Glycaemic control remained poor despite free insulin and supplies in Nairobi clinic
In a Kenyan tertiary diabetes clinic, type 1 diabetes mellitus control in children and adolescents was strikingly weak even in a setting supported by the Life for a Child programme. Among 160 participants cared for at Kenyatta National Hospital between September 2025 and May 2026, only 9 children and adolescents, or 5.6%, reached the HbA1c target of 7.5% or lower. The mean HbA1c was 11.02%, with an SD of 2.29, a level that signals prolonged hyperglycaemia rather than isolated lapses in day-to-day management.
The pattern extended beyond laboratory averages. Of 133 patients with glucometer data available, only 5, or 3.8%, had at least 70% of readings inside the target glucose range of 3.9 to 10.0 mmol/L. In practical terms, most of the cohort was living outside recommended glycaemic limits most of the time. For children and adolescents, that degree of persistent exposure creates the biological conditions for early microvascular injury, impaired growth, and long-term cardiometabolic risk.
Self-monitoring frequency carried the clearest signal
Two factors stood out on bivariate analysis: household income and self-monitored blood glucose frequency. Household income was associated with glycaemic control at p=0.026, and SMBG frequency was associated at p=0.018. The picture changed once the investigators moved to multivariable modelling designed for a small number of participants with good control.
Frequency of self-monitoring by glucometer emerged as the strongest independent factor. Children and adolescents who tested at least four times per day had lower odds of poor glycaemic control, with an adjusted OR of 0.19, 95% CI 0.04–0.84, and p=0.028. The household income effect weakened after adjustment, which suggests that economic status may matter partly through a more immediate pathway: the ability to test often enough to guide insulin dosing, food intake, and correction of hyperglycaemia.
This matters physiologically. Insulin therapy in type 1 diabetes only works well when it is paired with repeated feedback on current glucose exposure. Without testing, dose adjustments become guesswork, and the risks of both chronic hyperglycaemia and unrecognized hypoglycaemia rise. Frequent SMBG gives families a usable signal, turning insulin management from a static prescription into an adaptive process.
Programmatic support did not translate into adequate control
Kenya’s Life for a Child programme provides free insulin, glucose-monitoring supplies, and specialist education, yet glycaemic outcomes in this cohort remained poor. That gap suggests the central barrier is not simply whether a child has access to diabetes care on paper. Consistent use of supplies, sustained caregiver support, and daily self-management behaviors appear to be the real bottleneck.
The study was cross-sectional, so it cannot prove causality. Still, the association between frequent self-monitoring and better control fits clinical logic. More testing gives more opportunities to detect rising glucose early, adjust insulin before prolonged excursions accumulate, and reinforce food and activity decisions around measured rather than assumed values. When testing is infrequent, even well-intentioned families lose the ability to titrate treatment in a disciplined way.
The investigators used Firth’s penalized-likelihood logistic regression because only a small number of participants achieved good control. That analytic choice is appropriate when standard regression would be unstable in sparse data. It also reflects the underlying clinical reality: good control was rare enough that ordinary methods would struggle to estimate effects reliably.
Why income mattered, and why it did not explain everything
Household income likely reflects more than cash in the home. It can shape transport to clinic, food security, parental availability, device replacement, and the ability to sustain a regular testing routine. Families under financial pressure may also face competing priorities that disrupt structured diabetes care. Even so, income lost its independent association once SMBG frequency was entered into the model, indicating that the testing routine itself was the more proximal driver of better control in this cohort.
That finding should not be read as evidence that poverty is irrelevant. It suggests instead that the mechanisms through which poverty affects control may operate through daily management constraints, especially the ability to test glucose often enough to act on results. If test strips are scarce, broken glucometers go unreplaced, or routines collapse under logistical strain, insulin therapy becomes less precise and glycaemic exposure worsens.
For clinicians, the implication is concrete. Support for self-monitored blood glucose cannot be treated as a minor add-on to insulin provision. It is part of the treatment itself.
A narrow evidence base still leaves major unanswered questions
The study included one tertiary hospital, and the sample size was modest. It also relied on cross-sectional assessment, so it cannot separate cause from consequence. Children with better control may be more likely to test often because their families are already more organized, or frequent testing may help produce better control, or both may be true at once. The design cannot resolve that directionality.
The lack of peer review matters in the usual way for a preprint, but the clinical pattern is neither surprising nor trivial. It is consistent with the central role of glucose monitoring in paediatric diabetes care and with the known barriers that shape chronic disease management in resource-constrained settings. The very low proportion of children reaching target HbA1c suggests that access to insulin alone is not enough when monitoring is irregular.
This preprint points toward a simple but demanding priority for paediatric diabetes services in similar settings: protect the testing supply chain, teach families how to use the data, and build follow-up systems that support repeated SMBG rather than sporadic checks. The next step is not another slogan about access. It is longitudinal evidence that can show whether sustained monitoring support improves HbA1c and prevents the complications that start long before adulthood.