Reproductive Risk Factors and Coronary Heart Disease in the Women’s Health Initiative Observational Study

    loading  Checking for direct PDF access through Ovid

Abstract

Background—

Reproductive factors provide an early window into a woman’s coronary heart disease (CHD) risk; however, their contribution to CHD risk stratification is uncertain.

Methods and Results—

In the Women’s Health Initiative Observational Study, we constructed Cox proportional hazards models for CHD including age, pregnancy status, number of live births, age at menarche, menstrual irregularity, age at first birth, stillbirths, miscarriages, infertility ≥1 year, infertility cause, and breastfeeding. We next added each candidate reproductive factor to an established CHD risk factor model. A final model was then constructed with significant reproductive factors added to established CHD risk factors. Improvement in C statistic, net reclassification index (or net reclassification index with risk categories of <5%, 5 to <10%, and ≥10% 10-year risk of CHD), and integrated discriminatory index were assessed. Among 72 982 women (CHD events, n=4607; median follow-up,12.0 [interquartile range, 8.3–13.7] years; mean [standard deviation] age, 63.2 [7.2] years), an age-adjusted reproductive risk factor model had a C statistic of 0.675 for CHD. In a model adjusted for established CHD risk factors, younger age at first birth, number of still births, number of miscarriages, and lack of breastfeeding were positively associated with CHD. Reproductive factors modestly improved model discrimination (C statistic increased from 0.726 to 0.730; integrated discriminatory index, 0.0013; P<0.0001). Net reclassification for women with events was not improved (net reclassification index events, 0.007; P=0.18); and, for women without events, net reclassification was marginally improved (net reclassification index nonevents, 0.002; P=0.04)

Conclusions—

Key reproductive factors are associated with CHD independently of established CHD risk factors, very modestly improve model discrimination, and do not materially improve net reclassification.

Related Topics

    loading  Loading Related Articles