Estimated gains in life expectancy with use of postmenopausal estrogen therapy: a decision analysis.

BACKGROUND

Epidemiologic data are accumulating that suggest that postmenopausal estrogen therapy reduces the risk of developing coronary artery disease (CAD). Computer simulation by Markov analysis can be applied to current data to estimate the increase in life expectancy obtained from postmenopausal estrogen use and compare them with benefits from other therapies for CAD risk reduction. Decision-analysis techniques can also examine whether the benefits of unopposed estrogen regimens ever exceed those of combination therapy.

METHODS

In our analysis, hypothetical cohorts of postmenopausal women age 50 and 65 years with intact uteri were assigned either to estrogen and progesterone therapy or unopposed estrogens. The subjects were also defined by risk category for CAD. Outcomes were measured in terms of life expectancy for treatment cohorts compared with identical untreated cohorts.

RESULTS

Life expectancy benefits in combined therapy groups were found to be very substantial for all CAD risk categories. Cohorts who began therapy at age 50 years showed benefits ranging from 0.3 years of additional life for those at low risk of developing CAD to 2.3 years for those at high risk. Even though the addition of progestins may theoretically result in reduction of overall CAD benefits, impressive gains in life expectancy were still found even when a 40% reduction in estrogenic effect was considered. Overall, benefits were very favorable when compared with other accepted strategies for CAD risk reduction. Little additional benefit was found to justify use of unopposed estrogens given the potential added mortality from endometrial cancer.

CONCLUSIONS

Substantial increases in life expectancy may result from postmenopausal estrogen therapy. These may be equal to or possibly greater than benefits from other well-recognized risk-reduction strategies. Little advantage in additional life expectancy is found to justify use of unopposed estrogens.