Zhilong Cai , Shuoyu Rui , Jianhua Chen , Nanqu Huang , Yong Luo , Fei Feng
{"title":"Association between diabetes status and falls: a 9-year prospective cohort study using the China health and retirement longitudinal study","authors":"Zhilong Cai , Shuoyu Rui , Jianhua Chen , Nanqu Huang , Yong Luo , Fei Feng","doi":"10.1016/j.deman.2025.100282","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Falls represent a significant health burden among individuals with diabetes, yet the long-term relationship between diabetes status and fall risk remains inadequately characterized in Asian populations. This study aimed to investigate the association between diabetes status and incident falls among Chinese middle-aged and older adults using nationally representative longitudinal data.</div></div><div><h3>Methods</h3><div>Utilizing China Health and Retirement Longitudinal Study (CHARLS) 2011–2020 data, we included 9553 participants (aged ≥45 years) for a 9-year prospective cohort study. Diabetes was classified as normal glucose metabolism, prediabetes, or diabetes based on self-reported diagnosis, fasting plasma glucose (FPG), or HbA1c levels. Incident falls were assessed via self-reports across four survey waves from 2011 to 2020. Multivariable logistic regression models were to evaluate the independent association between diabetes and falls and subgroup/sensitivity analyses were conducted.</div></div><div><h3>Results</h3><div>The study included 7131 (74.6 %) participants with normal glucose levels, 1254 (13.1 %) with prediabetes, and 1168 (12.2 %) with diabetes. Mean age was 58.1 ± 9.0 years, with 46.9 % males. Fall incidence rates were significantly higher in the diabetes group (55.1 %) compared to prediabetes (48.3 %) and normal glucose groups (47.3 %) (<em>P</em> < 0.001). After full adjustment for potential confounders, diabetes was associated with a 27 % increased risk of incident falls (OR=1.27, 95 % CI: 1.11–1.45, <em>P</em> < 0.001), while prediabetes showed no significant association (OR=0.99, 95 % CI: 0.87–1.12, <em>P</em> = 0.817). Subgroup analyses revealed stronger associations in older adults aged ≥60 years (OR=1.45, 95 % CI: 1.18–1.78) compared to those <60 years (OR=1.17, 95 % CI: 0.98–1.39), with similar effects in both sexes. Sensitivity analyses confirmed the robustness of these findings.</div></div><div><h3>Conclusions</h3><div>Diabetes significantly increases the risk of incident falls among Chinese middle-aged and older adults, with a 27 % higher risk persisting after comprehensive adjustment. This association appears to be specific to established diabetes rather than prediabetes, suggesting a pathophysiological threshold effect. The findings support the integration of fall prevention strategies into routine diabetes care, particularly for older adults, and have important implications for clinical practice guidelines and public health policy in China's rapidly aging population.</div></div>","PeriodicalId":72796,"journal":{"name":"Diabetes epidemiology and management","volume":"19 ","pages":"Article 100282"},"PeriodicalIF":1.4000,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Diabetes epidemiology and management","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2666970625000307","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0
Abstract
Background
Falls represent a significant health burden among individuals with diabetes, yet the long-term relationship between diabetes status and fall risk remains inadequately characterized in Asian populations. This study aimed to investigate the association between diabetes status and incident falls among Chinese middle-aged and older adults using nationally representative longitudinal data.
Methods
Utilizing China Health and Retirement Longitudinal Study (CHARLS) 2011–2020 data, we included 9553 participants (aged ≥45 years) for a 9-year prospective cohort study. Diabetes was classified as normal glucose metabolism, prediabetes, or diabetes based on self-reported diagnosis, fasting plasma glucose (FPG), or HbA1c levels. Incident falls were assessed via self-reports across four survey waves from 2011 to 2020. Multivariable logistic regression models were to evaluate the independent association between diabetes and falls and subgroup/sensitivity analyses were conducted.
Results
The study included 7131 (74.6 %) participants with normal glucose levels, 1254 (13.1 %) with prediabetes, and 1168 (12.2 %) with diabetes. Mean age was 58.1 ± 9.0 years, with 46.9 % males. Fall incidence rates were significantly higher in the diabetes group (55.1 %) compared to prediabetes (48.3 %) and normal glucose groups (47.3 %) (P < 0.001). After full adjustment for potential confounders, diabetes was associated with a 27 % increased risk of incident falls (OR=1.27, 95 % CI: 1.11–1.45, P < 0.001), while prediabetes showed no significant association (OR=0.99, 95 % CI: 0.87–1.12, P = 0.817). Subgroup analyses revealed stronger associations in older adults aged ≥60 years (OR=1.45, 95 % CI: 1.18–1.78) compared to those <60 years (OR=1.17, 95 % CI: 0.98–1.39), with similar effects in both sexes. Sensitivity analyses confirmed the robustness of these findings.
Conclusions
Diabetes significantly increases the risk of incident falls among Chinese middle-aged and older adults, with a 27 % higher risk persisting after comprehensive adjustment. This association appears to be specific to established diabetes rather than prediabetes, suggesting a pathophysiological threshold effect. The findings support the integration of fall prevention strategies into routine diabetes care, particularly for older adults, and have important implications for clinical practice guidelines and public health policy in China's rapidly aging population.