Letter to the editor regarding “The utilization of health management information systems (HMIS) and associated factors in public and private health facilities in Mogadishu,Somalia”

To the Editor,

We read with interest the article titled “The Utilization of Health Management Information Systems (HMIS) and Associated Factors in Public and Private Health Facilities in Mogadishu, Somalia” recently published in SAGE Digital Health (2025). ¹ The topic is important and relevant to Somalia's health system. While we value the authors’ efforts, we would like to share a few methodological and analytical concerns that we believe need further clarification.

First, there is an inconsistency in the study's stated objective. While the abstract describes the aim as assessing HMIS utilization in both public and private health facilities in Mogadishu, the background and subsequent sections narrow the focus to only public facilities. This contradiction creates confusion about the actual scope of the study and undermines the clarity needed for methodological coherence and valid interpretation of the results. A consistently stated objective is essential to ensure alignment across the study design, sampling, and analysis.

Second, the geographical description of Mogadishu contains a factual error. This article incorrectly states that Mogadishu borders regions like “Middle Juba in the south” and “Middle Juba in the north,” which is geographically incorrect.

Additionally, the sampling design lacks clarity. The authors do not explain how the selected health facilities (six hospitals and two health centers) were determined to be representative of all public and private institutions across the Mogadishu municipality. The basis for proportional allocation of respondents and the sampling interval was also not described in sufficient detail, raising questions about the generalizability of the findings.

Moreover, the sample size determination section contains technical inaccuracies. The authors refer to a “10% response rate,” when it is evident from the context that they meant a “10% nonresponse rate.” These terms are not interchangeable and should be used precisely. Additionally, a design effect of 1.5 is mentioned but neither justified nor applied in the sample size calculation. Such inconsistencies undermine the rigor of the sampling approach. ²

The study's sampling procedure also lacks adequate explanation regarding how individuals were selected within facilities and whether sampling weights or stratification were applied. It is unclear how the investigators ensured randomness or prevented selection bias during respondent recruitment. ³

Turning to the results, this section contains errors in content and interpretation. The section is misleadingly titled “Factors Associated with the Infection Prevention Practice,” which is unrelated to the actual content on HMIS utilization. More importantly, several reported findings contradict logic and prior research.^4,5 For example, the study states that health professionals with data analysis skills were 91.1% less likely to utilize HMIS (AOR = 0.089; 95% confidence interval (CI): 0.042–0.185), which is implausible and suggests potential issues with variable coding or model specification. The authors also report a CI for HMIS utilization as 95% CI: 1.53–1.62, which is statistically invalid since CIs for proportions must fall between 0 and 1.

In addition, the study omits a rationale for excluding other health facilities in Mogadishu, which affects the comprehensiveness of the sample. Finally, the discussion section lacks reflection on the study's limitations regarding generalizability and potential confounders.