Reporters usually don’t write the headlines, but we wish someone would have changed the headline on this story. The body of the story conveyed important questions – about the number needed to treat – and about differences in treating the elderly – that perhaps were worthy of headline attention.
The story tried to capture why this is important in its opening sentence:
Despite recent debate about how well antidepressants really work in people with only mild or moderate depression, a new analysis of drug studies suggests they may have some benefit across the board.
But there is a nuance here. The previous patient level meta-analysis (JAMA) included fewer studies, but across a broader spectrum of depression (minor depression to major depression), and some with longer followup The current study and more included trials (good) restricted their analyses to studies enrolling patients with major depression (also good – this is the group for whom it appears treatment is most indicated) but only have short-term results (not so good). Although this level of detail is probably beyond what is needed for a general circulation story, pointing out WHY these 2 studies may have found different results and specifically noting that the results apply to those with major depressive disorder would be helpful
The costs of antidepressants weren’t discussed, but perhaps that is understandable in this kind of broad overview of a meta-analysis. Fluoxetine (prozac) and venalfaxine (effexor) are now both generic. So costs are pretty low.
The story gave detailed breakdowns of percent improvement in scores on depression tests and provided an estimate of the number needed to treat in order for one to benefit. The difference in benefits between older and younger patients was highlighted.
There was at least the acknowledgement that these drugs do carry side effects – but none was specified nor quantified.
Several of the limitations of the meta-analysis were mentioned in the story.
Of course, one key limitation in such cases is how they selected studies for the analyses (ones already available to them from the National Institute of Mental Health and from drug companies). This may not bias the results but may bias the interpretation (the authors say they only know these effects for prozac/effexor and can’t comment on other drugs; that could possibly lead a reader to incorrectly conclude that only prozac/effexor are shown to be effective.)
No disease mongering about depression here.
The story interviewed Irving Kirsch, frequently cited/quoted on antidepressant research issues.
It also did a good job of noting potential conflicts of interest.
The story at least noted alternatives including psychotherapy and exercise, but didn’t give information about the effects, pros/cons of these options.
The availability of antidepressants was not in question.
The story didn’t analyze the novelty claim that the researchers made:
“To our knowledge, this is the first research
synthesis in this area to use complete longitudinal
person-level data from a large set of published and
unpublished studies. The results do not support previous
findings that antidepressants show little benefit except
for severe depression.”
The second point was made in the story, but perhaps the story should have explained a bit more about the methodological reasons for why this analysis was – or what made it – different or important.
As to the first (questionable) part of the claim, what’s novel is this: more studies than the previous analysis, adolescents/adults/older adults included and analyzed by these groups.
The story did not rely on a news release.