Efficacy of delivery models in early intervention: Findings from Germany's nurse-family partnership on family and child welfare services and pediatric medical incidents in high-risk families

Background

This study examines the efficacy of two different delivery models of the ProKind intervention, a German adaptation of the Nurse-Family Partnership (NFP): The continuous model (CM) with visits exclusively delivered by NFP-trained midwives, and a tandem model (TM) with visits delivered by a midwife during pregnancy and a social worker postnatally.

Objective

To evaluate the effectiveness of different delivery models in the ProKind intervention on reducing surveillance by the child protection service (CPS), foster care placements, and pediatric medical incidents.

Participants and setting

755 pregnant disadvantaged women were randomized into intervention (n = 394) and control groups (n = 361).

Methods

Longitudinal multicenter RCT with two follow-up phases (phase I: birth to 36 months, phase II: approx. 7th birthday). Average treatment effects were estimated using augmented inverse probability weighting (AIPW) in combination with lasso to select relevant control variables.

Results

The CM significantly reduces hospital visits across both phases (ATE = −7.20, p < .001). Trends indicated reduced surveillance by the CPS (ATE = −5.24 PP, p = .10) and foster care placements (ATE = −2.25 PP, p = .09) in phase II. No significant differences were found for the TM for any outcome. The full sample analysis showed significant differences in hospital visits (ATE = −3.58 P < .005) in phase II.

Conclusions

The CM reduces hospital visits and shows potential in lowering the need for surveillance by the CPS and foster care placements. These findings highlight the importance of delivery models in home visitation programs and suggest that midwife-led interventions can achieve sustained positive outcomes for high-risk families.