JMIR Publications

ABSTRACT

Background:

Although global child mortality has declined considerably, it remains disproportionately high in many low-income countries. Identifying individuals at high risk is crucial for developing hypotheses that eventually lead to interventions.

Objective:

We aimed to discover sub-groups of children at high mortality risk in the capital of Guinea-Bissau, West Africa.

Methods:

We used data from a health and demographic surveillance site, the Bandim Health Project, in Guinea-Bissau (2003-2016) to identify baseline information registered before 6 weeks of age consistently associated with high levels of mortality between 6 weeks and 3 years of age. We analysed spatial differences, single risk factors, and clusters of multifactorial risk groups. Using a temporal split of the data, we assessed which sub-groups were consistent across time, and re-assessed the mortality risk using Targeted Maximum Likelihood Estimation.

Results:

Of factors identifying sub-groups with especially high child mortality, we identified one residential area (adjusted risk difference in deaths of 3.4% [95% confidence interval 0.3%; 6.5%]), having no prenatal consultations (5.8% [95% CI 2.6%; 8.9%]), and a subgroup of children of polygamous families born in the dry season (1.7% [95% CI 0.4%; 2.9%]). While these subgroups had higher risk of mortality across time, they constituted a small proportion of all children (constituting between 1-9% of children). Thus, common social and economic factors were linked to a larger share of the total child deaths.

Conclusions:

In areas with high child mortality, some children are at even higher risk than others emphasizing a potential for targeted interventions for vulnerable subgroups to complement universal public health approaches. The next step is understanding how to reduce the risk in the identified subgroups. We propose several empirical studies, which may further test, illuminate or develop the hypothesised phenomena. Clinical Trial: NA