Comparison bias and dilution effect in occupational cohort studies

Area of Science:

• Occupational health
• Epidemiology
• Biostatistics

Background:

• Comparing worker mortality to the general population is common for evaluating occupational exposures.
• This design is susceptible to dilution effect (DE) from unexposed workers and comparison bias (CB) from differing risk factors.

Purpose of the Study:

• To propose a theoretical model for the combined impact of DE and CB.
• To evaluate the influence of DE and CB on risk estimates in occupational cohort studies.

Main Methods:

• Developed a theoretical model for joint DE and CB effects.
• Applied the model to two occupational cohorts, analyzing specific causes of death.
• Used direct and indirect standardization to adjust for confounding factors.

Main Results:

• A linear relationship was observed between risk estimates and the biases.
• Higher risks for exposed workers were identified only after adjusting for DE and CB.
• Standardization methods alone were insufficient without bias adjustment.

Conclusions:

• Dilution effect and comparison bias can significantly distort findings in occupational cohort studies.
• Adjusting for DE and CB is crucial for accurate risk assessment.
• Cohort studies using external referent groups require careful consideration of these biases for reliable results.