Seventy-five local authorities across England are rolling out the government's Start for Life programme—a package of perinatal mental health support, breastfeeding services, and parent-infant bonding interventions—and this study will find out which parts actually work, for which families, and under what conditions. The problem is that Start for Life was launched rapidly with little evidence about what combination of services delivers the best outcomes. Without knowing which interventions drive impact, councils risk spending limited funds on approaches that don't help families. The study tracks 60 families in depth, surveys thousands more, and compares areas with and without the programme using NHS data on depression, anxiety, and hospital visits. If successful, this evaluation will tell policymakers exactly which services to fund and which to drop—saving public money while improving care for new parents. The findings will shape how 75 local authorities and others beyond them design perinatal support, potentially improving outcomes for tens of thousands of families each year without requiring new spending.
View original technical description
The proposed study will use a mixed-method multilevel triangulation design to assess the impact and effectiveness of SfL, and to determine which interventions appear to be having an impact, for whom, and in what contexts. The research comprises three workstreams with PPI involvement embedded across all workstreams. Workstream 1 (WS1) consists of a Process Evaluation, the aim of which is to understand the implementation of SfL, how this varies across contexts, and what works for who, where, and why. It has two components: i) Service level evaluation, which will involve the collection of aggregate management information data, surveys and interviews across all 75 LAs, alongside 12 deep dives in selected LAs at different stages of implementation; ii) Individual level evaluation will consist of a longitudinal case series study of 60 families across four LAs. Workstream 2 (WS2) involves an Impact Evaluation that aims to estimate the causal impact of SfL using a Synthetic (weighted) Control Method, and has two strands: i) Analysis of national aggregate (e.g. Fingertips OHID; Child and Maternal Health Data) and local individual and aggregate level datasets (e.g. Community Services Data Set; Maternity Services Data Set; Mental Health Services Data Set; Hospital Episode Statistics). All datasets, held by NHS Digital, will be linkable via an NHS identifier. We will also explore the possibility of accessing data from the Family Nurse Partnership Information System, and LA-level aggregates from a primary care data source, such as the RCGP data. ii) Analysis of local individual level data will be collected from the 12 deep dive LA s participating in workstream 1. Pre- and post-intervention data will be collected from families receiving more intensive perinatal mental health services, utilising a range of brief, standardised outcome measures of depression, anxiety, and parent-infant bonding. Counterfactual data will involve collection of data from sites not yet implementing SfL; or the use of wait-list controls. A brief cross-sectional survey of breastfeeding service user experience will be carried out in the same LA s. Future value for money evaluation will be supported through the collection of appropriate service-level data and through work with DHSC to put systems in place so that service use data being collected is fit for purpose. Workstream 3 consists of synthesising the data across the workstreams. Service level process data from WS1 will be combined with population impact data from WS2 to identify the configuration of services associated with the most impact, and to elicit best practices that can be applied across the 75 funded areas and beyond. The WS1 service level data will be used to identify the factors that drive and shape service delivery, and WS1 individual/family level data will be used to identify the acceptability of services to families, and stakeholder views regarding factors that affect uptake/impact. Finally, a realist synthesis will be used to test the overall Theory of Change for the SfL programme, drawing on all data sources to explain the observed impacts at programme and strand levels, and addressing synergies between SfL and Family Hubs.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know