Objective This study aimed to examine the association of visit-to-visit variabilities in metabolic factors with chronic kidney disease (CKD) in Shanghai community residents. Methods We used data from a cohort study of community residents who participated in three examinations in 2008, 2009, and 2013, respectively. Fasting plasma glucose (FPG) level, blood pressure (BP), and lipid levels were determined in 2,109 participants at all three visits, and CKD was evaluated between the second and the third visits. Visit-to-visit variabilities in metabolic factors were described by coefficients of variation (CV) at three visits. A variability score was calculated by adding the numbers of metabolic factors with a high variability defined as the highest quartile of CV. CKD was defined as the estimated glomerular filtration rate < 60 mL/min per 1.73 m2 or urinary albumin-to-creatinine ratio≥ 30 mg/g. Results A total of 200 (9.5％) participants had CKD at the third visit. Compared with the lowest quartile of CV, the highest quartile was associated with a 70％ increased risk of CKD for FPG [odds ratio, OR =1.70; 95％ confidence interval (CI) 1.06–2.72], 62％ for systolic BP (OR = 1.62, 95％ CI 1.04–2.50), and 85％for low-density lipoprotein cholesterol (OR = 1.85, 95％ CI 1.23–2.80). Furthermore, the risk of CKD increased significantly with an increasing variability score. Compared with participants with score 0, participants with scores of 1, 2, and 3 were associated with 58％ (OR = 1.58, 95％ CI 1.08–2.32), 121％(OR = 2.21, 95％ CI 1.40–3.49), and 548％ (OR = 6.48, 95％ CI 3.18–13.21) higher risks of CKD, respectively. Conclusion The visit-to-visit variabilities in metabolic factors were significantly associated with the risks of CKD in Shanghai community residents.