Diabetic ketoacidosis, commonly abbreviated as DKA, arises when the body cannot use glucose for fuel because insulin is absent or severely deficient. In response, the liver breaks down fatty acids into acidic molecules called ketones, which accumulate in the blood and lower its pH. The resulting biochemical cascade produces dehydration, rapid breathing, abdominal pain, and, in severe cases, cerebral edema, coma, and death. For children with type 1 diabetes, whose insulin-producing pancreatic beta cells have been destroyed by autoimmune attack, DKA is both a frequent presenting feature at diagnosis and a recurring threat throughout life, often triggered by missed insulin doses, infections, or interruptions in care.

The research team, led by Gashaw Tiruneh, Degefu Dana, Wondimagegn Paulos Kumma, and Tamirat Toma of Wolaita Sodo University, employed a census sampling technique, meaning that every eligible child with diabetes seen at the hospital during the ten-year window was included rather than a random subset. The investigators diagnosed DKA using a modified protocol adapted from guidelines issued by the International Society for Pediatric and Adolescent Diabetes, or ISPAD, along with other international recommendations tailored for resource-limited settings. A structured checklist allowed them to extract sociodemographic and clinical information from each record, and binary logistic regression was used to determine which factors were independently associated with the complication, with a p-value below 0.05 considered statistically significant in the final model.

The headline figure is stark: the overall prevalence of DKA was 42.0 percent, with a 95 percent confidence interval spanning 36.6 to 47.5 percent. In other words, the true rate in the underlying population of children served by the hospital most likely lies somewhere between roughly one-third and nearly half of all pediatric diabetes patients. Among the 139 children who developed DKA during the study period, the severity distribution was itself informative. Thirty-eight children, or 27.3 percent of cases, had mild DKA, while 88 children, or 63.3 percent, presented with moderate disease. Thirteen children, or 9.4 percent, suffered severe episodes, the category most closely linked to intensive care needs and long-term neurological consequences.

When the researchers interrogated the data for patterns, four factors emerged as significantly associated with DKA. Children aged ten to fourteen years faced dramatically elevated odds compared with younger age groups, with an adjusted odds ratio of 9.24 and a confidence interval of 4.04 to 21.11. This means that after accounting for other variables, adolescents in this bracket were more than nine times as likely to experience ketoacidosis, a finding consistent with the physiological and behavioral turbulence of adolescence, when growth hormone surges increase insulin resistance and emerging independence can disrupt adherence to insulin regimens.

Rural residence doubled the risk, with an adjusted odds ratio of 2.38 and a confidence interval of 1.43 to 3.79. The authors suggest that children living far from the tertiary hospital may encounter barriers to regular follow-up, insulin supply, and early recognition of warning signs. Distance to care is a recurring theme in pediatric diabetes research from low- and middle-income countries, where long travel times, transportation costs, and fragmented supply chains for insulin and blood glucose testing strips can turn a manageable chronic condition into a recurring emergency.

A previous history of DKA also predicted future episodes, with an adjusted odds ratio of 2.52 and a confidence interval of 1.21 to 5.27. This pattern reflects a well-documented clinical reality: children who have experienced one episode of ketoacidosis are at heightened risk of recurrence, particularly if the underlying causes, whether missed injections, inadequate monitoring, or psychosocial stressors within the family, remain unaddressed. The finding underscores the importance of structured education and follow-up after any DKA admission, rather than treating each episode as an isolated event.

Perhaps the most consequential risk factor was new diagnosis. Children who were newly diagnosed with type 1 diabetes mellitus had an adjusted odds ratio of 5.11, with a confidence interval of 2.65 to 9.84, meaning they were roughly five times more likely to develop DKA than those already established in care. This is a familiar pattern worldwide, since many children reach medical attention only after weeks of escalating thirst, urination, and weight loss have progressed to frank ketoacidosis. But the magnitude of the effect in this Ethiopian cohort points to delayed diagnosis as a major driver of the overall burden, a problem that community awareness campaigns and frontline health worker training could potentially mitigate.

The study’s ten-year span gives it unusual weight for a single-institution analysis. By capturing a full decade of admissions and follow-up visits, the design smooths out seasonal fluctuations and short-term disruptions that can distort shorter studies. The census sampling approach eliminates selection bias within the hospital population, although the authors and readers alike must remember the inherent limits of retrospective record review: the analysis depends on the completeness and accuracy of documentation, and the findings describe children who reached this particular tertiary hospital, not necessarily all children with diabetes in the surrounding region. Children who never made it to care at all remain invisible to any hospital-based study, a silent denominator that may conceal even greater unmet need.

The implications of the study are practical and urgent. The authors conclude that healthcare providers should pay particular attention to adolescents, children from rural areas, those with a previous history of DKA, and those with newly diagnosed diabetes. Targeted interventions could include intensified diabetes self-management education for teenagers, structured follow-up and insulin continuity programs for rural families, and community-level strategies to shorten the interval between symptom onset and diagnosis. As type 1 diabetes incidence continues to rise globally, including in sub-Saharan Africa, studies like this one provide the local evidence base that health systems need to allocate scarce resources where they will save the most young lives. The message from southern Ethiopia is clear: without deliberate attention to the highest-risk groups, diabetic ketoacidosis will remain a leading cause of preventable harm among children with diabetes.

Subject of Research: Prevalence and risk factors of diabetic ketoacidosis among children with diabetes mellitus in Southern Ethiopia

Article Title: Prevalence and associated factors of diabetic ketoacidosis among children with diabetes mellitus at a tertiary hospital in Southern Ethiopia: a 10-year retrospective study

Article References: Tiruneh, G., Dana, D., Kumma, W. P., & Toma, T. (2026). Prevalence and associated factors of diabetic ketoacidosis among children with diabetes mellitus at a tertiary hospital in Southern Ethiopia: a 10-year retrospective study. BMC Endocrine Disorders. https://doi.org/10.1186/s12902-026-02612-4

Image Credits: AI Generated

DOI: 10.1186/s12902-026-02612-4

Keywords: diabetic ketoacidosis, type 1 diabetes, children, Ethiopia, pediatric endocrinology, retrospective study, prevalence, risk factors, rural health, insulin, ISPAD guidelines, hospital-based study