Antidepressant Use Prospectively Relates to a Poorer Long-Term Outcome of Depression: Results from a Prospective Community Cohort Study over 30 Years

Antidepressant Use Prospectively Relates to a Poorer Long-Term Outcome of Depression: Results from a Prospective Community Cohort Study over 30 Years. Michael P. Hengartner, Jules Angst, Wulf Rössler. Psychotherapy and Psychosomatics, doi 10.1159/000488802

Despite marked increases in antidepressant drug prescriptions over the last 3 decades, the burden and disability attributed to major depression are still on the rise. This calls into question the sustainable clinical benefits attributed to antidepressants. The efficacy of antidepressants based on mostly industry-funded short-term trials has been challenged due to selective reporting and systematic method biases (e.g., unblinding of outcome assessors), and the long-term benefits of antidepressants have also been debated [1, 2]. Two long-term effectiveness trials of 1 year duration found no clinically important effect in terms of sustained remission for long-term antidepressant use (the sustained remission rate was < 6% [3, 4]), and a meta-analysis of long-term parallel-arm efficacy trials of 6–8 months treatment duration found no significant difference between antidepressants and placebo with respect to both remission and premature treatment discontinuation [5]. Some prospective observational studies with 1- to 9-year follow-ups even reported a poorer outcome in antidepressant users relative to non-users [6–8]. In contrast to these findings, discontinuation trials of commonly 6–18 months duration suggest that long-term antidepressant use may prevent relapses (reviewed in Hengartner [1]). However, the validity of these trials has been questioned, because they include only participants who respond well to the drugs and randomise some of them to have the drug withdrawn rapidly and replaced by placebo, which can cause severe withdrawal syndromes that mimic depression relapse [1, 2]. Since extended observation periods of 10 years and more are not feasible within a randomised placebo-controlled trial design, the aim of the present work was to test in a representative community cohort study over 30 years whether antidepressant use, relative to non-use, would relate to a poorer long-term outcome of depression.

[...]

In this community cohort of 591 adults followed from the age of 20/21 to 49/50 years, we found that, independently of illness severity and distress at baseline, antidepressant use prospectively relates to a poorer long-term outcome in depression. These findings are in line with a growing body of evidence from several naturalistic observational studies suggesting that (long-term) antidepresant use may produce a poor long-term outcome in people with depression [6–8]. A neurobiological mechanism that may causally explain these findings is the oppositional model of tolerance by Fava [10], which proposes that continued antidepressant use may recruit pharmacodynamic processes such as receptor sensitization that results in loss of clinical effect and in some patients even in an increased vulnerability to depression relapse.

The present work is not without limitations. Most importantly, participants were not randomised to antidepressant use, which precludes causal conclusions. Antidepressant use relied on self-report, and we do not have data on the specific drug prescribed, its dosage, and on the duration of medication. Considering these limitations, our findings raise the possibility that antidepressants may worsen the long-term course of depression. It remains to be established whether this effect is causally related to antidepressant use or to uncontrolled aetiological factors differentiating antidepressant users from non-users, such as for instance coping skills, social support, or personality characteristics.