Medical statistics and Data Science: Epidemiology

E-value in observational research

     

News:

A new book “An introduction to Directed Acyclic Graph (DAG) for health researchers” has been published in Amazon in the 21st December 2024

An introduction to Directed Acyclic Graph (DAG) for health researchers

The book description:

"Directed acyclic graph (DAG) is increasingly used in modern epidemiology, especially guide researchers to implementing causal inference in observational studies. Casual DAG visually presents causal knowledge and assumptions between variables. Once one can manage the rules, it can facilitate many tasks, such as using DAG makes it easier to understand many concepts for example direct and indirect causal effects, mediation analysis, collider stratification bias, selection bias, and information bias, etc. It also makes easier to recognize and avoid mistakes in analytic decisions such as using the backdoor criterion to select variables to be adjusted."

"More advanced texts on DAGs are readily available in textbooks and in scientific papers, but a simple and comprehensive introduction to DAG is lacking."

"The book thoroughly introduces DAG in a plain language from the scratch, step by step with more simple and accessible language explaining the concepts, terminologies, rules, and potential applications. The book will pave the way for researchers using DAG."

  • Link to the book

    • Introduction

    • According to the reference1, "The E-value is the minimum strength of association, on the risk ratio scale, that an unmeasured confounder would need to have with both the treatment and outcome, conditional on the measured covariates, to explain away a treatment–outcome association."
    • Both the Stata command (evalue) and web E-value calculator implement the formula given by the table 1 and table 2 form the reference1.
    • You are encouraged to read comments2,3 and tecniques4,5 about E-value
    • We will periodically update the reference list.

    Stata commands

    We have developed a Stata command (evalues) to carry out the calculation. Type and run "ssc describe evalues" in the Stata command window. You should be able to view and download the command.

    E-value calculator

    1. E-value for an observed risk ratio, odds ratio, and hazard ratio when the outcome is relatively rare (for example the prevalence ≤ 15%) (

      2. The observed risk ratio:
      The low limit of 95%CI (optional):
      The up limit of 95%CI (optional):

    2. E-value for an observed odds ratio when the outcome is relatively common (for example the prevalence > 15%)

      3. The observed odds ratio:
      The low limit of 95%CI (optional):
      The up limit of 95%CI (optional):

    3. E-value for an observed hazard ratio when the outcome is relatively common (for example the prevalence > 15%)

      4. The observed hazard ratio:
      The low limit of 95%CI (optional):
      The up limit of 95%CI (optional):

    References

    1. VanderWeele TJ, Ding P. Sensitivity Analysis in Observational Research: Introducing the E-Value. Ann Intern Med. 2017 Aug 15;167(4):268-274
    2. Ioannidis JPA, Tan YJ, Blum MR. Limitations and Misinterpretations of E-Values for Sensitivity Analyses of Observational Studies. Ann Intern Med. 2017 Aug 15;167(4):268-274
    3. VanderWeele TJ, Mathur MB, Ding P. Correcting Misinterpretations of the E-Value. Ann Intern Med. 2017 Aug 15;167(4):268-274
    4. Ding P, VanderWeele TJ. Sensitivity Analysis Without Assumptions. Epidemiology. 2016 May;27(3):368-77.
    5. VanderWeele TJ, Ding P, Mathur M. Technical Considerations in the Use of the E-value. J Causal Inference. 2019;7.