PolicyLab

To compare hospital-based utilization for early childhood injuries between program recipients and local-area comparison families following statewide implementation of an evidence-based home visitation program, and to describe site-level program variation. Propensity score matching on baseline characteristics was used to create a retrospective cohort of Nurse-Family Partnership (NFP) clients and local area matched comparison women. The main outcome, a count of injury visit episodes, was enumerated from Medicaid claims for injuries examined in an emergency department or hospital setting during the first 2 years of life of children born to included subjects. Generalized linear models with a Poisson distribution examined the association between injury episode counts and NFP participation, controlling for other non-injury utilization and stratifying by individual agency catchment area in a fixed effects analysis. The children of NFP clients were more likely in aggregate to have higher rates of injury visits in the first 2 years of life than the children of comparison women (415.2/1,000 vs. 364.2/1,000, P < 0.0001). Significantly higher rates of visits among children of NFP clients for superficial injuries (156.6/1,000 vs. 132.6/1,000, P < 0.0001) principally accounted for the attributable difference in injury visit rates between groups. Among more serious injuries, no significant difference in injury visit rates was found between NFP clients and comparison women. The proportion of children with at least one injury visit varied from 14.5 to 42.5% among individual sites. Contrary to prior randomized trial data, no reductions in utilization for serious early childhood injuries were demonstrated following statewide implementation of an evidence-based home visitation program. Significant program variation on outcomes underscores the challenges to successful implementation.