Data Quality and Associated Technological, Organisational, and Behavioural Factors in the Routine Health Information System of Osun State, Nigeria: A Cross-Sectional Study
This cross-sectional study of 22 healthcare facilities in Osun State, Nigeria, reveals that while routine health information system data quality is primarily constrained by accuracy rather than completeness or timeliness, organizational factors influence quality only indirectly by shaping staff behavior and technological adoption, rather than through direct effects.
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 any functioning health system, decisions about where to send doctors, how much medicine to buy, and which diseases to fight depend entirely on the information flowing from clinics to the government. This information is not just a list of numbers; it is the foundation of public health. For these decisions to be sound, the data must be reliable. If a clinic reports that it treated a hundred patients for malaria, but the records are wrong, the government might send too few drugs, leaving people sick, or too many, wasting scarce resources. In many parts of the world, including Nigeria, the systems used to collect this daily information often struggle. The challenge is not just having a system in place, but ensuring the data inside it is accurate, complete, and arrives on time. Researchers have long suspected that the quality of this data depends on three things: the tools people use, the rules and support they receive from their bosses, and the personal habits and attitudes of the staff themselves.
A team of researchers set out to test these ideas in Osun State, Nigeria, by looking directly at how health workers in different types of clinics handle their daily reporting. They visited twenty-two facilities, ranging from small primary health centers to large teaching hospitals, and spoke with nearly three hundred staff members who are responsible for collecting and reporting health data. Instead of just asking the staff what they thought, the researchers also checked the actual paper records and digital reports to see if the numbers matched reality. They wanted to know which factors truly mattered: was it the computers and internet connection, the management style of the hospital, or the willingness and skill of the individual workers?
The study revealed a surprising gap between what people said and what was actually happening. When health workers were asked to rate their own data, they were quite optimistic. They reported that their records were usually complete and arrived on time. However, when the researchers looked at the actual files, a different picture emerged. While the reports were indeed mostly complete and mostly on time, the accuracy of the numbers was a major problem. In the objective check, only about one in five facilities had monthly reports that were fully accurate. Many records contained missing data or numbers that did not match the original patient logs, particularly for common conditions like malaria and prenatal care. This suggests that while the system is good at filling out forms, it is not always good at getting the facts right.
The researchers then investigated why this was happening by looking at the three main areas they had planned to study. They found that the quality of the data was most directly linked to the behavior of the staff and the technology they used. Staff who felt their work was meaningful, who were punctual, and who had a positive attitude toward their tasks produced better data. Similarly, having reliable computers and internet access helped, but this was only a direct factor in the smaller, primary clinics. In the larger, more complex hospitals, having better technology did not automatically lead to better data quality. The study also found that the management style of the facility, such as how supervisors supported their teams, did not directly improve the data.
This last finding was the most complex part of the puzzle. At first glance, it seemed that management support didn't matter at all. But when the researchers looked deeper, they discovered that management support was actually very important, just not in the way they expected. The influence of good management did not reach the data directly. Instead, good management improved the data quality by changing how the staff behaved and by ensuring they had the technology they needed. In other words, a supportive boss helps by making staff more motivated and by making sure the computers work, and it is those changes in behavior and tools that actually fix the data. Without those intermediate steps, simply changing the rules or the organizational structure would not improve the numbers.
The study also highlighted that the needs of different clinics are not the same. The small primary clinics, which often have the least amount of technology, actually produced the best overall data quality. This was because the staff there had strong habits and were closely supervised. The large tertiary hospitals, despite having the best computers and the most resources, produced lower quality data. This suggests that simply giving more expensive equipment to big hospitals will not solve the problem. The large hospitals have more complex workflows and higher patient volumes, which can make data entry difficult even with good tools. The solution for them is not just more hardware, but simplifying the way they work.
For the health officials in Osun State, the path forward is now clearer. The research indicates that the biggest weakness is not in the timeliness of reports or their completeness, but in their accuracy. To fix this, health leaders should focus on specific activities that check the numbers against the original records, rather than just counting how many forms were submitted. They should also invest in technology, but specifically for the smaller clinics where it will have the most immediate impact. Most importantly, they should recognize that organizational changes, like new supervisory rules, will only work if those changes actually make the staff more motivated and better equipped to use their tools. The data quality problem is not just a technical glitch; it is a human one, solved by understanding how people, tools, and leadership interact in the daily rhythm of a clinic.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.