Postmenopausal hormone therapy initiation before and after the Women's Health Initiative in two French cohorts Running title: Evolution of HT use in France

Abstract : The current evaluation of the benefit/risk ratio associated with menopausal hormone therapy (MHT) use is largely based on clinical trials which investigated the effects of oral treatments. Would MHT with transdermal estrogens be associated with a more favourable benefit/risk ratio? We reviewed the available epidemiologic evidence on that question. Epidemiologic studies were considered if they provided risk estimates of conditions which carry an important weight among menopausal women, and for which epidemiologic evidence of a possible link with MHT use is convincing: cardiovascular diseases, breast cancer, diabetes, colorectal cancer and hip fracture. We did not include studies with only surrogate measures. We found that the available information on the potential impact of the route of administration of MHT on the risk of our selected outcomes is limited. To date, epidemiologic data suggest that it has no impact on the risk of breast cancer and hip fracture. Results on the risk of coronary heart disease and colorectal cancer are inconsistent. Studies on stroke and diabetes risk are too few to allow meaningful conclusions. There is a suggestion that transdermal MHT may be less deleterious than oral MHT regarding venous thromboembolism which needs to be confirmed. The issue of the route of administration of MHT should remain an active area of research as part of an attempt to identify treatment modalities that would have the least potential for exerting adverse effects.