CONTEXTUAL AND INDIVIDUAL LEVEL FACTORS INFLUENCING CHILD HEALTH IN LOW AND MIDDLE INCOME COUNTRIES

Abstract

<p>This thesis examined contextual and individual-level determinants of child health in low and middle income countries, with specific focuses on slum residency, maternal use of cesarean section (CS) during delivery and malaria prevention campaigns. It also examined parental HIV preventive behaviors that could ultimately be associated with child health outcomes (although this was not studied in this thesis). The thesis is presented as four separate projects that examine the following: (1) associations of urban slum residency with infant mortality and child stunting in 45 countries and factors modifying these associations; (2) relations between individual-level CS and neonatal mortality in 46 countries and the moderating influence of country-level CS rates on these relations; (3) determinants of consistent condom use based on the perspectives of both husband and wife where at least one partner is HIV infected and resident in three sub-Saharan African countries with high HIV prevalence; and (4) evaluation of the associations between long-lasting insecticidal nets (LLIN) mass distribution campaigns led by different organizations (the World Bank, UNICEF, or Global Fund) and malaria among children in Nigeria. Data were from nationally representative Demographic and Health Surveys (DHS) for the first three projects and Nigeria Malaria Indicator Survey (NMIS) for the fourth project. The research objectives were addressed using multilevel modeling for projects 1, 3, and 4 and propensity score matching, meta-analysis and meta-regression for project 2.</p> <p>Results of each separate project of this thesis showed that: (1) after controlling for the socio-economic circumstances of individual families, residency in a slum community was associated with infant mortality. This association was attenuated among children born to women who had received antenatal care from a health professional. Finally, residency in a slum community exacerbated the risk of stunting among older versus younger children; (2) individual-level CS increased the risk of neonatal mortality and that this risk was the highest in countries with the lowest CS rates; (3) couples were more likely to have used condoms consistently if the husband was HIV positive, and was the only one in the couple who knew his HIV test result. This association was stronger if the HIV-positive husband was aware of his HIV test result and his HIV-negative wife aware of her test result. No corresponding associations were observed for HIV-positive wives; (4) compared with children living in areas with no LLIN mass distribution campaigns, those in the World Bank Booster Project areas were significantly less likely<strong> </strong>to test positive for malaria but no significant differences in child malaria infections were found between other campaigns and non-campaign areas. Results also showed that community-level wealth, community-level maternal knowledge regarding malaria prevention, and child-level use of insecticidal nets were negatively associated with child malaria.</p> <p>Taken together, the findings suggest that improving the material circumstances of slum neighborhoods and increasing antenatal care coverage among women living in these neighborhoods could help reduce stunted child growth and infant mortality associated with slum residency. Results also indicate the need of improving the quality of maternal and newborn health care services, especially in countries with low CS rates in order to reduce neonatal mortality associated with CS. In addition to LLIN mass distribution campaigns, improving maternal knowledge on malaria prevention at the community level might be helpful in preventing child malaria. Finally, results show that increasing HIV testing and awareness of results among husbands could help reduce HIV transmission among couples.</p>