literature. Brit. Med. J., 313, pp. 1229-1232.
Nicol-Smith presents a review of 94 studies from 1966 to 1996, all of which failed to show a causal relationship between menopause and depression. She was, however, able to find 17 articles during this time period which did show an association.
Nicol-Smith made a wise decision in selecting as her null hypothesis (i.e. that there is no relationship) that menopause does not cause depression. She chose this because she believes it would be more destructive to the patient to ascribe depression to menopause and not treat it than to do the reverse. The author points out that it is very difficult to get a completely unbiased sample when searching data banks and reference sources as most authors will select articles which support their point of view. This is expressed at the beginning of the paper so that the reader is aware that there may be some bias in any of the studies she includes in her survey.
Nicol-Smith used a standard set of criteria for deciding whether a particular factor is a cause of or merely associated with depression. This is a fairly rigorous set of criteria. She did not look at experiments specifically designed to look at the relationship between depression and menopause, but included all experiments dealing with the two factors. She did not include studies specifically designed to add weight to the argument. She also looked at consistency of observations, specificity of criteria, and timing factors. Also included in her selection of articles was the biological plausibility of the arguments made and the coherence of the results. All these factors help to establish that this review was very carefully carried out, with every attempt made to avoid the author's bias in the conclusions drawn. In this aspect, this is a very fair and well-executed review of the available literature at the time.
After describing her methodology, Nicol-Smith goes on to consider t...