Background The association between occupational physical activity (OPA) and cardiometabolic risk factors remains controversial, potentially due to differences in the associations between OPA and various cardiometabolic indicators, as well as the lack of a clearly defined optimal OPA range for multiple-indicator synergistic benefits. Objective To investigate the relationship between OPA and cardiometabolic risk factors in individuals at high risk of cardiovascular disease (CVD) in Hubei Province, and to explore an optimal OPA range for multi-indicator improvements. Methods Data were derived from the Hubei Province dataset of the China Health Evaluation And Risk Reduction Through Nationwide Teamwork from 2015 to 2023, including 19028 high-risk CVD individuals. OPA (metabolic equivalent·h·d⁻¹, MET·h·d⁻¹) was assessed using the China Kadoorie Biobank physical activity questionnaire, anthropometric measurements (height, weight, waist circumference) and cardiometabolic indicators (blood pressure, fasting blood glucose, lipid profiles) were measured. Multivariate logistic regression was used to analyze the association between logOPA quartiles (Q1-Q4) and abnormal cardiometabolic indicators, while restricted cubic spline (RCS) models were employed to explore their dose-response relationships. Subgroup and interaction analyses were also conducted by gender. Results The median OPA of all participants was 12.86 MET·h·d⁻¹ (male: 14.40 MET·h·d⁻¹, female: 11.14 MET·h·d⁻¹). After adjusting for confounders, compared with the Q1 group, logOPA in the Q4 group was associated with 20% (OR=0.80, 95%CI: 0.72, 0.89), 14% (OR=0.86, 95%CI: 0.78, 0.96), and 13% (OR=0.87, 95%CI: 0.79, 0.95) reduction in the risk of abdominal obesity, low high-density lipoprotein cholesterol (HDL-C), and metabolic syndrome (MS), respectively, and 33% (OR=1.33, 95%CI: 1.18, 1.49), 33% (OR=1.33, 95%CI: 1.18, 1.50), and 38% (OR=1.38, 95%CI: 1.21, 1.57) increase in the risk of high total cholesterol (TC), high low-density lipoprotein cholesterol (LDL-C), and high non-HDL-C, respectively. Similar trends were observed in males and females, with significant positive associations of logOPA with high TC and high non-HDL-C in males (P_interaction<0.05). The risk of low HDL-C decreased rapidly when logOPA was more than 2.213 (OPA>9.14 MET·h·d⁻¹), showing an inverted L-shaped trend. The risk of high TC, high LDL-C, and high non-HDL-C increased beyond logOPA thresholds of 3.244, 2.476, and 3.377 (OPA >25.64, 11.89, and 29.28 MET·h·d⁻¹), respectively, showing a J-shaped trend. However, logOPA exhibited a U-shaped nonlinear dose-response relationship with blood pressure and fasting blood glucose abnormalities (P_non-linear<0.05) , with minimal risks at logOPA 2.969 and 2.372 (OPA of 19.47 and 10.72 MET·h·d⁻¹). Conclusion OPA exhibits heterogeneous dose-response patterns across cardiometabolic indicators. Integrating the inverted L-, J-, and U-shaped association patterns, we suggest an optimal OPA range of 9.14−25.64 MET·h·d⁻¹ for multiple cardiometabolic indicators in total individuals at high-risk of CVD, with ranges of 9.61−31.34 MET·h·d⁻¹ for males and 8.97−22.87 MET·h·d⁻¹ for females. These findings provide critical evidence for developing OPA interventions strategies targeting high-risk population for CVD.