Organized session
Digital Technologies for Health Financing: Novel Evidence on Opportunities and Challenges on the Path to Universal Health Coverage
Talks
Evaluating the mobile phone-based renewal service for health insurance in Ghana
Laura Nübler, Technische Universität Berlin
Introduction
Background Since the Ghanaian National Health Insurance Scheme (NHIS) was introduced in 2004, active membership had remained low, despite affordable premiums and payment exemptions for minor, senior, poor and pregnant individuals. While 70% of the population had previously registered with the NHIS, many individuals do not complete annual renewal process on time and thus incur coverage gaps or lose insurance altogether. In order to simplify the time-consuming, annual health insurance renewal process at the NHIS offices, a mobile renewal service was rolled out in 2019, through which members can complete their annual insurance renewal and premium payment easily via mobile phone. We evaluate the effectiveness of this intervention in reducing insurance loss and coverage gaps. Methods We use a full sample of anonymized NHIS routine data for the years 2017-2021. We investigate the demographics of active members, determinants of mobile renewal service usage, using logistic regressions to compute odds ratios. Further, we estimate Weibull survival regressions controlling for sociodemographic determinants, and further illustrate differences in renewal patterns with Kaplan-Meier survival curves. Results Active membership demographics show a stark loss of insurance coverage among males who turn 18 and lose the payment exemption status for minors, while coverage among females of the same age remains high. We observe that the mobile renewal service was quickly accepted, accounting for 87% of renewals of those eligible to use it by 2021. Preliminary results show that mobile renewal users were 33% less likely to default on their insurance, and had significantly shorter coverage gaps when they did. Minors and older individuals were most likely to renew via mobile. Young adults, men, and informal workers were less likely to use the service and more likely to experience coverage loss and gaps. Conclusions An evaluation of the mobile renewal service showed that Ghanaian NHIS members were largely willing and able to take advantage of the service, which lowered their likelihood of defaulting on insurance and shortened their coverage gaps. Encouragingly, elderly members (or their family members) and rural residents were able to use the service to complete their renewals at high rates. However, young adults, males, and informal workers are at a high risk for coverage loss and are less likely to use mobile renewal, which suggests that these demographics in particular may need to be further aided to renew through additional interventions.
The impact of a mobile money-based intervention on maternal and neonatal health outcomes in Madagascar: a cluster-randomized controlled trial
Lisa Bogler, Department of Economics and Centre for Modern Indian Studies, University of Goettingen, Goettingen, Germany.
Introduction
Background Financial barriers to accessing skilled care during pregnancy persist in many low-resource settings. With increasing use of mobile phones, mobile money services appear as promising tool to address this concern. Maternal healthcare is particularly suitable for a mobile money savings program due to pregnancy’s predictable timing and costs. The Mobile Maternal Health Wallet (MMHW) developed by the non-governmental organization Doctors for Madagascar, aimed to alleviate out-of-pocked expenses for maternal healthcare by providing an accessible savings tool. Users could save with the MMHW, pay for services in participating health facilities, and received additional benefits such as a free obstetric ultrasound check. Methods We used a stratified cluster-randomized hybrid effectiveness implementation trial to assess the impact of the MMHW on maternal and neonatal health outcomes in the Analamanga region of Madagascar. We randomized 63 public-sector primary-care health facilities (CSBs) to either receive the intervention or not. In intervention CSBs and their catchment areas, sensitization campaigns encouraged pregnant women to register with the MMHW. We estimated intention-to-treat effects (ITT) and contamination-adjusted ITT effects. The primary outcomes included (i) delivery at a health facility, (ii) antenatal care visits, (iii) total healthcare expenditure. Between March and Dec. 2022, 6483 women who had been pregnant between July 2020 and Dec. 2021 and were surveyed in the catchment areas of treated and control CSBs. Findings Sensitization and availability of the intervention led to significantly higher uptake by women living in the catchment areas of intervention CSBs. Among women in these areas, 39% had heard of the MMHW and 37% of those registered with the tool, compared with 13% and 27% in catchment areas of control CSBs respectively. There was considerable variation in uptake even across intervention CSBs. The contamination-adjusted ITT suggests a positive impact of the MMHW among those using it. However, this study was only powered to detect significant differences in the primary outcomes assuming uptake among more than 50% of women in the intervention areas. Due to lower than anticipated uptake and resulting reduced power, we did not find statistically significant effects on delivery in facility, number of ANC visits, or total expenditure for health. Conclusion The estimated contamination-adjusted ITT effects suggest that the MMHW carries potential to improve access to maternal care. The lower than anticipated uptake and large differences across CSBs raise concerns about implementation fidelity. Imperfect implementation of the intervention in CSBs therefore might have reduced impact on potential beneficiaries.
Scaling-up digital health financing technologies in Sub-Saharan Africa: A cross-country comparison of the relevance of reminders and automatic renewal in Kenya, Nigeria, Rwanda, Tanzania, Uganda
Verena Verena Struckmann, Technische Universität Berlin
Introduction
Background: The Research to Improve Resilience in Major African Cities through Universal Health Coverage (ReachUHC) project has co-conceptualized a mobile phone-based intervention consisting of regular renewal reminders for the Ghana National Health Insurance Scheme (NHIS), combined with an automatic renewal option to address the challenge of coverage gaps. We aim to assess the potential for similar mobile phone-based interventions to be implemented in health insurance systems in Kenya, Nigeria, Rwanda, Tanzania, and Uganda, while learning from existing implementation experiences in these countries. More specifically, our objectives are to describe, analyze and compare information from these countries about (1) the availability of digital technologies for health financing (DTHF), (2) the role and effects of DTHF in supporting different health financing functions; and (3) the regulatory context that has supported (or hindered) the development and implementation of DTHF in these countries. Methods: We are conducting semi-structured in-depth interviews with various stakeholders on factors influencing the implementation process and the role of contextual factors. In addition, focus groups with end-users are conducted to better understand usability and acceptability of existing DTHF. The collected data will be analysed to assess the feasibility of implementing mobile phone-based DTHF to improve coverage in selected countries. Comparing contextual and regulatory differences across countries will support the development of a better understanding of the relevance of these factors for the implementation of DTHF and their potential to support UHC objectives. Results: The study will provide results on the implementation and use of mobile phone-based health financing interventions to reduce coverage gaps in Sub-Saharan Africa. Implementation and scale-up of DTHF depend on a thorough understanding of contextual factors, including predisposing characteristics, needs, and enabling resources as determinants of user satisfaction and intervention uptake. Conclusions: Working in close collaboration with national experts, research results will inform plans for future national scale-up of the ReachUHC intervention in Ghana, and they may facilitate the implementation of similar interventions in selected countries. Our cross-country analysis may further support the development of a better understanding of conditions that support the implementation and scale-up of DTHF in Sub-Saharan Africa.
Benefits and risks of artificial intelligence and machine learning for universal health coverage: the need for health financing specific guidance and regulations
Inke Mathauer, World Health Organisation, Health Financing Unit
Introduction
Background and objective: There is growing use of machine learning (ML) for the health financing functions (revenue raising, pooling, purchasing and benefits design), yet a lack of evidence of the effects on universal health coverage (UHC). This paper provides a synopsis of the use cases of ML and their potential benefits and risks. Methods: Building on a previous literature review, we identify the main domains of health financing in which ML approaches are used in countries or for which theoretical modelling exercises and model comparisons have been undertaken. We then extract the benefits and risks from various review papers and deductively derive the implications in relation to UHC objectives. Findings: The assessment reveals that the various use cases of ML for health financing have the potential to affect all the intermediate UHC objectives – the equitable distribution of resources (both positively and negatively), efficiency (primarily positively), and transparency (both positively and negatively). There are also both positive and negative effects on all three final UHC goals, i.e. utilization in line with need, financial protection and quality care. When the use of ML facilitates or simplifies health financing tasks that are counterproductive to UHC objectives, there are various risks, for instance risk selection, cost reductions at the expense of quality care, reduced financial protection, or over-surveillance. Whether the effects of using ML are positive or negative depends on how and for which purpose the technology is applied. Discussion and conclusion: Therefore, there is need for health financing specific guidance and regulations, particularly for (voluntary) health insurance, including risk mitigation measures. In order to inform their development, the paper proposes a number of key policy and research questions. More rigorous research and systematic evidence should accompany the application of ML so as to better understand its effects for health financing oriented towards UHC.