Psychiatry, Homeostasis, and Regression to the Mean

Are atypical antipsychotics overprescibed?  This question was raised in a recent article on the Al Jazeera English website, and has been debated back and forth for quite some time on various blogs, including this one.  Not surprisingly, their conclusion was that, yes, these medications are indeed overused—and, moreover, that the pharmaceutical industry is responsible for getting patients “hooked” on these drugs via inappropriate advertising and off-label promotion of these agents.

However, I don’t know if this is an entirely fair characterization.

First of all, let’s just be up front with what should be obvious.  Pharmaceutical companies are businesses.  They’re not interested in human health or disease, except insofar as they can exploit people’s fears of disease (sometimes legitimately, sometimes not) to make money.  Anyone who believes that a publicly traded drugmaker might forego their bottom line to treat malaria in Africa “because it’s the right thing to do” is sorely mistaken.  The mission of companies like AstraZeneca, Pfizer, and BMS is to get doctors to prescribe as much Seroquel, Geodon, and Abilify (respectively) as possible.  Period.

In reality, pharmaceutical company revenues would be zero if doctors (OK, and nurse practitioners and—at least in some states—psychologists) didn’t prescribe their drugs.  So it’s doctors who have made antipsychotics one of the most prescribed classes of drugs in America, not the drug companies.  Why is this?  Has there been an epidemic of schizophrenia?  (NB:  most cases of schizophrenia do not fully respond to these drugs.)  Are we particularly susceptible to drug marketing?  Do we believe in the clear and indisputable efficacy of these drugs in the many psychiatric conditions for which they’ve been approved (and those for which they haven’t)?

No, I like to think of it instead as our collective failure to appreciate that patients are more resilient and adaptive than we give them credit for, not to mention our infatuation with the concept of biological psychiatry.  In fact, much of what we attribute to our drugs may in fact be the result of something else entirely.

For an example of what I mean, take a look at the following figure:

This figure has nothing to do with psychiatry.  It shows the average body temperature of two groups of patients with fever—one who received intravenous Tylenol, and the other who received an intravenous placebo.  As you can easily see, Tylenol cut the fever short by a good 30-60 minutes.  But both groups of patients eventually reestablished a normal body temperature.

This is a concept called homeostasis.  It’s the innate ability of a living creature to keep things constant.  When you have a fever, you naturally perspire to give off heat.  When you have an infection, you naturally mobilize your immune system to fight it.  (BTW, prescribing antibiotics for viral respiratory infections is wasteful:  the illness resolves itself “naturally” but the use of a drug leads us to believe that the drug is responsible.)  When you’re sad and hopeless, lethargic and fatigued, you naturally engage in activities to pull yourself out of this “rut.”  All too often, when we doctors see these symptoms, we jump at a diagnosis and a treatment, neglecting the very real human capacity—evolutionarily programmed!!—to naturally overcome these transient blows to our psychological stability and well-being.

There’s another concept—this one from statistics—that we often fail to recognize.  It’s called “regression to the mean.”  If I survey a large number of people on some state of their psychological function (such as mood, or irritability, or distractibility, or anxiety, etc), those with an extreme score on their first evaluation will most likely have a more “normal” score on their next evaluation, and vice versa, even in the absence of any intervention.  In other words, if you’re having a particularly bad day today, you’re more likely to be having a better day the next time I see you.

This is perhaps the best argument for why it takes multiple sessions with a patient—or, at the very least, a very thorough psychiatric history—to make a confident psychiatric diagnosis and to follow response to treatment.  Symptoms—especially mild ones—come and go.  But in our rush to judgment (not to mention the pressures of modern medicine to determine a diagnosis ASAP for billing purposes), endorsement of a few symptoms is often sufficient to justify the prescription of a drug.

Homeostasis and regression to the mean are not the same.  One is a biological process, one is due to natural, semi-random variation.  But both of these concepts should be considered as explanations for our patients “getting better.”  When these changes occur in the context of taking a medication (particularly one like an atypical antipsychotic, with so many uses for multiple nonspecific diagnoses), we like to think the medication is doing the trick, when the clinical response may be due to something else altogether.

Al Jazeera was right: the pharmaceutical companies have done a fantastic job in placing atypical antipsychotics into every psychiatrist’s armamentarium.  And yes, we use them, and people improve.  The point, though, is that the two are sometimes not connected.  Until and unless we find some way to recognize this—and figure out what really works—Big Pharma will continue smiling all the way to the bank.


12 Responses to Psychiatry, Homeostasis, and Regression to the Mean

  1. doctorz says:

    Atypicals are clearly overpriced, considering their incredible expansion of indications over the years. And no question that antispsychotics are useful drugs in taming mania, managing various behavioral disorders associated with dementia, autism, and even the obscure condition of intractable hiccups! (an indication for chlorpromazine). And look at what we’re calling bipolar these days (formerly known as ODD, impulse control disorders, P-C Relational problems in kiddies and Cluster B syndromes in adults), undoubtedly due to the unholy alliance between Big Pharma and the enlightened Biedermans of the world. But these drugs were priced initially based on the incidence of schizophrenia (1% of population), which in part, explains the reason that they cost more than antidepressants…ie, much smaller target population. On that note, look at all of the indications for some of the SSRIs, covering almost the entire spectrum of anxiety disorders, which have a lifetime prevalence of 22% of the population. So potentially, a drug such as Paxil while under trade status could have been indicated in ~30% of the US population (MDD, GAD, SAD, PTSD, PD, etc, etc). Not favoring government intervention into price fixing, but perhaps an exception should have been or could be made in these intstances, whereby a price reduction is mandated of the manufacturer upon additional approvals granted. That could put an abrupt halt to these drugs being promoted as mental health panaceas.

  2. Rob Lindeman says:

    As you correctly point out, Steve, regression to the mean and homeostasis are usually the correct interpretations for symptomatic improvement (along with placebo effect). When people erroneously attribute their recovery to a drug, they are committing the familiar logical error called “post hoc ergo propter hoc” (after this, therefore because of this). Thank you for your take on the problem

  3. Tom says:

    Nice post. But the corrupt alliance between the drug industry and psychiatry is surely a major factor here. When I was in medical school many moons ago, we were taught that all the data indicated major depressive episodes remitted on their own (i.e. homeostasis at work). But as soon as Pharma paid off the KOL’s in psychiatry and hired psychiatrists to be drug reps, all of a sudden things changed! Why we (and our patients) were told that you had to stay on medications for the rest of our lives (perhaps “augmented” by another drug) or else that dreaded disease would return! its all a bunch of nonsense.

  4. Steve says:

    Don’t forget the off label use of these meds. People in my clinic won’t prescribe any benzos or sleep meds. Guess what the pts. get…Seroquel, Trazadone, you name it. You can tell the people with sleep problems because they get fat from the atypicals…a dirty little secret the prescribers don’t tell them. Come in cute, leave obese.

    Psychologist Steve

    • A Non says:

      I commented last thread about my (now-ex) pdoc, the closet addiction specialist. Well, he was a big BIG fan of atypicals, and tried session after session to get me on one, despite my having a four-page printout of my bloodwork showing I was borderline diabetic as is. So, old story. But the comment from Tom about being on them forever prompted me to recall an exchange early in our relationship:

      (NB: I like the bootstrap model of recovery: enough pills to get me to the gym, followed by enough visits to the gym to get me stable, taper off, happiness and thrivery.)

      Me: “You can’t stay on these things FOREVER!”
      Him: “Why not?”

      For one of the few times in my life, I was speechless.

  5. Gary says:

    Nice post. “Until and unless we find some way to recognize this—and figure out what really works—Big Pharma will continue smiling all the way to the bank.” The question for me–before asking what works–is asking what does “works” mean. I want to know the Number Needed to Treat and the Number Needed to Harm. Does a 2-point reduction on a MADRS qualify as “works”? Does any non-random improvement beyond the placebo effect justify using the medicine?

  6. David says:

    Just to be clear, there really isn’t a difference in your description of homeostasis and regression to the mean. The property of homeostasis is what ensures that there is a mean value for the biological property or trait being measured. The source of fluctuations in a statistical sample may have underlying causes, just like your example of fever patients. For example, those with an extreme score on some psychological survey very likely experienced some event that caused them to score that way. They eventually revert to the mean. In this case, most likely due to homeostasis. In other words regression to the mean is universal for any fluctuating property that has a mean value or trend. This includes most living systems as they exhibit homeostasis and therefore have mean values.

  7. Gemma says:

    Genuinely when someone doesn’t understand after that its up to other
    visitors that they will help, so here it takes place.

  8. The Maryland woman and her boyfriend had fitted the plastic
    phallic device directly over the saw blade. Maybe, Sweden will likely be the first
    country to officially outlaw religion in the future, to know…There you’ve got it folks, the 25 most Atheist
    nations in the world. Customer services facilities are crucial so that you are capable to require support or help if ever you encounter some technical difficulties.

  9. It’s genuinely very complex in this busy life to listen news on Television, thus I only
    use the web for that reason, and take the most up-to-date information.

  10. dead trigger 2 marvel avengers alliance hack tool no survey says:

    Asking questions are genuinely good thing if you are not understanding something completely, except this piece of writing gives good understanding yet.

  11. Hi, after reading this remarkable article i am as well cheerful to
    share my know-how here with friends.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: