Exposure in Epidemiology Class 1

Exposure in Epidemiology

• Exposure can be difficult to define and may fluctuate over time, even within the same person.
• True exposure is often not measurable, and a surrogate is needed, balancing sensitivity, specificity, accuracy, patient burden, and cost.

Sources of Exposure

• Exposure can occur through many sources, such as environmental factors, lifestyle habits, and biological agents.
• Example: Exposure to aluminum welding fumes can cause pulmonary fibrosis.

Biological Agent of Interest

• Isolating the biological agent of interest helps measure exposure accurately.
• Specify the active agent to make the measurement of exposure more precise.
• Example: If caffeine is the exposure of interest, exclude decaffeinated coffee and include all sources of caffeine, such as sports gels and medication.

Examples of Exposure

• Vaccination and autism due to exposure to mercury from thimerosal in vaccines.
• Eating meat and the risk of colon cancer due to naturally occurring chemicals, added nitrates, and nitrites.
• Smoking and prostate cancer, considering different types of smoking and toxic components.

Considerations for Selecting Exposure Measurement

• How does exposure relate causally to the outcome of interest?
• What is the best way to represent the amount or dose of exposure?
• Example: Serum 25(OH)D as a biomarker of exposure to vitamin D.

The Exposure-Disease Pathway

• Exposure to disease is a pathway of multiple steps or causal events.
• Exposures can be measured at multiple steps or states.
• Disease can be measured at multiple steps or states.
• As you move down the exposure-disease pathway, what one measures is closer to the causal agent, making it more likely to show a non-biased association between exposure and disease.

Exposure Measurement in Epidemiological Studies

• In case-control studies, behavioral exposures should only be included before symptoms occur, and a reference date should be established for both cases and controls.
• Example: 2 years for colon cancer.

Timing of Exposure

• Timing of exposure is crucial in determining disease risk, and different methods can be used to relate timing of exposure to disease risk.
• Methods include:
  • Latency analysis or exposure lagging: exclude exposures that occurred in the induction or latent period.
  • Exposure window analysis: only count exposures that occurred in the etiological window, excluding those before and after.
  • Example: a cumulative dose of asbestos for the time window 10 to 25 years before diagnosis/reference date had the greatest power to detect an association with lung cancer.

Etiological Window

• The etiological window is the critical exposure window during which the exposure has the greatest causal effect on outcomes.
• Including exposures outside of the etiological window results in exposure measurement error.
• Examples of etiological windows include:
  • Prenatal exposure (different effects in 1st, 2nd, and 3rd trimesters)
  • Exposures during different phases of childhood
  • Peripubertal period (e.g., cannabis exposure and schizophrenia-like deficits)
  • Time in relation to menopause (e.g., postmenopausal women and cardiovascular disease)

Additional Methods

• Time since first exposure or age at first exposure: important for causes that are early on the sequential cause chain.
• Example: age at immigration and risk of melanoma.
• Regency analysis: considers only recent exposure and works for exposures that are causal in the later part of the sequential cause chain.
• Example: smoking in the past year is the best exposure window for the association with myocardial infarction.

Statistical Approaches

• Examining cumulative exposure dose for different non-overlapping time periods in the same model.
• Examining multiple overlapping periods in separate models and comparing models for goodness of fit.
• Using weighted cumulative exposure, with each time window carrying different weights.
• Caution: exploratory modeling tends to overestimate the effect size, and components of dose and time are correlated, so need to adjust for other exposure windows.

Measuring Exposure

• Step 1:
  • Identify research question
  • Identify biological agent of interest
  • Determine where in the exposure-disease pathway to measure exposure
• Step 2:
  • Identify ideal and feasible method of measurement
  • Ensure measurement method is sensitive and specific
  • Determine timing of exposure to include
• Step 3:
  • Determine duration of exposure
  • Determine frequency of exposure

This quiz explores the concept of exposure in epidemiology, discussing its definition, fluctuations, and measurement. It also touches on the importance of choosing a surrogate when true exposure is not measurable.