SSRIs show weak advantage over bupropion for anxious depression
Last year, we reviewed a meta-analysis implying that bupropion is as effective as SSRIs for the treatment of patients with mixed anxiety and depression (TCPR, Aug 2007). In a new paper, these same researchers have sliced and diced the data a bit differently in order to answer this topic more confidently. In a pooled analysis of 10 trials, researchers found that patients with anxious depression had a higher response rate when taking an SSRI as opposed to bupropion for both depression (65.4% vs. 59.4%) and anxiety (61.5% vs. 54.5%), as measured by the Hamilton Rating Scale for Depression (HAM-D) and the Hamilton Rating Scale for Anxiety (HAM-A). These differences were statistically significant. There was also a statistically significant, but very small, difference favoring SSRIs over bupropion on the mean HAM-D score. For nonanxious depression, there were no differences in treatment outcomes between patients taking bupropion and those taking SSRIs. (Papakostas GI et al., J Clin Psychiatry; Published online ahead of print).
TCPR’s Take: For anxious depression, this is the largest comparison between bupropion and SSRIs. The clinical relevance of the results is certainly questionable, as the average patient taking an SSRI fared less than a single point better on the HAM-D relative to the average patient taking bupropion. There were no differences between SSRI and bupropion in terms of remission rates for depression or anxiety. Based on these findings, SSRIs appear to possess, at most, a very slight advantage over bupropion for anxious depression. All trials included in the analysis were sponsored by GlaxoSmithKline, manufacturer of bupropion.
Computer as psychotherapist?
Cognitive behavior therapy (CBT) is widely acknowledged to be the most well-researched, and possibly the most effective therapy option for a wide variety of disorders. The problem is that it is not widely available, particularly in the treatment of patients with substance abuse disorders. In this new study, researchers developed a computer-based training in CBT skills (cutely labeled CBT4CBT) specifically for the treatment of substance abuse. This program consisted of six lessons based on a CBT manual published by the National Institute on Drug Abuse. Each lesson begins with a movie illustrating a vignette in which a substance abuser fails to use appropriate skills (for example, is unable to refuse a drug when offered). This is followed by a description of the cognitive skills required and then by a repetition of the videotaped vignette in which the drug user is able to successfully use the relevant skill. Other components of each lesson include interactive assessments of symptoms, instructions on how to generalize CBT principles to situations other than substance abuse, and a videotape of a patient completing a homework assignments based on the lesson. Participants are then given the same assignment to take with them.
The participants were treatment-seeking outpatients recruited from a substance abuse clinic in Connecticut. Most were polysubstance abusers, with cocaine being the primary drug problem for 58%. The 77 subjects were randomly assigned to either treatment as usual (TAU – weekly individual and group therapy sessions) or TAU plus access to the CBT4CBT computer program in a small private room in the clinic. After 8 weeks of treatment, the CBT4CBT participants had a lower proportion of drug-positive urine tests (34% vs. 53%, statistically significant) and a longer number of abstinent days (22 days versus 15 days, just short of statistical significance) (Carroll KM et al., Am J Psychiatry 2008;165:881-888).
TCPR’s Take: While the study needs to be replicated, the results are pretty impressive, indicating that simply adding a computer-based therapy to standard substance abuse treatment can yield clinically significant benefits. In an accompanying editorial (Am J Psychiatry 2008;165:793-795), John Greist (who discloses financial interests in computer therapies other than CBT4CBT) points out some of the inherent advantages of computer therapy, including the fact that it is always available, proceeds at the patient’s own pace, is unaffected by variations in the quality of therapist training, and is much cheaper to develop and administer than either medications or human-based therapy.