Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya
| dc.contributor.advisor | Leibbrandt, Murray | |
| dc.contributor.author | Kamundia, Susan | |
| dc.date.accessioned | 2025-02-25T12:24:48Z | |
| dc.date.available | 2025-02-25T12:24:48Z | |
| dc.date.issued | 2024 | |
| dc.date.updated | 2025-02-25T12:21:00Z | |
| dc.description.abstract | Maternal and child health outcomes are an important indicator of a country’s well-being. Kenya has less maternal and child mortality levels compared to Sub-Saharan Africa averages, but it still performs worse compared to the rest of the world. Some of the reasons behind this are the prohibitive costs of accessing maternal health care, lack of access to quality health care and longer distances to health facilities. One of the ways that have been touted as a way of improving these outcomes is the utilisation of maternal health care from skilled providers. In this thesis, I seek to explore the maternal and child health outcomes in Kenya in relation to accessibility, spatial distribution, equality and free maternal care in Kenya. The second chapter tackles the description of data, construction of variables relevant to the rest of the thesis, maps the health facilities offering maternal health care in Kenya and presents the framework within which I conduct my analysis. The data used throughout my thesis are drawn from the Kenya demographic and health surveys (DHS) collected in 2003, 2008/09 and 2014. DHS surveys have a rich set of information on individuals and households. However, it lacks information on health facility characteristics which are pertinent in determining maternal and child health outcomes. I, therefore, endeavored to introduce supply-side factors which are not traditionally included in DHS data into the analysis of issues surrounding maternal and child health. The constructed asset index shows an increase in asset wealth between 2003 and 2014. A mapping of the health facilities offering maternal health care in Kenya shows a higher density of health facilities in the western, central and southeastern parts of the country. However, the pattern of health facility service provision also follows the population density with more densely populated areas having a higher density of health facilities. The third chapter serves a three-fold purpose. First, I trace the trend of utilization of three maternal health care services. The utilization shows a significant increase between the 2003 and 2014 surveys. Secondly, a logit regression model is used to examine the factors that determine the utilization of maternal health care when supply-side factors are controlled for. As expected, utilization of maternal health care is found to increase with increases in maternal education levels, household wealth, mother’s age at the time a child is born, facility level and alternative supply of health. The utilization is lower for women in rural areas as compared to those in urban areas, for women with more than four children compared to those with less than four children, for married women/living with a partner as compared to those who are single/living alone and for women living further away from a health facility. Spatial dependence is shown to exist in the utilisation of maternal health care and geographically weighted regression (GWR) models are used to analyze the factors explaining the utilization of maternal health care when spatial dependence is taken into account. GWR models are also used to explore the possibility of non-stationarity existing in the factors that explain the utilization of maternal health care. The largest positive effects on the utilization of maternal health care are in clusters with low maternal education levels, younger mothers and a lower alternative supply of health facilities. The largest reductions in utilization of maternal health care are in areas with mothers who have more children and in rural areas with a lower density of health facilities. Mothers are also more likely to utilize higher-level health facilities in areas with a higher density of higher-level health facilities. High inequalities have been shown to exist in the utilisation of maternal health care services due to differences in socioeconomic, demographic and access to health facilities. In the fourth chapter, I, therefore, aim to analyse the inequality arising from socioeconomic factors, more specifically from differences in asset wealth. I describe the evolution of inequality in the utilisation of maternal health care using Wagstaff concentration indices and assess the differences in inequality arising from the introduction of the free maternal care (FMC) program in Kenya. I also decompose the factors that explain the inequality in the utilisation of maternal health care between the poor and non-poor using recentered influence functions (RIFs). The results show the presence of substantial inequalities which favour the non-poor. The main contributor to this inequality is the maternal level of education. The introduction of the FMC program saw an increase in the utilisation of maternal health care by both the poor and nonpoor groups. However, the difference in inequality levels was not significant. One of the mechanisms used by governments to encourage women to seek maternal health care services from skilled providers is the reduction or removal of user fees. In chapter five, I review the free maternal care (FMC) program introduced in Kenya in June 2013 which made antenatal, delivery and postnatal care free in public health facilities. I seek to analyse the impact of free maternal care on the levels of neonatal mortality in Kenya by comparing the births that occurred in the 16 months before and after the start of the FMC program using the matching differencein- differences (MDID) method of estimation. The births in public health facilities make up the treated group while the births in private health facilities and at home and other places make up the control group. The results show a significant increase in the utilisation of maternal health care services after the start of the FMC program. However, the MDID results show an increase in neonatal mortality when MDID estimation is done by comparing births in public health facilities and those at home and other places. This indicates that while the cost of maternal health care services is an important determinant of utilisation, other key factors exist that also determine utilisation. In sum, I introduce supply-side factors to the analysis of maternal and child health outcomes using demographic and health surveys. I find that spatial dependencies exist in the utilisation of maternal health care and utilise an appropriate spatial model to analyse the factors explaining the utilisation of maternal health care. I also find that inequality in the utilisation of maternal health care exists and the decomposition of the factors explaining this inequality shows that maternal education levels and place of residence are the key determinants. Finally, an investigation of the free maternal health care program shows a curious result with neonatal mortality increasing which is contrary to expectations. | |
| dc.identifier.apacitation | Kamundia, S. (2024). <i>Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya</i>. (). University of Cape Town ,Faculty of Commerce ,School of Economics. Retrieved from http://hdl.handle.net/11427/41006 | en_ZA |
| dc.identifier.chicagocitation | Kamundia, Susan. <i>"Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya."</i> ., University of Cape Town ,Faculty of Commerce ,School of Economics, 2024. http://hdl.handle.net/11427/41006 | en_ZA |
| dc.identifier.citation | Kamundia, S. 2024. Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya. . University of Cape Town ,Faculty of Commerce ,School of Economics. http://hdl.handle.net/11427/41006 | en_ZA |
| dc.identifier.ris | TY - Thesis / Dissertation AU - Kamundia, Susan AB - First and foremost, to my supervisor, Professor Murray Leibbrandt, thank you for your mentorship, invaluable support and immense patience throughout the development of my thesis. I appreciate the timely feedback, dedication and guidance that have led to the successful completion of this thesis. Thank you for believing in me and helping me build confidence in my research. To my family, thank you for being my pillar of support throughout this PhD journey. Thanks to my mum Agnes for being my rock, always encouraging and uplifting me through it all. To my brothers, Waweru and Gerald and sisters Lydia, Charity and Grace, thank you for always motivating me, to my nephews Kwame, Faraji, Lemaiyan and Jelani, thank you for always putting a smile on my face even when the going got tough. I'm eternally grateful to have a family that provided me with financial, emotional and mental support that kept me balanced throughout the ups and downs of doing a PhD and I remain forever grateful and indebted to you. Thank you to my friends and colleagues Diana, Bongai, Betty, Sam, Malefeu, Godfrey, Arindam, Nyasha and Grace for the amazing comments and feedback on my thesis and for giving me a sounding board for my ideas and sometimes just a forum to vent and clear my mind. Thank you to the School of Economics staff especially Paula Bassingthwaighte, Haajirah Esau and Thembisa Nyamakazi for the support for making my stay at the school as comfortable as possible by giving timely assistance when needed. I would like to appreciate the African Economic Research Consortium (AERC) for providing the funds that enabled me to start my PhD studies. Thanks to my supervisor for providing additional funding through the South African Labour and Development Research Unit (SALDRU) and African Centre of Excellence for Inequality Research (ACEIR) without which the final leg of the PhD would not have been completed. Special thanks to the group A AERC 2019 June and December biannual conferences participants for their constructive comments, especially Professor Femi Ayadi who took a lot of care to advise me during the conceptual stages of my thesis and helped me fine-tune my ideas. Special thanks to the African Centre of Excellence for Inequality Research (ACEIR) and Professor Richard Harris from the University of Bristol for organising training on mapping and modelling geographical data. This training was very enlightening and provided much more clarity in my third chapter. I owe immense appreciation to the United Nations University – World Institute for Development Studies (UNU-WIDER) for having given me an opportunity to participate in their prestigious PhD fellowship program in September-November 2021 during which I completed chapter four of my thesis and for inviting me to the WIDER Development Conference on Reducing Inequality in 2022 which provided a forum for me to present my paper and get constructive feedback. More specifically, I owe my gratitude to Dr Simone Schotte for the guidance, remarkable input and mentorship during and after the duration of my fellowship. Thank you to Dr Carlos Gradin for the valuable comments on my chapter. I am grateful to the University of Cape Town Digital Library Services especially geographical information systems (GIS) officer Thomas Slingsby, for his invaluable support in acquiring some of the data used in this thesis. My appreciation also goes to the University of Cape Town high performance computing (HPC) team especially Andrew Lewis for giving me access to their resources and guidance on the best way to utilise them which made the computer-intensive analysis on geographically weighted regressions in chapter three and recentered influence function decompositions in chapter four possible. Last but not least, I thank God for giving me the graces and strength to complete this PhD journey. I do not take it for granted that I was able to come this far. DA - 2024 DB - OpenUCT DP - University of Cape Town KW - Economics LK - https://open.uct.ac.za PB - University of Cape Town PY - 2024 T1 - Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya TI - Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya UR - http://hdl.handle.net/11427/41006 ER - | en_ZA |
| dc.identifier.uri | http://hdl.handle.net/11427/41006 | |
| dc.identifier.vancouvercitation | Kamundia S. Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya. []. University of Cape Town ,Faculty of Commerce ,School of Economics, 2024 [cited yyyy month dd]. Available from: http://hdl.handle.net/11427/41006 | en_ZA |
| dc.language.rfc3066 | Eng | |
| dc.publisher.department | School of Economics | |
| dc.publisher.faculty | Faculty of Commerce | |
| dc.publisher.institution | University of Cape Town | |
| dc.subject | Economics | |
| dc.title | Maternal and child health outcomes in relation to accessibility, spatial distribution, inequality and free maternal care in Kenya | |
| dc.type | Thesis / Dissertation | |
| dc.type.qualificationlevel | Doctoral | |
| dc.type.qualificationlevel | PhD |