BMI with waist circumference threshold upgrades mortality prediction in postmenopausal women

Further stratification of BMI categories by waist circumference (WC) thresholds can improve prediction of mortality risk among postmenopausal women, suggests a recent study, noting the association of larger WC with greater mortality.

Overall, 139,213 postmenopausal women aged 50 to 79 years were included in a development cohort (n=67,774) and two external validation cohorts. The first validation cohort had high prevalence of overweight or obesity (n=48,335), while the second included diverse, geographically separate centres (n=23,104). Their height, weight, and WC were measured at enrolment.

The authors assessed mortality annually and supplemented this with serial National Death Index queries. They compared 10- and 20-year mortality prediction models that included BMI categories to those with BMI categories stratified further by WC thresholds using c-statistics and continuous net reclassification improvement (NRI).

Of the participants, 69,297 died over a median follow-up of 24 years. Women in BMI categories with large WC had consistently higher multivariable-adjusted mortality risk than those with normal WC.

Women with normal (hazard ratio [HR], 1.17, 95 percent confidence interval [CI], 1.12–1.21) or overweight BMI (HR, 1.19, 95 percent CI, 1.15–1.24) but large WC had a similar risk to those with obesity 1 but normal WC (HR, 1.12, 95 percent CI, 1.08–1.16) relative to women with normal weight and WC.

Furthermore, the mortality risk associated with obesity 1 and large WC (HR, 1.45, 95 percent CI, 1.35–1.55) was comparable to that with obesity 3 and normal WC (HR, 1.40, 95 percent CI, 1.28–1.54).

Prediction models with BMI-specific WC thresholds improved discrimination and risk stratification at 10 years for the first validation cohort. The c-statistics improved by 0.7 percent (95 percent CI, 0.3–1.0) to 61.3 percent (95 percent CI, 60.2–62.5). Continuous NRI was 20.4 percent (95 percent CI, 17.3–23.6).

On the other hand, the second validation cohort showed mixed results. Risk stratification improved (continuous NRI, 12.3 percent, 95 percent CI, 8.5–16.0), but discrimination did not.

Results at 20 years were similar.