Association between vitamin D deficiency and the risk of prevalent type 2 diabetes and incident prediabetes: A prospective cohort study using data from The Irish Longitudinal Study on Ageing (TILDA)

Background It is hypothesized that vitamin D contributes to the aetiology of type 2 diabetes mellitus (diabetes). This study’s objective was to examine the relationships between baseline vitamin D status (as measured by plasma 25- hydroxyvitamin D concentration) and both prevalent diabetes and prospective risk of developing diabetes, including prediabetes, in a population with historically low levels of vitamin D.

Methods In this prospective cohort study, data from The Irish Longitudinal Study on Ageing (TILDA), a nationally representative cohort of adults aged ≥50 years residing in Ireland were analysed, including wave 1 (October 2009 −June 2011) (n = 5272) and wave 3 (March 2014−October 2015) (n = 3828). Those aged <50 years at baseline or who did not complete the health assessment were excluded. Logistic regression models examined the associations between baseline vitamin D concentration (nmol/L) with prevalent diabetes status and incident diabetes/prediabetes collected at a 4-year follow-up. Models were adjusted for age, sex, education, body mass index, smoking history, physical activity, use of statins, and the season in which the vitamin D concentration was sampled. 

Findings Deficient baseline vitamin D concentration was cross-sectionally associated with an increased likelihood of having prevalent diabetes (Relative Risk Ratio [RRR] 1.5, 95% CI: 1.03, 2.18; p = 0.037). In longitudinal analyses evaluating diabetes status 4 years later, there was a 62% increased likelihood (RRR: 1.62, 95% CI: 1.12, 2.35; p = 0.011) of developing prediabetes for those with vitamin D <30 nmol/L compared to those with ≥75 nmol/L. The rate of progression from prediabetes to diabetes between wave 1 and 3 was observed to be 32.5%. 

Interpretation Those with lower concentrations of vitamin D, as measured by 25-hydroxyvitamin D, may have different risk profiles with regards to their glycaemic status. Our study had limited power due to the low incidence of diabetes but showed strong associations with incident prediabetes, so further research is required. Optimising vitamin D status at a population level may significantly reduce diabetes.