The role of vitamin D in type 2 diabetes risk : a retrospective study

Abstract

Background The incidence of Type 2 Diabetes Mellitus (T2DM) in the world rose drastically in the past ten years. Identification of risk factor of T2DM is important to predict and/or prevent the disease. Serum vitamin D level has been proposed as a protective factor against T2DM. However, the majority of the evidences were obtained in population from Western countries. The objective of this study is to examine the association of serum vitamin D level with T2DM in Hong Kong.

Methods This was a retrospective analysis of 1417 participants of the Hong Kong Osteoporosis Study,1157 participants of them were analyzed (Figure 4). The diabetes status at baseline was ascertained from a structured questionnaire, whereas the incident diabetes was ascertained using the electronic medical record (EMR): Clinical Data Analysis and Reporting System (CDARS) of the Hong Kong Hospital Authority. The relationship between vitamin D and T2DM was evaluated using multivariable binary logistic regression and Cox proportional hazard regression. Subgroup analysis was also performed by vitamin D quartile, sex, BMI status, vitamin D insufficiency status and vitamin D deficiency status.

Results The median follow-up of the study participants was 10.7 years. In binary logistic regression, serum vitamin D level was inversely associated with T2DM in fully adjusted model inall studied participants. In vitamin D insufficient subgroup analysis, serum vitamin D level was inversely associated with T2DM in crude model and fully adjusted model in vitamin D sufficiency subgroup. In Cox proportional hazard ratio, serum vitamin D level was associated with incident T2DM in the highest vitamin D quartile. However, the result became insignificant after further adjustment.

Conclusion Inconsistent results were observed between serum vitamin D level and T2DM. Thus, future study should focus on people with vitamin D sufficiency and it is required to re-examine the findings. The discrepancy between findings from the Cox proportional hazard regression and binary logistic regression could be due to the fact that serum vitamin D is a marker of the presence of T2DM, rather than a predictor; or the time of T2DM ascertainment was not accurate. Future study in other Chinese population is required to provide more information on the role of vitamin D in T2DM.