OBJECTIVE: To identify sociodemographic and behavioral variables potentially associated with abdominal obesity in women attending a health care unit. METHODS: A total of 827 women aging from 12 to 65 years and living in the metropolitan area of Belo Horizonte, Brazil, were investigated using a cross-sectional design. Abdominal obesity was defined as waist circumference >88cm.The association between abdominal obesity and the co-variables was assessed by unconditional logistic regression. This technique estimates the odds ratio and the respective confidence intervals of 95%. RESULTS: The highest prevalence of abdominal obesity was observed in women aging >51 years (63.3%) and women with >5 deliveries (53.8%). The following variables remained associated with abdominal obesity in the final model: age (Odds Ratio varied from 2.53 to 27.64), schooling <8 years (Odds Ratio=1.49, confidence intervals 95%: 1.02 - 2.18), parity 1 - 4 deliveries (Odds Ratio=1.80, confidence intervals 95%: 1.04 - 3.12), >5 deliveries (Odds Ratio=3.20, confidence intervals 95%: 1.48 - 6.90) and light physical activity at work (Odds Ratio=1.63, confidence intervals 95%: 1.01 - 2.64). Age and body mass index were also included in the final model to control for confounding. CONCLUSION: Schooling and parity are important predictors of abdominal obesity in the studied group and studies on obesity should take them into account.
anthropometry; educational status; risk factors; women; obesity; parity
