RESUMO
OBJECTIVE: An innovative low-cost parenting intervention, implemented through health services in Jamaica showed benefits to children's cognitive development at 18 months and parent's attitudes concerning childcare. We assessed the impact of the intervention on child and parent outcomes at 6 years of age. METHODS: A cluster randomized trial of 2 parenting interventions was conducted through 20 health centers in Jamaica. Interventions were implemented from age 3 to 18 months and each intervention benefited cognitive development at 18 months (effect size 0.34-0.38 standard deviation). Children were reassessed at 6 years (n = 262, 80.1% of those assessed at 18 months) to determine any benefits to cognition, behavior, and parenting behavior. Loss to follow-up was not significantly different by treatment. Inverse probability weighting and Lee bounds were used to adjust for loss to follow-up, and multilevel regression analyses conducted with random effects at the health center level. RESULTS: There were no significant benefits to any child outcomes at age 6 years or to parenting behavior. Results are robust using the wild cluster bootstrap procedure and using Lee bounds for attrition. The initial trial benefits were reproduced with the current sample and methods. CONCLUSION: Lack of sustained benefits may be related to the initial effect size and low intensity of the intervention that ended very young at age 18 months. It may also be related to lack of initial impact on home environment and fade-out of effects in a country with near universal preschool. The findings have implications for intervention design and targeting.