24 July 2007

The Reassurance Work-up

I had an attending in my training program who took a dim view of human nature; I recall that once told me that "if it weren't for alcohol and human stupidity we'd all be out of jobs." While I agreed at the time, I can now attest that he had it only half-right. To be sure, I still see many alcohol-fueled misadventures bringing folks into the ER. But I have come to the conclusion that if it weren't for anxiety and its attendant disorders, we would be out of jobs.

I blogged the other day about a rambling lady with, as they say in medicine, "multiple somatiform complaints" which is Med-Speak for "absolutely friggin' nothin'." Scalpel commented:

Patients like these are frustrating, simply because we know after about a minute that nothing is seriously wrong with them, but we also are often inclined to do a megaworkup to CYA for all of their potentially serious complaints.

Which is completely accurate. It's a terrible thing with these anxious types, because you absolutely cannot blow them off, since even anxious people get ill (and being sick makes them even more anxious). I recall a couple of years ago one of my partners blew off a guy with anxiety and chest pain (he had a history of neither). It turned out he had a thoracic aortic dissection which we did diagnose, but only after a substantial delay, and the patient died. So I hate doing the Review of Systems on the anxiety-driven cases, since they almost always have some complaint related to their chest which might be some sort of atypical angina. We joke that in these cases the RoS is "positive," meaning that for every system we ask about, the patient has a positive finding ( Cardio - Chest Pain and palpitations, Respiratory - Shortness of breath, Neuro - Headache and dizzy, GI - abd cramps and constipation, Skin - itchy and tingly, Eyes - blurry visions and spots, etc etc etc), and you just pray for the complaint of "palpitations" rather than "chest pain" since you can get away with a mini-workup for palpitations, but not for chest pain.

My trick is, however, to ask a lot more questions about their chest pain, because the more I ask, the more details they invent -- that it's not really pain per se, but like ants marching across the skin of the chest, or the pain in their chest shoots around the back and down the legs, for example. If they wind up getting so far off the beaten path that I can make a compelling case that the symptoms are non-physiologic and completely atypical for any cardiovascular complaint, then I can with some degree of comfort document that and defer a work-up. It's a dangerous game to play though, since some of these folks are professional consumers of healthcare services and know the "classic" symptoms of angina pectoris and will recite them as if from a textbook (despite their negative nuc stress and cath within the last two months). And it's hard to document that sort of history and then let them go home.

Fortunately, for the a significant majority of anxiety-driven complaints, I am able to tease out enough history that, on clinical grounds alone, I can conclude that there is no acute life-threat. But that still leaves the question -- what do I do with/for the patient? Experience has taught me that these folks are unwilling to leave the ER without a whole battery of comprehensive tests, so convinced are they that there is something seriously wrong. I have devised a simple, cheap and quick "reassurance work-up" for these folks which consists of:
an ECG, an i-Stat, a D-dimer, and a troponin. Sometimes I add a chest x-ray if it seems helpful. (We are lucky in that most of these tests can be done in the ED's stat lab with a turn-around-time of about 15 minutes.) Then I sit down with the patient and invest a few minutes telling him or her about all the tests we did and all the Bad Things we ruled out. I list each electrolyte separately, the normal blood sugar (we ruled out diabetes), normal blood counts (rules out anemia), ruled out heart attack, blood clots, aneurysm, etc etc. It's interesting how well patients respond to that. The long list of things "you don't have" seems to really be effective in reassuring patients. Then a quick laugh -- I ask the question for them: "Great, doc, you told me what I don't have, so what do I have? Well, I can't tell you what is causing your symptoms, but there are only x number of Bad Things that can cause symptoms like yours, and you don't have any of those Bad Things, so I know it is safe for you to go home, we will keep an eye on it, and I expect that it will go away on its own."

Again, it's surprising how happy this approach makes people. I get genuine thank-yous and (generally) they leave smiling and reassured. And my patient satisfaction scores stay high and the bed is quickly opened up for the next poor soul languishing in the waiting room.


  1. Howdy. I know you're doing the workup more for the patient than for yourself, but for my own personal knowledge... per the docs I've talked with a single biomarker isn't good enough to rule in/out PE or MI... So aren't you putting yourself at risk by doing just a single d-dimer or troponin? A negative test leaves you open to malpractice since your clinical suspicion warranted a test, but ultimately a half assed test? Thanks for any insight you can lend... Have a pleasant day.

  2. Rory,

    You are right that a single negative troponin is almost worthless in ruling out an acute coronary syndrome. So you need to document the test along the lines of "I considered MI/ACS, but in my clinical judgment the probability of this is low, given that the patient is young, has no risk factors, and a very atypical presentation for myocardial ischemia. Additionally, an ECG and single troponin were normal, which further reinforces my clinical assessment that this presentation is unlikely to represent an MI/ACS."

    Now if the patient goes home and croaks, you are *of course* at risk for liability, and nothing will change that. However, the plaintiff's lawyer will have trouble depicting you as negligent; wrong, maybe, but you're allowed to be wrong, but not negligent. The point is you're basing the decision on clinical judgment and using the test as supporting data.

    If the patient had significant risk factors or a typical story, one troponin would not be appropriate - but those aren't the patients I'm writing about here.

    However, with PE, a negative D-dimer pretty much excludes PE in low-risk patients. There is some sketchy evidence that it might be useful in ruling out Aortic Dissection, though the evidence on that is not so good.

  3. Shadowfax-

    Keep those troponin's coming! We've diagnosed more "troponin-secreting tumors" at our institution than MI's... and the consults generated are keeping our general cardiologists hopping and beds loaded! Echo's, treadmills, nuc studies... aaahhhhhh.....

    Man I love defensive medicine!!!!

  4. You don't need serial troponins. If someone's chest pain is constant for the past 3 days and has a negative troponin then you know that pain isn't cardiac.

  5. I have a very similar approach. It's amazing how many of these patients there are. I don't get the ddimer though, because of its notorious lack of specificity (ie it will frequently come back positive and then what do I do?!) So I reserve the PE workup for people who have a significant chance of PE. My 2 ED's also don't have istat--I never understood the utility of a stat troponin. If positive, then there was some delay in presentation. If negative and the patient is infarcting, it's equally useless (except as part of your reassurance workup).

  6. Whisperer,

    The utility of the iStat troponin is time-to-dispo, really. If someone with risk factors comes in with pain suggestive of angina, I get a negative ECG and negative istat troponin, and the patient is off to the Chest Pain Unit or Cardiac Tele (depending on history) within ten minutes of my first seeing them.

    Obviously, this is only for simple/uncomplicated chest pain patients, but they really are the majority, aren't they?

    Ditto if the troponin is positive. I've got results in 10 minutes of arrival, and can directly call cardiology and post the admission.

  7. Even your reassurance workup is probably $3-4K. And if the d-dimer is not negative (there is no positive, only "not negative") add a couple of thousand for CT angiogram. If they are admitted for observation, their bill will certainly be over $10K.

    But it's still cheaper to make 100 patients pay $10K each than for us to pay either a million dollar settlement or an extra million dollars in malpractice insurance premiums over the next few years if we miss a 1/100 pulmonary embolism or MI, for example.

  8. The odds of any of you paying a million dollar settlement are infinitesimal, as are the odds of your premiums being a million dollars in the course of a few years. That's a pretty lame excuse.

  9. A missed MI in a 35 year old attorney, for example, who makes 200K per year and has 30 expected years of lost wages would equal $6,000,000 in lost income alone which his grieving spouse and children would eagerly seek from the hospital, myself, and any other insured individuals whose names were on the chart.

    The combined insurance increases over all of the insured and sued entities over our next 30 years of practice could easily reach a million dollars. It isn't like I'm getting the entire $10K from the bill either....my bill is the same whether I do the extra CT scan and arrange for the admission or not...maybe $500 max, of which I'll actually receive much less.

    Yeah, maybe the comparison is a stretch, but where is the incentive to not overtest? Where is the incentive to use clinical judgment? There is none.

    Admittedly, not everyone is a young attorney with a large income, but the younger patients with longer life expectancies and decreased likelihood of actual illness are the ones that require more testing and defensive medicine costs.

    Somewhere, there is a cutoff point where thousands of dollars in unnecessary tests aren't necessary. How do we find it? How many missed MIs and pulmonary embolisms are acceptable?

  10. Anon above, there have been 3 million dollar payouts in my community in the last year alone. What are you basing your comment on?

  11. as for the comment about documentation...

    It's SO important how you document! I can't tell you the number of charts I've seen where the decision-making/course part says no more than "pt stable" with a list of labs, no interpretation, and home they go.

    If you do the "one troponin..." explain it, as above. Don't say "don't think this is ACS, pt had negative troponins." What does that MEAN? Say "3 days of pain in low risk patient, negative troponin, normal EKG...highly unlikely this is ischemic chest pain."

    I tell this to the other residents all the time when I train them on our documentation system...MDM is where you explain to ANYONE READING YOUR CHART why you did what you did. That includes other doctors, in addition to the lawyers, jury, etc in any lawsuit.

  12. "Somewhere, there is a cutoff point where thousands of dollars in unnecessary tests aren't necessary."

    If I want the test, and I'm willing to pay for it, why do you care? Arguably, all the options on most cars are "unnecessary", but if I want them, why is that a bad thing?

    "What are you basing your comment on?"

    I have no doubt there are million dollar payouts. I just doubt that a single dollar ever comes out of a physician's pocket to pay them.

    As for the claim regarding the increased insurance over 30 years, that's pure speculation. Your premiums are going to go up due to inflation regardless. To attempt to separate the cost out because of one claim is impossible unless you're the insurer. And they're not exactly eager to give you a straight answer.

  13. Anon 8:15

    To be sure, I as a front-line working doc personally don't much care if I order an extra test. But on a system-wide, policy level, it is problematic. It's wasted resources, and unnecessary tests often wind up resulting in unnecessary follow-up tests, some of which are invasive and carry risk. For example, I get a D-dimer to reassure you that you don't have a blood clot. Well, it comes back positive, which happens sometimes since the test is not real specific. So now I more or less *have* to get a CT scan of your chest, which requires radiation and IV contrast dye, both of which carry a small risk. And maybe that CT scan shows a small lymph node which "looks funny" (radiologists love that sort of finding) and you need a lung biopsy. And maybe you just get a needle stuck in your lung, and maybe you also get a complication. Hooray!

    Yeah, the odds are low, but shit like that happens all the time, and when you have 115 million Ed visits annually, the more unnecessary tests we do, the more people it does happen to.

    And as for the malpractice point, you are gravely mistaken. Our insurance premium (over $1 million annually, BTW) is determined by our loss ratio over the past ten years or so. Our actuaries can tell us exactly how much each case increases the perceived risk and how much that costs us. So you are right that when I get a million-dollar judgment against me (knock on wood), I don't pay it out of pocket, but the group of doctors sure does collectively, over time.

  14. Again, though, if I WANT the test, after you've explained the risks and benefits to me, and told me you don't think I need it, how is it a "wasted resource" if I'm willing to pay for it? Isn't that my choice? And if not, why not? Aren't you just attempting to ration what the customer wants in the same way you negatively imply the govt. will do so in the event of single payer/universal health care?

    As for how your premiums are determined, yes, payouts are a factor. But so is the bond market, so is the financial health of other insurers in the market, so is the reinsurance market. Your actuaries cannot, or more accurately likely will not, tell you that. When was the last time you heard an insurer attribute an increase to a disastrous investment? Or attribute a refund or decrease to the fact that they overestimated loss reserves for the previous years? It's pretty rare, but the fact is those things all matter.

  15. I went through several rounds of ER visits after treatment for hypothyroidism.

    The worst was when a dr told me I was having an anxiety attack because my hands were shaking. He seemed to be missing the slurred speech, loss of coordination, and mental confusion. I couldn't hold the pen to write or walk down the hall without falling.

    He refused bloodwork and wanted to sedate me. I left.

    On another visit I was so dehydrated my skin was standing in folds and I had had strange spasms all over the left side of my chest. I had no circulation in my hands and feet. I was left in the waiting room for three hours and told I couldn't drink anything.

    Most of the other visits were more vague and I was treated much the way you treat your patients. I didn't understand the "treat em and street em"role of the er at the time and was trying to find help somewhere. This was made worse by the nature of endocrine hormonal cycles which fall to the lowest levels at night-when I would become the sickest.

    Perhaps these folks are not just worried.

    perhaps they have subtle problems that, while not ideally treated at the er level, are very real. You guys (drs in general)are great at passing the buck when you don't know what's wrong. That combined with the lack of communication among physicians treating the same patients is worrisome.

  16. Again, though, if I WANT the test, after you've explained the risks and benefits to me, and told me you don't think I need it, how is it a "wasted resource" if I'm willing to pay for it?

    It's only wasted because it's unnecessary. You may note from the original post that *all* of these tests are more or less unnecessary, because I have a clinical impression that the patient does not have a medical emergency. But I do them because it's the quickest and easiest way to make the patient happy and facilitate their discharge.

    Now if you want a test that's more invasive or limited availability, then I am more likely to say no, regardless of your ability to pay for it. An unnecessary Abdominal CT scan can take four hours with oral contrast, and if I have patients in the waiting room and hallways, or a long line of sick people waiting for the scanner, then I am much more likely to decline to order the test and explain why it is not needed (or failing that, order it on an elective basis as an outpatient).

  17. I think you illustrate how far physicians have fallen away from the "free market" they claim to want. The concept of "the customer is always right" is foreign to you.

    While the engineer does not have to prepare a faulty building because the customer asks, if the customer wants five different drawings/designs, or five different ways of engineering the bridge when the engineer knows which one is the best, so be it. The customer gets, and pays for it. If the lawyer can tell his client what the likely outcome of the case is up front, and encourage a client to take a settlement offer which he believes is reasonable, and the client still wants to hear what the jury says, so long as the client will pay and it doesn't violate the law, so be it. It's the client's case.

    Likewise, with medicine, so long as the patient understands the risks/reward and is giving informed consent, then what's the harm if they will pay? That's the free market.

    I think you have been insulated from it so long you've forgotten that basic concept of the customer is always right. Even if there are exceptions for professionals.

    It's also why I don't believe many physicians would survive very long without the third party payment system.

  18. Anon,

    In our ER we have a huge focus on customer service. "The Customer Comes First," is the motto. But not, "The Customer is Always Right," because they're not. There's a huge enough asymmetry of information that in most cases the decision-making model has to be by necessity paternalistic. There are exceptions -- occasionally I see someone who may have a rare disease and they may know more about it than I do. I respect that. Some patients are medically sophisticated enough to have reasonable preferences about treatment options and in that case I do have a more collaborative approach. But those are exceptional cases; the vast majority of cases I am going to have to tell the patient what the best course is.

    Also, you keep bringing up this notion of you are "paying for it." Unless you are a cash customer, you are not paying for it; you are paying your deductible, and the other members of your insurance pool are paying the expense. Further, I try very hard not to let the economic class of my patients influence how I care for them.

    And again, please remember the setting. If you come in to the ER and ask for a full-body CT scan to rule out cancer, that sort of service is not appropriate for the ER. If you want that, there are radiology boutiques that will do that on demand (cash on the barrel). But that sort of thing is just not my mission.

  19. Do I not pay my insurance? I pay my premiums whether I use my coverage or not, so yes, I am paying for it. You complain about the third party payers, but because of them you have forgotten where your paycheck comes from. My premium.

    It may not be your mission, and that's fine. But that's because you have forgotten what the free market you claim to want to be a part of is like.

  20. 1:45,

    You might be paying a premium, that is quite different from you "paying for the test".

  21. How so? If you are sued for malpractice and lose, do you not think you have paid the judgment?

  22. DoctorAudrey4/28/2012 1:12 PM

    One of the great things about working in Africa - same amount of bullshit comes in, but due to lack of resources and patients by and large not being able to sue, at least in public sector, the number of tests we are allowed to do is whittled down to the fewest possible needed to make a good diagnosis. (And then sometimes even those are unavailable.) And its amazing what a good ol' Vitamin B shot does, especially when administered with a nice big needle.

    Being sued is very unlikely, and then you really have to foul up. And the legal system is so clogged up that even if you could be sued, it takes a very determined patient indeed to continue to press charges.

    Except when working with rich white people. Then medicine more closely resembles US medicine. So there are still many docs left in govt sector, because freedom from worry is worth a lot.


Note: Only a member of this blog may post a comment.