LAUSR

OBJECTIVES We investigate the association between child dietary diversity and anthropometric status and failure in Sri Lanka while accounting for other child and household factors by employing multivariable logistic and linear regression analyses. METHODS Using a nationally representative survey of children ages 6 mo to 59 mo, child dietary diversity was based on the 24-h recall of a child's food intake across seven food groups. The minimum dietary diversity was a score of four or above. Anthropometric status (i.e., height for age (HAZ), weight for age, and weight for height z-scores) and failure (i.e., stunting, wasting, and underweight) were calculated. RESULTS The prevalence of stunting, wasting, and underweight was 15%, 21%, and 26%, respectively. The prevalence of inadequate dietary diversity was 9%. Child dietary diversity was positively associated with HAZ (b = 0.02; standard error = 0.01; P = 0.04) but not with any indicator of anthropometric failure. However, low birth weight, wealth, and location were strong risk factors for anthropometric status and failure. Analyses stratified by child age indicated that dietary diversity was positively associated with anthropometric status for children ages 24 mo to 59 mo (HAZ: b = 0.03; standard error = 0.01; P = 0.02). Mixed associations were found for children ages 6 mo to 12 mo and there were no associations for children ages 12 mo to 24 mo. CONCLUSIONS Child dietary diversity predicted anthropometric status among children ages ≥24 mo. Interventions to address both proximal and distal risk factors for anthropometric status may be necessary in Sri Lanka to reduce anthropometric failure among infants and young children.