Development and Validation of a Cardiovascular Disease Risk Prediction Model for the Japanese Working Population: The...

Area of Science:

• Occupational health
• Cardiovascular disease epidemiology
• Risk prediction modeling

Background:

• Cardiovascular disease (CVD) poses a significant public health challenge.
• Accurate risk prediction is crucial for effective CVD prevention strategies.
• Occupational health data offers a valuable resource for developing population-specific risk models.

Purpose of the Study:

• To develop and validate a cardiovascular disease (CVD) risk prediction model.
• To utilize routine health checkup data from a large occupational cohort for model development.
• To assess the 10-year CVD risk prediction accuracy and applicability in the workplace.

Main Methods:

• A cohort of 96,117 Japanese employees (aged 30-64, no baseline CVD) was analyzed.
• Cox proportional hazards regression models were used to identify significant CVD risk predictors.
• Model performance was evaluated using discrimination (Harrell's C-statistic) and calibration metrics, with internal validation for overfitting.

Main Results:

• A total of 422 incident CVD cases were recorded during a mean 6.7-year follow-up.
• The final model incorporated age, smoking, diabetes, systolic blood pressure, and lipid levels (LDL, HDL).
• The model demonstrated strong predictive ability (C-statistic: 0.796) and excellent calibration, with minimal overfitting.

Conclusions:

• The developed model accurately predicts 10-year CVD risk in an occupational setting.
• Its basis in routine health checkup data facilitates easy workplace implementation.
• Further research is needed to confirm external validity and transferability of the CVD risk model.