Racial/ethnic disparities in the diagnosis of otitis media in infancy

Objective: Otitis media (OM) is an extremely common pediatric diagnosis. Several risk factors have been associated with OM, but the relationship between OM and race/ethnicity remains controversial. We sought to define the relationship between OM diagnosis and race/ethnicity in infants. Methods: By multivariable logistic regression, we evaluated the association between OM diagnosis and race/ethnicity in 11,349 non-low-birthweight infants who were participants in a prospective cohort study of infant care practices. Results: As in previous studies, breastfeeding was associated with a decreased risk of OM diagnosis while other factors were independently associated with a substantially increased risk of OM diagnosis: out-of-home daycare, multiple children living in the home, and mother's multiparity. Daycare was associated with a "dose effect" in that the risk of OM diagnosis increased with an increasing number of children in the daycare. While the crude analysis suggested little relation of OM diagnosis and race/ethnicity, the association was confounded by several covariates including maternal marital status, number of children living in the home, breastfeeding status, and maternal age. After adjustment for relevant confounders, Black (OR 0.74; 95% CI 0.61-0.89) and Asian infants (OR 0.77; 95% CI 0.57-1.0]) were less likely to be diagnosed with OM than White infants. Conclusions: This large prospective study confirms previous risk factors for OM and demonstrates a strong "dose effect" of the size of daycare centers on OM. The study also demonstrates that the association between race/ethnicity and OM diagnosis is confounded by social factors. After adjusting for such factors, Black and Asian infants are less likely to be diagnosed with OM than White infants. The reason for this racial disparity remains unknown.