These records are essential for monitoring health services and identifying where support is most urgently needed.

However, collecting and reporting data does not automatically mean that the information is being used.

This distinction inspired our study, “Utilization of routine health information system and its determinants among healthcare professionals in public health facilities of Banadir region, Somalia: A cross-sectional study,” published in BMC Health Services Research.

Our research asked a practical question: Are healthcare professionals using the routine information they collect to guide decisions and improve health services?

Why we undertook the study

Somalia has made important progress in rebuilding its health information system after decades of conflict, institutional disruption and dependence on fragmented humanitarian reporting arrangements. The country established a dedicated health information system function and adopted the District Health Information Software 2, commonly known as DHIS2, as its national routine reporting platform.

These developments have made it increasingly possible to bring health information from different facilities into a common system. Yet the existence of registers, reporting forms, computers and digital platforms does not guarantee that data will influence decisions.

A health worker may complete a monthly report and submit it on time without ever analysing the figures. A facility may report the number of children vaccinated without examining how many children failed to complete their vaccination schedule. A hospital may record maternal and newborn outcomes without comparing its current performance with previous months.

When information flows upwards only for reporting, its potential value at the facility level is lost.

As researchers familiar with Somalia’s health system, we recognized that reporting completeness was receiving increasing attention, but much less was known about what happened after the reports were produced. There was limited published evidence on whether healthcare professionals interpreted routine information, discussed it with colleagues or used it to plan services, allocate resources and identify performance gaps.

Our study was designed to address that evidence gap.

Looking beyond computers and reporting forms

We conducted the study between March and June 2024 in 32 public health facilities in the Banadir Region, which includes Mogadishu. The facilities consisted of six hospitals and 26 health centres located across eight randomly selected districts.

A total of 405 healthcare professionals participated.

We used questionnaires and observational checklists based on the Performance of Routine Information System Management, or PRISM, framework. PRISM examines health information systems through three connected dimensions.

The first is technical: are appropriate tools, indicators and analytical methods available?

The second is organizational: do health institutions provide supervision, feedback, resources and clear responsibilities?

The third is behavioural: do healthcare professionals have the knowledge, confidence, motivation and attitudes needed to use information?

This approach was important because weaknesses in health information systems are often described as technology problems. In reality, an information system can fail even when computers and software are available. Staff may not have been trained, indicators may be interpreted differently, supervisors may focus only on report submission, or workers may not believe that analysing data will lead to meaningful action.

What we found

Overall, 68.1% of the participating healthcare professionals demonstrated good routine health information system utilization.

This finding showed that routine information was already supporting several functions within public facilities in Banadir. Participants reported using information for patient treatment, planning, disease prioritization, medicine procurement, resource allocation, performance monitoring, community mobilization and other decisions.

Nevertheless, the result also showed that almost one-third of the participants did not meet the study’s threshold for good utilization.

The figure of 68.1% must be interpreted carefully. It does not mean that 68.1% of all data collected in Banadir were used. It also does not mean that every decision made by these professionals was evidence-based. It represents the proportion of surveyed professionals who met the study’s defined criteria for good use of routine health information.

Several factors clearly distinguished stronger data users from weaker ones.

Healthcare professionals who had not received health information system training were much less likely to use routine information effectively. Lack of training in data analysis was another significant barrier. This demonstrates that knowing how to complete a register is not the same as knowing how to interpret trends, calculate indicators, identify gaps and translate findings into action.

Insufficient supervision was also associated with lower information utilization. Supervision is most useful when it goes beyond checking whether forms have been completed. Healthcare workers need opportunities to review their data with supervisors, correct errors, understand performance patterns and agree on actions.

Non-standardized indicators presented another barrier. When indicator definitions are unclear or applied inconsistently, healthcare professionals may lose confidence in the information and struggle to compare results across departments, facilities or reporting periods.

Attitude also mattered. Professionals with positive attitudes towards data utilization were approximately 2.8 times more likely to demonstrate good use of routine information. Clear understanding of roles and responsibilities was also associated with stronger utilization.

We additionally found that professionals working in hospitals were less likely to demonstrate good utilization than those working in health centres. This finding requires further investigation. Hospitals generally manage larger patient volumes, more departments and more complex reporting requirements. These conditions may increase workloads and create fragmented responsibility for data, but our cross-sectional study could not determine the exact reasons.

The central lesson

The strongest lesson from our research is that digitalization alone will not create a data-use culture.

Computers, dashboards and electronic reporting platforms are necessary, but they are only part of the solution. Effective information use also requires people who understand the indicators, institutions that support analysis and leaders who expect decisions to be justified with evidence.

Health information system performance should therefore not be judged only by the number of reports submitted or the percentage of facilities reporting on time. Health authorities should also ask:

What did the facility learn from its data?

Which performance gap was identified?

What decision was made?

Was the decision implemented?

Did service delivery improve?

These questions move the health system from reporting numbers to using evidence.

What should happen next

Our findings support continuous, competency-based training for healthcare professionals. Training should cover DHIS2 use, standard indicator definitions, basic data-quality checks, trend analysis, data visualization and the application of findings to local decisions.

Supportive supervision and feedback mechanisms should also be strengthened. Facility teams should regularly review their own performance data rather than waiting for regional or national authorities to analyse it for them.

Data-review meetings should result in documented actions, responsible persons and follow-up dates. Indicators should be standardized, and healthcare workers should have clear responsibilities for data collection, validation, analysis and use.

Digital tools must also be accessible and functional. However, equipment should always be accompanied by training, technical support and an organizational expectation that data will be used.

Recognizing the study’s limits

To our knowledge, this was the first study of its kind in Somalia and provides an important baseline. Nevertheless, it was conducted only in public facilities in Banadir, meaning that its findings cannot automatically represent private facilities or other regions of Somalia.

Because the study was cross-sectional, it identified associations but could not prove that training, supervision or positive attitudes directly caused better data use. The self-administered questionnaire may also have been affected by participants providing socially desirable answers.

Future studies should include private and humanitarian facilities, expand to other regions and use qualitative interviews to explore why healthcare professionals use or do not use information. Intervention studies are also needed to test whether strengthened training, supervision and feedback produce sustained improvements.

Every number represents a person

Routine health information is sometimes treated as an administrative product: a form to complete, a deadline to meet or a report to send.

But behind every number is a person.

A vaccination figure represents a child who was protected—or missed. An antenatal-care figure represents a pregnant woman who received care—or did not. A disease notification may be an early warning that allows authorities to act before an outbreak spreads.

The true value of health data is not found in the register, spreadsheet or dashboard where it is stored. Its value is realized when reliable information reaches people who can act on it.

Somalia has made progress in building the structures needed to collect health information. The next challenge is to ensure that the information is consistently understood, trusted and used.

Read the published article: Utilization of routine health information system and its determinants among healthcare professionals in public health facilities of Banadir region, Somalia: A cross-sectional study