Prevalence of vitamin D deficiency and its association with diabetes in a South-Asian population

Abstract

Introduction: Sri Lanka has been experiencing rapid urbanization, with ~30% of the population residing in urban areas. We report the age and sex-specific prevalence of dysglycaemia and vitamin D (VitD) status, along with the association between the two in an urban community in Sri Lanka. Methodology: Using a stratified random sampling method, 369 subjects (116 men and 253 women) aged 18 years and above, were tested for 25OH VitD3, 75 g OGTT, and HbA1c. Demographic, anthropometric, educational, and social details were recorded using a standard proforma. Results: The age and sex adjusted prevalence of VitD deficiency (<20 ng/ml) was 57.2% and VitD insufficiency (20–30 ng/ml) was 31%. The cumulative prevalence of VitD deficiency and insufficiency was 88.2%. Age and sex adjusted overall prevalence of diabetes was 26.9% and pre-diabetes was 32.3%. The cumulative prevalence of diabetes and pre-diabetes was 59.2%. Although, not statistically significant, the highest prevalence of VitD deficiency was found in the young adults of 18–40 years (64.2%). Females had a significantly higher prevalence of VitD deficiency at 63.7% ( P <0.000). There was no significant difference in the different income groups. However, people of Sinhalese ethnicity had a higher prevalence of VitD deficiency (62%). Bivariate analysis using ANOVA t -test, to detect correlation between socio-demographic factors and VitD status found statistically significant association with female sex and VitD deficiency ( P <0.01) Age and income status showed no statistical correlation with the VitD status. VitD status did not show a statistically significant correlation to type 2 diabetes mellitus or pre-diabetes ( P =0.977 and P =0.972). Conclusion: This study shows a high prevalence of dysglycaemia, VitD deficiency/insufficiency in urban Sri Lanka. Dysglycaemia was seen in half the population, with a large pool of subjects with pre-diabetes. Only 11.8% of the population was VitD replete. Females had a significantly higher prevalence of VitD deficiency. We could not detect a statistically significant correlation between VitD deficiency and dysglycaemia.