Using routinely collected health data to assess health facilities’ ability to provide child health services in sub-Saharan Africa: a case study from Ghana’s Eastern region
This study demonstrates that routinely collected health data from Ghana's Eastern region can be effectively used to construct a multidimensional child health service provision index, enabling the ongoing identification of underperforming facilities and offering a replicable, cost-effective alternative to traditional cross-sectional assessments across sub-Saharan Africa.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the vast and often under-resourced landscapes of sub-Saharan Africa, the health of a child frequently depends on the distance they can travel and the specific capabilities of the clinic they reach. For decades, health officials have relied on periodic, expensive surveys to understand which clinics are ready to treat patients. These surveys, known as health facility assessments, involve teams visiting a small sample of buildings to check for medicines, equipment, and staff. While useful, these snapshots are slow to update and often miss the day-to-day reality of thousands of facilities. In contrast, every time a patient visits a clinic, a record is created. In many African nations, these routine records are now collected digitally into a central system, creating a continuous, living stream of information about how health services are actually used. The challenge has been learning how to turn this massive, dry stream of numbers into a clear picture of which facilities are truly performing well and which are struggling, without needing to send a team to every single door.
A team of researchers turned to this stream of routine data to solve a specific puzzle in Ghana's Eastern region. They wanted to know if they could build a reliable scorecard for child health services using only the records already being generated by the clinics themselves. Instead of waiting for a new survey, they looked at the digital logs from 450 health facilities over the course of 2020. Their goal was to create a single number for each facility that reflected how many different types of child health services it was actually providing. They focused on a wide range of care, from routine vaccinations and nutrition checks to treating diarrhea and managing hospital admissions for young children. By analyzing whether a facility provided at least one of these services in every single quarter of the year, they could determine if the service was being offered consistently or if it was sporadic and unreliable.
The researchers used a method that groups similar patterns together to simplify the complex data. They found that the ability to provide care naturally separated the facilities into two main groups: hospitals and smaller, community-level clinics. Hospitals, which are larger and better equipped, consistently provided a wider array of services, including inpatient care for children, which smaller clinics simply cannot offer. This difference was the most significant factor in their scoring system. However, the study revealed that the story did not end there. Even among facilities of the same type, such as the many small community clinics scattered across the region, there were clear differences in performance. Some clinics were providing a broad range of care, while others were falling short, offering only a fraction of the services they were expected to provide.
The resulting scorecard, which ranged from zero to about 23 points, showed a clear divide. Hospitals generally scored high, with a median score of nearly 20, reflecting their comprehensive role in the health system. In contrast, the smaller facilities, including community health compounds and clinics, clustered around a median score of roughly 11. This gap was not just about the size of the building; it highlighted a variation in what was actually happening inside. The researchers identified that while most facilities were doing a decent job, a specific group of community clinics and health centers fell into the bottom ten percent of their own type. These were the facilities that, despite being open and operational, were not consistently delivering the full range of child health services expected of them.
This approach offers a powerful new way for health officials to monitor the system. Unlike traditional surveys that might take years to repeat, this method uses data that is already being collected every month. It allows decision-makers to see, in near real-time, which facilities are underperforming and need help. The study suggests that by simply looking at what services are being reported as used, officials can identify gaps in care without needing to visit every site. This is particularly important in a region where resources are tight and every dollar spent on monitoring must count. The researchers found that this routine data could explain more than half of the differences in service provision across the entire network, proving that the daily records of patient visits hold a deep truth about the health system's capacity.
The findings also challenge the idea that all clinics of the same type are equal. In the past, maps of health facilities often assumed that every clinic in a district offered the same level of care. This study shows that assumption is incorrect. Even within the same category of small clinics, some are providing a robust set of services while others are barely functioning. By pinpointing exactly which facilities are lagging, health leaders can target their support more effectively. They can send medicines, training, or resources specifically to the clinics that need them most, rather than spreading resources thinly across the board. This precision is a step toward ensuring that every child, regardless of where they live, has access to the care they need.
The study was conducted during 2020, a year marked by the early stages of the global pandemic, yet the data remained robust enough to reveal these patterns. The researchers noted that while the pandemic caused some fluctuations in patient numbers, the consistency of service provision remained a reliable indicator of a facility's capability. They also acknowledged that the system is not perfect; some private clinics and public facilities were excluded because they did not report their data consistently. This highlights that for this method to work everywhere, every facility must be part of the reporting system. However, where the data exists, it provides a clear, objective view of the health landscape.
Ultimately, this work demonstrates that the digital footprints left by patients visiting clinics can tell a richer story than periodic surveys alone. It shows that by listening to the routine data, health systems can move from guessing about performance to knowing it. The ability to generate these scores continuously means that the health system can adapt quickly, identifying problems as they emerge rather than waiting for the next survey cycle. For the children in the Eastern region of Ghana, and potentially for millions more across sub-Saharan Africa, this means a future where health services are not just available on a map, but are verified, monitored, and strengthened based on the reality of what is happening every day.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.