Agitating Research

Aripiprazole (Abilify) is now available in injectable form and is allegedly a great way to calm down patients quickly. Here’s a piece from the press release touting the release of this medication: “ABILIFY Injection provides rapid control of agitation in adults with schizophrenia or bipolar mania at primary endpoint (2 hours).” Let’s see if the numbers actually support this statement…

From one abstract (Andrezina et al 2006) : “At 2 h, mean improvements in PEC scores with IM aripiprazole (-8.0) were significantly greater versus IM placebo (-5.7; p less than or = 0.01).” That is a difference of 2.3 points. The Abilify group changed by 29% more than the placebo.

From another abstract (also Andrezina et al. 2006 in a slightly different study): "Mean improvement in PEC at 2 h was significantly greater for IM aripiprazole (-7.27) vs placebo (-4.78; p less than 0.001)." A difference of 2.49 points. The Abilify group changed 34% more than placebo.

What is the PEC, you might wonder. It’s the Positive and Negative Syndrome Scale Excited Component, and I’m willing to venture a guess that a difference of less than three points versus placebo is not particularly compelling clinical evidence. It has five questions, each of which is scored on a 7-point scale.

But wait, there’s more! The first study I mentioned was “A sub-population analysis” of 325 patients with agitation. I’m betting that the second study was just the first study with a few more patients thrown in whose initial scores on the PEC were not as high as in the second study. Looking at the abstracts quickly, one might think they were entirely different studies, but I highly doubt this is the case. I’ll do more research and correct myself if I am wrong, although if an enlightened reader has this information readily available, please let me know.

So, coming back to efficacy, we have a large sample size, which helps make this apparently small difference between drug and placebo turn up as statistically significant. Way too many people are under the impression that statistical significance is the most important thing since sliced bread. All it shows in this case is that there really is a difference between Abilify and placebo – but it does not say if the difference was teeny-tiny or if it was gigantic. To put this in perspective, remember that the PANSS-EC is a 35-point scale. So a difference of 2.3 points is nothing to write home about.

Don’t worry, as is highly common in pharmaceutical press releases discussing research, an academic spokesperson chimed in: “"ABILIFY Injection controls agitation independent of sedation,” according to Michael H. Allen, MD, Director, Emergency Psychiatry, Associate Professor, University of Colorado at Denver and Health Sciences Center's School of Medicine. What is not mentioned in the press release is that Dr. Allen is a member of the Scientific Advisory Board, has received research support from, and is a member of the speakers’ bureau for (you guessed it) Bristol Myers Squibb, distributor of Abilify. It is always helpful to place a little academic credibility onto your press release. Dr. Allen has relationships with several other pharmaceutical companies as well, and his opinion on Abilify may indeed be separate from his financial relationship with BMS. It is, however, certainly misleading for the press release to not mention that Dr. Allen has a significant relationship with the company, making him appear as if he were only an independent academic researcher.

Thus, the question for today is as follows: Does Abilify reduce agitation to a meaningful degree more than placebo or does this marketing campaign make you significantly agitated?