The use of questionnaire data to classify peri- and premenopausal status

We assessed how well questionnaire data could classify peri- vs premenopausal status in 280 women, ages 38-49 years. Sixty-seven per cent of those who reported use of hormone replacement therapy begun when menstrual periods were irregular, or who had hot flashes at least once a day, or whose last menstrual period was at least 60 days in length, were perimenopausal by hormonal assay (early follicular phase serum follicle-stimulating hormone > or = 15 international units per liter), with a sensitivity of 0.56 [95% confidence interval (CI) = 0.41-0.72] and a specificity of 0.95 (95% CI = 0.93-0.98). In the classification of premenopausal status, the criterion of no menstrual cycle length changes in the past 5 years yielded the highest sensitivity (0.61; 95% CI = 0.54-0.67) and specificity (0.81; 95% CI = 0.68-0.93). With these few items, researchers can improve the classification of peri- and premenopausal status when longitudinal or hormonal assessments cannot be obtained.