The Relationship between Red Blood Cell Distribution Width and Incident Diabetes in Chinese Adults: A Cohort Study

Background. Previous studies reported the controvertible association between red blood cell distribution width (RDW) and diabetes. The aim of this study is to explore whether RDW is associated with incident diabetes. Methods. We performed this cohort study in 16,971 Chinese adults (9,956 men and 7,015 women, aged 43.3 +/- 12.8 years). The level of RDW was measured at baseline (2014). All the participants were further classified into four quartile groups based on baseline RDW. Fasting blood glucose (FBG) and glycated hemoglobin A1c (HbA1c) were measured annually during follow-up (2014-2019). Diabetes was diagnosed if either FBG >= 7.0 mmol/L or HbA1c >= 6.5%. We used the Cox proportional hazards regression model to evaluate the association between baseline RDW and incident diabetes. Results. We identified 2,703 new cases of diabetes during five-year follow-up. The incidence was 15.9%. Comparing with participants in the lowest quartile group (reference group), the adjusted hazard ratios (HR) for the risk of diabetes were 1.31 (95% CI: 1.16, 1.48) for the highest quartile group (p trend<0.001), after adjustment for potential confounders. Further adjusting baseline FBG and HbA1c did not materially change the association between RDW and incident diabetes. Each unit increase of RDW was associated with a 16% higher risk of incident diabetes (HR=1.16, 95% CI: 1.06, 1.26) in a fully adjusted model. Sensitivity analysis generated similar results with prospective analyses after excluding aged participants, participants who are overweight and with obesity, participants with elevated blood pressure, participants with decreased eGFR, and those with anemia at baseline. Conclusions. High RDW was associated with high risk of developing diabetes in Chinese adults. As RDW is an inexpensive, noninvasive, and convenient indicator, RDW might be considered for inclusion in the risk assessment of high-risk groups of diabetes.