This study sought to examine changes in the distribution of public and overall health spending (public, donor, and OOPE) for curative services and institutional deliveries as UHC reforms were being implemented in Zambia. The study makes an important contribution to the literature on UHC, being the first to assess the changes in the distributional incidence of public and overall health spending over time and also differentiating between curative and maternal care services in Zambia. Given the complexity of attributing change to individual UHC policies, and the data available, our study falls short of being able to attribute the distributional patterns to any specific UHC reform, but nonetheless examines changes overtime in relation to these reforms. Overall, we observe that public and overall health spending on curative services tended to benefit the poorer segments of the population while public and overall health spending on institutional delivery tended to benefit the least-poor. For both curative services and institutional deliveries, health spending at higher levels of health care (public hospitals) benefited the least-poor more than the poor while at lower levels of health care (health centres) and mission health facilities, the poor benefited more.
Zambia removed user fees in all rural areas in 2006, in peri-urban areas in 2007, and across the entire primary health care level in 2012 [21, 23] to address inequalities in access and utilization of health services. Three systematic reviews on user fees removal in LMICs by Qin et al. , Dzakpasu et al. , and Lagarde & Palmer  suggest that removing user fees has the potential to increase the utilization of both curative and maternal health services, especially for the poor. Our findings confirm results from previous studies in Zambia [18, 21, 23] which revealed that the removal of user fees in Zambia has contributed to increased utilization of curative services by the poor in Zambia. Public and overall spending on curative services benefited more the poor than the least-poor overtime. Given that most of the public health facilities providing primary health care are located in rural areas where the majority of the poor live and where about 90% of patients seek care in public facilities ; the removal of user fees has contributed to increased utilization of curative services among the poor. This pro-poor distribution of benefits from health spending on curative services is positively surprising, considering that Zambia has not adopted any specific policy to protect the ultra-poor from informal payments for healthcare. This evidence is inconsistent with evidence from Malawi, a neighbour country of Zambia, which has never introduced user fees but has high OOPE associated with using curative services that hinder the poorer segments of the population from using curative services ([34, 35]. For Zambia, Masiye and colleagues  observe that patients incur informal payments for health services that should be offered at free of charge. This presents a financial barrier for the poor segments of the population to use formal care .
Contrary to curative services, our findings on institutional delivery reveal that the overall distributional incidence for the relevant public and overall health spending is in favour of the least-poor. These results are consistent with findings by Chama-Chiliba & Koch  who conclude that removal of user fees has not fully removed barriers to utilisation of delivery services at public facilities in Zambia. Findings from Burkina Faso also question the fidelity of the free care policy in Zambia in ensuring free access to institutional deliveries . A study by Sochas  further reveals that health facility rules in Zambia can influence women's behaviour during pregnancy and childbirth, and create inequities against women with fewer financial resources. As part of the rules, pregnant women are required to purchase items needed for the delivery at a health facility such as bleach, a bathing tub, bucket, plastic sheet, gloves, nappies, and cotton wrapper, among others. In addition, costs for transport and new clothes for the babies and mothers are incurred (Scott et al., 2018). Consequently, inability to cater for costs associated with childbirth leads to low institutional deliveries in Zambia, especially for women from poor households . Kaonga and colleagues  also show that female-headed households bear the highest financial burden of healthcare payments in Zambia. This suggests that the costs associated with seeking care are still an important barrier to institutional deliveries among poor women in Zambia. The decrease of the inequality in public and overall spending on institutional deliveries between 2007 and 2014 implies that the removal of user fees may have had a positive effect, but was not fully effective in removing all the financial burden among poor women who would wish to deliver at a health facility . Other than affordability and as observed in other LMICs [41, 42], there are other dimensions of the health system environment in Zambia such as geographical accessibility, cultural beliefs, availability, and perceived quality of care that can negatively affect institutional deliveries . Therefore, to eliminate the inequality in the distribution of health spending on institutional deliveries, the Zambian government needs to implement strategies aimed at removing financial and non-financial barriers associated with childbirth at a health facility, especially for the poor segments of the population
Consistent with previous studies in LMICs [12, 44–46], inequalities in health spending on both curative services and institutional deliveries remain high for higher levels of care (i.e., inpatient care and deliveries at hospitals). This implies that UHC policies are not very effective at public hospitals. This could be because the user fee removal policy in Zambia is only applicable at lower levels of the public healthcare delivery system. In line with a study from India  and Zambia ; our findings indicate that health spending for both curative services and institutional deliveries at public health centres and mission health facilities, which operate at a lower level of healthcare and mostly in rural areas, tended to be more pro-poor following the user fee removal policy. The performance-based financing scheme, which was implemented between 2012 and 2014 at public health centres in some districts with a focus on maternal and child services—could have also contributed to greater equality of health benefits at the lower level of healthcare provision [19, 25]. Contrary to lower level of healthcare, individuals who access hospital services directly incur bypass fees or pay to access high-cost schemes and hospital prepayment medical schemes which are unaffordable to the poor. Except for emergency cases, a bypass fee is charged to patients who present themselves for treatment at a hospital without being referred from a health centre. Individuals from richer households can afford to pay the bypass fee and register for hospital prepayment schemes but this is not the case with poorer households. The existence of these charges at public hospitals in Zambia could explain why there are still disparities in the financing and utilization of healthcare services in Zambia . The other reason public and overall health spending favour the least-poor at public hospitals is that most of the tertiary and general hospitals are located in urban areas while the majority of the poor segments of the population live in rural areas where there are mostly public health centres and mission health facilities. As observed by Hjortsburg  and Eckman , the cost of providing health care in Zambia is skewed towards the urban areas, while access and consequences are concentrated among the rural areas and poorer socio-economic groups. Furthermore, there is an erratic supply of delivery kits, drugs, and other medical supplies at public hospitals as compared to public health centres . The scarcity of healthcare resources presents a high financial burden for the poor at higher levels of healthcare [36, 51]. As the core goal of UHC is that all people get access to needed high-quality healthcare regardless of one’s ability to pay , our findings call for specific actions by the Zambian Government to lift the financial and non-financial barriers that are still hindering the poor from using services at higher level of the healthcare delivery system.
Notwithstanding the value of this study, we need to note some limitations. Firstly, LCMS, DHS and ZHHEUS household surveys classify individuals across socioeconomic groups differently. Therefore, the socioeconomic groups may not be fully comparable across these surveys and we need to acknowledge bias that may arise from the use of different socioeconomic status measures. Secondly, based on the data at our disposal, having applied the constant unit subsidy/cost assumption, we might have masked differences in financial health benefits accruing to people of different socioeconomic groups at different health facilities or in different geographical settings. Thirdly, this study focused on the distribution of benefits from using curative services and institutional deliveries, expressed in monetary terms, without looking at health need and healthcare quality. Therefore, even if curative care and institutional deliveries were pro-poor at both public health centres and mission health facilities, it is difficult to tell if the services which the clients received were of high quality. Further analysis taking into consideration the health needs, quality and demand for healthcare could be undertaken.