Interior Health Learning how to learn from Patient Safety Events A journey in adoption of the New View of Human Error Wrae Hill CSRT Meeting St. John NFLD May 15/16.2010 Vision: To set new standards of excellence in the delivery of health services in the Province of British Columbia. Human Factors & Systems Safety Holistic approach to understanding and solving problems that sees relationships and interactions and departs from the old view mechanics of reductionism, decomposing wholes into individual components. Understand how systems maintain themselves in a changing environment by balancing available resources against the pressures, goals, constraints and opportunities……. concepts are those of emergence and of resilience…how we maintain safety amid dynamic complexity. Challenge the common language of incident management and describe the problems of old view Newtonian assumptions of symmetry between cause and effect. Use a new lens to understand human error as a symptom of complex systems. Sidney Dekker – Lund Universitat 2010 PART ONE Check your paradigm A brief history of human factors Selected topics ; 1. 2. 3. 4. 5. The New View of Human Error Hindsight Bias On counting errors Cause is something you construct Making “recommendations” The Field Guide to Understanding Human Error Sidney Dekker 2006 PART TWO Applied Learning from Patient Safety Events Selected topics ; 1. Healthcare Human Factors 2. Reporting culture ( our experience with PSLS) 3. Patient Safety Investigation (PSI) 4. Just Culture - Incident management , disclosure 5. Closing the loop & communicating actioned / audited improvements 6. Examples of learning from adverse events (CILS) Influence of 17th Century Scientism ORGANIC WORLD VIEW 1000-1500 RENAISSANCE 1500-1600 Leonardo da Vinci - Archeotype of the renaissance man. The first complexity / systems theorist SCIENTIFIC REVOLUTION 1600-1700 Rene Descartes 'Cogito, ergo sum,' 'I think, therefore I exist ..created the conceptual framework for seventeenth-century science. Sir Issac Newton‘s laws created a comforting symmetry between cause & effect and the basis for the scientific method. (Determanism) ENLIGHTENMENT 1800-1900 There is no world and man beside it, only the world and man within it (Neitszche) Perspectivism NEW PHYSICS - CONFLICT & COMPLEXITY 1900 – 1950s World Wars ,The New Physics, Cold War Science, Philosophy, Geo politics, Sociology , Psychology merge (Artistry) HUMAN FACTORS 1950-2010 The birth of Human Factors has military aviation roots, Fitts & Jones 1947 Alphonse Chapanis (1917-2002) Father of Human Factors / Ergonomics Fitts, P.M., & Jones, R.E. (1947a). Analysis of factors contributing to 460 ―pil ot error‖ experiences in operating aircraft controls (Report No. TSEAA-694-12). Dayton, OH: Aero Medical Laboratory, Air Materiel Command, U.S. Air Force B-17 Superfortress Bomber 1944 3 % Reach 5% Unintentional activation 6% Unintentional reversal of controls 18% Forgetting to do something 18 % Adjustments 50% Substitution ( Confusing one control for another ) GEAR FLAPS First use of human factors to adapt the technology to the needs of the operator (Pilot’s Perspective) Fitts, P.M., & Jones, R.E. (1947a). Analysis of factors contributing to 460 “pilot error” experiences in operating aircraft controls (Report No. TSEAA-694-12). Dayton, OH: Aero Medical Laboratory, Air Materiel Command, U.S. Air Force Wheels-up C-17 crash caused by pilot error By Bruce Rolfsen - Staff writer Posted : Tuesday May 12, 2009 11:42:17 EDT Pilots of a C-17 Globemaster failed to lower the transport‘s landing gear, forcing them to make a crash landing at Bagram Airfield in Afghanistan, an Air Mobility Command investigation concluded. None of the six onboard were injured; the repair bill for the $200 million aircraft, however, totaled $19 million Anaesthesia Critical Incident Studies www.qualityhealthcare.com • In 1978 Cooper and colleagues published their landmark paper on the application of the critical incident technique, adapted from uses in aviation and other fields, to examine the causes—and later prevention strategies—for adverse anesthesia outcomes. • Following on 20 years of rudimentary anesthesia mortality studies, this was a brilliant approach that gave anesthesia clinicians new insights on which we could act. • As the report told us: ― . . . factors associated with anesthetists and/or that may have predisposed anesthetists to err have, with a few exceptions, not been previously analyzed. Furthermore, no study has focused on the process of error—its causes, the circumstances that surround it, or its association with specific procedures, devices, etc—regardless of final outcome.‖ Human Factors in Anaesthesia Ohmeda Tec 4, 5 With the center vaporizer removed (if three are mounted side by side), one can activate both outer vaporizers simultaneously (in machines manufactured after 1995, this fault is corrected). Vaporizer outlet has check valve. 1. Br J Anaesth 2008;100:333-343 Human factors in anaesthetic practice: insights from a task analysis Phipps ,D. 2. J Am Med Inform Assoc 2006;13:635-642 The Evaluation of a Pulmonary Display to Detect Adverse Respiratory Events Using High Resolution Human Simulator Wachter, S.B. 3. Qual Saf Hlth Care 2002; 11: 227-282 Preventable anesthesia mishaps: a study of human factors, Cooper, J.B. 4. Br. J, Anaesthesia 1987 Human Factors in Accidents Allnutt,M.F 5. Anesthesiology. 1978 Dec;49(6):399-406. Preventable anesthesia mishaps: a study of human factors. Cooper, J.B (Classic paper) 1) The New View of Human Error Cause or Symptom ? NEW VIEW OLD VIEW Why did that make sense at that time ? what goes wrong ‗Bad Apple theory‘ What goes wrong  • Human error is the cause of trouble • To explain failure you seek failures  • You must discover people‘s failures in decisions, judgments, assessments  How to make it right • Complex systems are basically safe  • Unreliable, erratic humans undermine rules, procedures , regulations & defenses  • To make systems safer, tighten procedures, more automation, better supervision • Find and control the ‗badapple‖   Human error is a symptom of deeper problems Do not try to find where they ‗screwed up‘ Understand how their decisions, judgments, assessments made sense to them at the time. How to make it right Complex systems are not inherently safe Complex systems are dynamic tradeoffs between multiple irreconcilable goals (safety/efficiency) People create safety through practice at all levels Human error is not the conclusion of an investigation…it is the beginning The Field Guide to Understanding Human Error S.Dekker 2006 2) Hindsight Bias Retrospective – we are able to select the sequence of events we think occurred based on our knowledge of the outcome Counterfactual – Lays out detail the operators could have or should have done. Biases your investigation to items you now know were crucial, and causes you to artificially narrow your assessment to only a piece of ―w hat was going on at the time.‖ Judgmental - Judges people‘s actions /inactions from your context & perspective, not theirs Proximal - Focus on those people who were closest in time and space to the event, may ignore unit, system, societal factors To understand failure you must first understand your reaction to failure The more you react to failure, the less you will understand The Field Guide to Understanding Human Error S.Dekker 2006 Hindsight Bias One of the most reproducible of all psychological phenomena Fischhoff, B. 1977 “ The future is implausible & the past is incredible ―David Woods Reinforces Newtonian /Cartesian symmetry between cause and effect WYLFYWF • • • • • • • • • • • • • OLD VIEW INVESTIGATIONS NEW VIEW INVESTIGATIONS Down & In Up & Out Tend to be rushed (First story) Societal pressure to seek the cause Hindsight bias confuses our reality with that of the participants Singles out ill performing individuals Finds evidence of erratic, wrong behavior Bring to light bad decisions, inaccurate assessments, deviations from written procedure / guidelines Investigators have limited human factors knowledge RECOMMENDATIONS Numerous , many in-actionable Focus is on retraining Procedural over-specification (less safe) Increased automation ** Remove the bad apples PERSONAL RESPONSIBILITY Individual responsibility may not be well calibrated to individual authority             Second stories takes much longer (Friendly Fire - Snook 2002) Much harder. Resists simple ―ans wers‖ Focus on local rationality – The view from inside…People do what is reasonable given their ; point of view, focus of attention , knowledge & objectives Context & history considered at every step Questions; who gets to decide ? Objectivity is an arrogant illusion Follow the goal conflicts…. Investigators have detailed human factors knowledge RECOMMENDATIONS May be quite high level (Moshansky – Air Ontario) Make the right things easier to do Fewer, simpler, brief ,flexible guidelines Focus on learning and share generously TEAM – CONTEXT FOCUSED The Field Guide to Understanding Human Error Sidney Dekker Lund University – Sweden 2006 3) On counting “Errors” • You can‘t count errors • Error classification systems look unsuccessfully for simple answers to the sources of trouble and sustain the myth of a stubborn 70% human error. • This also makes artificial distinctions between human and mechanical failure. Sidney Dekker 2006 4a) Cause is something you construct Organizational learning ? CDN 75 y/o person dies of Med Error.. Non-diabetic received insulin that was not prescribed . Pt died, Nurse not fired , no smoking gun found ….. Family involved in disclosure and investigation . Detailed RCA found 4 potential causes . 4b) Cause is something you construct Organizational learning ? • USA 16 y/o Mom to be dies of Med Error. • Infusion intended for epidural route was given intravenously. • Pt died, baby lived, Nurse fired , Detailed RCA found 4 proximate « causes ».. • Shaping systems for better behavioral choices: lessons learned from a fatal medication error. Smetzer J, Baker C, Byrne FD, Cohen MR. J Qual Patient Saf. 2010 Apr;36(4):152-63. • We have Newton on a retainer: reductionism when we need systems thinking. Dekker SW. Jt Comm J Qual Patient Saf. 2010 Apr;36(4):152-63. 5) Actionable Recommendations • Recommendations must be realistic and locally actionable : Specific, Measurable, Appropriate, Realistic, Timely. • Make sure you assign accountability to a specific local person (with their permission) , who has both responsibility & the required authority , resources and a specific timeline. • Are these easy to audit & verify ? • Always consider applied human factors* and the hierarchy of effectiveness: 1. Forcing functions ( MOST EFFECTIVE ) 2. Automation / computerization 3. Simplification / standardization 4. Reminders, checklists, double checks 5. Rules and policies 6. Education 7. Information ( LEAST EFFECTIVE ) From Cdn Root Cause Analysis Framework PART TWO Applied Learning from Patient Safety Events Selected topics ; 1. Healthcare Human Factors 2. Reporting culture ( our experience with PSLS) 3. Patient Safety Investigation (PSI) 4. Just Culture - Incident management , disclosure 5. Closing the loop & communicating actioned / audited improvements 6. Examples of learning from adverse events (CILS) Healthcare Human Factors - WHO 1. 2. 3. 4. 5. Organizational Safety Culture Managers‘ Leadership Communication Team (structures and processes) Team Leadership (supervisors) 6. Situation Awareness 7. Decision Making 8. Stress 9. Fatigue 10. Work Environment Rhona Flin, Jeanette Winter, Cakil Sarac, Michelle Raduma Industrial Psychology Research Centre, University of Aberdeen, Scotland Report for Methods and Measures Working Group of WHO Patient Safety. Geneva: World Health Organization April 2009 Improving the quality of quality of care reviews Incident Management Policy / Process Training & Audit Incident Management Follow Up Improved Systems of Incident Management Building Capacity in Patient Safety Investigation & Follow-up Reporting Patient Safety Investigation (PSI) Improving the quality, of quality of care reviews and merges Medical QI , Risk Mgmt & Quality Improvement Intensive three day learning lab. 12-18 month continuous mentorship, learning and support 1) Utilizes a new lens for understanding human error as an organizational problem of complexity 2) Ensures timely and thorough [best practice] investigations using appropriate tools and sensitivity 3) Results in audited follow up with actions and shared improvements utilizing IH tools and policy Incident Reporting 1° Disclosure Unexpected Severe Harm or Death ? Use critical incident checklist Severity Decision Investigation Follow up Disclosure Recommendations Implement changes Audit changes Share learning Supportive & Just Culture Support for Health Care Staff Interior Health is committed to promoting a just and trusting culture of safety in which it‘s health care providers can readily report Incidents in order to learn and work to improve the safety of patient care. While professional standards are known and understood, when there has been a failure in the provision of care, Interior Health commits to: 1) Appropriate care and support for patients, families and healthcare providers 2) Evaluating all systemic factors, that may have contributed to failure. 3) Following established fair, processes for evaluating actions and behaviors Critical Incident Learning Summaries • Learning from Adverse events & sharing de-identified case summaries. • The demonstrable outcome of a thoroughly reviewed critical incident • This is what our staff , on patient safety culture surveys , want to know. Closing the loop on system improvements 29 Human factors approach to a recurrent clinical technical problem. MAV solution created by clinical RRTs. Published in CJRT CSRT/ CSA Technical Ctte / Health Canada are interested 8 months work 31 Automation ―surpri se‖ in Meditech made it possible to magnify patient ID risks Thankfully, an easy fix 4 months work Management of Anticoagulants with over the counter cough medications 1 months work Common narcotic adverse event. Corrected at source, and well communicated 2 weeks work Easier to resuscitate a patient who is alive Brindley Have you retooled your code team into a ; MET, RRT, CCOT ? Survival following CPR in hospital has not changed in 40 years Survival - witnessed 48.3% resuscitated, 22.4% to discharge, and 18.9% made it home Survival – Unwitnessed 21.2% resuscitated, but only 1 patient (1.0% ) to hospital discharge and was able to return home Brindley, P. CMAJ 2002 Implementing Rapid Response teams (collaboratives) Hill,W. CJRT 2005 Insanity: Doing the same thing over and over again and expecting different results. Albert Einstein A simple tool for clear, concise communication in urgent circumstances with appropriate assertion….. SBAR Situation the punch line 5-10 seconds Background the concise context, objective data Assessment what is your assessment of the situation? Recommendation what do we need to do now ? Nine Steps to Move Forward on Safety D. Woods / R.I. Cook 1. Pursue Second Stories underneath the surface to discover multiple contributors. 2. Escape the hindsight bias. 3. Understand work as performed at the sharp end of the system. 4. Search for systemic vulnerabilities. 5. Study how practice creates safety. 6. Search for underlying patterns. 7. Examine how change will produce new vulnerabilities and paths to failure. 8. Use new technology to support and enhance human expertise. 9. Tame complexity through new forms of feedback. Leonardo da Vinci Center for Complexity and Systems thinking LUND UNIVERSITY , SWEDEN www.lusa.lu.se Wrae Hill BSc.RRT Corporate Director Quality Improvement & Patient Safety Interior Health ,Kelowna, BC. V1Y4N7 P: 250-870-5893 E: [email protected]