The Case of the Unexpected Outcome presented an attending trying to deal with a poor outcome in one of their patients despite competent medical management. No matter how good of a clinician we are, odds are that at some point this will happen to all of us. In addition to being potentially emotionally devastating, a serious miss can make us question our competence and shift our practice patterns from evidence- to anecdote-based. Check out the ALiEM community’s discussion of the case.
This follow-up post includes
- The responses of our medical education experts, Drs. David Marcus (@EMIMDoc) and Ryan Radecki (@EMLitofNote)
- A summary of insights from the ALiEM community derived from the Twitter and blog discussions
- Freely downloadable PDF versions of the case and expert responses for use in continuing medical education activities
David Marcus, MD
Hofstra-North Shore LIJ School of Medicine
Melissa dropped the ball. She let the patient with a massive PE walk out of her department. And yet, assuming the patient was well appearing, not hypoxic, and comfortable at the time of discharge, most observers would probably agree that she had done everything right. She correctly applied validated, commonly used decision rules, combined them with her clinical judgment, and acted appropriately. Melissa did nothing wrong. Why then is she being so hard on herself?
Well… wouldn’t you?
We all go into medicine in order to help people; we want to make them better. We like seeing patients walk out of the ED and despise the thought of someone rolling back in in worse shape – especially if they are sicker because of our mistake. But her distress is about more than just error: we are acculturated from an early age to believe in the all-knowing, infallible physician, and we feel very uncomfortable admitting error . Whether she made a mistake or not, Melissa feels that she should have known that her patient had a PE.
Scarred for life
Emergency Physicians (EPs) are expected to make critical decisions in rapid sequence, all day, every day, and without error. Decisiveness and confidence are essential personality traits. Experiences such as the one Melissa is experiencing can have a lasting impact on a physician’s practice patterns. Self-doubt can evolve into a nagging clinical uncertainty which snowballs into over-testing. This kind of “defensive medicine” is associated with increased costs, unnecessary testing, and worse patient outcomes [2-4]. Insecurity and self doubt can be career ending for an EP in a busy Emergency Department (ED).
As friends, we must be there for emotional support. As fellow physicians and friends, we are also in a position to help Melissa work through the moral distress she is feeling at having fallen victim to the realities of Evidence Based Medicine (EBM). This case serves as a reminder that even the most rigorous, validated clinical decision rules never attain 100% predictive value. Positive or negative, they all carry a calculated miss rate. Even when applied to the “ideal” patient 1.8% of those screened using the Pulmonary Embolism Rule-out Criteria (PERC) will have a PE. And yet, across a population, the risks of harm from additional testing and treatment (both indicated and un-indicated treatment) when PERC are negative is greater than the risk of actually having an undiagnosed PE. Objectively, there was no indication that this patient was at an increased risk for a PE and there was no reason for this EP to act any differently.
These rational arguments, however compelling, make little difference to a doctor who is convinced that her action (or inaction) nearly led to the death of a patient. Melissa is simply facing her own humanity. She is no longer the infallible physician she thought she was. This highly emotional state might be compared to PTSD in the sense that this one traumatic event could lead to permanent behavioral changes associated with similar patient presentations. In over-testing Melissa is already showing evidence of a new, maladaptive, practice pattern. But she may be able to gain insight, and perhaps even reverse course, if we apply the principles of Motivational Interviewing (MI), a coaching technique which has been successfully applied for incremental behavioral modification (smoking cessation, weight loss, medication adherence, etc) . If MI can be used for patient wellness, it might be applied to physician wellness too.
Table 1. The OARS Motivational Interviewing Skills (adapted from Bilich et al. )
|O||Open Ended Questions||What are you feeling? Why are you upset? Why does this bother you? How will this impact you? What do your colleagues say about what happened?|
|A||Affirmation||That makes sense. Of course you’re conflicted. I’ve known other people who’ve been in the same situation. Many of your colleagues have felt that they’ve missed a diagnosis. It is completely normal to initially feel you are to blame. Something similar happened to me.|
|R||Reflective listening||I see that… I understand that… Tell me more about… So you’re mad at… It sounds like you think this will interfere with how you function in the ED… In other words…|
|S||Summarizing Joint Plan||Recap of plan: “Let’s go over the plan…”|
One Day at a Time
MI is a counseling technique that “creates an empathetic environment”  by allowing the discussant to do most of the talking so that they find the motivation for change, or in this case a motivation to avoid change where it is not needed. The goal is to “assist in raising the patient’s awareness to their behaviors and to understand how their activities may be at odds with their desired goals.”  In other words, the aim would be for Melissa to identify how drawing a d-Dimer of performing a CT angiogram for even the mildest suspicion of PE conflicts with the goal of providing quality patient care.
During the conversations with her it will be important to legitimize Melissa’s feelings while empowering her to find her own solution. Using the OARS micro-counseling skills [5,6], elicit from Melissa her specific concerns through open ended questions. Try to get to the source of her distress. One can use reflective listening techniques to allow her to further explore the events and her current behavior. Affirm her feelings while reassuring her analytic EP mind that she did the right thing. She did not make a mistake. And though some sources  suggest that it is better to avoid “Righting” (proving to the discussant that they are wrong about something) while working with patients, in this case it may help to remind Melissa of the increased risks of over testing and the rationale behind the Well’s Score, the PERC decision rule, and EBM in general. By maintaining a normal practice pattern she will be keeping her patients out of harm’s way.
Finally, help your friend Melissa devise her own plan that provides some scaffolding as she gradually regains her confidence. Perhaps she could employ a cognitive forcing strategy to avoid over-testing.(8) For example, Melissa might commit to physically calculating the Well’s Score after each chest-pain patient and reflecting for 2 minutes before ordering a d-dimer. Or maybe she could – for a time – consult with a colleague before sending the blood test or performing a CT. She might even engage in some self-auditing and keep track of all the chest pain patients she sees for a period. Conducting a chart review with a senior colleague may provide her with some external metrics of frequency and probability. This way the validity of her practice will be reinforced and she might again regain the confidence that she is not in any way subjecting her patients to unnecessary risks.
Using the plan she has devised, Melissa should be able to move forward without becoming “that doctor who scans everyone”. We all know a few. This is a very difficult situation, one that we will all have to face – either in the role of Melissa or in that of her friend – without guidance. There are no physician peer-support “best-practices”. We must simply do what feels right in order to help our colleague maintain a rational medical practice. Motivational Interviewing provides a structure where one is lacking by creating a framework for coaching a distressed physician. With peer support and careful reflection in-action and reflection on-action Melissa will be empowered to overcome this unexpected outcome.
- Moskop et al. Emergency Physicians and Disclosure of Medical Errors . Ann Emerg Med. 2006;48:523-531.
- Robin, ED. Overdiagnosis and Overtreatment of Pulmonary Embolism: The Emperor May Have No Clothes. Ann Intern Med. 1977;87(6):775-781.
- Hoffman JR, Cooper RJ. Overdiagnosis of Disease: A Modern Epidemic. Arch Intern Med. 2012;172(15):1123-1124.
- Manner PA. Practicing defensive medicine—Not good for patients or physicians. AAOS Now. Jan/Feb 2007. Accessed online on 3/22/14: http://www.aaos.org/news/bulletin/janfeb07/clinical2.asp
- Kline JA, et al. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost. Aug 2004;2(8):1247-1255.
- Billich et al. Small Changes Counseling. MedEdPORTAL; 2014. Available from: www.mededportal.org/publication/9702 Accessed online on 3/22/14
- Miller W. An Overview of Motivational Interviewing. http://www.motivationalinterview.org/Documents/1%20A%20MI%20Definition%20Principles%20&%20Approach%20V4%20012911.pdf Accessed online on 3/22/14
- Crosskerry, P. Cognitive Forcing Strategies in Clinical Decisionmaking. Ann Emerg Med. 2003 Jan;41(1):110-20.
Ryan Radecki, MD
Assistant Professor of Emergency Medicine, the University of Texas Medical School at Houston
Coping with poor outcomes and medical errors is a challenge ubiquitous to medical practice. No specialty is immune from cognitive errors and the resultant patient harms. Recognition of the impact of errors on physician well-being and decision-making is widely documented – approximately half of physicians involved with a serious medical error reported increased anxiety for future errors, decreased confidence at their job, decreased job satisfaction, and insomnia . Incidence of these same adverse effects occurred with only slightly lower frequencies for both minor medical errors and near-misses.
Unfortunately, formal support networks lag behind needs. While many different strategies have been proposed, no consensus regarding effectiveness or appropriateness exists. Institution-based responses to incidents and medical errors may not prioritize the physician’s well-being, nor provide the level of support necessary for individuals under stress. Suggested strategies appropriate for this case include referrals to an Employee Assistance Program, one-to-one follow-up with a colleague, or professional counseling.
The successful practice of medicine depends on rational recognition of the limitations of knowledge and testing. The advantage of “evidence-based medicine,” where applicable, is the explicit recognition of non-zero rates of unanticipated poor outcomes. The application of Bayes’ Theorem to estimate patient-specific disease likelihood does not generate a simple absolute result. These estimates, and the explicitly recognized uncertainty, provide a context for which to judge the harms of testing and treatment. For example, as estimated by Kline et al., 1.8% of patients undergoing testing and treatment for pulmonary embolism will ultimately be harmed by the test and subsequent anticoagulation . Therefore, by applying the PERC Rule – as Melissa does in this case – she has identified this patient as belonging to a cohort for which testing will generate greater net harms than benefits, despite having a non-zero risk for pulmonary embolism. Recognition of this sound decision-making process, despite the outcome, may provide reassurance.
Melissa’s subsequent reactive practice of over-testing is grounded in several recognized cognitive biases. These include “outcome bias,” the tendency to judge a decision based on its outcome, rather than the quality of evidence initially available, and “availability bias,” the inordinate weighting of recent or emotionally charged events in memory. Her behavioural and practice changes are consistent with those observed in other physicians following medical error . There is, unfortunately, no universal, validated approach for restoring confidence in medical decision-making. In general, with time and support tailored to her individual needs, her practice patterns should return to baseline.
- Waterman AD, Garbutt J, Hazel E, et al. The emotional impact of medical errors on practicing physicians in the United States and Canada. Jt Comm J Qual Patient Saf 2007;33(8): 467–476.
- White AA, Waterman AD, McCotter P, Boyle DJ, Gallagher TH. Supporting health care workers after medical error: Considerations for health care leaders. J Clin Outcomes Manage 2008;15(5):240-247.
- Kline JA, Mitchell AM, Kabrhel C, Richman PB, Courtney DM. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost. 2004 Aug;2(8):1247-55.
- Charles SC, Warnecke RB, Wilbert JR, et al. Sued and non-sued physicians. Satisfaction, dissatisfactions, and sources of stress. Psychosomatics 1987;28:462–8.
The following are some themes that emerged from our discussion in the case comments for this past week.
While poor outcomes resulting from appropriate, evidence-based management are appropriately rare, our participants noted that if enough patients are seen they are bound to happen. Most were able to relay personal anecdotes about similar experiences. As Seth Trueger noted, we never hear about “the patient from last night” that ended up doing just fine, so we remember when they do not.
If bad outcomes will happen despite the best care that we are capable of providing, as emergency physicians we need to learn to tolerate risks. Clinical decision rules are good tools for risk-stratifying patients, but they cannot bring the risk down to nothing and if we investigate too extensively we are likely to cause more harm. Often, mistakes can happen not because of personal inadequacy, but because, as Daniel Cabrera noted, our understanding of medicine is incomplete.
Experiencing an adverse event
Judging by the community response, we have substantial room for improvement in helping healthcare professionals to cope with adverse events. The importance of recognizing the healthcare professional as a “second victim” was highlighted. Daniel Cabrera noted that despite often being beyond our control, we feel responsible for them [1,2]. Hans Rosenberg noted that we classically cycle through emotions such as denial, rationalization, despair, and fear as we work through the event.
Susan Shaw found that in general that medicine’s ability to create safe spaces to address these events are “pretty lousy” with R.S. Sahsi finding that the most frequent response is some variation of “shake it off” or “get back on the horse.” While physicians can participate in institutional debriefing following critical incidents, it rarely addresses the physician on an individual level. As noted by Eric Holmboe, the Agency for Healthcare Research and Quality has started online morbidity and mortality rounds .
Moving past adverse events
Several participants noted that no physician can get past a devastating outcome by themselves. Daniel Cabrera noted that the evidence supports the development of institutional infrastructure to help health care professionals through these events . Doing so may require introspection, discussion with trusted mentors, and mobilizing resources and education to prevent future adverse events. Hans Rosenberg noted that it is a long road to get through these events and encouraged taking the time to work through the associated emotions and prepare ourselves for future events.
Rob Woods noted the explicit importance of supportive senior staff for young physicians going through this process. Amy Walsh noted that when approached for support we should attempt to normalize their emotions and share events that have affected us while avoiding questions geared towards ensuring that we would not have made a similar mistake.
The after effect
While we would never want to base our practice on a study with an n of 1, singular bad experiences can color our judgement in the future. Amy Walsh and Justin Stowens shared personal stories about how they were impacted.
- Amy was helped by a counselor who asked her “If a friend came to you and told you about this case, what would you think of them? How would you counsel them?” This question helped her to realize that we are harder on ourselves than we are on others.
- Justin discussed the event with a senior physician who stated quite simply “Well… that’s why we always tell people to come back if they feel worse…” While it was a very matter-of-fact statement, it made him realize the importance of that common discharge instruction in acknowledging the imperfection of our science.
Heather Murray noted that “Hindsight is easy, medicine is not,” in regards to our language when discussing the decisions of other physicians. It is much easier to retrospectively come to a different conclusion than we would have had we seen the patient ourselves. Judging another physician does not fix the problem. It is important to remember that, in all likelihood, at some point we will be that physician.
Thank you to all of the people that participated in this discussion both on Twitter and the blog: Felix Ankel (@felixankel), Daniel Cabrera (@CabreraERDR), Teresa Chan (@TChanMD), Anton Helman (@EMCases), Justin Hensley (@EBMGoneWild), Eric Holmboe (@boedudley), Khurram Jahangir (@AcademicEM), Philip Lederer (@philiplederer), Heather Murray (@HeatherM211), Hans Rosenberg (@hrosenberg33), Rupinder Sahsi (@hotSahs), Susan Shaw (@drsusanshaw), Jonathan Sherbino (@sherbino), Justin Stowens (@JStowens), Nicole Swallow (@doc_swallow), Seth Trueger (@MDAware), Ankur Verma, Amy Walsh (@docamyewalsh), and Rob Woods (@robwoodsuofs).
- Saving the “Second Victim” Helping the Clinicians Involved in Patient Safety Events. Joint Commission: The Source. 2014;12(1):1–9.
- Scott, S. D., Hirschinger, L. E., Cox, K. R., McCoig, M., Brandt, J., & Hall, L. W. (2009). The natural history of recovery for the healthcare provider “second victim” after adverse patient events. Quality and Safety in Health Care, 18(5), 325-330.
- Morbidity and Mortality Rounds on the Web. Agency for Healthcare Research and Quality.
- Wu, A. W. (2000). Medical error: the second victim. Western Journal of Medicine, 172(6), 358.
Case and Responses for Download
Click Here (or on the picture below) to download the case and responses as a PDF.