Thanks.

I guess that with the study you mentioned I'm concerned that if the experimental group and the control group weren't matched for other factors that could be relevant to prevent relapse (such as social support, therapy, time spent with a p-doc, a p-doc who had faith that the medication / placebo would make a significant difference etc) then the differences that were found could have been due to differences in those factors rather than differences in whether the person was taking medication or not.

Of course this may have been controlled for (I haven't read the study). It is just that I'm especially wary of studies that are done on medications because of the significant investment that drug companies have in sponsoring the studies (and that researchers have in finding favourable outcomes for the drug companies with respect to future funding opportunities).

Of course it is wise to think critically about experimental findings more generally... But I'm especially sceptical about ones involving pharmacuticals...

If there isn't empirical evidence that taking medication prevents relapse then there wouldn't seem to be a problem with stating this to the client. Many people find that they have nasty side effects of taking medication. Therapy would seem to have less nasty side effects and hence if the studies haven't been done on whether anti-depressants (all or some of them?) can help prevent relapse then it would seem to me that therapy would be a better option. I could be wrong... But I thought the studies had been done on cognitive behaviour therapy and it was found that people who attended a course of CBT were less likely to relapse compared to a control group who had not attended a course of CBT. Though I don't know that the groups were matched for medication and for social supports more generally...