McCarthy · EClinicalMedicine 2022 · prospective cohort study · n=5272

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).

Cited 37 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective observational cohort study with 4-year follow-up

PubMed 36147626 · doi:10.1016/j.eclinm.2022.101654 · record verified 2026-08-30

What was done

This prospective cohort study analysed data from The Irish Longitudinal Study on Ageing (TILDA) in community-dwelling adults aged ≥50 years residing in Ireland. Plasma 25-hydroxyvitamin D concentrations were measured at Wave 1 baseline (2009–2011, n = 5272). Participants who completed health assessments were re-evaluated at Wave 3 (2014–2015, n = 3828; 4-year follow-up). Logistic regression models evaluated associations between baseline vitamin D status and both prevalent diabetes at baseline and incident diabetes or prediabetes at follow-up, adjusting for age, sex, education, body mass index, smoking history, physical activity, statin use, and season of blood sampling.

What was found

Cross-sectionally, deficient baseline vitamin D was associated with higher prevalence of diabetes (Relative Risk Ratio [RRR] 1.5, 95% CI: 1.03 to 2.18; p = 0.037). In prospective analyses evaluating status 4 years later, individuals with baseline vitamin D <30 nmol/L had a 62% increased likelihood of developing prediabetes compared to those with ≥75 nmol/L (RRR 1.62, 95% CI: 1.12 to 2.35; p = 0.011). The rate of progression from prediabetes to diabetes between Wave 1 and Wave 3 was 32.5%. The study was underpowered to detect effects on incident diabetes directly due to low event incidence.

Why it matters

The study shows a longitudinal link between low vitamin D concentrations and incident prediabetes in an older European population, highlighting a potential target for preventive metabolic strategies.

Limits

The observational design cannot establish causality, and residual confounding cannot be ruled out. The study was underpowered for incident diabetes, attrition was substantial between Wave 1 (n = 5272) and Wave 3 (n = 3828), and findings from this cohort of Irish adults aged ≥50 may not generalize to younger or non-Caucasian populations.

Cited by