Evaluation of exposure-specific risks from two independent samples: A simulation study
2011

Evaluating Exposure-Specific Risks: A Simulation Study

Sample size: 1000 publication Evidence: high

Author Information

Author(s): William M Reichmann, David Gagnon, C Robert Horsburgh, Elena Losina

Primary Institution: Brigham and Women's Hospital

Hypothesis

The study aims to evaluate the existing methodology for calculating exposure-specific risks and propose an improved estimator.

Conclusion

The revised product-based estimator provides much improved accuracy over the existing simple product-based method.

Supporting Evidence

  • The revised product-based estimator exhibited little observed relative bias compared to the simple product-based estimator.
  • Coverage probabilities for the 95% CI of the revised estimator were above 95% across most scenarios.
  • The existing methodology performed well only when the exposure probability was low and the relative risk was small.

Takeaway

This study shows that a new way to calculate disease risk based on exposure is better than the old method, especially when the exposure is low.

Methodology

A simulation study was conducted to test the performance of two estimators and their associated confidence intervals.

Potential Biases

The existing methodology showed significant bias in certain scenarios, particularly with higher exposure probabilities.

Limitations

Simulation studies cannot be considered proof, and the validity of estimates may be questionable with small sample sizes.

Statistical Information

Confidence Interval

95%

Digital Object Identifier (DOI)

10.1186/1471-2288-11-1

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication