The use of theories, models and frameworks to institutionalise routine data-driven decision-making in sub-Saharan African health facilities: a scoping review
This scoping review of 44 studies across fourteen Sub-Saharan African countries reveals that while determinant frameworks dominate efforts to institutionalize routine data-driven decision-making, current applications remain largely operational and lack the dynamic, systems-thinking approach needed to integrate data use into strategic and clinical decision-making over time.
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 hospitals across Sub-Saharan Africa, a quiet revolution has been taking place over the last two decades. Governments and health organizations have poured resources into building digital systems, installing computers, and creating electronic records to track patient care. The goal was simple: if doctors and nurses could easily access accurate information about who was sick, what medicines were needed, and how many patients were being treated, they could make better decisions to save lives. This shift from paper notebooks to digital databases promised to turn raw numbers into a powerful tool for improving health. However, having the data is not the same as using it. Just because a computer holds a record does not mean a manager looks at it to change how a clinic operates. The challenge lies in "institutionalization," a word that simply means making the use of data a natural, unthinking part of daily work, like brushing one's teeth, rather than a special task done only when an inspector arrives.
To understand why this transition has been so difficult, a team of researchers from the KEMRI-Wellcome Trust Research Programme and the University of Oxford set out to examine the tools scientists use to study this problem. They looked at how experts try to explain why some hospitals successfully use their data while others struggle. These experts rely on "theories, models, and frameworks," which are essentially structured maps or checklists that help researchers identify the specific reasons behind success or failure. These maps might point to things like staff training, the quality of the internet connection, or the way hospital managers give orders. The researchers wanted to know: which of these maps are being used most often in Sub-Saharan Africa, and are they actually helping hospitals learn how to make data a permanent part of their decision-making process?
The team conducted a massive search, scanning six major scientific databases for studies published between 2007 and 2025. They narrowed their focus to 44 specific research papers that took place in fourteen different countries across the region. These papers covered a wide mix of methods, including interviews with health workers, surveys, and statistical analyses. The researchers then carefully read through these studies to see how the authors used their theoretical maps. They found that the vast majority of these studies relied on "determinant frameworks." These are specific types of maps designed to list the factors that influence an outcome, such as whether a health worker is motivated, whether the software is easy to use, or whether the hospital leadership supports the effort. The most common maps used were the PRISM framework and the CFIR framework, which together accounted for nearly two-thirds of all the studies reviewed.
What the researchers discovered was a significant gap between identifying problems and solving them. While these maps were excellent at listing the barriers—such as poor internet, lack of training, or heavy workloads—they rarely explained how these barriers interacted with each other over time. For instance, a study might note that a computer is broken and that staff are tired, but it often failed to show how the broken computer makes the staff tired, which then leads to them entering data incorrectly, which in turn makes the managers lose trust in the numbers. The researchers found that most studies treated these factors as a static list of obstacles rather than a dynamic system where one problem feeds into another. Consequently, the evidence showed that while hospitals were getting better at collecting data, they were not getting better at using it to make strategic changes. The data was mostly being used for basic operational tasks, like counting how many patients were seen or checking if drug supplies were running low, rather than for deeper clinical decisions or long-term planning.
The review highlighted that the way data is used in these facilities is often driven by external pressure rather than internal need. Health workers frequently fill out reports because they are required to by national authorities, not because they believe the information will help them treat their patients better. This creates a cycle where data is collected to satisfy a reporting requirement, but it is never returned to the frontline staff in a way that helps them improve their daily work. When the data is poor quality or arrives late, it erodes trust, and staff stop believing that the numbers matter. The researchers noted that simply introducing new technology, like electronic health records, does not fix this. In many cases, the new digital systems were just layered on top of old paper systems, creating extra work without changing the underlying habits or decision-making structures.
Ultimately, the study suggests that the path forward requires a shift in thinking. Instead of just checking off a list of barriers, health systems need to understand the complex web of relationships between technology, people, and rules. The researchers argue that for data to become truly useful, it must be woven into the daily rhythm of the hospital. This means that the people who collect the data must see how it helps them, and the people who make decisions must rely on it to guide their actions. The current evidence shows that while the tools to understand these problems exist, they are not yet being used to their full potential to create lasting change. The future of health care in the region depends not on generating more data, but on building a system where that data is trusted, understood, and acted upon by everyone, from the nurse at the front desk to the director at the top.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.