The validity of a parent report motor scale for infants

Previously we reported poorly nourished infants had slower gross motor development as measured by trained workers’ observations. This study examined whether a parent report scale could also detect nutritional influences on motor skills in the same group of children. The sample was 780 Zanzibari infants 5–9 mo of age enrolled in a 12 mo randomized micronutrient trial. Trained workers documented infants’ motor development twice monthly with a picture scale; parents responded to verbal questions about their child's motor skills every 3 mo. Convergent validity of the parent report scale was assessed with Chronbach's α and by correlating observers’ and parents’ reports of the child's most advanced motor skill. At baseline, the mean (SD) age (mo) was 12.03 (4.01), 82% were anemic and/or iron deficient (ID) (Hb < 100 g/L and/or ZPP ≥μmol/mol heme) and 35% were stunted. The convergent validity was good (α = .92 at baseline; α = .85 at study end) and the correlations for observer and parent reported scores were high (r = .90 at baseline; r = .69 at study end). A mulitvariate logistic regression model (adjusted for age, sex and SES) showed HAZ increased (ß = 1.92, p. < 0.000) and anemia/ID decreased (ß = 0.42, p. = 0.014) the odds of reported walking. These coefficients were very similar for observed walking (ß = 2.07, p. < 0.000 for HAZ; ß = 0.34, p. = 0.001 for anemia/ID). We conclude that the parent report scale was valid and capable of detecting the effects of nutritional status on motor development in infants. Funded by Gates Fn & USAID Ofc of Hlth & Nut.