Monday, September 25, 2006

Drugs or Therapy for Adolescent Depression? TADS...

Admittedly, this post gets an ‘F’ for timeliness, since the study was published in 2004! However, since this site is relatively new, I ask for your forgiveness.

The Treatment for Adolescents With Depression Study examined depressed adolescents who were assigned to one of four groups: fluoxetine (Prozac) alone, placebo, cognitive-behavior therapy (CBT), or pill placebo.

Treatment lasted for 12 weeks. At first glance, the results seem clear. The CBT + fluoxetine group had the best outcomes, improving to a statistically significantly greater degree than each of the other conditions. Further, fluoxetine alone was significantly more effective than CBT alone, which was not more effective than a placebo. Case closed – right?

Not exactly. First of all, people who were receiving fluoxetine alone had the lowest rate of comorbidity (i.e., having additional psychiatric diagnoses) – their rate of comorbidity was 43.1% compared to 58% for CBT alone, 51% for placebo, and 55.6% for CBT + fluoxetine. Additionally, those on fluoxetine had the lowest rate of dysthymia, a longstanding mild to moderate form of depression, (5.5%) compared to the other conditions – CBT alone (15.5%), placebo (10.7% ), CBT + fluoxetine (10.3%). Thus, one could argue that the fluoxetine group was the easiest to treat, as their overall psychiatric condition appeared to be somewhat less severe than that of participants in the other treatment groups.

Treatment was not always blind. Those receiving fluoxetine + CBT were fully aware that they were receiving mediation whereas those on CBT were fully aware that they were not taking medication. This is a huge confound given that it is quite clear that the placebo effect in depression is huge. Simply put, the placebo effect likely put fluoxetine + CBT at a substantial advantage over CBT alone.

Safety. Harm-related adverse events occurred in 11.9% of the fluoxetine group, 8.4% of the fluoxetine + CBT group, 4.5% of CBT participants, and 5.4% of placebo-treated patients. The relative risk of greater than 2 when comparing fluoxetine to CBT suggests that psychotherapy is notably safer than SSRI treatment. Comparing SSRI treated patients to non-SSRI treated patients indicated a significant difference in the direction of SSRIs being more likely to induce harm-related adverse events. Nobody committed suicide in the study, but let’s keep in mind that these patients are likely monitored much more carefully (thus, protecting from actual suicide) compared to patients in the real world. In addition, there was one psychiatric (non-harm related) adverse event in the CBT group, 16 in the CBT + fluoxetine group, and 23 in the fluoxetine alone group. ‘Nuff said on safety.

My View: This study was generally well-done but the confounding issues described above leave one to wonder about its results. Evidence shows that, considering both safety and effectiveness, psychotherapy is not much better than a placebo for adolescent depression and that medication is problematic at best.

Oh, and one more thing. The press release from the National Institute of Mental Health indicated that TADS was a $17 million study. Good study, but it sounds like the researchers were getting paid on one of those infamous Halliburton contracts!

No comments: