Background: Astrological or Zodiac (star) sign has been shown to be a statistically significant factor in the outcome of a variety of diseases, conditions, and phenomena.
Methods: To investigate its relevance in the context of a stem cell transplant (SCT), we examined its influence in chronic myeloid leukaemia, a disease with well-established prognostic factors. Data were collected on 626 patients who received a first myeloablative allogeneic SCT between 1981 and 2006. Star sign was determined for each patient.
Results: Univariate analyses comparing all 12 individual star signs showed considerable variation of 5-year probabilities of survival, 63% for Arians, to 45% for Aquarians, but without significance (P=.65). However, it was possible to pool together star signs likely to provide dichotomous results. Thus, grouping together Aries, Taurus, Gemini, Leo, Scorpio, and Capricorn (group A; n=317) versus others (group B; n=309) resulted in a highly significant difference (58% vs 48%; P=.007). When adjusted for known prognostic factors in a multivariate analysis, group B was associated with an increased risk of mortality when compared with group A (relative risk [RR], 1.37; P=.005).
Conclusion: In this study, we show that, providing adequate care is taken, a significant relationship between patient star sign and survival post SCT for CML can be observed. This is, however, a completely erroneous result, and is based on the pooling together of observations to artificially create a statistically significant result. Statistical analyses should thus be carried out on a priori hypotheses and not to find a meaningful or significant result.
Copyright © 2010 Elsevier Inc. All rights reserved.