METHOD: Data from the National Health and Morbidity Survey 2016 conducted by Ministry of Health was analysed. This nationwide survey involved 15,188 children below five years old. The survey was carried out using a two-stage stratified sampling design to ensure national representativeness. The Questionnaire from UNICEF's Multiple Indicator Cluster Survey (MCIS) was adapted to suit local requirements. Analysis was done using SPSS Version 23. Descriptive followed by multiple logistic regression were done to identify relevant factors.

RESULT: The prevalence of diarrhoea among children under five in Malaysia was 4.4% (95% CI: 3.8,5.2). Analysis using logistic regression indicated that only ethnicity and usage of untreated water were significantly associated with diarrhoea among children after controlling for relevant factors. By ethnicity, children in the 'Other Bumiputera' group had 2.5 times the odds of having diarrhoea compared to children of Malay ethnicity. Children of Indian ethnicity were also at higher risk, at almost double the odds, as well as other ethnic groups (1.5 times). Children who used untreated water supply were two times more likely to develop diarrhoea.

METHODS AND FINDINGS: We searched the major electronic databases Medline, Embase, and Google Scholar (January 1990-October 2018) without language restrictions. We included cohort studies on term pregnancies that provided estimates of stillbirths or neonatal deaths by gestation week. We estimated the additional weekly risk of stillbirth in term pregnancies that continued versus delivered at various gestational ages. We compared week-specific neonatal mortality rates by gestational age at delivery. We used mixed-effects logistic regression models with random intercepts, and computed risk ratios (RRs), odds ratios (ORs), and 95% confidence intervals (CIs). Thirteen studies (15 million pregnancies, 17,830 stillbirths) were included. All studies were from high-income countries. Four studies provided the risks of stillbirth in mothers of White and Black race, 2 in mothers of White and Asian race, 5 in mothers of White race only, and 2 in mothers of Black race only. The prospective risk of stillbirth increased with gestational age from 0.11 per 1,000 pregnancies at 37 weeks (95% CI 0.07 to 0.15) to 3.18 per 1,000 at 42 weeks (95% CI 1.84 to 4.35). Neonatal mortality increased when pregnancies continued beyond 41 weeks; the risk increased significantly for deliveries at 42 versus 41 weeks gestation (RR 1.87, 95% CI 1.07 to 2.86, p = 0.012). One additional stillbirth occurred for every 1,449 (95% CI 1,237 to 1,747) pregnancies that advanced from 40 to 41 weeks. Limitations include variations in the definition of low-risk pregnancy, the wide time span of the studies, the use of registry-based data, and potential confounders affecting the outcome.

CONCLUSIONS: Our findings suggest there is a significant additional risk of stillbirth, with no corresponding reduction in neonatal mortality, when term pregnancies continue to 41 weeks compared to delivery at 40 weeks.