Association of frailty index with new-onset diabetes: from the China Health and Retirement Longitudinal Study (CHARLS).

Background: To investigate the association of frailty status and its changes with new-onset diabetes.

Methods: A total of 4638 participants from the China Health and Retirement Longitudinal Study (CHARLS) were included. Frailty status was assessed by the frailty index (FI) and categorized as robust, pre-frail, and frail. Changes in frailty were assessed based on frailty status at wave1 and wave3 of CHARLS. New-onset diabetes was identified by self-reported physician-diagnosed or diagnosed by glucose or glycosylated haemoglobin during follow-up period. Logistic regression was used to examine the association of frailty and outcomes.

Results: 51.6%, 38.1% and 10.3% of the individuals were respectively classified as robust, pre-frail, and frail at baseline. The risk of new-onset diabetes was significantly higher in the pre-frailty [odds ratio (OR) (95% confidence interval (CI)): 1.326 (1.101-1.597), p = 0.003)] and frailty [OR(95% CI): 1.721 (1.304-2.271), p < 0.001)] than the robust. A total of 3145 participants were included in the changes of frailty status analyses. Compared with the stable robust individuals, individuals who developed from robust to frailty status [OR (95%CI): 3.752 (1.647-8.547), p = 0.002] had an elevated risk of new-onset diabetes. In addition, participants who ever had a robust status in the two surveys had a significantly lower risk of new-onset diabetes compared to those who never had a robust status [OR (95%CI) 0.666 (0.483-0.920), p = 0.014)].

Conclusion: Frailty status is significantly associated with the risk of new-onset diabetes. The progression from robust to frailty or pre-frailty increased the risk of new-onset diabetes.