Using health insurance to improve equitable access to quality services in low-and-middle income countries

Universal health coverage (UHC) has received tremendous attention over the last few decades, particularly after its central placement as part of Sustainable Development Goal 3. UHC means all people have access to high-quality services without financial burden. Earlier evidence before this thesis suggested that health insurance programs in LMICs increase service utilization and may reduce financial burden to an extent. However prior to this thesis, there was little evidence on the relationship between health insurance and quality of care in LMICs. Furthermore, there was also very little systematic evidence on inequities in uptake of health insurance in LMICs or the effects of insurance on quality of care in LMICs. The broad goal of the research was to examine health insurance and quality of care in LMICs guided by frameworks for quality of care with case studies in Tanzania and Zambia to understand the implementation of their national health insurance programs.
The first part of the research is an overview of the evidence on the impact of health insurance programs on UHC goals - equity, service utilization and financial protection - and then presenting the rationale for the research on equity and quality of care. The results from the systematic review and meta-analysis on equity show that, on average, vulnerable populations (the poorest and least educated groups) are less likely to enroll in health insurance than better-off groups, despite exemptions and subsidization policies by governments and health insurance agencies to increase uptake among these groups. Only one health insurance program in Colombia that relied on the existing social security database reported higher enrollment among vulnerable groups. While the findings of the review may seem obvious, it fills an important gap in the literature and contributes to the equity debates surrounding the scale-up of health insurance in LMICs. Regarding the impact of health insurance on quality of care in low-income countries, we found few studies that used rigorous study designs or evaluated the effects of health insurance on structural inputs and processes of care. The evidence from these studies indicates that health insurance is not associated (positively or negatively) with structural quality, and its effects on processes of care remain mixed. In regards to the outcome dimension, the evidence suggests that health insurance is linked to improved anthropometric measures for children and biomarkers, such as blood pressure and hemoglobin levels. Therefore, we suspect that the improvements in health outcomes from health insurance were driven mainly by increases in access to care rather improvements in quality.
In the subsequent chapters, the research examined the implementation of health insurance programs in Tanzania and Zambia and their ability to influence the quality of care. In Zambia, we assessed the health system factors that could affect its national health insurance, which offers only hospital services from providers in the public and private sectors. The results showed that in Lusaka, most adult patients do not use primary care facilities for non-emergency care and heavily rely on pharmacies and drug shops. In terms of their confidence in the health system and insurance enrollment, the findings show that among the informal sector population, confidence in the care provided by the public sector is low compared to confidence in the private sector. Confidence in the health system was found to be a significant determinant of health insurance uptake. While confidence in the public sector was only weakly associated with enrollment, confidence in the private sector was strongly associated with enrollment. In examining the implications of the health system context and the purchasing arrangements of the Zambian National Health Insurance (NHI) on the insurance goal of improving quality of care, the results showed how some of the challenges within the health system could affect the insurance’ s ability to influence the quality of care. The challenges include the low public funding for health that has deteriorated the quality of care, particularly at primary healthcare levels. Moreover, weak regulations on health professionals, medicines, and health facilities have also contributed to poor-quality inputs. The findings also shed light on the purchasing arrangements of the Zambia NHI that can influence the quality of care. The health insurance attempted to mitigate some of the challenges in the health system by providing public hospitals with advanced payments for the procurement of medicines and minor renovations. While this may improve some structural inputs for quality of care, the revenue from insurance may not be sufficient for prepaying larger infrastructure projects for hospitals, and they may still require government support through other financing mechanisms. Another finding was that the design to improve the care experiences of members, through short waiting periods and designated services, might not be equitable and unsustainable as coverage increases.
However, the purchasing arrangements of the insurance may also have negative implications for high-quality care. First, the current referral policy does not promote coordination between the public and private sectors. This decreases the opportunities for integration to ensure the continuum of care. Second, the provisional benefits may not be equitably distributed geographically, as the rural areas have fewer private providers and higher-level hospitals than urban areas. The inclusion of private providers was intended to mitigate the challenges in the public sector, but it may further exacerbate the pro-urban pattern of the distribution of health benefits. Third, not all facilities included in the health insurance, particularly those in the public sector, met the quality criteria set by the insurance, thus compromising access to benefits and quality. However, this could create a path dependency where public facilities may not be motivated to uphold the same quality standards as the private sector. Fourth, its supervision and accreditation checklists are heavily focused on structural indicators, and the only dimension of processes of care is care experiences, neglecting other components of quality of care that could assess the quality of care. Fifth, the limited resources in health facilities and the incentive by health insurance for providers to improve the care experiences of its members may jeopardize the care experiences of the uninsured, who are often the poorest populations. Finally, the low payment rates for first-level hospitals, the bottom of the insurance service delivery system, may create incentives for unnecessary referrals to high levels of care and may worsen the bypassing challenges.
While Tanzania has many years of experience implementing its national health insurance scheme, we found that the country faces similar challenges to those that Zambia faces in the design phase of its health insurance scheme. In both countries, we found that delays in reimbursements are a significant burden that affects inputs for quality of care. Some of the contributing factors for the delays are mechanisms for claims processes (electronic vs. paper-based) and lack of competent staff for claims processes. We also found that higher-level health facilities benefit more from health insurance due to members’ preferences for higher levels of care. There is also a strong focus on improving members’ care experiences through an extensive selection of public and private providers, but its benefits are distributed inequitably across geographical areas. Similar to in Zambia, health insurance in Tanzania has improved access to high-cost services. However, unlike Zambia, Tanzania’s NHIF payment mechanism has incentivized adherence to the national clinical guideline by reimbursing only treatments that follow these guidelines. However, the reduction in NHIF benefit entitlements over the years has dissatisfied its beneficiaries. Although the NHIF had its challenges, the quality of services and benefits are perceived to be much better than the improved community health fund (iCHF), which targets the informal sector. We found that the negative perception of iCHF was due to governance factors, such as the failure of the insurance design to support greater access to medicines and weak accountability of revenue generated from premiums.
The overall findings also informed concrete guidance in the areas of financing, governance and service delivery to countries considering using health insurance programs to make progress towards equitable access to quality services. Importantly, the results offer insights on how countries with existing health insurance programs can design their purchasing arrangements to monitor and improve quality of care. Health insurance programs should balance the different dimensions of quality of care to ensure providers are not incentivized to focus on improving structural inputs of care, which may not lead to a higher quality care. There is also a strong need to use data, such as data from claims and routine health information systems, to monitor the quality of care and use them as learning vehicles to redesign insurance programs for high-quality care and to change providers’ behavior.
In addition, the research of this thesis provided the foundation for future research work on various aspects of health insurance and equitable access to quality care. Furthermore, the research showed the need for robust study designs suitable for determining the effectiveness on health insurance and quality of care and more data sources to enable measurement of the different dimensions of quality of care.