Mental Health Blog - antidepressant treatment review

Recently, many psychologists, psychiatrists, the general public, and the media have been becoming more aware of bias in pharmaceutical research.  The most recent work found that researchers design the study in order to favor the medication that is sponsored by the funding source.  In this post, I will talk about some of the problems with pharaceutical study research design.

The Placebo Washout Phase

Nearly all studies of pharmaceuticals, at least psychotropics, include an initial phase in which people who respond (read benefit) from a sugar pill are excluded from the study.  So, if in the initial phase, they get better on placebo, they are excluded from the study.  Purportedly, this is to remove an unimportant variable.  But what this does is amplify the effects of the medication in the actual study.  So, the drugs end up looking like they have a greater effect than they actually have.  You end up with a group of folks who purportedly do not respond well to placebo taking an active drug versus a sugar pill.  If you are going to run this type of study, the results need to state, "For people who are poor responders to placebo, the drug proved to be moderately efficacious."  That would be a more honest way to present the results, but I have never seen this occur.  The exclusion of 'placebo responders' amplifies the effect of the 'drug.' 

Use of Percentages in Results

I have seen a number of popular drugs tested against placebo and the only statistics you find in the article to be "percent who achieved remission."  You read the whole article and the only thing presented is percentages and statistics to tell you if the percentages who fall into certain groups differ.  This is also done so that the drug looks better than placebo.  As an example, please see my previous analysis of on a study of antidepressants (Prozac) in children.

The Double Blind is not Blind

The double-blind research design is considered to be the 'gold standard' in pharmaceutical research.  It is purported to be the most objective and best way of determining whether or not a drug has efficacy.  The double-blind experiment is accurately described on Wikipedia:

Double-blind describes an especially stringent way of conducting an experiment, usually on human subjects, in an attempt to eliminate subjective bias on the part of both experimental subjects and the experimenters. In most cases, double-blind experiments are held to achieve a higher standard of scientific rigour.

So, the assumption is that the researchers do not know which group the study participants have been assigned, and the participants do not know which group they have been assigned to.  Ah....but there's a problem here.  Previous research with antidepressant medication, reveals that both patients and doctors are pretty good at figuring out which group they are in, most likely because of side effects.

Some Solutions to the Problem

If you are reading pharmaceutical research that does not include a check to see if the blinding actually worked (in other words ask the researchers or physicians what group they think the participant is in and ask the participant what group they think they are in), uses a placebo washout phase, and only uses percentages or stats to tell if the percentages differ, be very skeptical of this research.  Irving Kirsch has proposed a model for research that includes an 'active placebo,' in other words a drug that produces side effects, but in theory does not act in the same way as the drug.  This should help to actually insure the 'blind' of the study.  Finally, when research is funded by a pharmaceutical company, you need to be very skeptical of the results.  If researchers are going to receive pharmaceutical industry funding, they need to be blind to the 'who' is funding them.