In recent years, there has been some controversy about whether or not antidepressants are as effective as many had thought. One group – mostly on the outside of the psychiatric profession – argues that antidepressants are only slightly more effective than placebo, if at all. Another group – in the psychiatric professional mainstream – believes that antidepressants are effective to a marked degree.
The PL view is that both groups are mistaken.
The antidepressant (AD) critics argue that there was only a small benefit in a meta-analysis of all AD randomized clinical trials (RCTs) in major depressive disorder (MDD). This small benefit didn’t mask a large benefit in a subgroup, they argue, because there was very little benefit in the majority of patients when divided by severity of illness: mild, moderate, and severe. The mild and moderate depression groups had little to no benefit with AD over placebo, and they were 75% of the patients in the RCTs. The severe depression group had benefit, but it was only 25% of all patients. So, they conclude: most patients don’t benefit with ADs in MDD. Benefit seen reflects a placebo effect; the main authors are psychologists, hence they argue that these patients should receive psychotherapy rather than medications
Some of the PL editors reanalyzed the above meta-analysis based on correcting from a statistical problem called the “floor effect”: in milder depression, there is less room for improvement on depression rating scales, compared to more severe depression. Hence it’ll seem like there is less improvement. We corrected for this problem by assess % improvement, not absolute scores. Conducted this way, there was benefit for moderate depression, not just severe depression. In this analysis, 75% of patients improved with ADs; hence the majority improved, with only a minority of 25% with mild depression showing no benefit over placebo.
In fact, even the mild depression group improved, in both placebo and AD groups. So there was improvement, it just wasn’t because of the pharmacological properties of the ADs. Rather than assuming this effect can be induced by psychotherapies, we raise the possibility that it could reflect the natural course of mild depression. Such patients could have neurotic depression, which is mild, and which has brief exacerbations which naturally resolve quickly over weeks to months (at most).
The PL view thus contradicts the AD critics based on their own data.
However, the mainstream view that ADs are effective for all kinds of “MDD” is also contradicted. The mild depression group, representing perhaps neurotic depression, doesn’t appear to need ADs to improve. Further, the assumptions that ADs are needed or effective long-term is not proven by these studies. PL has some concerns about the validity of the research designs used to claim long-term effectiveness of psychotropics in general, whether ADs in MDD or antipsychotics in bipolar illness, as described elsewhere.