The use of hypothesis testing baseline imbalance after randomization as a means to evaluate the effect of the randomization is clearly based on a misunderstanding of hypothesis testing, and it has been criticized by several authors, see for example (1,5,7). However, this is still a common a mistake.

Knol’s et al. (6) results from 2012 can be compared with similar review results in articles published in 2000 (3) and in 1990 (2). The comparison, based on reviews of articles in New England Journal of Medicine (NEJM), British Medical Journal, the Lancet and Journal of the American Medical Association (JAMA) or Annals of Internal Medicine, shows a slowly declining frequency from 58% in 1990, to 48% in 2000, and to 35% in 2012. This is, of course, encouraging. However, the two American journals NEJM and JAMA did not improve at all, with 54% and 58% respectively in 2012.

This stands in striking contrast to Senn’s (7) statement about hypothesis testing of baseline imbalance in randomized trials that “this practice is philosophically unsound, of no practical value, and potentially misleading”, as well as to CONSORT statement’s (4) guideline that baseline testing should not be performed. The latter is somewhat surprising since both JAMA and NEJM require their authors to comply with the CONSORT statement.

References

  1. Altman DG. Comparability of randomised groups. Statistician 1985;34:125-36.
  2. Altman DG, Doré CJ. Randomisation and baseline comparisons in clinical trials. Lancet 1990;335:149-153.
  3. Assmann SF, Pocock SJ, Enos LE, Kasten LE. Subgroup analysis and other (mis)uses of baseline data in clinical trials. Lancet 2000;355:1064-1069.
  4. http://www.consort-statement.org/consort-statement/13-19—results/item15_baseline-data/ (accessed January 2013).
  5. Roberts C, Torgerson DJ. Understanding controlled trials: baseline imbalance in randomised controlled trials. Br Med J 1999;319:185.
  6. Knol MJ, Groenwold RHH, Grobbee DE. P-values in baseline tables of randomised controlled trials are inappropriate but still common in high impact journals. Eur J Prev Cardiol 2012;19:231-232.
  7. Senn S. Testing for baseline balance in clinical trials. Stat Med 1994;13:1715-1726.