A randomized clinical trial of home intervention for children with failure to thrive

Abstract

Objective. To evaluate the efficacy of a home-based intervention on the growth and development of children with nonorganic failure to thrive (NOFTT). Design. Randomized clinical trial. Participants. The NOFTT sample included 130 children (mean age, 12.7 months; SD, 6.4) recruited from urban pediatric primary care clinics serving low income families. All children were younger than 25 months with weight for age below the fifth percentile. Eligibility criteria included gestational age of at least 36 weeks, birth weight appropriate for gestational age, and no significant history of perinatal complications, congenital disorders, chronic illnesses, or developmental disabilities. Children were randomized into two groups: clinic plus home intervention (HI) (n = 64) or clinic only (n = 66). There were no group differences in children's age, gender, race, or growth parameters, or on any of the family background variables. Most children were raised by single, African-American mothers who received public assistance. Eighty-nine percent of the families (116 of 130) completed the 1-year evaluation. Interventions. All children received services in a multidisciplinary growth and nutrition clinic. A community-based agency provided the home intervention. Families in the HI group were scheduled to receive weekly home visits for 1 year by lay home visitors, supervised by a community health nurse. The intervention provided maternal support and promoted parenting, child development, use of informal and formal resources, and parent advocacy. Measurements. Growth was measured by standard procedures and converted to z scores for weight for height and height for age to assess wasting and stunting. Cognitive and motor development were measured with the Bayley Scales of Infant Development, and language development was measured by the Receptive/Expressive Emergent Language Scale. Both scales were administered at recruitment and at the 12- month follow-up. Parent-child interaction was measured by observing mothers and children during feeding at recruitment and at the 12-month follow-up, and the quality of the home was measured by the Home Observation Measure of the Environment 18 months after recruitment. Analyses. Repeated-measures multivariate analyses of covariance were used to examine changes in children's growth and development and parent-child interaction. Analyses of covariance were used to examine the quality of the home. Independent variables were intervention status and age at recruitment (1.0 to 12.0 vs 12.1 to 24.9 months). Maternal education was a covariate in all analyses. When changes in developmental status and parent-child interaction were examined, weight for height and height for age at recruitment were included as covariates. Results. Children's weight for age, weight for height, and height for age improved significantly during the 12-month study period, regardless of intervention status. Children in the HI group had better receptive language over time and more child-oriented home environments than children in the clinic-only group. The impact of intervention status on cognitive development varied as a function of children's ages at recruitment, with younger children showing beneficial effects of home intervention. There were no changes in motor development associated with intervention status. During the study period, children gained skills in interactive competence during feeding, and their parents became more controlling during feeding, but differences were not associated with intervention status. Conclusions. Findings support a cautious optimism regarding home intervention during the first year of life provided by trained lay home visitors. Early home intervention can promote a nurturant home environment effectively and can reduce the developmental delays often experienced by low income, urban infants with NOFTT. Subsequent investigations of home intervention should consider alternative options for toddlers with NOFTT.