Characteristics of intervention studies on family-centred care in neonatal intensive care units: a scoping review of randomised controlled trials.

Background: Different definitions of family-centred care (FCC) exist in the newborn setting, and many FCC interventions have been tested, while a comprehensive review synthesising characteristics of existing intervention studies is still lacking.

Objective: This review aims at summarising the characteristics of randomised controlled trials (RCTs) on FCC interventions in neonatal intensive care units.

Methods: We searched PubMed, Embase, Web of Science and the Cochrane Library up to 31 January 2022, and reference lists of included studies and other reviews. Interventions were grouped into five categories according to a previous Cochrane review: (1) family support, (2) educational, (3) communication, (4) environmental interventions and (5) family-centred policies. Subgroup analyses by time period (RCTs published before vs after 2016) and by country income (based on the World Bank Classification) were conducted.

Results: Out of 6583 retrieved studies, 146 RCTs met the eligibility criteria, with 53 (36.3%) RCTs published after 2016. Overall, 118 (80.8%) RCTs were conducted in high-income countries, 28 (19.1%) in middle-income countries and none in low-income countries. Only two RCTs were multicountry. Although mothers were the most frequent caregiver involved, fathers were included in 41 RCTs (28.1%). Very few studies were conducted in at-term babies (nine RCTs); siblings (two RCTs) and other family members (two RCTs), maternity care units (two RCTs). The role of health professionals was unclear in 65 (44.5%) RCTs. A large variety of intervention combinations was tested, with 52 (35.6%) RCTs testing more than 1 category of interventions, and 24 (16.4%) RCTs including all 5 categories.

Conclusion: There is a large and rising number of RCTs on FCC interventions in neonatal intensive care units, with specific research gaps. The large variety of FCC interventions, their high complexity, the need to tailor them to the local context and major gaps in implementation suggest that implementation research is the current priority.