Hospital admissions associated with body mass index in Canadian adults

Objective: To examine the associations between excess weight and health-care utilization in general Canadian population. Design: A national survey with a multistage stratified sampling design. Subjects: A total of 113 603 adults 20 years of age or more in 2003. Measurements: Self-reported overnight stay as an in-patient during the past 12 months, length of stay, height and weight. Results: The 12-month cumulative incidence of being an in-patient was 6.6% for men and 10.4% for women, with the length of stay in median of 3 nights for the patients. The adjusted odds ratio for obesity associated with being an in-patient was 1.24 (95% confidence interval: 1.109, 1.40) for men and 1.25 (1.14, 1.37) for women. The association between obesity and being an in-patient was significant or marginally significant in all categories of length of stay with an exception of the '1 night' category for men. Underweight men and women had an increased risk of approximately 30%. Underweight was only significantly associated with being an in-patient with a stay of '47 nights' in both sexes, and only small proportion of participants were underweight. Obesity contributed slightly more than 4% of being an in-patient in the Canadian adults. The population attributable fraction was very small for underweight because of low proportion of cases in that category. Conclusion: Obesity burdens the universal health-care system in Canada.