Assessing the effect of intensity when exposure varies over time

Area of Science:

• Epidemiology
• Occupational Health
• Biostatistics

Background:

• Assessing exposure-response relationships in epidemiology is challenging due to variable exposure intensities over time.
• Existing summary exposure metrics often assume a linear relationship between exposure intensity and risk, which may not always hold.
• Accurate exposure quantification is crucial for understanding disease etiology and risk.

Purpose of the Study:

• To develop and illustrate a flexible approach for constructing summary exposure measures.
• To incorporate alternative assumptions about the impact of exposure intensity and timing on health outcomes.
• To evaluate how different exposure quantification methods affect exposure-response analyses and risk estimates.

Main Methods:

• Developed a framework for creating summary exposure metrics with varied intensity and timing functions.
• Utilized conditional logistic regression to fit these novel exposure metrics to data.
• Applied the methodology to a case study of lung cancer mortality in asbestos miners and millers.

Main Results:

• Demonstrated that the choice of exposure quantification function significantly influences exposure-response analysis.
• Showcased how different metrics can lead to varying risk estimates for lung cancer.
• Highlighted the importance of considering non-linear or time-dependent exposure effects.

Conclusions:

• The proposed method offers a more adaptable way to summarize complex exposure histories in epidemiologic studies.
• Flexible exposure metrics can provide more accurate risk assessments compared to traditional time-weighted approaches.
• This approach enhances the ability to study the impact of exposure intensity and timing on disease risk.