post by Bill Gardner
Many of us are concerned about overtreatment and overdiagnosis in psychiatry, and that the drug companies have significantly exaggerated the benefits of anti-depressant medications for moderately depressed patients.
If you are interested in this discussion, I strongly urge you to read Sandra Tannenbaum's essay on her personal experience with depression and antidepressants.
Depression is sometimes confused with sadness. Many depressed people are very sad, as I was, but the essence of my depression was feeling dead among the living. Everything was just so hard. William Styron describes depression as “a storm of murk.” Andrew Solomon’s atlas of depression is titled Noonday Demon. I too found depression to be fierce, wrapping me in a heavy woolen blanket and mocking my attempts to cast it off. The self-loathing was palpable; it felt like I was chewing glass.
Depression is like being at the bottom of a well at midnight, like falling past the event horizon, like pain radiating from your bone marrow and intra-cellular matrix.
Tannenbaum is also a health services researcher. Her point, in a sentence, is that the critiques of anti-depressants and their manufacturers are in large part true. And yet one of the drugs, Cymbalta (Duloxetine), had a profoundly beneficial effect for her. Unfortunately, the widespread view that anti-depressants are ineffective have had a cost for her:
Most parties to the debate agree that antidepressants can be effective for severely depressed patients such as me, but selfishly I fear the rhetoric of antidepressant uselessness will influence the pharmacy policies of my health plan. At present I am charged an inflated copayment for Cymbalta because my health plan claims it is no more effective than generic antidepressants. I am not privy to the basis for this determination; I do not know if it is based on average treatment effects, the preferences of plan professionals, or an overriding concern for cost. I do know that it does not include my experience, and when I queried the plan about an appeal, I was told I could appeal but should not bother: there are no successful appeals. The plan representative was unmoved by my savings on psychiatry, rheumatology, and hospitalization. She intimated that it is just too hard to satisfy individuals and that the plan has enough to do managing costs.
My story is similar but different from Tannenbaum's. I have been in treatment in one modality or another since adolescence. However, things improved fundamentally during a period of mid-adulthood after I started taking Zoloft (Sertraline). It's less clear to me than it is to Tannenbaum that the drug is what mattered. I was also in psychotherapy. And, as my life improved I remarried, started competing in triathlons, and deepened my Buddhist practice. These changes transformed my diet, sleep, and relationships; as these took hold I left Zoloft behind. Did the drug really matter in all of that? Was it an indispensable catalyst to the other changes? If it was, would I have done just as well with a placebo?
So how do we reconcile the small treatment effects of anti-depressants (relative to placebo) with testimony like Tannebaum's? It is possible that Tannenbaum is mistaken about the effects of Cymbalta: she might have recovered just as well in a counterfactual world in which she had not taken the drug. It's also plausible that the effects of anti-depressants are small in experiments because a particular drug is effective for only a small subset of patients. When you average lake of effect for many with a powerful effect for a few you get a weak experimental effect. We have no idea whether this story is true.
For patients, this means that you must, as Tannenbaum and I have, search over a large space of medications and other therapies to discover what works for you.
For society, it means... I'm not sure what it means. I absolutely believe that Tannenbaum's well-being is worth whatever it costs (even if it was only a placebo effect). But what about the other patients for whom Cymbalta (or Zoloft) has produced great cost, significant side effects, and little discernible beneft? How do we craft insurance policies that cover extensive searches across a broad formulary of drugs to find the one that works for you?