Essays on health and health care utilization in Botswana

Mmopelwa, David (2021) Essays on health and health care utilization in Botswana. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (2MB)


The role of health in both economic growth and development has been established in the literature. Consequently, public health policies aim to ensure an equitable provision of quality curative health services and promotion of preventive care. However, some national survey-based health indicators in Botswana show that both child and adult health deteriorated over time, which warrants investigation that is now timely given that both the current health policy and the integrated health service plan are nearing their end. Therefore, this three-essay thesis attempts to provide insights into deteriorating health.

Following an increase in the proportion of low birth weight (i.e., less than 2.5kg) children, during a period in which the country was among the worst performers in the WHO African region and the upper-middle-income category, the first essay assesses the role of prenatal care utilization on infant`s health. A joint treatment effect model for care utilization and infant outcome is estimated, accounting for the endogeneity of care utilization. As not all infants have birth weight reported, the essay also estimates the sample selection model to control for potential sample selection bias. Results show that adequate prenatal care utilization is positively associated with infant health and that failing to account for endogeneity reduces its effect. On average, birth weight improves by 0.67, 0.73 and 0.64 kg in full, urban and rural samples, respectively. These findings imply the need to audit care quality since utilization rates are high, but outcomes have been deteriorating.

The second essay investigates the effect of household size and child`s birth order on child nutrition as measured by the three anthropometric indicators: height for age, weight for age and weight for height. The essay is motivated by one of the theoretical predictions of the resource dilution hypothesis relating to the impact of family size on child capital. The hypothesis postulates that large family sizes and high child`s birth order are more likely to have a negative effect on child capital. However, there are arguments that the assumption of a fixed and narrow flow of resources from parents underpinning the theory may not always hold. We estimate the random effects model to explore the within and between household effect. We find that household size is negatively associated with child nutrition and high birth order children (i.e. later born) are more likely to fare worse than their lower birth order counterparts. A variation decomposition analysis shows a higher variation in child nutrition across than within households. Of the total variation, a large share is accounted for by the unexplained component.

For all the three indicators the between household explained shares are in the range 5-8%, while for within household explained shares are less than 1%.

The third essay investigates whether sexual behaviour (indicators of activity and use of a condom) is responsive to both objective risk (HIV prevalence rate) and perceived risk (concern on the likelihood of being infected by HIV due to ART). Using three of the four waves of the Botswana AIDS Impact Survey (BAIS) collected by Statistics Botswana, we account for the endogeneity of both risk indicators through the method of instrumental variables. HIV prevalence rate is instrumented by the distance from respondent`s area of residence to the district with the highest HIV rate, while the perceived risk is instrumented with duration (in months) between antiretroviral therapy (ART) rollout and the survey date. Overall, results suggest there is more responsiveness to subjective rather than objective risk indicators. Compared to those who indicate to have not changed their level of concern about being infected by HIV due to ART, the less (more) concerned are less likely to abstain and use a condom (more likely to use condoms). The HIV prevalence rate has no effect on both the number of sexual partners and use of condoms. From a policy perspective, the non-behavioural response to HIV prevalence could be a cause for concern if there are unobserved fatal behavioural patterns. Thus, because programmes such as ART prolongs the lives of the infected and affects perception about the likelihood of infection risk, there could be some negative externalities as revealed by the results.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Morrissey, Oliver
Owens, Trudy
Keywords: Prenatal care, Utilization, Botswana; Children, Nutrition; Households; Sex; HIV infections; Antiretroviral agents
Subjects: R Medicine > RA Public aspects of medicine
Faculties/Schools: UK Campuses > Faculty of Social Sciences, Law and Education > School of Economics
Item ID: 64220
Depositing User: Mmopelwa, David
Date Deposited: 04 Aug 2021 04:40
Last Modified: 04 Aug 2021 04:40

Actions (Archive Staff Only)

Edit View Edit View