Useless thought experiments during medical school and residency

I recently admitted a patient with a pulmonary embolism. Before heparin drip was started, my attending ordered a hoard of eccentric, non-indicated hypercoagulable workup in the hope of avoiding the effect of heparin on these test results, including phosphatidylserine antibody and methylenetetrahydrofolate reductase DNA. I watched in horror as the nurse drew out approximately 13 tubes of blood, since each test needs its own tube.

On rounds, the attending of course pimped us about all the possible tests you could draw on a patient who comes in with a clot. I zoned out in my head since I know that regardless of the test results, pulmonary embolism is life-threatening enough that this patient will need life-long anticoagulation – the tests will not change anything. Even if the patient tested positive for genetic clotting diseases, there is no point screening family members or even considering anticoagulation in them unless they develop clots themselves. We went on a long and useless thought experiment. What bothers me the most is that we get a pat on the back when we try to draw these tests just because they are vaguely related to hypercoagulability, but when you actually stop to ask yourself what you would do if the tests come back positive, the answer is usually a shoulder shrug . Imagine the phosphatidylserine antibody came back positive – what does that even mean? and do we even care? I personally do not as the patient is going home with anticoagulation no matter what.

Useless thought experiments, unindicated lab tests and interventions happen on a regular basis, both in my medical school and residency program. We transfuse blood when there is no good evidence that it improves any outcome, just because doctors are nervous that patients will have heart damage from demand ischemia. Consulting surgeons would like to follow lipase level in my patients with pancreatitis, and my attendings let them even though there is no evidence that its correlates with symptoms just because we don’t want to get in a fight. I was told to perform extensive workup in a patient with acidemia, when the obvious cause of kidney failure stares us right in the face – I couldn’t convince my attending to wait until the kidney failure resolves to see if the acidemia goes away, at which point no workup will be needed. We looked for zebras even though we know it’s a horse, only because my attending “didn’t want to miss anything.”

In my opinion, most of what we do in medicine is not backed up by good evidence. A lot of the guidelines are biased by group interests and potential profits. This is part of the reasons why some of the tests/procedures that used to be indicated are no longer. Mammogram used to be recommended every year, but now yearly screening may lead to unnecessary breast procedures and once US Preventive Services Task Force tried to recommend increased screening interval, the backlash from groups that stand to profit from mammograms and breast procedures was overwhelming.

But a lot of it stems from how we are trained in medical school and residency. If we get a pat on the back every time we draw useless tests, we will continue to perform unindicated workup when we become attendings. If we are deemed unknowledgeable every time we refuse to do extensive lab work because we have a good explanation of what the abnormality is, we will spend more money without improved outcomes, which is happening in the US as we speak.

I wonder if my patient knew that the test for phosphatidylserine antibody costs $194 at our hospital, and that in the end the test will leave her with nothing better than less blood in her veins.

 “angienadia” is an internal medicine physician who blogs at Primary Dx.

Submit a guest post and be heard on social media’s leading physician voice.

Comments are moderated before they are published. Please read the comment policy.

  • Terence Ivfmd Lee

    Brilliant nonconventional thinking! In my area, frustrated infertility patients sometimes make the request that “I want to be tested for everything! I read about this test. Why didn’t you order that for me?”

    I share with them my thinking that the decision to order a test is balanced by the following questions:

    1. What is the cost of the test? Not just the financial, but also the invasiveness, inconvenience and risk. One hidden cost is the follow up if it is a false-positive. There have been many cases of asymtompatic patients getting unnecessary surgery as a result of the medico-legal “necessity” of following up positive screening tests.

    2. How likely is this test to be positive in this patient?

    3. Will knowing the results of this test allow us to make better decisions?

    The answers to this question change dynamically over time, so whereas for one patient, I may assess that it’s not time to do a specific test after one miscarriage, it changes and becomes warranted if she, for example, has three miscarriages.

  • Anonymous

    Sounds like you would be happier in family medicine :)

    I think it’s great that there are so many different fields of medicine.  The advantage of primary care is we can follow people over time.  When you see someone one time in the hospital, there is an extra incentive to figure out as much as possible on the first visit rather than waiting to see if things get better on their own.

    It’s great that medicine draws smart and curious people.  And sometimes that curiosity and out-of-the-box thinking leads to brilliant diagnoses and breakthroughs in research.  I’ve seen so many people who “had symptoms for years and years and then I saw this one brilliant doctor who ordered this one test that explained all of my problems.”  We all want to be that one doctor.  But better to work through the zebras over time rather than doing the “pan-man-scan” that is often necessary for a patient you’ve never seen before.

    Great post on the problem with ordering tests just because we can.  Thanks for sharing your thoughts!

  • Anonymous

    1. Pulmonary embolus in the absence of genetic or acquired metabolic risk factors especially in the presence of an inciting event is NOT an absolute indication for lifelong anticoagulation therapy. 2. Some (not all) of the hypercoag studies indicate genetic predispositions that may be of interest to the patients offspring (or future offspring) because they might require testing to see if they might need lifelong anticoagulation. 3. Bragging that you turned off your brain because you know more then the attending is beyond arrogant and moves into the Dunning-Kruger realm. Save this kind of stuff for the call room.

    • June Sangarlangkarn

      Thank you for your comment. Your first comment is correct – unprovoked PE does not necessarily require lifelong anticoagulation. However, in her case, regardless of the test results, she WILL receive coumadin for at least 3 months then get reassessed (if not lifelong – she has no bleeding problems) – the test results do not change anything at all. For your second comment, genetic testing is really useless, because unless the offspring develops clots, there is no point testing the offspring, since we don’t anticoagulate people who’s never had a clot (if nothing is broken, why fix it?). For your third comment, I never said or thought that I knew more than the attending – he knows the same studies I do, but what we decide to do with that knowledge is different. All the people I mentioned who wanted the tests that seem unnecessary to me know everything I know and more, but they still order these tests out of habit, or as I mentioned, to get out of fights. In the end we never cared about the results of the tests I mentioned – we just ordered it and forgot about it, or we looked at the number, shrugged the shoulders and went about our lives. In the case of the PE patient, a new attending came on the next day and discharged the patient on anticoagulation – none of the fancy test results came back since they took a few days to run.

      • bladedoc

        I’m sorry but your response shows an even further lack of understanding of the disease process and general medical practice. In order

        1. Yes, she will receive coumadin for 3-6 months no matter what however 3-6 months is not the same as “lifetime anticoagulation” the results of the tests WILL matter in regards to that decision. This also speaks to your final point, no one expects the studies to come back during the patient’s hospital stay, they are for follow-up decision making for either you/your team (if they remain your patient) or their primary care doctor. Furthermore, in the absence of risk factors and inciting events an “unprovoked” episode of DVT/PE should result in screening for occult carcinoma.

        2. Your understanding of asymptomatic screening is faulty. There are a number of genetic diseases that require prophylactic therapy prior to the development of symptoms (for example BRCA1 and 2, and all of the hereditary colon cancers). The incidence of DVT/PE in genetic protein C deficiency for example is 75% and as the presenting symptom of PE is occasionally sudden death, I would consider prophylactic anticoagulation in this population. At the very least knowing that you are at higher risk because of genetic abnormality would allow you to determine if prophylaxis is necessary for risk events (long airplane rides, surgical procedures, etc.).

        I write not to change your mind for as Dale Carnegie says “Those convinced against their will are of the same opinion still.” but instead because you’ve made very authoritative statements about issues that frankly are not nearly as clear cut (to be kind) as you have made them out to be.For reference Thomas review of hypercoaguable states 2001

  • Andrew Schutzbank

    I especially like your comments about ordering tests “to avoid a fight.” I cannot think of a more clear way we have abdicated our responsibility to patients then by jabbing someone with a needle and withdrawing their blood, running expensive tests that have no clear merit, all so another human being will not speak loudly at us…  Our patients need us to be courageous on their behalf, and our colleagues need to not fight…

  • Anonymous

    Anyone ask why tests were ordered?


    agree entirely about wasteful testing, evidence-based practicing and challenging questioning of leadership. do think you should turn brain back on and reconsider the who, what and how long for anticoagulating patients with pulmonary embolization (6 months, maybe, not lifetime) and identification of subgroup with hypercoagulable state

  • Patricia G Dunegan

    This reminds me of what happened to my mother who was in the hospital with end-stage cancer, with a diagnosis of only days to live.  No one, including the doctors and nurses, thought otherwise.  She was in considerable pain and even the morphine didn’t help.  They came in every day to take her blood pressure, which caused her much discomfort.  Another day they moved her by ambulance to another hospital for a MRI – she was screaming in pain.  When I arrived from across the country, my question to them was always “Why?”.  I banned them from taking her blood pressure and from doing anymore testing of any kind.  After all, it was useless in this situation.  Forget the money — what about the patient distress with all the useless tests?

  • joe

     totally agree with blade doc on this.Though the genetic testing can be done long after the fact, Functional studies and studies such as the DVV IS AFFECTED BY ANTICOAGULATION. If your argument is the “wait on blood draw to start anticoagulation” then bring it up with admin to make things happen faster.  Protein C/S/Anti-thrombin is hard to interpret in the acute clot setting but a normal level is reassuring. It is a judegment call to order it depending on clinical setting.  Remember protein C/S is not interpretable on anticoagulation. So is an elevated DVV (dilute viper venom). These levels can argue towards or against retesting after anticoagulation and lengthof anticoagulation. The bigger question is CLINICAL CONTEXT which you did not provide in your rant. Was this a first DVT that was clearly provoked  in which case no testing is indicated outside of coags and CBC/creat OR is this an repeat and/or unprovoked DVT in the setting of a family history, in which the attending was right. More info is critical. IF it was the former than you should have asked the attending about the need for ordering a1-2 K hypercoag w/u that wasn’t indicated. Also, though you may not have “looked at the results” ( I actually doubt most were back yet). I can guarantee you  if “clinically indicated” the outpatient internist or hematologist will be. Sadly, you are fallen into inpatient resident mode in which the only focus is moving the patient. There is a life and clinical decisions to be made for this patient after you discharge them. It’s not just about you.  You will see what I mean in a couple of year in the real world when the RRC isn’t limiting your work hours. I also suggest you take a good look at the latest edition of the CHEST anticoagulation guidelines. It is on the internet. Yes there is alot of waste in medicine, based on the (granted limited ) info you presented you didn’t pick a good example. Of course what do I know, I am just a hematologist.
    PS: One thing you are right about is that the MTHFR is test that has been found NOT to have an increased risk of associated thrombosis. I still see get consults on these patients who get “hypercoag panels” from lab mills such as labcorp. A complete waste of $$.

  • Anonymous

    In my daughter’s case, knowing that she had inherited a genetic clotting disorder from her father (who had a PE in April of 2011) allowed her to discontinue BCPs before developing a life threatening clot and allowed her to have knowledge of possible high-risk pregnancy before pregnancy. Knowing about an inherited clotting disorder in the family also pointed a possible cause of early stroke/heart  and unexplained death in men in my husband’s previous two generations, and helped our sons know how to decrease their risk of clotting occurring, whether they choose to have a work up or not.

    • Anonymous

      Thank you for that perfect example.

  • jamesp

    Though bladedoc may have some valid points, the theme of the article is sound- In US hospitals (offices/clinics too), there are far too many tests ordered which will not change the treatment. Day in and out I as a PCP get asked by patients, “I want to be tested for everything.’ Whether they need it or not.

    When will we as a nation realize this approach is costly (in terms of money and suffering) and wasteful????

    Joe, thanks for your insights-”lab mills!”   LOL!

    As long as it’s been since I was a resident, maybe it’s better to have today’s residents “moving” folks and letting the seasoned PCP manage their cases for the long term.

     Would the author’s argument be more valid if she’d picked a test such as MRI for low (nonradicular) back pain?

Most Popular