 1690 Cases collected

• All collected data analysed

• Some data lost (batteries & breakages)


Phase 1 Project

• Supervision – 62% level 1; 22% level 2

• 27% out of hours

• 11% remote locations

• Operative Speciality

• 480 procedures logged for Cusum analysis


Phase 1 Project

• 42 critical incidents

• 2.5% of total anaesthetics logged

• 19 uneventful; 8 minor;14 major; 1 death

• 64% “near miss” reporting

• cf 50% event reporting by LAOS

• 21 airway respy events; 17 cardiovascular


Phase 2 Project

• 715 anaesthetics recorded

• 17 critical incidents

• 2.4% of total anaesthetics logged

• 7 uneventful; 8 minor;2 major

• 88% “near miss” reporting

• 7 airway; 4 procedure; 2 cardiovascular


• Relatively objective.• Easy and quick to collect.• Provides early feedback.• Provokes specific action early.• Consistent with the “continuing quality

improvement” paradigm.• Allows for ongoing, self directed learning• Phase 2 better than Phase 1


“The Future Now”• Personal Professional Monitoring• Real Time Prospective Reporting• Numerator Data & Denominator Data• “Near Miss” Reporting• Critical Incident Analysis• Targeted Incident Reporting• Immediate Feedback of generic data• Policy & Procedural change as a result


The Vision• Reduction in Adverse Events• Elimination of attributable Adverse Events• Improved Health Care Practice• Reduced Health Care Costs (Au$4-6 billion)• Improved Patient Outcomes• Reduced Legal Costs• Safest Hospital status


• www.ppm.com.au– Articles on CUSUM

– Product details

– Program tour

• www.syncint.com– Look under what’s

new

http://www.ppm.com.au/



http://www.syncint.com/



Professional Monitoring & Cultural Change

A/Prof Stephen Bolsin, Dr Mark Colson,

Dr Peter Stow, Dr Peter Tolley, Mr Morteza Mohajeri, Mr James Kenny, Dr Rory Wolff

Depts Perioperative Medicine, Cardiac Surgery & ICU

The Geelong Hospital

Barwon Health

Dept Epidemiology & Preventive Medicine

Monash University

Complication - Second 500

-20.00

-15.00

-10.00

-5.00

0.00

5.00

10.00

15.00

20.00

500 525 550 575 600 625 650 675 700 725 750 775 800 825 850 875 900 925 950 975

Pump Number

Cu

su

m


Inexplicable change in performance

Complication - Third 500

-20.00

-15.00

-10.00

-5.00

0.00

5.00

10.00

15.00

20.00

1000 1025 1050 1075 1100 1125 1150 1175 1200 1225 1250 1275 1300 1325

Pump Number

Cu

su

m


Protocol Changes

New Surgeon Xmas Holidays

Review Meeting


• Overall bleeding rate 5.5%

• Important variables on univariate analysis– Emergency category– Renal failure (Pre-op creatinine >120mol/L) – Cardiopulmonary bypass time (10% : 10 min)– Surgeon

Complication - Third 500

-30.00

-20.00

-10.00

0.00

10.00

20.00

30.00

1000 1025 1050 1075 1100 1125 1150 1175 1200 1225 1250 1275 1300 1325 1350 1375 1400 1425 1450 1475 1500

Pump Number

Cu

su

m


Review MeetingProtocol Changes

Xmas holidays

New Staff Changed Performance


• Further analysis required

• Renal failure implicated

• Bypass time implicated

• Emergency surgery implicated

• Surgeon in part explanatory

• Surgeon effect remediable

• Decrements in performance explicable

• Increments in performance inexplicable


• We are being expected to do better

• We could do better

• We should do better

• Our patients would want us to do better

• Our patients would benefit

• We would benefit

• “All Win” Medical Management

The Lessons of Bristol (Models for the Future)

Documents

health carewolff

health care wolff

health carequality

health carestimulus

health caredetecting

hospital adverse events

adverse eventsrequires

wolff mja