A Longitudinal Birth Cohort Study in Coast Province, Kenya

In 2010, 7.6 million children under the age of five died worldwide, mainly from preventable and treatable conditions (Liu et al., 2012). Notably, the burden of under-five mortality varies dramatically by country. The majority of child deaths are seen in Africa (3.6 million) and southeast Asia (2.1 million deaths), compared to 0.16 million and 0.28 million under-five deaths in Europe and the Americas, respectively (Liu et al., 2012). In Kenya, over 120,000 under-5 deaths were estimated in 2010 and approximately 35% of these deaths occurred in the neonatal period.

A thorough understanding of the etiology of child mortality is necessary to guide research efforts aimed at tackling this important global health problem. Importantly, in 2010, the causes of only 2.7% (0.205 million) of all deaths in children under the age of five were medically certified (Liu et al., 2012), highlighting the need to gather data on the causes of mortality.

Prospective longitudinal birth cohort studies present an opportunity to examine temporal relationships between early-life exposures (i.e. prenatal, pregnancy, and early postnatal exposures) and multiple health and non-health related outcomes including mortality, morbidity, and socioeconomic circumstances. It is well documented that exposures that occur early in life, including genetic, environmental, socioeconomic, and lifestyle factors, may have long-lasting effects on growth, development, and health outcomes throughout an individual's entire life course (Lynch & Smith, 2005). Thus, data on exposures during pregnancy and early childhood are valuable and may provide clues to the etiology of long-term outcomes.

Additional value can be gained through cross-cohort collaborations and comparisons (Larsen et al., 2013)(Paternoster et al., 2012)(Brion et al., 2011). Notably, by pooling data from multiple cohort studies, causal inferences can be made with greater confidence. For example, if a similar relationship is observed across multiple populations, each with their own distinct set of confounding variables, it is less likely that the observed association is being driven by confounders. Similarly, cross-cohort comparisons enable researchers to investigate patterns associated with health, social, and economic outcomes in distinct regions of the world. These types of analyses may provide valuable insight into the underlying causes of global health inequalities.

The objective of this study is to implement a longitudinal prospective birth cohort study in Kenya to obtain extensive information on early-life exposures and health and non-health related outcomes in the first year of life.