[Hormone replacement therapy: thoroughness in method and interpretation of results. Regarding the Women's Health Initiative--WHI].

The randomised clinical trial of the Women's Health Initiative, conducted in women who had a natural menopause, has confirmed the absence of benefit of a regime of equine conjugated estrogens and medroxy-progesterone acetate on the occurrence of coronary heart disease. In addition, the treatment induced an increase in the incidence of invasive breast cancer that lead to its premature stop. However, as over 15'000 women have been followed for an average of 5.2 years in this well conducted trial, the results can be used with confidence. Furthermore, these results are similar to those of other recently published clinical trials. The expected benefits of hormone therapy on cardiovascular health corresponded to a premature conclusion derived from observational cohort studies that did not adjust for misclassification bias. This most plausible bias was due to inappropriate measurement of the socio-economic and education category, as higher socio-economic and education is associated with a lower risk of chronic disease such as coronary heart disease. Observational studies that allow appropriate adjustment for this factors produced results similar to those of randomised clinical trials that confer, if well designed and conducted, the highest degree of evidence of effectiveness of therapeutic and preventive interventions. Indeed clinical trials should be used to show the effectiveness of the new postmenopausal substitutive therapies that may be proposed.