LAUSR

AIMS To investigate the seasonal variations of HbA1c values in a semi-arid climate among type 2 diabetic patients throughout 11 years of care. METHODS A retrospective cohort study of 2860 patients with 61,187 HbA1c measures. We calculated the averages of HbA1c values of all patients and sub-groups defined by age, gender, BMI, ethnicity, insulin use, smoking status and region of residence, matched to the mean temperature and diurnal range over the 30 and 60 days prior to the HbA1c test date. RESULTS We found a sinusoidal seasonal pattern with statistically significant HbA1c levels gaps between the spring peak and the autumn trough (F = 18.4, P < 0.001). The lowest values were in October-November and the highest in March or August with a difference of 0.17-0.31% (2-4 mmol/mol). We found different relations between the mean temperature and HbA1c depending on season, gender, age, ethnicity and smoking habits. CONCLUSION The study identified specific subgroups in which the seasonal influence is particularly marked. A possible explanation is that during hot periods people tend to change their diet and decrease their physical activity, effecting indirectly diabetic imbalance. Recognition of the seasonal variations of HbA1c levels in this geographical region could help physicians with diabetes care and HbA1c control.