Aims: High physical capability reduces risk of diabetes, but the association of its changes with risk of diabetes and glycemic control is unclear in older adults. This study aimed to quantify their association. Methods: A total of 1,667 participants without diabetes and aged >= 60 years from the China Health and Retirement Longitudinal Study were included and followed over 4 years. Physical capability was objectively measured at baseline and 4-year later. Logistic regression models were used to assess the association of changes in physical capability with risk of diabetes. Results: During follow-up, 160 participants developed diabetes. None of the changes in physical capability expressed in continuous scales or in tertiles was associated with risk of diabetes in unadjusted or adjusted models (all P > 0.05), and no favorable joint effects were observed. Changes in physical capability were not associated with changes in fasting plasma glucose or hemoglobin A1c. Yet increases in walking speed or grip strength were related to reduced metabolic score for insulin resistance (both P < 0.03). Conclusions: Increases in objectively-measured physical capability were not associated with reduced risk of diabetes but may ameliorate insulin resistance in older adults. Studies with longer follow-up periods are required to confirm these findings. (C) 2021 Elsevier B.V. All rights reserved.