Hospital care and parent groups delivered measurable gains in two cluster trials
Across 16 hospitals in four African countries, the ALERT maternity-care package was associated with 22 per cent lower odds of early perinatal death, while the fresh-stillbirth difference was not significant. In wartime Ukraine, participants in the 12-session Hope Groups programme reported 52.7 per cent less violence against children than a waitlist group. That second result relies on self-report and cannot exclude expectancy effects.
Science··Evening
Early perinatal-death odds were lower in the ALERT clinical study
In a stepped-wedge cluster-randomised trial across 16 hospitals in Benin, Malawi, Tanzania and Uganda, the ALERT package was associated with 22 per cent lower odds of early perinatal death. The adjusted odds ratio was 0.78, with a 95 per cent confidence interval of 0.65 to 0.94. The outcome combines fresh stillbirth with newborn death within 24 hours of birth. Fresh stillbirth on its own produced a ratio of 0.88 and an interval from 0.68 to 1.13, so that difference was not significant. The five six-month periods covered 134,630 women and 139,300 newborns; the researchers list the inclusion of only 16 hospitals as a limitation.[1]
The Hope Groups trial recorded a measurable difference
A cluster-randomised trial following 510 Ukrainian parents and caregivers found that those who completed the 12-session Hope Groups programme reported 52.7 per cent less physical or emotional violence against children than the waitlist group. The same comparison recorded 57.4 per cent lower depression and anxiety symptoms and 33.1 per cent stronger protective parenting practices, with no harms or adverse events identified. Every outcome came from participant self-report. A waitlist design leaves open the possibility that expectations created by joining the programme account for part of the measured difference.[2]
Cluster designs tested institutions and groups
Both clinical studies assigned the intervention through hospitals or parent groups rather than person by person. ALERT combined co-design with local sites, training, quality-improvement support and leadership mentoring in one care package. The Hope Groups trial established that regular sessions could be delivered and health and parenting outcomes measured under difficult field conditions. Their percentages are not directly comparable because the outcomes differ. One co-primary outcome in the maternity trial was not significant, while self-report and a waitlist comparator set important limits on the size of the effect claimed for the parenting programme.[1], [2]
