Circannual variation in glycohemoglobin in Singapore

BackgroundGlycohemoglobin is routinely used for the monitoring of glycemic control in patients with diabetes mellitus. The yearly pattern of glycohemoglobin was examined in a tropical country to assess the effect of festivities without the confounding effect of seasonal variation in temperature and climate.MethodsDetails of all outpatient HbA1c measurements with identified ethnicity over 3 y were examined with respect to date of sampling and compared to the dates of public holidays.ResultsThere were 40267 records available. There was a circannual pattern with a peak in February/March, a minor peak in July/August and a nadir in November/December with an amplitude of 0.5%. Date of sampling was a significant factor in predicting HbA1c > 8.0% despite controlling for race, sex and age. Odds ratios (vs. Quarter 4) were: Quarter 1 1.34 (1.26-1.43); Quarter 2 1.10 (1.03-1.17); Quarter 3 1.11 (1.04-1.19). There was a positive correlation between the number of festive days in the preceding 3 months and the monthly mean HbA1c value. Malay and Indian patients had significantly higher HbA1c values than Chinese.ConclusionsThere is a circannual pattern of HbA1c values in adult outpatients in Singapore, reflecting the celebration of holidays in the preceding 3 months. Clinicians seeking to improve glycemic control should remember the effect of festive occasions.