ERRORS AND PATIENT SAFETY
| A COGNITIVE AUTOPSY OF THINKING ERRORS —James R. Hubler, MD, Clinical Assistant Professor of Surgery,
Department of Emergency Medicine, University of Illinois College of Medicine, Peoria
|
| Why diagnostic errors made: thinking—key to any diagnosis; guides multiple decisions required at every stage to
arrive at diagnosis; 3 major drawbacks of models for analysis and prevention of medical errors—decisions not made
under optimal conditions by rational beings who understand and apply principles of probability; models do not consider
prevailing intrinsic or extrinsic systemic error-producing conditions (eg, noisy environment) and violation-producing behaviors
(incomplete appreciation of heuristics (shortcuts that mind uses); heuristics driven by cognitive disposition to respond
(CDR) to particular patient in particular situation
|
| Error-producing conditions: intrinsic—CDR; high diagnostic uncertainty; high decision density; high cognitive
load; narrow time windows; multiple transitions of care; multiple interruptions; low signal-to-noise ratio (low likelihood
of critical diagnosis vs benign diagnosis); surge phenomenon; circadian desynchronicity; systemic—inexperience; high
communication load; overcrowding; production pressures; high noise levels; inadequate staffing; poor feedback (patients
lost to follow-up); inadequate supervision; resource availability continuous quality improvement trade-off (RACQITO;
trade-off quality to get patients through system); violation-producing behaviors—risk-taking behavior (more likely in
men than in women); normalization of deviance; maladaptive group pressures; maladaptive coping behaviors; underconfidence
(may lead physician to order tests that cause patient injury); overconfidence (may cause physician to underorder
tests, resulting in missed diagnoses); maladaptive decision styles; authority gradients (between physicians and, eg, pharmacists,
paramedics); likelihood of detection
|
| Sources of error: no fault—unreliable information from patient; misrepresentation of illness (malingering); somatization;
insufficient information about new disease; patient refusal; silent presentation of comorbid illness; provide some absolution
for health care provider (least culpable); experience limits chance for errors; systemic—error-producing
conditions; trade-offs that sacrifice quality; laboratory error; poor follow-up; time delays; unavailable services (eg, late at
night); poor patient follow-up; systemic errors not under control of emergency department (ED) physician; ED continuous
compromise of system failings, and many physicians and nurses come to live with them, resulting in acceptance of
systemic problems; cognitive—most due to CDRs
|
| Cognitive dispositions to respond: anchoring—tendency to lock onto significant features of patient’s initial presentation
too early and failing to adjust initial impression in light of new information; compounded by confirmation bias;
diagnosis momentum—diagnostic labels attached to patients tend to become stickier and stickier; what starts as possibility
gains momentum until it becomes definite, to exclusion of all other possibilities; feedback sanction—ignorance
trap or time-delay trap; making diagnostic error may carry no immediate consequence; only 1 in 17 patients with compensable
injury sues ED physician; time may elapse before error discovered; poor system feedback; fundamental attribution
error—tendency to be judgmental and blame patient for illness (dispositional cause) rather than examine
circumstances possibly responsible; frequently seen in psychiatric patients, minorities, and other marginalized groups
(eg, smokers, alcoholics); hindsight bias—knowing outcome may profoundly influence perception of past events and
prevent realistic appraisal of what occurred; may compromise learning and evaluation of case through underestimation
(illusion of failure) or overestimation (illusion of control) of decision-maker’s ability; omission bias—tendency to inaction
rooted in principle of nonmalfeasance; belief that events that occur through natural progression of disease more acceptable
than those health care provider might do that would cause injury; overconfidence bias—tendency to believe we
know more than we do; gives rise to acting on incomplete information, intuition, or hunches; too much faith placed on
opinion, instead of carefully gathered evidence (seen in some veteran paramedics and “burned out” ED physicians or
nurses; posterior probability error—gambling on continuation of sequence of events; premature closure—powerful
CDR accounts for high proportion of missed diagnoses; tendency to apply premature closure to decision-making process;
accepting diagnosis before fully verified; psych-out error—leads to delay in diagnosis because patient’s symptoms attributed
to psychiatric illness or condition; representativeness restraint—looking for prototypical manifestations of
disease; search satisficing—universal tendency to call off search once something found; comorbidities, second foreign
bodies, other fractures, and coingestants in poisoning may be missed; triage cueing—triage theory “geography is destiny”;
unpacking principle—failure to elicit all relevant information; if patients limit their history-giving or physicians
limit history-taking, unspecified possibilities may be discounted (eg, fall in elderly patient could be due to syncope; physician
must ask); visceral bias—influence of affective sources of error on decision making; as with countertransference,
negative and positive feelings toward patients may result in diagnoses being missed
|
| Case 1: patient fell 15 ft from ladder onto concrete; brought by ambulance to ED without immobilization; x-rays of both
ankles showed fractured right ankle (swollen and deformed); referred to orthopedist who put on walking cast; 6 wk later,
cast removed (patient complaining of foot pain); 4 wk later, foot x-ray (ordered by another physician) revealed calcaneal
fracture; orthopedic shoe ineffective; bone fusion required
|
| Analysis of case 1: considerations in fall from height—calcaneal fracture; injuries to lumbar spine; jump (suicide attempt);
errors—deficient knowledge (about falls from heights); search satisficing (ankle fracture found, no search for
other injuries; 15%-20% of traumatic injuries missed on initial examination); diagnostic anchoring (orthopedist fixated
on ankle fracture, despite further information about foot pain); diagnosis momentum (diagnostic label of ankle fracture
stuck to patient; premature diagnostic closure (once diagnosis made, no investigation of further complaints)
|
| Case 2: patient presented to ED with stroke-like symptoms; stroke protocol started; computed tomography (CT) appeared
normal; ED physician makes diagnosis of conversion reaction and gives psychiatric referral; symptoms improved, then
worsened next day, and she presented to same ED; subtle findings and positive Babinski reflex; diagnosed with conversion
disorder again and referred to psychiatric institute; ED physician noted patient “medically cleared and otherwise stable,”
but requested electroencephalography (EEG) and neurology consult (canceled by staff psychiatrist at psychiatric
institute); no further tests ordered; repeated falls, urinary incontinence, and confusion during 3 wk as psychiatric inpatient;
discharged to outpatient mental health facility despite poor ambulation; 18 days later, patient had severe disabling
stroke; neurologist diagnosed stroke secondary to thrombotic thrombocytopenic purpura (TTP; cause of psychiatric
symptoms)
|
| Analysis of case 2: psych-out error—all problems attributed to psychiatric diagnosis (patient had no previous psychiatric
problems); term “medical clearance”—misleading and medicolegally dangerous; implies no organic basis for patient’s
condition; better to describe patient as stabilized; caveat—physician should never rely on another physician to do
his or her work; ED physician should have obtained neurologic consult himself; error to think problem neurologic, yet
admit patient to psychiatric facility; other errors—diagnosis momentum; premature diagnostic closure (conversion disorder
diagnosis of exclusion); once patient in psychiatric care, one-half of major medical diagnoses missed, so ED physician
must rule out medical causes for delirium or confusion; errors of staff psychiatrist—vertical line thinking
(concentrated on conversion disorder despite patient’s ataxia and urinary incontinence); overconfidence (canceled neurologic
consult and EEG); diagnostic anchoring; premature closure
|
| Case 3: patient (39-yr-old woman) has history of 9 ED visits for complaints of chest pain during 6 mo following motor vehicle
accident (MVA); no steps taken to diagnose chest pain; patient repeatedly told pain result of MVA trauma; patient
had family history of risk factors (not obtained) and was smoker; patient had fatal cardiac arrest; $2 million awarded at
verdict
|
| Analysis of case 3: fundamental attribution error—patient labeled “frequent flyer”; problem attributed to disposition
of patient rather than to underlying medical condition; malingering, secondary gain, possible drug-seeking behavior, all
attributed to patient over series of visits; negative countertransferance—ED physician develops dislike of patient because
of repeated visits; diagnostic anchoring—also fostered by repeat visits; posterior probability error—seventh,
eighth, and ninth physicians to see patient continue diagnosis of chest wall pain from MVA; other errors—diagnostic
momentum; premature diagnostic closure; possible solutions—ECG at triage; risk factor analysis; observation unit; low
threshold for stress testing in patients complaining of chest pain
|
| How to think of error: not individual failing; only minority of errors due to negligence or misconduct (majority misdiagnoses);
designating error as failure can make problem worse, as it may lead to hiding error; must involve physicians
and nurses in process and not be punitive; malpractice suit or error may harm physician inadvertently (second victim); in
study by American College of Legal Medicine, sued ED physicians 4 times more likely to be sued in following year (physician
distracted; being sued stressful event that also labels physician)
|
| How to prevent medical errors: reduce complexity (more steps equals more errors); optimize information processing
(use, eg, checklists, reminders, prompted charting, protocols); use constraints; mitigate unwanted side effects of change
|
| Building safe health care system: develop policies to prevent errors—reporting should include qualified indemnity
and confidentiality; separate data collection from disciplinary procedures; disciplinary systems should differentiate
between acceptable and unacceptable behavior and involve peers; develop procedures—training in recognition of errors;
feedback on recurrent error patterns; awareness that procedures and protocols cannot cover all circumstances;
metacognition—stepping back and reflecting on one’s thinking
|
| Debiasing strategies to reduce diagnostic errors: provide descriptions of CDRs; force consideration of alternatives;
improve accuracy through cognitive aids (eg, mnemonics, clinical practice guidelines, algorithms, Breslow tapes;
handheld computers; electronic data information source [EDIS] systems; simulation model that uses heuristics and
CDRs); minimize production pressures and understaffing
|
| Summary: CDRs must be incorporated into clinical teaching; morbidity and mortality rounds should be restructured to
promote cognitive autopsies; risk management through cognitive feedback; individual feedback changes behavior
|
| LIVING WILLS AND DNR ORDERS: IS PATIENT SAFETY COMPROMISED? —Ferdinando L. Mirarchi, DO, Chairman,
Department of Emergency Medicine, Hamot Medical Center, Erie, PA
|
| Living wills: state and national recognition but never evaluated for patient safety
|
| Patient’s code status: patients with living wills often thought to be “do not resuscitate” (DNR); DNR does not equal do not
treat; alarming that many DNR orders get written for patients with living wills when DNR order may not be appropriate or understood
|
| Problems with living wills: template; often not read and not understood; by law, whoever has primary responsibility
for patient is attending physician (could be family doctor, intensivist, ED physician); terminal condition—in eyes of
law, terminal condition means does not respond to sound medical treatment; “do not” section—when choice required
in seconds to minutes, someone may scan document and not institute care; understanding of document important;
terms—“effective” means living will valid and exists; “enacted” means living will effective and activated by listed triggers
(eg, persistent vegetative state, terminal illness, permanent unconsciousness); living wills—not activated because
they exist; do not state “do not treat me if I suffer from critical illness”
|
| Physician involvement in advanced care planning: physicians—do not address issue of end-of-life care or living
wills because they do not feel compensated for effort; avoid issue for fear of causing pain and being bearer of bad
news; lack knowledge of laws on advanced directives and training in delivering bad news; view death as enemy to be defeated;
anticipate disagreement with patient and/or family; have medicolegal concerns and feel threatened by such discussions;
speaker thinks DNR equates to death penalty; only reprieve when appropriateness of DNR order questioned;
DNR misunderstood or misinterpreted to mean patient would not wish to receive any lifesaving or supporting interventions
or care
|
| DNR scrutinized: 1983 Presidential Commission on Deciding to Forego Life-Sustaining Treatment—any DNR
policy should ensure that order not to resuscitate has no implications for any other treatment decisions; DNR defined in literature
as designation not to intervene if patient found pulseless or apneic (dead); nowhere stated in medical literature that
DNR means “do not treat” or reduced level of care; published data (≈30 yr) describe less aggressive care and treatment for
DNR patients; Beach et al—study confirmed that physicians less aggressive with DNR patients (less likely to transfuse,
transfer to intensive care unit [ICU], order diagnostic studies, utilize critical care monitoring and procedures, and perform
intubation); Keenan et al—reviewed influence of DNR order on patients admitted to surgical ICU at cancer center; concluded
that DNR order resulted in fewer medical interventions and chart documentation but nursing care remained unchanged;
Bedell et al—showed that DNR orders frequently entered on patient charts without discussion with patient or
informed consent, even if patient competent to participate in such discussion
|
| Nursing care: survey showed 47% of critical care nurses failed to distinguish DNR from other end-of-life decisions;
72% felt DNR patients should not receive aggressive interventions, and 65% felt patient with DNR designation should
not be admitted to ICU; Hennenman et al—confirmed that nurses significantly less likely to perform basic care for critically
ill patient with DNR; possibility of misinterpreting DNR to mean more than “no cardiopulmonary resuscitation
(CPR)”; Shelly et al—used case scenarios and found that increasing age as well as DNR order significantly decreased
aggressiveness of nursing care; nurses delay notifying physician of significant change in clinical status of patient deemed
DNR
|
| Hospital administrations: many have policies that limit DNR patients from being admitted to ICUs; yet, severely depressed
patients, when treated, often change their minds (ie, feel better and willing to be treated more aggressively)
|
| Speaker’s survey:question 1—involved showing living will and asking for interpretation without clinical information;
64% of physicians, 79% of nurses, and almost 90% of prehospital personnel after looking at living will, felt patient
DNR; question 2—looked at understanding of DNR; 50% of physicians thought DNR meant comfort care or end-of-life
care, and 50% thought it meant full care; 68% of nurses and almost 90% of prehospital personnel thought it meant comfort
care and end-of-life care; physicians felt responsibility to talk over advanced directives but did not feel compensated
for this; overall, ≈80% of participants, after looking at living will, thought patient DNR (statistically significant), regardless
of prior training or education
|
| Summary: speaker thinks patient safety compromised by living wills and DNR orders; living wills need further patient
safety data; patients’ wishes should trump DNR order, living will, and proxy
|
Suggested Readings
Beach MC et al: The effect of do-not-resuscitate orders on physician decision-making. J Am Geriatr Soc 50:2057,
2002; Bedell SE et al: Do-not-resuscitate orders for critically ill patients in the hospital. How are they used and what is
their impact? JAMA 256:233, 1986; Croskerry P: The importance of cognitive errors in diagnosis and strategies to minimize
them. Acad Med 78:775, 2003; Dyer C: Bill clarifies gap in law over living wills. BMJ 328:1516, 2004; Dyer C:
Code sets out framework for "living wills". BMJ 332:623, 2006; Dyer C: Living wills will have to specify treatments that
patient is refusing. BMJ 328:1035, 2004; Gordon R et al: Cognitive underpinnings of diagnostic error. Acad Med
78:782, 2003; Gorman TE et al: Residents' end-of-life decision making with adult hospitalized patients: a review of the
literature. Acad Med 80:622, 2005; Graber M et al: Reducing diagnostic errors in medicine: what's the goal? Acad Med
77:981, 2002; Hanchett JM: Advance care planning. N Engl J Med 350:1470, 2004; Jackson EA et al: Do-not-resuscitate
orders in patients hospitalized with acute myocardial infarction: the Worcester Heart Attack Study. Arch Intern
Med 164:776, 2004; Keenan CH et al: The influence of do-not-resuscitate orders on care provided for patients in the
surgical intensive care unit of a cancer center. Crit Care Nurs Clin North Am 12:385, 2000; Landrigan CP et al: Effect
of reducing interns' work hours on serious medical errors in intensive care units. N Engl J Med 351:1838, 2004; Lynn
J et al: Advance care planning for fatal chronic illness: avoiding commonplace errors and unwarranted suffering. Ann Intern
Med 138:812, 2003; Schoenbaum SC et al: Malpractice reform must include steps to prevent medical injury. Ann
Intern Med 140:51, 2004; Upadya A et al: Patient, physician, and family member understanding of living wills. Am J
Respir Crit Care Med 166:1430, 2002; Volpp KG et al: Residents' suggestions for reducing errors in teaching hospitals.
N Engl J Med 348:851, 2003.
Cultural and Linguistic Resources
In compliance with California Assembly Bill 1195, Audio-Digest Foundation offers selected cultural and linguistic resources on
its website. Please visit this site: www.audio-digest.org/ CLCresources.
Educational Objectives
| The goals of this program are to reduce medical errors and minimize the risks to patient safety in the presence of living wills
and do-not-resuscitate (DNR) orders. After hearing and assimilating this program, the clinician will be better able to:
|
 | 1. Recognize the error-producing conditions and sources of error involved in making a diagnosis.
|
| 2. Distinguish the cognitive dispositions to respond responsible for diagnostic errors.
|
| 3. Apply policies and procedures that promote a safe health care system.
|
| 4. Describe problems associated with living wills and DNR orders.
|
| 5. Recognize appropriate circumstances for enacting a living will.
|
Faculty Disclosure
In adherence to ACCME Standards for Commercial Support, Audio-Digest requires all faculty members to disclose relevant
financial relationships within the past 12 months that might create any personal conflicts of interest. Any identified
conflicts were resolved to ensure that this educational activity promotes quality in health care and not a proprietary business
or commercial interest. For this program, the following has been disclosed: Dr. Hubler is a risk consultant for The Sullivan
Group.
Acknowledgements
Dr. Hubler was recorded at Downstate Emergency Medicine Conference, held February 15, 2007, in Peoria, IL, and sponsored
by the Illinois College of Emergency Physicians and American College of Emergency Physicians. Dr. Mirarchi was
recorded at Grand Rounds, held April 12, 2007, in Erie, PA, and sponsored by Hamot Medical Center. The Audio-Digest
Foundation thanks the speakers and the sponsors for their cooperation in the production of this program.
|
