OBJECTIVES:

To determine whether there are racial/ethnic differences in the prevalence of obesity among preschool children and to determine whether these differences are explained by socioeconomic factors.

DESIGN:

Cross-sectional assessment.

SETTING:

Twenty large US cities, from 2001 to 2003.

PARTICIPANTS:

Of the 4898 children enrolled at birth in the Fragile Families and Child Well-being Study, we analyzed data for the 2452 who, at the age of 3 years, had their height and weight measured during a maternal survey.

MAIN EXPOSURES:

Three socioeconomic indicators were the main exposures-maternal education, household income, and children's food security status, as assessed by the US Household Food Security Survey Module.

MAIN OUTCOME MEASURE:

Obesity, defined as a body mass index at the 95th percentile or higher for age and sex.

RESULTS:

Of the mothers, 41.0% had education beyond high school, 52.9% of households had an income above the federal poverty threshold, and 79.5% of the children were food secure. The prevalence of obesity was 25.8% among Hispanics (any race), 16.2% among blacks, and 14.8% among whites. Compared with whites, the odds of obesity were significantly higher in Hispanics (odds ratio, 2.00; 95% confidence interval, 1.46-2.73), but not in blacks (odds ratio, 1.10; 95% confidence interval, 0.82-1.48). Neither of these odds ratios changed meaningfully after adjusting for all 3 socioeconomic indicators (Hispanics: odds ratio, 1.86 [95% confidence interval, 1.33-2.60]; and blacks: odds ratio, 1.07 [95% confidence interval, 0.78-1.47]).

CONCLUSION:

In a sample of preschool children drawn from 20 large US cities, the high prevalence of obesity among Hispanics relative to blacks or whites was not explained by racial/ethnic differences in maternal education, household income, or food security.