First published on Mad in America on 14/04/25
Antidepressant Trials Last Eight Weeks, So Why Do We Take Them for Years?
The studies are of short duration and are riddled with methodological issues like unblinding, nonstandard assessments, and failure to assess withdrawal and adverse effects.
In a new study, researchers question the long-term use of antidepressants, given that clinical trials of the drugs typically last eight weeks.
“Substantial discordance exists between the typical 8-week duration of clinical trials and the median 5-year real-world use of antidepressants,” they write.
According to the researchers, this is especially problematic because those trials also fail in other ways, including using nonstandard measures, unblinding, and not assessing withdrawal and adverse effects.
“This gap, compounded by inadequate monitoring for withdrawal effects and post-treatment outcomes, raises important questions about the evidence supporting current long-term prescribing practices,” they write.
The study was conducted by William Ward at Ottumwa Regional Health Center, Iowa, and Alyson Haslam and Vinay Prasad at the University of California, San Francisco. It was posted as a preprint (before peer review) on the website medRxiv.
According to the CDC, 13.8% of American adults had an antidepressant prescription in 2018 (18.6% of women; 8.7% of men). This has likely only increased over the intervening years, especially with the boom in psychiatric drug prescriptions during the pandemic. For instance, data shows that almost a quarter of women and 11% of men were being treated for depression in 2023 (this may include psychotherapy; however, antidepressants are common; this may also be an undercount since the drugs are also prescribed for other reasons, including anxiety and chronic pain).
Clinical trials typically find that antidepressants are slightly better than placebo, at least in the short term. However, many researchers argue that this difference—about two points on a 52-point depression scale—is clinically imperceptible. In real life, less than a quarter of people get better after depression treatment, even when they are given multiple drugs, hospitalization, and add-on psychotherapy.
And antidepressants have been found to worsen outcomes in the long-term, even after accounting for baseline severity and other factors. Beyond worsening depression through a process known as “tardive dysphoria,” the drugs are also harmful for cardiovascular and metabolic health.
This stands in stark contrast to untreated depression. Studies have documented that as many as 85% of people recover spontaneously from depression in the absence of treatment.
The Current Study
Given the large number of people taking antidepressants, and the concerns about long-term use, Ward, Haslam, and Prasad wanted to determine how long people are actually prescribed the drugs—and whether the evidence base supports this use.
The researchers looked at the randomized, placebo-controlled trials (RCTs) of the top 10 antidepressants (in terms of current prescriptions) for either depression or anxiety. They obtained a representative sample by picking RCTs on these drugs that were published at five-year intervals between 1978 and 2023. (Thus, they did not include all RCTs of all antidepressants.)
They used data from the National Health and Nutrition Examination Survey (NHANES) to determine real-life usage of antidepressant drugs. The most recent NHANES survey they were able to use was from 2011-2014. (Thus, their data on usage was about a decade old.)
In total, they looked at 52 RCTs including 10,116 participants (48 on depression, 4 on anxiety). The median duration of RCT was 56 days (eight weeks). Based on the NHANES data, almost all (94.2%) of those taking antidepressants use them for longer than that. The median duration of antidepressant use was five years.
Methodological Issues
Only six trials (11.5%) were longer than 12 weeks. Half of those trials used nonstandard assessment measures. Four of them were specifically testing patients with acute coronary syndrome, breast cancer, alcohol use disorder in remission, or Alzheimer’s disease, so only two trials were longer than 12 weeks and focused only on depression. No trials lasted more than a year.
No trials used an active placebo, which provides side effects similar to the active drug to make sure the participants can’t guess whether they received the drug or not. Thus, the researchers conclude that the risk of finding a positive effect due to unblinding is large.
Only two trials assessed for withdrawal symptoms, and only two assessed post-treatment outcomes. A quarter of the trials assessed for sexual side effects, which are common in antidepressant treatment.
“Withdrawal and relapse monitoring at treatment completion would help validate the optimal duration of treatment in clinical trials. However, these features are missing from clinical trials of all durations, even though the risk of withdrawal has been known for 30 years,” Ward, Haslam, and Prasad write.
In short, the evidence base for the long-term use of antidepressant drugs does not hold up to scrutiny, according to the researchers. The studies are of short duration and are riddled with methodological issues like unblinding, nonstandard assessments, and failure to assess withdrawal and adverse effects.
“Publicly funded randomized controlled trials comparing antidepressants to placebo with long duration, which also monitor for withdrawal, sexual side effects and relapse upon treatment discontinuation are necessary to determine the optimal duration of therapy,” write Ward, Haslam, and Prasad.
****
Ward, W., Haslam, A., & Prasad, V. (2025). Antidepressant trial duration versus duration of real-world use: A systematic analysis. medRxiv. Published online on February 28, 2025. https://doi.org/10.1101/2025.02.27.25323057 [Full text]