Long-term impact of community-based participatory women's groups on child and maternal mortality and child disability: follow-up of a cluster randomised trial in rural Nepal.

BACKGROUND: Community-based women's groups practising participatory learning and action (PLA) can reduce maternal and neonatal mortality in low-income countries. However, it is not clear whether these reductions are associated with subsequent increased or decreased rates of childhood death and disability. We assessed the impact on child deaths and disability beyond the perinatal period among participants in the earliest trial in Nepal 2001-2003. METHODS: Household interviews were conducted with mothers or household heads. At cluster and individual levels, we analysed disability using pairwise log relative risks and survival using multilevel logistic models. FINDINGS: From 6075 children and 6117 mothers alive at 4 weeks post partum, 44 419 children (73%) were available for interview a mean 11.5 years later. Rates of child deaths beyond the perinatal period were 36.6 and 52.0 per 1000 children in the intervention and control arms respectively. Rates of disability were 62.7 and 85.5 per 1000 children in the intervention and control arms respectively. Individual-level analysis, including random effects for cluster pairing and adjusted for baseline maternal literacy, socioeconomic status and maternal age, showed lower, statistically non-significant, odds of child deaths (OR 0.70 (95% CI 0.43 to 1.18) and disability (0.64 (0.39 to 1.06)) in the intervention arm. CONCLUSION: Community-level exposure to women's groups practising PLA did not significantly impact childhood death or disability or death beyond the perinatal period. Follow-up of other trials with larger sample sizes is warranted in order to explore the possibility of potential long-term survival and disability benefits with greater precision.

Prenatal and perinatal risk factors for disability in a rural Nepali birth cohort.

BACKGROUND: Improving newborn health remains a global health priority. Little however is known about the neurodevelopmental consequences for survivors of complications in pregnancy, labour and the neonatal period in in low-income countries outside of small selective and typically urban facility studies. We ask which antenatal, birth and neonatal factors are associated with disability in childhood in a large community birth cohort from rural Nepal. METHODS: 6436 infants were recruited during a cluster randomised control trial (RCT) of participatory women's groups (ISRCTN31137309), of whom 6075 survived beyond 28 days. At mean age of 11∙5 years (range 9.5-13.1), 4219 children (27% lost to follow-up) were available for disability screening which was conducted by face-to-face interview using the Module on Child Functioning and Disability produced by the Washington Group/UNICEF. Hypothesised risk factors for disability underwent multivariable regression modelling. FINDINGS: Overall prevalence of disability was 7.4%. Maternal underweight (OR 1.44 (95% CI 1.01-2.08)), maternal cohabitation under 16 years of age (OR 1.50 (1.13-2.00)), standardised infant weight at 1 month (OR 0.82 (0.71-0.95)) and reported infant diarrhoea and vomiting in the first month (OR 2.48 (1.58-3.89)) were significantly associated with disability adjusted for trial allocation. The majority of hypothesised risk factors, including prematurity, were not significant. INTERPRETATION: Proxies for early marriage and low birth weight and a measure of maternal undernutrition were associated with increased odds of disability. The lack of association of most other recognised risk factors for adverse outcome and disability may be due to survival bias.