Reprinted from British Journal of Psychiatry, 1990, 157, 300

Newcombe (Journal, February 1990, 156, 282), in his letter on double-blind trials, continues to defend randomised controlled trials in psychiatric research (Newcombe, 1988), without apparently recognising their limitations (Kramer & Shapiro, 1984). In particular, he does not accept the fallibility of the double-blind and criticises Oxtoby et al's (1989) suggestion that the ability of participants to guess their drug status should be used as a retrospective criterion to exclude certain results. Dr Newcombe's caution about this latter suggestion is justified, although the recording of guesses of whether patients were taking active drugs or placebo can be of value. As a recent example, Marks et al (1988) found that assessors' guesses after the end of treatment were mostly right. They did not assess the blindness of patients but suggested that such checks might yield similarly sobering data. Moreover, Oxtoby et al (1989) advocate reworking results when the double-blindness has been disproven, and the question then arises of why the trial was not double-blind.

The breaking of the double-blinding on occasions has been interpreted as the explanation for a positive trial result. For example, Karlowski et al (1975) found that ascorbic acid seemed to reduce the duration of a common cold, but these differences were eliminated when taking into account the correct guesses of medication. Practically, it is difficult to know whether the issues that Dr Newcombe raises about upsetting the randomisation and similar problems would be relevant to this study.

Perhaps a more appropriate conclusion to draw from this debate is that clinical trials are unlikely to be definitive in the scientific sense that Dr Newcombe would like. Interpretation of results is inevitably important, which may explain why there is so much controversy about the effectiveness of psychiatric treatment.

KARLOWSKI, T.R., CHALMERS, T.C., FRENKEL, L.D. et al (1975) Ascorbic acid for the common cold. Journal of the American Medical Association, 231, 1038-42.

KRAMER, M.S. & SHAPIRO, S.H. (1984) Scientific challenges in the application of randomised trials. Journal of the American Medical Association, 252, 2739-45.

MARKS, I.M., LELLIOTT, P., BASOGLU, M., et al (1988) Clomipramine, self-exposure and therapist-aided exposure for obsessive-compulsive rituals. British Journal of Psychiatry, 152, 522-34.

NEWCOMBE, R.G. (1988) Evaluation of treatment effectiveness in psychiatric research. British Journal of Psychiatry, 152, 696-7.

OXTOBY, A., JONES, A. & ROBINSON, M. (1989) Is your 'double-blind' design truly double-blind? British Journal of Psychiatry, 155, 700-1.