Civil Society Organization–Led Home Visit Intervention to Reduce Unintentional Child Injury in Rural China: A Cluster Randomized Clinical Trial

Importance Unintentional injuries are a leading cause of childhood death and disability, but evidence on scalable home visit interventions in rural settings is limited. Objective To evaluate whether a civil society organization–led home visit intervention can reduce unintentional injuries among children and improve caregiver and household safety. Design, Setting, and Participants This outcome assessor–blinded cluster randomized clinical trial was conducted in 24 rural communities in Guizhou and Shaanxi, China, from May 6 to October 1, 2025. After baseline assessment, communities were randomized as clusters in a 1:1 ratio. Participants were households with children aged 3 to 12 years. Interventions Control communities received routine social worker home visits; intervention communities received routine services plus 3 monthly structured visits with home safety and first aid tool kits and a follow-up visit at month 4. Main Outcomes and Measures The primary outcome was caregiver-reported cumulative child unintentional injury incidence during 4 months, a subjective outcome with caregivers unblinded to group allocation. Secondary outcomes were caregiver knowledge, safety attitudes, and supervisory behaviors, with scores ranging from 0 to 40, 0 to 8, and 0 to 80, respectively, with higher scores indicating better knowledge, more positive safety attitudes, and more appropriate supervisory behaviors, respectively, and household environmental hazards, with scores ranging from 0 to 230 and higher scores indicating more household environmental hazards and a less safe home environment. Generalized estimating equations accounted for clustering. Results Among 24 randomized clusters, caregivers (mean [SD] age, 45.4 [13.0] years; 271 [75.3%] female) and children (mean [SD] age, 8.6 [2.3] years, 198 [55.0%] female) from 360 households were enrolled, with 180 each randomized to intervention and control groups. After baseline, 9 households withdrew with missing outcome data; all 360 households were included in the intention-to-treat analyses with multiple imputation. Injury occurred in 21 of 178 households (11.8%) with observed outcomes in the intervention group and 33 of 173 (19.1%) in the control group. The intervention was associated with lower injury odds (adjusted odds ratio, 0.36 [95% CI, 0.18-0.73]; P = .005). At month 4, adjusted mean differences favored the intervention for supervisory behaviors (17.08 [95% CI, 14.71-19.46] points) and household environmental hazards (−36.66 [95% CI, −42.87 to −30.46] points). Conclusions and Relevance In this cluster randomized clinical trial of households with children aged 3 to 12 years, a civil society organization–led home visit intervention reduced caregiver-reported unintentional injuries and improved caregiver supervisory behaviors and household safety during the 4-month intervention period. Longer follow-up is needed to assess durability of the response. Trial Registration Chinese Clinical Trial Registry Identifier: ChiCTR2500098928