User Tools

Site Tools


medical_statistics

Medical statistics

Uncertainty is as fundamental for medical research as in other fields of science. Medical research is typically performed on small groups of subjects, e.g. 32 healthy volunteers included in a randomized trial, a cohort of 587 asbestos workers followed 18 years, or some 83761 patients having been treated with a specific drug and registered in a patient register. Whatever the sample size, the unavoidable question is whether the observed outcomes are relevant for others than those studied? Are the results generalizable to other, future patients? Or do the results only reflect sampling variation, or bias? The inferential uncertainty is usually presented in terms of a probability (p-value), or an interval of plausible values (confidence interval).

Given the importance of statistical science in medical research, one would expect medical researchers to be especially qualified and experienced in using statistical methods. The sad truth is, however, that wrong techniques often are used, either willfully or in ignorance (1), and that many published results are misleading. For example, Gore et al. (2) showed 1976 that of 62 articles published in the British Medical Journal, 32 (52%) had statistical errors and 18 (29%) were seriously flawed. Similar errors appear today, but, as this reference collection shows, the problem is now even greater because the number of scientific publications is greater; the pressure to publish is greater, and the availability of advanced computing techniques is greater. Complicated calculations can easily be made without any methodological understanding and many user-friendly statistical software packages are available. Even completely erroneous research results have good chances of being published as long as they appear convincing, a problem that is exacerbated by the common exaggerations and frequent spin that now characterize many scientific publications and much of the media's presentation of research news. Critical thinking is crucial when reading scientific literature.

References

  1. Altman DG. The scandal of poor medical research. Br Med J 1994;308:283.
  2. Gore SM, Jones IG, Rytter EC. Misuse of statistical methods: critical assessment of articles in BMJ from January to March 1976. Br Med J. 1977;1:85-87.
medical_statistics.txt · Last modified: 2020/02/16 18:46 by ranstam