GENETIC AND BEHAVIORAL DETERMINANTS OF WAIST-HIP RATIO AND WAIST CIRCUMFERENCE IN WOMEN TWINS

Objectives: This study examines genetic and behavioral determinants of waist-hip ratio (WHR) and waist circumference among women twins. Rese arch Methods and Procedures: Six hundred eighty participants from the second examination of the Kaiser-Permanente Women Twins Study (1989 to 1990) were included. Women ranged in age from 31 to 90 years, and inc luded 59% monozygotic and 41% dizygotic twins. Heritabilities of WHR a nd waist circumference were estimated (range = 0 to 1) using three dif ferent statistical methods. Linear regression models that adjusted for the lack of independence within twin pairs were used to assess associ ations between behavioral variables: WHR, and waist circumference. Res ults: Age and body mass index-adjusted heritability estimates ranged f rom 0.36 to 0.61 for WHR and 0.72 to 0.82 for waist circumference. Whe n considered individually, after adjusting for age and body mass index , all behavioral characteristics, except calorie-adjusted fat intake, were significantly (p<0.10) associated with waist circumference and WH R. Greater central adiposity was associated with lower education, high er alcohol consumption, lower physical activity, current cigarette smo king, higher parity, and post menopausal status without hormone replac ement therapy. In multivariate models, these associations persisted, e xcept neither educational attainment nor alcohol intake was significan tly associated with waist circumference. In longitudinal analyses, bot h measures were positively associated with current or recent-past smok ing; infrequent or inconsistent physical activity; and long-term, dail y consumption of alcohol. Discussion: These cross-sectional and longit udinal associations are consistent with genetic and behavioral predict ors of waist circumference and WHR. Whereas the evidence for genetic i nfluences is stronger for waist circumference, both body fat measures may be similarly influenced by the behavioral factors considered.