BMJ 1999; 318 doi: (Published 22 May 1999) Cite this as: BMJ 1999;318:1427
  1. Douglas Carnall
  1. BMJ This week's subject illustrates the classic problem of medical information on the internet: a surfeit of sites of varying quality. On page 1375 we publish details of a systematic review of trials of treatment of the premenstrual syndrome with vitamin B-6 (pyridoxine) that suggests that, although existing trials are of low quality, the vitamin may be useful in relieving premenstrual symptoms, especially depression.

    The need for rigorous evaluation is clear: premenstrual syndrome is as much a lay diagnosis as a medical one, which will tend to increase the quantity of information directed at a lay audience. It is also a condition for which there are many treatments, usually a fair sign that nothing works particularly well. Worse still, some of these are vitamins, a certain way to (how can I be polite?) attract practitioners with unusual health beliefs. And behind every treatment is a company with something to sell (see, for example, Add in the variability that a prefixed word can cause in a digital search, and the result is entanglement.

    The Cochrane Library would be a good place to start; abstracts of its systematic reviews are free on the web (at, as are protocols in progress (including the one of the research reported this week), but subject coverage remains limited. Omni ( wasn't much help: two links, one dead, and one a brief summary. You could pick hundreds of sites with information about the syndrome from a raw search on any search engine. The best that I came across was HealthAnswers, which has good quality, well organised information aimed at consumers.

    However, the problems of the medical encyclopaedia persist into the virtual world: you have to know what you've got before you can look it up.

    View Abstract

    Log in

    Log in through your institution


    * For online subscription