Group antenatal care compared with standard antenatal care for Somali-Swedish women: a historically controlled evaluation of the Hooyo Project.

Abstract

Comparing language-supported group antenatal care (gANC) and standard antenatal care (sANC) for Somali-born women in Sweden, measuring overall ratings of care and emotional well-being, and testing the feasibility of the outcome measures. A quasi-experimental trial with one intervention and one historical control group, nested in an intervention development and feasibility study. Midwifery-led antenatal care clinic in a mid-sized Swedish town. Pregnant Somali-born women (<25 gestational weeks); 64 women in gANC and 81 in sANC. Language-supported gANC (2017-2019). Participants were offered seven 60-minute group sessions with other Somali-born women led by one to two midwives, in addition to 15-30 min individual appointments with their designated midwife. Primary outcomes were women's overall ratings of antenatal care and emotional well-being (Edinburgh Postnatal Depression Scale (EPDS)) in gestational week ≥35 and 2 months post partum. Secondary outcomes were specific care experiences, information received, social support, knowledge of pregnancy danger signs and obstetric outcomes. Recruitment and retention of participants were challenging. Of eligible women, 39.3% (n=106) declined to participate. No relevant differences regarding overall ratings of antenatal care between the groups were detected (late pregnancy OR 1.42, 95% CI 0.50 to 4.16 and 6-8 weeks post partum OR 2.71, 95% CI 0.88 to 9.41). The reduction in mean EPDS score was greater in the intervention group when adjusting for differences at baseline (mean difference -1.89; 95% CI -3.73 to -0.07). Women in gANC were happier with received pregnancy and birth information, for example, caesarean section where 94.9% (n=37) believed the information was sufficient compared with 17.5% (n=7) in standard care (p<0.001) in late pregnancy. This evaluation suggests potential for language-supported gANC to improve knowledge acquisition among pregnant Somali-born women with residence in Sweden ˂10 years. An adequately powered randomised trial is needed to evaluate the effectiveness of the intervention. ClinicalTrials.gov Registry (NCT03879200).