Background Provision of home safety education and equipment prevents home injuries in children under 5 in socio-economically disadvantaged communities. We co-produced a low-cost home safety intervention (Stay-One-Step-Ahead, ‘SOSA’) with parents and health and early years family practitioners in a UK city, trained 142 practitioners in its use, and evaluated implementation fidelity over 2 years. SOSA comprised a home safety checklist used in routine child health reviews, monthly safety messages, safety activities and safety weeks.

Methods Implementation fidelity was assessed by triangulating data from parent and practitioner questionnaires, interviews and observations of parent/practitioner interactions. Quantitative data were analysed using logistic regression and thematic analysis was used for qualitative data.

Results Overall, 361 intervention and 401 control parents participated in the evaluation. Significantly more intervention parents reported receiving home safety advice (Odds Ratio (OR) 2.59 (95% CI 1.59–4.21), advice from ≥2 practitioners (OR 5.09; 1.34–19.33) and home safety leaflets (OR 1.90; 1.11–3.23). More intervention practitioners reported signposting parents to safety agencies (Intervention: 92%; Control: 20%). Whilst 4,859 home safety checklists were completed (86% of reviews), they were often not used as intended. Safety weeks tended to be shorter than planned. Monthly safety message use varied by type of practitioner from 38%-75%. Home safety activities were undertaken in all 22 observed contacts.

Conclusions The SOSA intervention increased home safety advice received by parents but aspects of the intervention were often modified.

Learning outcomes Monitoring intervention fidelity is important when interpreting programme effectiveness results. Strategies to increase fidelity should be integral to intervention design.