Introduction: Mortality in early age provides a primary measure of health in a population. Knowledge of levels, trends and differentials in mortality rates are useful for public health monitoring and program evaluation. Indonesia is a country where early age mortality levels have declined in recent decades. Despite the fall in early age mortality in Indonesia, under-five mortality rate remains high in Nusa Tenggara Timur (NTT). The
infant mortality rate in NTT was 57 per 1,000 live births for the period 1998-2007, derived from the national demographic and health survey for Indonesia. It is difficult to obtain valid and reliable local data. There is a need for district level mortality estimates for local area health monitoring and planning. To that extent, this study aimed to identify the level and determinants of under-five mortality risk in Ende District, NTT.
Method: Cross-sectional design was applied to measure early age mortality through complete retrospective birth histories. This study is a household based survey which was conducted between March until May 2010 in Ende, NTT. The survey sample included random selection of 32 villages covering about 14% of the district population, with estimated approximately 8,000 households. This survey used individual questionnaire for ever-married women aged 15-49 years to collect complete retrospective birth histories and related health care variables. A household questionnaire was also administered to collect selected socio-economic and environmental variables. Field data collection was preceded by pilot test and focused training of survey staff. We measured early age mortality rates using direct methods by place of residence and various socio-economic categories. Multilevel regression was performed for analyzing determinants associated with under-five mortality. Results: A total of 7,454 households participated in this study. There were 7,324 births from 2000 to 2009. Descriptive analysis of survey data identified that a traditional birth attendant was the main birth assistant in the North and East by delivering more than half of births. Also, fifty-seven per cent of delivery took place at home. In particular, the North and East regions have lower formal education relative to the other regions in Ende. About 52.95% of births from 2000-2009 were in households who used piped water as the main source of drinking water. In the East, West and North only 50% or less of births were in households with a private toilet. Under-five mortality rates have declined from 63.2 in the period of 2000-2004 to 55.9 per 1,000 live births in 2005-2009, similarly for infant mortality (51.8 to 44.7 per 1,000 live births). Furthermore, infant and under-five mortality rates were three times higher in rural than urban areas. Overall, the point estimate of under-five mortality rate suggested a declining trend from 2000-2009. However the overlapping of 95% confidence interval indicated that the declines are not statistically significant. The odds of infant mortality were higher for mother whose delivery helped by traditional birth attendant than those helped by doctor/midwife/nurse (OR=1.61, 95% CI= 1.12, 2.31). The 95% confidence interval shows that infant mortality risk is statistically significant. The odds of under-five deaths were also higher for children living in North region than those in Ende Town region (OR=2.0, 95% CI=1.37, 2.93). Similar results also applied for those who live in East region (OR= 1.74, 95% CI=1.20, 2.51). The analysis results also show the same result for neonatal mortality. A child whose mother completed junior high school or above has half the risk of dying compared with those whose mother never attended or incomplete primary school. Child under 5 years who lived in a house without private toilet type was statistically not significant protected under-five mortality death (OR=1.58, 95% CI= 1.15, 2.10). Conclusion: The level and trend of early age mortality in Ende is remained high but it is necessary to do similar survey for other districts in NTT by obtaining greater sample. There are inequalities to access to health services across Ende in relation to ANC. This study also found that lower education level of mother’s leads to a higher mortality rate with poor child will have poor health status that lead to death at anytime. This study recommends that the government should take further action to deliver accessible health care for mother and child, equally distributed professional health workers (doctor/nurse, midwife), give regular health education for ever-married women aged 15-49 years about child survival and improving hygiene and sanitation in remote areas such as in North and East region. Those interventions would reduce under-five mortality in Ende and NTT.