Conclusion
In our cohort, the health characteristics of newly postmenopausal women changed similarly between the pre-WHI period and the post-WHI period among women who started HT and those who did not, with the exception of troublesome hot flushes and (although not significantly so) psychological symptoms. During the two periods, physicians, rather than women, mainly made the decision about HT. These factors imply that the substantially greater selectivity in prescribing HT was independent of women's health characteristics, contrary to what we expected. Although our results suggest that hot flushes had to be more severe for HT to be prescribed in the post-WHI period, almost one third of prescriptions were still written for reasons not linked to hot flushes. On the other hand, the recommendation that HT not be used as a first-line treatment of osteoporosis seems to have been taken into account because osteoporosis prevention became a far less frequent reason for HT use in the post-WHI period.