Citation: Barnes J, Stuart J, Allen E, Petrou S, Sturgess J, Barlow J, et al. Results of the First Steps study: a randomised controlled trial and economic evaluation of the Group Family Nurse Partnership (gFNP) programme compared with usual care in improving outcomes for high-risk mothers and their children and preventing abuse.Public Health Res 2017;5(9)


Family Nurse Partnership (FNP) is a home-based nurse home-visiting programme to support vulnerable parents. Group Family Nurse Partnership (gFNP) has similar aims and materials and was demonstrated to be feasible in implementation evaluations.


To determine whether or not gFNP, compared with usual care, could reduce risk factors for maltreatment in a vulnerable group and be cost-effective.


A multisite randomised controlled parallel-group trial and prospective economic evaluation, with eligible women allocated (minimised by site and maternal age group) to gFNP or usual care.


Community locations in the UK.


Expectant mothers aged


Forty-four sessions of gFNP (14 during pregnancy and 30 in the first 12 months after birth) were offered to groups of between 8 and 12 women with similar expected delivery dates (the difference between the earliest and latest expected delivery date ranged from 8 to 10 weeks depending on the group) by two family nurses (FNs), one of whom had notified her intention to practise as a midwife.

Main outcome measures

Parenting was assessed by a self-report measure of parenting opinions, the Adult Adolescent Parenting Inventory Version 2 (AAPI-2), and an objective measure of maternal sensitivity, the CARE-Index. Cost-effectiveness was primarily expressed in terms of incremental cost per quality-adjusted life-year (QALY) gained.

Data sources

Interviews with participants at baseline and when infants were aged 2, 6 and 12 months. Cost information from nurse weekly logs and other service delivery data.


In total, 166 women were enrolled (99 to the intervention group and 67 to the control group). Adjusting for site and maternal age group, the intention-to-treat analysis found no effect of gFNP on either of the primary outcomes. AAPI-2 total was 7.5/10 [standard error (SE) 0.1] in both arms [difference also adjusted for baseline 0.08, 95% confidence interval (CI) –0.15 to 0.28; p?=?0.50]. CARE-Index maternal sensitivity mean: intervention 4.0 (SE 0.3); control 4.7 (SE 0.4) (difference –0.76, 95% CI –1.67 to 0.13; p?=?0.21). The sensitivity analyses supported the primary analyses. The probability that the gFNP intervention was cost-effective based on the QALY measure did not exceed 3%. However, in terms of change in AAPI-2 score (baseline to 12 months), the probability that gFNP was cost-effective reached 25.1%. A separate discrete choice experiment highlighted the value placed by both pregnant women and members of the general population on non-health outcomes that were not included in the QALY metric.


Slow recruitment resulted in smaller than ideal group sizes. In some cases, few or no sessions took place owing to low initial group size, and small groups may have contributed to attrition from the intervention. Exposure to gFNP sessions was below maximum for most group members, with only 58 of the 97 intervention participants receiving any sessions; FNs were experienced with FNP but were mainly new to delivering gFNP.


The trial does not support the delivery of gFNP as a means of reducing the risk of child abuse or neglect in this population.

Future work

A randomised controlled trial with modified eligibility to enable first-time mothers aged

Trial registration

Current Controlled Trials ISRCTN78814904.