Evaluation of the real-world implementation of the Family Nurse Partnership in England: an observational cohort study using linked data from health, education, and children's social care

Abstract

BackgroundThe Family Nurse Partnership (FNP) is an early home visiting service supporting young mothers. A randomised controlled trial of FNP in England found no effect on short-term primary outcomes or maltreatment in children up to age 7 years, but positive impacts on some educational outcomes. We report preliminary results of a national evaluation of FNP using linked administrative data. MethodsWe constructed a cohort of all mothers in England aged 13–19 years who gave birth between April 1, 2010, and March 31, 2019, to their firstborn child or children, using linked administrative data from hospital admissions (Hospital Episode Statistics) and education and social care (National Pupil Database). We evaluated differences in a range of policy relevant child and maternal outcomes, comparing mothers who were enrolled in FNP with those who were not, using propensity score matching. FindingsOf 110 960 mothers in our linked cohort, 26 290 (24%) were enrolled in FNP. FNP mothers were younger, more deprived, and more likely to have adversity or social care histories than mothers not enrolled. Compared with mothers not enrolled in FNP, those in FNP did not have fewer unplanned hospital admissions for injury or maltreatment in children by age 2 years, lower rates of children looked after in out-of-home care by age 7 years, or improved maternal outcomes, but were more likely to achieve a good level of development at school entry. We present findings among subgroups of younger maternal age (13–15 years), increased deprivation according to quintile of Index of Multiple Deprivation, and adversity and social care history. We also present sensitivity analyses that aim to minimise confounding. InterpretationOur study supports findings from previous trials of FNP showing little benefit for measured child maltreatment and maternal outcomes, but some evidence of benefit for school readiness. Interpretation of results needs careful consideration of the impact of residual confounding due to unmeasured or undisclosed factors (eg, family violence) linked to targeting of FNP to higher risk mothers, and surveillance bias. FundingNational Institute for Health and Care Research.