Home-based intervention for non-Hispanic black families finds no significant difference in infant size or growth: results from the Mothers & Others randomized controlled trial.

ABSTRACT

BACKGROUND: Non-Hispanic black (NHB) infants are twice as likely as non-Hispanic white infants to experience rapid weight gain in the first 6 months, yet few trials have targeted this population. The current study tests the efficacy of "Mothers & Others," a home-based intervention for NHB women and their study partners versus an attention-control, on infant size and growth between birth and 15 months. METHODS: Mothers & Others was a two-group randomized controlled trial conducted between November 2013 and December 2017 with enrollment at 28-weeks pregnancy and follow-up at 3-, 6-, 9-, 12-, and 15-months postpartum. Eligible women self-identified as NHB, English-speaking, and 18-39 years. The obesity prevention group (OPG) received anticipatory guidance (AG) on responsive feeding and care practices and identified a study partner, who was encouraged to attend home visits. The injury prevention group (IPG) received AG on child safety and IPG partners only completed study assessments. The primary delivery channel for both groups was six home visits by a peer educator (PE). The planned primary outcome was mean weight-for-length z-score. Given significant differences between groups in length-for-age z-scores, infant weight-for-age z-score (WAZ) was used in the current study. A linear mixed model, using an Intent-To-Treat (ITT) data set, tested differences in WAZ trajectories between the two treatment groups. A non-ITT mixed model tested for differences by dose received. RESULTS: Approximately 1575 women were screened for eligibility and 430 were enrolled. Women were 25.7 ± 5.3 years, mostly single (72.3%), and receiving Medicaid (74.4%). OPG infants demonstrated lower WAZ than IPG infants at all time points, but differences were not statistically significant (WAZdiff = - 0.07, 95% CI - 0.40 to 0.25, p = 0.659). In non-ITT models, infants in the upper end of the WAZ distribution at birth demonstrated incremental reductions in WAZ for each home visit completed, but the overall test of the interaction was not significant (F2,170 = 1.41, p = 0.25). CONCLUSIONS: Despite rich preliminary data and a strong conceptual model, Mothers & Others did not produce significant differences in infant growth. Results suggest a positive impact of peer support in both groups. TRIAL REGISTRATION ClinicalTrials.gov, NCT01938118 , 09/10/2013.