Longitudinal changes in glycemic control and associated factors in patients with type 2 diabetes mellitus in a public referral hospital in Peru

Objectives To evaluate longitudinal changes in glycemic control, measured by glycated hemoglobin (HbA1c), among patients with type 2 diabetes mellitus treated at a national hospital in Peru, and to identify factors associated with persistent poor glycemic control (HbA1c ≥ 7%). Methods We conducted a retrospective cohort study including 741 patients aged 18–65 years with T2DM who had at least two HbA1c measurements (baseline and one-year follow-up) recorded between 2015 and 2019. Changes in the proportion of patients with HbA1c ≥ 7% were assessed using McNemar’s test, and individual baseline-to-follow-up HbA1c changes were evaluated. Factors associated with persistent poor glycemic control over time were examined using generalized estimating equation (GEE) models with binomial distribution and logit link. An exchangeable correlation structure and robust standard errors were applied. Progressive models were constructed according to the percentage of missing data, along with a complete-case sensitivity analysis. Results The proportion of patients with HbA1c ≥ 7% decreased from 66.6% to 60.7% (absolute difference: 5.9%; 95% CI: 2.4–9.4; p < 0.001). Overall, 14.3% of patients improved (from ≥7% to <7%), whereas 8.4% worsened (from <7% to ≥7%), yielding a paired odds ratio of 1.71 (95% CI: 1.23–2.37). The median individual baseline-to-follow-up change in HbA1c was 0.0%; however, 35.6% experienced a reduction >0.5%, and 28.3% an increase >0.5%. In GEE models, the odds of HbA1c ≥ 7% were lower at follow-up (adjusted OR: 0.73; 95% CI: 0.61–0.88). Longer diabetes duration (≥10 years), insulin use, and hypertriglyceridemia were associated with higher odds of poor glycemic control, whereas reduced estimated glomerular filtration rate was associated with lower odds. Conclusions Among patients with type 2 diabetes treated at a public hospital in Peru, a modest reduction in the proportion of poor glycemic control was observed after one year of follow-up. Although this reduction was statistically significant, most patients remained above recommended HbA1c targets. These findings underscore persistent gaps in diabetes management and highlight the need for strengthened clinical and public health strategies to improve glycemic control in routine care settings in Peru.