Joseph Mfutso-Bengo

This paper provides an overview of the Thanzi La Mawa (TLM) survey, data collection methods and procedures, data modules and core variables, and quality assurance measures and expected analytical outputs from the TLM survey. The TLM survey is part of the Thanzi la Onse (TLO) model initiative. This is the first comprehensive health system model developed for any country, and aims to enhance understanding of healthcare delivery and resource allocation by capturing real-world data across the levels of healthcare delivery. The study seeks to support informed decision-making and future implementation of comprehensive healthcare system models in similar settings. The study uses a cross-sectional, mixed-methods approach to collect data on healthcare workers use of time, patient experiences, facility resources, and quality of care. The TLM dataset is based on a sample of 30 health facilities across Malawi comprising facility audits, patient interviews, and health worker use of time collected through time and motion studies. Data collection was conducted from January-May 2024. The analyses from this dataset provide insights into health workers’ time allocation to different activities, availability and use of healthcare resources at facilities, patient satisfaction, and overall service quality. This data is crucial for enhancing the TLO model’s capacity to answer complex policy questions related to health resource allocation in Malawi. The study also offers a structured framework that other countries, especially in East, Central, and Southern Africa that have expressed strong demand for the TLO model, can adopt as part of efforts to improve healthcare systems. By documenting methods and datasets, this paper provides guidance and tools for researchers and policymakers interested in healthcare system evaluation and improvement. Given the formal adoption of the TLO model in Malawi, the TLM dataset serves as a foundation for ongoing analyses into quality of care, healthcare workforce efficiency, and patient outcomes.

Non-profit, faith-based providers (FBPs) play a major role in primary care delivery across Sub-Saharan Africa and are increasingly integrated into national health systems, yet evidence on how the quality of faith-based primary care compares to that of public providers remains limited. Using healthcare worker observations, patient exit interviews, and patient follow-up data from Malawi, we compare consul-tation duration, clinical content, and patient-reported outcomes across faith-based and government facilities. Ordinary least squares estimates controlling for facility-, provider-, and shift-level characteristics, as well as patient health conditions, indicate that FBPs conduct significantly longer consultations, by around 1.9 minutes (75% of the sample mean), and perform a greater number of key clinical processes, including physical examinations and diagnostic tests. Patient-reported data corroborate these findings. Patients attending FBPs report receiving more tests and examinations and, at follow-up, are more likely to state that their treatment is working. These findings provide new evidence that publicly supported FBPs de-liver higher-quality primary care on both process measures and patient-reported outcomes, underscoring the importance of accounting for provider ownership when examining variation in access to quality primary care in Sub-Saharan Africa.

Objectives Modelling approaches that consider system-wide delivery platforms rather than single diseases are increasingly recognized as crucial for the economic evaluation of policy and investment choices and can be instrumental in forward-looking policy formulation. This paper develops a costing approach tailored to one such model, the Thanzi La Onse (TLO) model of Malawi’s healthcare system, to estimate healthcare delivery costs under varying health system assumptions. Methods We developed a mixed-method costing approach to estimate the total cost of healthcare delivery in Malawi using the TLO model, from a healthcare provider perspective. Through an iterative adjustment of key costing parameters, we attempted to align our model-based estimates with real-world expenditure and budget data. Costs were estimated over an 8-year period (2023–2030) under alternative scenarios of health system capacity, including supply chain performance and the size of the health workforce. Results A detailed comparison of our cost estimates against expenditure and budget data demonstrates the reliability of our costing method and assumptions, for the conditions and resources captured by the model. Under current health system capacity, the total cost of healthcare delivery in Malawi between 2023 and 2030 was estimated at $1.53 billion [95% confidence interval, $1.51b -$1.54b], which translates to an average annual cost of $309.83 million [$306.17m -$313.56m]. The estimation of costs under alternative scenarios demonstrates the importance of capturing feedback effects to correctly forecast healthcare costs. Conclusion Mixed-method costing used within health system models, such as TLO, is a feasible and robust method for estimating healthcare delivery costs. This approach can provide valuable insights for health sector planning and resource allocation.

Background Decisions need to be made in all healthcare systems about the allocation of available resources with the aim of improving population health. Evidence is needed for these decisions, which can have enormous consequences for population health, especially in lower-income settings. Methods We address this need using an individual-based simulation model of healthcare need and service delivery that we have developed for Malawi, drawing on demographic, epidemiological and routine healthcare system data (on facilities, staff, and consumables). We compare the model’s simulated outputs with available data and estimate the impact that the healthcare system is having currently. We analyse the effects of improvements in healthcare access, clinician performance and consumables availability. Findings Malawi’s healthcare system averted 40 million Disability-Adjusted Life-Years (DALYs) in the five-year period to end-2019, which is half of the total DALYS that the population (total size: 19 million in 2020) would otherwise incur. This impact is strongly focussed on young children (mediated largely by programmes addressing respiratory infections, HIV/AIDS and malaria) and also by the HIV/AIDS and TB programmes (among adults). More services seem to be delivered than would be expected based on the number of staff and expected time needed for services. Nevertheless, the additional services that are provided (through service times being reduced or additional HCW hours worked) account for half the impact of the healthcare system (i.e., ∼20 million DALYS averted). If system improvements gave ill persons immediate access to healthcare, led to optimal referral and diagnosis accuracy, and eliminated consumable stock-outs, the overall impact of the healthcare system could increase by up to ∼30% (12 million more DALYS averted). Conclusions The healthcare system in Malawi generates substantial health gains to the population with very limited resources. Strengthening interventions could potentially increase these gains considerably and so should be a priority for investigation and investment. A detailed individual-based simulation model of healthcare service delivery is a valuable tool for healthcare system planning and for evaluating proposals for healthcare system strengthening.

Objectives. Preterm infants are at risk of hypothermia. This study described the available infant warming devices (IWDs) and explored the barriers and facilitators to their implementation in neonates in Malawi. Methods. A qualitative descriptive study was conducted among 19 health care workers in Malawi from January to March 2020. All interviews were digitally recorded, transcribed, and managed using NVivo and analyzed using a thematic approach. Results. The warming devices included radiant warmers, Blantyre hot-cots, wall-mounted heaters, portable warmers, and incubators. Inadequate equipment and infrastructure and gaps in staff knowledge and capacity were reported as the main challenges to optimal IWD implementation. Caregiver acceptance was described as the main facilitator. Strategies to optimize implementation of IWD included continuous practical training and adequate availability of equipment and spare parts. Conclusion. Implementation of warming devices for the management of neonatal hypothermia is effective when there are adequate human and material resources.

Although community engagement is increasingly promoted in global health research to improve ethical research practice, there is sometimes a disconnect between the broader moral ambitions for community engagement in the literature and guidelines on the one hand and its rather narrower practical application in health research on the other. In practice, less attention is paid to engaging communities for the ‘intrinsic’ value of showing respect and ensuring inclusive participation of community partners in research design. Rather, more attention is paid to the use of community engagement for ‘instrumental’ purposes to improve community understanding of research and ensure successful study implementation. Against this backdrop, we reviewed the literature and engaged various research stakeholders at a workshop to discuss ways of strengthening ethical engagement of communities and to develop context-relevant guidelines for community engagement in health research in Malawi.

Although community engagement is increasingly promoted in global health research to improve ethical research practice, there is sometimes a disconnect between the broader moral ambitions for community engagement in the literature and guidelines on the one hand and its rather narrower practical application in health research on the other. In practice, less attention is paid to engaging communities for the ‘intrinsic’ value of showing respect and ensuring inclusive participation of community partners in research design. Rather, more attention is paid to the use of community engagement for ‘instrumental’ purposes to improve community understanding of research and ensure successful study implementation. Against this backdrop, we reviewed the literature and engaged various research stakeholders at a workshop to discuss ways of strengthening ethical engagement of communities and to develop context-relevant guidelines for community engagement in health research in Malawi.