THS reader Chad sent in this question:
Antidepressants – effective or placebo?
The use of antidepressant medication has become so widespread and commonly accepted that it seems almost sacrilegious to question it. But alas, questioning is the name of the game here at The Healthy Skeptic!
What that means is that in most of the trials reviewed, patients who took a sugar pill recovered from depression just as often as those who took the active drug. This study may come as some surprise to both physicians and the general public, whose faith in the efficacy of these drugs has led to over 118 million prescriptions in 2007 and over $16 billion in sales.
But should this really come as a surprise? Antidepressant drugs are thought to act by altering levels of brain neurotransmitters; however, it takes several weeks before these changes can be measured. Yet patients often report symptomatic relief within hours or days of receiving an antidepressant.
Available data suggests, in fact, that SSRIs are no more effective than placebos and have considerable adverse effects and risks, including increased suicidality amongst both children and adults. Sapirstein and Kirsch conducted a meta-analysis of 3,000 patients who received either antidepressants, psychotherapy, placebo or no treatment at all. They found that 27% of therapeutic responses were attributable to drug activities, 50% to psychological factors, and 23% to “non-specific” factors. In other words, 73% of the response to the drug was unrelated to its pharmacological activities – and antidepressants may be no better or more specific than placebos.
This of course raises grave questions about why the National Institute for Health and Clinical Excellence (NICE) still recommends that antidepressants should the be first line treatment for moderate or severe depression. Their message is identical to that of the Defeat Depression Campaign in the early 90s, which contributed to the 253% rise in antidepressant prescribing in 10 years.
In a review published in the British Medical Journal in February of 2006, researchers Joanna Moncrieff and Irving Kirsch point out that the NICE recommendations ignore even their own study data. Although the NICE meta-analysis of placebo controlled trials of SSRIs found statistically significant differences in levels of symptoms, these were so small that the effects were deemed “unlikely to be clinically important.”
After analyzing several published studies and reviews, Moncrieff and Kirsch reached the following conclusions:
- SSRIs have no clinically meaningful advantage over placebo
- Claims that antidepressants are more effective in more severe conditions have little evidence to support them
- Methodological artifacts may account for the small degree of superiority shown over placebo
- Antidepressants have not been convincingly shown to affect the long-term outcome of depression or suicide rates
The response to a drug or placebo in a clinical trial for depression is often measured using the Hamilton rating scale, a multiple choice questionnaire which doctors use to rate the severity of a patient’s condition. The questionnaire rates the severity of symptoms observed in depression such as low mood, insomnia, agitation, anxiety and weight-loss; it is considered to be a highly reliable physician-rated scale and has been reported to be more sensitive than patient-rated scales to drug/placebo differences. (Murray, 1989)
In the NICE meta-analysis, the difference between drug and placebo groups was one point. The most commonly used 17 item version of the Hamilton scale has a maximum score of 52. It is highly unlikely that a difference of one point on a 52-point scale is clinically significant, a fact that the FDA has admitted in memoranda (Laughren, 1998; Leber, 1998) reviewed by Moncrieff and Kirsch.
Other studies have yielded similar results. A study by Khan et al. found a 10% difference in levels of symptoms between placebo and active drugs in two different meta-analyses. In a more recent review, Kirsch et al. invoked the Freedom of Information (FOA) act to obtain access to previously unpublished studies (the drug companies are under no requirement to publish a study they have sponsored if the results don’t suit them). The overall difference between drugs and placebos in that analyses was 1.7 points on the Hamilton scale.
Moncrieff and Kirsch also point out that the Hamilton scale contains seven items concerning sleep and anxiety, with each item on sleep scoring up to six points. Therefore any drug with some sedative properties, including many antidepressants, could produce a difference of two points or more without exerting any specific antidepressant effect.
Follow-up studies that track patients for a significant length of time have also shown very poor outcomes for people treated for depression both in the hospital and in outpatient settings, and the overall prevalence of depression is rising despite increased use of antidepressants. Suicide rates have increased in some groups and some countries, despite increased prescribing of antidepressant, and there are continuing concerns that SSRIs may increase the risk of suicidal behavior in obht cildren and adults.
In children, the balance of benefits to risks in antidepressant treatment is already recognized as “unfavorable”. The analyses performed by Moncrieff and Kirsch strongly suggests that the same is the case for adults, and that the ongoing uncertainty about the possible risk of increased suicidality as well as the adverse effects of antidepressant drugs warrant a “thorough re-evaluation of our current approach” to treating depression.
I couldn’t agree more. One question the authors failed to pose, which I believe to be at the root of the matter, is why are so many more children and adults depressed now than before? You might not be surprised to learn that I have some thoughts about this. But I’ll save them for another post.