Are antidepressants an effective treatment for major depression? According to Professor Gordon Parker AO, a professor of psychiatry at UNSW and the founder of the Black Dog Institute, the answer is irrelevant—because we’re asking the wrong question.
In a British Journal of Psychiatry editorial published this month, Parker challenges the findings of several studies that analysed the outcomes of multiple clinical trials of antidepressants. Collectively, these trials spanned hundreds of thousands of patients who received either antidepressants or placebos to treat a diagnosis of major depression.
The aggregated trial data were interpreted in the earlier studies as indicating that antidepressants were no more effective than placebos, leading to widespread public confusion about the validity of these drugs, while the most recent analysis argued for their effectiveness.
The problem, Parker argues, is that trialling antidepressants against a target condition of major depression produces, if not guarantees, flawed results. ‘Major depression’ is an umbrella term that refers to a range of depressive illnesses with varying causes, trajectories and treatment responses, rather than to a single condition.
“The conclusions that these studies have drawn are, in a sense, beside the point,” Parker says.
“Basically, the target diagnosis of major depression captures multiple types of depressions—some biological, some psychological, some social—and not all would be expected to respond to medication.
“By bundling those with depression due to social and psychological causes in with the biological conditions, the signal benefits of antidepressant medication are swamped and effectively lost.”
Parker likens using an antidepressant to treat all types of depression as equivalent to using a Ventolin inhaler to treat all types of lung conditions, regardless of whether a patient has asthma, where an inhaler may be effective, or pneumonia or a pulmonary embolus, where an inhaler would be ineffective and an entirely different treatment might be required.
“For patients with depression, if you narrow down to those who have a biologically-based depressive sub-type, the antidepressants are distinctly effective,” he says.
“For example, biological depression or melancholia shows a 60 per cent response to medication, compared to only a 10 per cent response to placebo. If, however, you include other depressive disorders with no primary biological basis—as occurs, if not dominates, in most studies—antidepressants appear to be ineffective.”
Parker’s editorial also raises serious questions about the design of antidepressant clinical trials, in which patients with serious conditions—such as suicidal ideation, drug and alcohol problems or personality disorders—are ruled out in favour of those with mild symptoms.
Despite meeting the major depression criteria, these low-acuity participants are highly likely to go into spontaneous remission or to respond to placebos, thereby skewing the trial results and subsequent analyses of a drug’s efficacy.
“This approach doesn’t respect the real world, where there are multiple types of depressive conditions,” Parker says.
“It should not be a simplistic question—are antidepressants effective or ineffective? No treatment should be tested as if it has universal application for multiple differing depressive disorders.”
“It’s all about saying, what are the circumstances in which they’re effective, and when are they irrelevant?”
The editorial was published on August 1 in The British Journal of Psychiatry, Volume 213/Issue 2.
The full research article is available here: