Interaction between psychotherapy and psychopharmacology
/I was recently asked the following question: "What research is there on the paradox that psychotherapy doesn't add to the effect of (psycho-)pharmacological treatment?" This is a big topic without complete or satisfactory answers. I comment here on a few aspects of it, and list some references for who wants to read more.
The question in more detail
The question itself is assuming a bit much. There has been good research either way, i.e. examples where it appears that combined psychotherapy and pharmacology produce better results, studies where there seemed to be no difference, and well-documented studies where the outcome of combined therapy and drugs was worse than that of only psychotherapy. Most of this research just reported the results; there does not seem to be a good understanding why the studies come out the way they do.
Interaction of variables considered in statistics
There is a considerable literature in the field of statistics, espcially statistics as used in social sciences and in epidemiology, that explains that there are no valid generalisations as to how variables will interact. Simple additive interaction is possible, sometimes there is multiplicative interaction, other effects are possible. This means that in any real-life situation or experiment the data have to be interrogated as to what interaction there may be. A start on some of the issues here can be found in wikipedia, in general and with a statistical focus.
A good source on psychopharmacology and psychotherapy
In the latest edition of Michael Lambert's Handbook of Psychotherapy and Behavior Change there is an excellent summary chapter on this: chapter 18. pp. 735 - 774, by Forand, DeRubeis, & Amsterdam, with a long list of source articles, including key research examples of the combination being better and worse than the monotherapy. You will also find there some good speculation as to possible causes of the unpredictable interaction.
Moderation and mediation
Another statistical terminology that may be relevant to this problem, are the concepts of moderating and mediating variables. There are a number of (mostly modern) tests to try to make sense of this. I can't be certain, but have a sense that it is at least possible that this might also play a role in the understanding of the impact of drugs and counselling. The same Lambert Handbook also has a few good descriptive pages about this, pp. 36-38 in Ch.2, which is worth reading, with further references, especially to chapter 15 by MacKinnon et al. in the new Comer and Kendall (Eds.) Oxford Handbook of Research Strategies for Clinical Psychology, just out in the USA, bot not yet in the UK. Also the wikipedia article on this is quite rich.
Don't ignore psychopharmacology!
Overall, I personally feel that we psychotherapists only ignore the possibilities and impact of pharmacology at our peril. There is a range of up to date books on this topic, most of which are American (I rather like Preston et al. and Sinacola et al.). In the UK there is the excellent book by David Healy, Psychiatric Drugs Explained, helpful, as it uses British names for the drugs. I refer here to the fifth edition, which is now about five years old. The sixth edition is on its way and should come out this year or in 2014.
But these books mainly talk about the pharmacology, and don't really focus on the interaction with psychotheraypy - possibly because there is not too much hard knowledge on that topic.