A simple diagnostic model for ruling out pneumoconiosis among construction workers

Area of Science:

• Occupational Health
• Pulmonary Medicine
• Diagnostic Modeling

Background:

• Silica dust exposure in construction poses a risk of silicosis, even at low levels.
• Current health surveillance lacks specific diagnostic criteria, leading to potential overuse of chest x-rays.
• Accurate identification of workers without pneumoconiosis is crucial for efficient healthcare resource allocation.

Purpose of the Study:

• To develop a simple diagnostic model for estimating pneumoconiosis probability in workers.
• To utilize questionnaire and spirometry results for accurate exclusion of non-affected individuals.
• To minimize unnecessary diagnostic procedures in occupational health surveillance.

Main Methods:

• Cross-sectional study of 1291 Dutch construction and natural stone workers with high quartz dust exposure.
• Development of a multivariable logistic regression model using chest x-ray (ILO profusion category >= 1/1) as the reference standard.
• Validation of the model's calibration and discriminative ability (ROC area), with internal validation via bootstrapping.

Main Results:

• Key predictors identified: age >= 40, current smoker, high-exposure job, >= 15 years in construction, self-reported "unhealthy" status, and FEV1.
• The model demonstrated good calibration and discriminative ability (ROC area 0.81, 95% CI 0.74-0.85).
• An optimized cut-off point with high negative predictive value allows efficient exclusion of workers with low pneumoconiosis probability.

Conclusions:

• The developed diagnostic model is effective for ruling out pneumoconiosis in construction workers.
• Implementation in health surveillance can significantly reduce redundant chest x-ray investigations.
• This approach enhances the efficiency and accuracy of occupational health monitoring for silica-exposed workers.