Interval estimation of the attributable risk for multiple exposure levels in case-control studies

Area of Science:

• Epidemiology
• Biostatistics

Background:

• Risk factor exposure often occurs at multiple levels in case-control studies.
• Estimating attributable risk at each exposure level is crucial for public health.

Purpose of the Study:

• To review existing methods for estimating attributable risk in 2 x 2 and 2 x k tables.
• To propose a novel method for calculating confidence intervals for attributable risk in 2 x k tables.
• To evaluate the proposed method's performance through a Monte Carlo simulation.

Main Methods:

• Review of estimation techniques for dichotomous and multi-level exposure in case-control studies.
• Development of a confidence interval calculation for attributable risk in 2 x k tables.
• Application of the method to stratified data (2 x k x s tables).
• Monte Carlo simulation to assess coverage probabilities.

Main Results:

• Existing estimation methods for 2 x 2 and 2 x k tables are discussed.
• A new method for confidence intervals in 2 x k tables is presented.
• The Monte Carlo study confirmed satisfactory agreement between nominal and actual coverage probabilities for practical use.

Conclusions:

• The proposed method provides reliable confidence intervals for attributable risk in multi-level exposure scenarios.
• The method is applicable to both unadjusted and adjusted (stratified) estimates.
• This contributes to more accurate risk assessment in epidemiological research.