Aims/hypothesis: Serum albumin concentrations may be associated with future risk of type 2 diabetes, but the epidemiological evidence is limited and uncertain. We prospectively assessed the association between baseline values of serum albumin and incident type 2 diabetes risk in the Kuopio Ischaemic Heart Disease population-based cohort study.
Methods: We analysed the data of 1,785 men aged 42-61 years with no known history of diabetes at baseline. Participants' serum albumin concentrations were measured at baseline. HRs and 95% CIs for type 2 diabetes events were subsequently assessed.
Results: During a mean follow-up of 20.4 years, 382 participants developed diabetes. Serum albumin concentrations were weakly correlated with several established risk factors for diabetes. Serum albumin was approximately linearly associated with type 2 diabetes risk. In analyses adjusted for several conventional risk factors, the HR for type 2 diabetes per 1 SD increase in serum albumin was 1.15 (95% CI 1.03, 1.28; p = 0.016), which persisted after further adjustment for triacylglycerol, C-reactive protein, γ-glutamyltransferase, estimated glomerular filtration rate and total energy intake (HR 1.15; 95% CI 1.02, 1.29; p = 0.018). The findings were generally consistent across several clinical subgroups. Addition of information on serum albumin to a diabetes risk prediction model containing conventional risk factors led to no significant change in C-index (0.0126; 95% CI -0.0055, 0.0306; p = 0.17).
Conclusions/interpretation: A near linear, positive and independent association was found between serum albumin and type 2 diabetes, but this did not improve event discrimination. Further work is warranted to evaluate the causal relevance of these findings.