This paper presents the methods of using routinely collected health, education and social care data to evaluate the Family Nurse Partnership (FNP) in Scotland using a natural experiment methodology (case-cohort design). FNP is a licensed US programme which offers intensive, structured home visiting support to first-time young mothers delivered by a specially trained nurse, from early pregnancy until the child’s second birthday. The evaluation was commissioned by the Scottish Government.
The evaluation is structured around a well-specified programme logic model and the availability of routine data mapping onto key short, medium and long-term outcomes. The overarching aim of the evaluation is to compare first time young mothers enrolled and receiving the FNP programme (FNP Clients) to a comparable group of Controls over a range of outcomes covering maternal health, child health and development, and parental life course. Delivering the evaluation depends upon approval and successful linkage to routinely collected electronic health, education and social care records to evaluate a complex intervention already available in the population.
We have so far established the required model of data linkage to routine Scottish data in order to evaluate FNP using a natural experiment. Approvals were obtained to access data on over 3,000 FNP Clients and these cases were mapped to routine health data with little loss of records. The quality of the data used to identify the potential control sample was found to be high and had limited missingness. Over 5,000 Controls were successfully identified using routine health data. Both FNP Clients and Controls were linked to health data to characterise these young mothers.
The assessment of effectiveness in the evaluation is necessarily limited to outcomes available from routinely collected data. Study outcomes have been selected by matching routinely collected administrative data to the three main aims of the Scottish FNP logic model, which is based on the underlying programme theory. These aims are 1) to improve pregnancy outcomes, 2) to improve child health and development, and 3) to improve parents’ economic self-sufficiency. The logic model translates these aims into outcomes, which were then matched to the routine data. This resulted in matches for about 50% of the outcomes detailed within the model. Therefore, the included outcomes have been selected on the basis that they are outcomes FNP aims to influence and for which there is routine data, rather than a set of specific outcomes where research indicates the most significant contribution.
This report therefore, describes the principal study objectives, the broad evaluation design including the detailed data flows, the required governance approvals and the identification of both cases (FNP Clients) and the control group of young mothers. It describes the identification of study outcomes against available data sources and how they will be regarded in analysis. The proposed analysis is described in the main body of the report and the detailed statistical analysis plan is provided in an appendix. Progress in establishing the study database and descriptive summary statistics for the study population are included in this report. Full results will be reported in mid-2020.