Long term health of patients vs. Short term risk to doctors

What is your real motivation when making medical decisions? Is it ‘what’s in the patient’s best interest’ or is it  ‘what will keep me from getting sued’? The reflexive answer is, of course,  the former, but if you really do some soul-searching, there’s probably a bit of the latter as well. In this episode, Mike Weinstock, author of Bouncebacks, and Bouncebacks Pediatrics, discusses why we sometimes have our priorities misaligned with the patient’s and how that doesn’t need to be the case.

Direct Download


Along with Amal Mattu and Erik Hess, Mike has recently published an article titled How Do We Balance the Long-Term Health of a Patient With the Short-Term Risk to the Physician? (link). The study’s conclusion is this:  A test/intervention should be done if the risk of a missed diagnosis or adverse outcome is greater that the risk of the test/intervention. Involving the patient in the decision-making process may help to shift the management balance from the physician’s short-term concern of their own risk to the patient’s long term health. 

An example of where this sort of thinking comes into play is with the evaluation of patients with chest pain. Should they be admitted to the hospital? Are they safe to go home? Before we can answer either of those questions, we first need to address the elephant in the room…


What is an acceptable miss rate for chest pain?

Than, M., et al. “What is an acceptable risk of major adverse cardiac event in chest pain patients soon after discharge from the Emergency Department?: a clinical survey.” International journal of cardiology 166.3 (2013): 752. PMID:23084108

  • Survey of 1029 emergency department clinicians.
    • 395 comfortable with a 1-2% miss rate
    • 267 comfortable with < 1%
    • 331 only comfortable w < 0.1%

Putting that last number in perspective, that’s admitting 1000 patients to find one with cardiac disease. Is that a good return on investment or wise stewardship of health care resources? I think the answer is no. There are several reasons for this. The first is that being admitted to the hospital is not a benign event.


What is the chance that a patient will be harmed by hospitalization?

James, John T. “A new, evidence-based estimate of patient harms associated with hospital care.” Journal of patient safety 9.3 (2013): 122-128. PMID: 23860193

  • Estimates a lower limit of over 200,000 deaths per year related to the deleterious effects of hospitalization


It’s no secret that hospitalization can be dangerous, but it can also be extremely helpful in the properly patient. So in a patient with chest pain who has two negative troponins and a non ischemic EKG, what is the short term risk of clinically relevant adverse cardiac event (CRACE)? In other words, if we hospitalize patients with these factors (negative EKG and enzymes) what is the likelihood that something bad will happen in the next few days that could be mitigated by hospitalization?

Weinstock, Michael B., et al. “Risk for clinically relevant adverse cardiac events in patients with chest pain at hospital admission.” JAMA internal medicine 175.7 (2015): 1207-1212. PMID:25985100

  • This study sought to determine the risk of life-threatening arrhythmia, inpatient ST-segment elevation myocardial infarction, cardiac or respiratory arrest, or death during hospitalization in patients with negative serial troponin, non concerning initial ED vital signs, and nonischemic, interpretable EKG.
  • Over 7,000 chest pain patients admitted (non ischemic EKG, negative serial troponin)
  • 0.06% incidence of CRACE during the hospitalization
  • So over the few days that a patient with non corning EKG and negative enzymes is in the hospital, there was a 6 in 10,000 chance of a catastrophic outcome
  • This is different than what’s being asked by the HEART score. HEART is looking for the risk of major adverse cardiac event (MACE) in 6 weeks.


Taking all of this into account, how does Dr. Weinstock approach shared decision making?

  • If the ED workup shows a non-ischemic EKG and there are two negative serial troponins, he presents the option of an outpatient workup. A caveat to this is that access to rapid outpatient evaluation must be readily available
  • He advises the patient that the possibility of a CRACE is one in several thousand and, while being hospitalized may seem like the safest course of action, hospitalization itself is not without risk
  • The Weinstock Credo: Don’t practice defensive medicine. Document “defensibly”


Also mentioned in this episode

  • Brown, Terrence W., et al. “An epidemiologic study of closed emergency department malpractice claims in a national database of physician malpractice insurers.” Academic Emergency Medicine17.5 (2010): 553-560. Full article link
  • Have a good time, all the time


Editor’s note: Show notes are meant to complement the podcast and do not represent a complete synopsis of what is contained in the audio.



  1. EM Basic


    As always- great episode- lots to think about here. Whenever I hear about some pushing this idea of the Triple Aim of Healthcare (improved patient satisfaction, improved outcomes, and reduced costs)- I throw on the breaks on that last “aim” and ask a simple question:


    Why should I endeavor to reduce costs in the healthcare system? Why should I take on unnecessary risk to myself, my assets, my sanity (from a lawsuit), and my future career with the goal of reducing costs? Also- since we know that people respond well to incentives- where is my incentive for doing so outside of a moral obligation to my patients?

    Whenever this idea gets tossed around its usually in the context of reducing low or zero value testing. But why should that be on me to take on this battle when on the other side of it is an undefinable risk of being thrown into an adversarial system that leaves doctors various levels of beaten, battered, and (in some cases) completely broken? Where is my incentive to reduce unnecessary testing? Where is my safe harbor? (And for me- this is totally theoretical paranoia because, knock on wood, I have not been sued yet)

    While I strive to do the best for every patient and reduce risks that may come from unnecessary testing, I don’t think I will go out on a limb to do so. I will use clinical decision rules when I can and evidence based practice that is defensible but every EM doc knows that there are so many scenarios that fall outside EBM and clinical rules/guidelines that a lot of times you are stuck.

    I am so happy that we now have things like the HEART score to help us out with chest pain. I’m a young buck only 5 years out of training, 8 years removed from my internship. When I tell current residents that 8 years ago, if you got a single troponin, you were virtually committing the patient to an admission for a stress test, they look at me like I have two heads. I cringe when I think of all the healthy 40 year olds with histories not consistent with anything close to ACS and the arguments just to get them admitted to a chest pain unit run by the cardiologists. The HEART score has done wonders in defining what a low risk patient looks like and I, myself, am ok with a less than 2% risk of ACS if the patient has a low HEART score. Same thing with PERC as I am more than ok with the less than 2% risk of PE because it is defensible care.

    However, outside of those clinical decision rules, I feel like you are hanging on a thread. A young female with pleuritic CP- not that concerning of a history and no physical findings worrisome for PE but she is on OCPs. Now comes the d-dimer that we know will likely be positive and will likely lead to a CTPE. I would love to just say with my gestalt “this is not a PE” but that is taking on unnecessary legal risk for me and it’s not something I want to do. Back to those pushing to reduce “unnecessary testing”- where is my safe harbor in this situation? I want to do the right thing for the patient but it shouldn’t come in the context of increased malpractice risk to myself. While this may seem to be the perfect case for shared decision making, I don’t think it’s unreasonable to say that I would feel better if the patient got the CT scan because it would give me certainty in the situation.

    Another example outside of any validated decision rules- a young patient says they have “tearing” chest pain, 12 out of 10, laying prone on the stretcher, propped up on elbows, texting. Physical exam normal, not Marfanoid, BPs and pulses normal throughout. Nothing in chart to suggest drug seeking behavior. Those that seek to reduce testing would say “doctor, use your judgment and don’t CT this patient”. If I was in a country without malpractice burden, then the patient would probably be handed their discharge instructions. I, on the other hand, wonder what “tearing CP” would look like blown up to 100 point font and order the CT dissection protocol. There are just some things that you can’t get around because of what the patient tells you.

    I am certainly not advocating for lots of testing and CTs for all but I do like what Mike said about defensible charting and this is what I teach the residents to do. However, at some point, it all really comes down to what you as a clinician are willing to tolerate as far as malpractice risk. Those that are more conservative are not weak clinicians or not educated in the latest and greatest- they just want to look out for themselves, their livelihood, and their sanity and I don’t think that should be looked down upon. I don’t think that you and Mike were doing that in this episode but thats’ what I feel like we do whenever we push this narrative that doctors need to reduce costs.


    1. Post

Awesome article, I know - please share your erudite thoughts...