Association of waist-hip ratio with diabetes mellitus. Strength and possible modifiers.

Abstract

To characterize the association between central obesity, as measured by the waist-hip ratio (WHR), and non-insulin-dependent diabetes mellitus (NIDDM), considering the effects of sex, age, overall obesity, and family history of diabetes. Case-control study nested within a community-based survey. We selected 151 subjects with NIDDM and 301 nondiabetic control subjects as a systematic sample of survey screening negative individuals. Odds ratios for NIDDM, comparing a high WHR (greater than or equal to 0.926 for men, greater than or equal to 0.83 for women) to a low WHR were 4.72 with a 95% confidence interval of 2.39-9.34, and 2.17 with a 95% confidence interval of 1.03-4.58, for women and men, respectively, controlling for age, overall obesity, and a family history of diabetes. Women with high WHRs in the presence of these risk factors are notably at risk for diabetes. Central obesity, as measured by the WHR, is importantly and independently associated with NIDDM.