A non-randomised controlled trial of a pilot early intervention family support service designed to improve outcomes for children and families in Northern Ireland

Introduction:The concept of early intervention has received considerable support nationally and internationally, and particularly the relationship between early intervention, parenting and early childhood development. In Northern Ireland, a newly designed service, the Early Intervention Support Service (EISS), provides services to families with emergent problems. A 4-tiered model (the Hardiker model, 1991) is used to determine the level of family need. Targeted at tier 2 families, with one service based in each of the five Health and Social Care Trusts in Northern Ireland, and based on voluntary engagement, the EISS delivers services to families. Using mixed methods, the effectiveness and parental experiences of the service were evaluated. Method:The main trial employed a non-randomised wait-list controlled design. Each of the early intervention support services identified potential participants and providing them with information on the study. Those who consented became part of the study. Initially 109 families and their children were included. However, attrition resulted in 80 families completing pre-and post-test measures. Reliable and valid psychometric measures of family functioning, parental self-efficacy, parenting stress and assessment of child behaviour were used to assess outcomes. Pre-and post-test data were collected via telephone and directly uploaded to a secure server with an iPad. Main effects analysis was carried out using regression models to compare mean scores for intervention and control groups. A process evaluation was also undertaken alongside the main trial and analysed thematically to inform findings. Results: Findings from the process evaluation revealed that the service was well received by families. However, the quantitative analysis showed no overall effects.Conclusions:In exploring the lack of significant impact on outcomes, three main issues are discussed: service design and logic models; fidelity; and outcome measures and evidence. The paper explores these in detail and ideas for future service design and evaluation highlighted. KEYWORDSEarly intervention; family support; non-randomised controlled trial; logic model; fidelity; outcome measures