Home-based Intervention Among Non-Hispanic Black Families Finds No Significant Difference in Growth at 15 Months: Results from the ‘Mothers & Others’ Randomized Trial (P04-187-19)

Abstract

ObjectivesMothers and Others (M&O) was an efficacy trial of a home-based intervention designed to prevent obesity in the first year of life. The primary outcome was infant growth, as assessed by differences in weight-for-age z-scores (WAZ) between the obesity prevention group and an attention-control group on infant safety. MethodsThe study design was a two-group randomized control trial among 428 non-Hispanic black (NHB) women recruited at 28 weeks’ pregnancy. The primary delivery channel for both groups was 6 home visits by a peer educator (PE). Mothers in the intervention group received anticipatory guidance (AG) on infant feeding and care behaviors related to obesity prevention. Mothers in the attention-control group received AG on child safety. Mothers in both groups identified a study partner, with partners in the control group only completing study assessments. Infant weight was assessed at birth by maternal self-report and at 3, 6, 9, 12 and 15 months of age by direct measurement. Infant WAZ scores were created using the World Health Organization 2006 international growth standards. The primary efficacy analysis was a linear mixed model (LMM) on an intention-to-treat (ITT) dataset with WAZ score at birth, 3, 6, 9, 12 and 15 months as the dependent variable and treatment group, age and their interaction term as the independent variables. ResultsEnrolled women were 25.76 ± 5.3 years at baseline and the majority were single (72.3%), receiving Medicaid (74.4%), and expecting their first child (56.1%). Baseline characteristics and visit completion rates did not differ by treatment group. At all time points, infants in the intervention group were smaller than those in the control group (e.g., mean WAZ at 15 months was 0.39 ± 1.04 among intervention infants and 0.53 ± 1.07 among control infants), but these differences were not significant. Results of the primary efficacy model yielded no difference in WAZ between infants in the two groups (beta for intervention = −.07, P = 0.659). ConclusionsM&O was designed to fill several gaps in early life obesity prevention trials at the time of its inception, including a focus on infant temperament and inclusion of support partners. Despite rich preliminary data and a strong conceptual model, the intervention did not produce a significant difference in infant growth. Funding SourcesNational Institute of Diabetes and Digestive and Kidney Diseases.