Asset Details
Do you wish to reserve the book?
Testing group differences for confounder selection in nonrandomized studies: flawed practice
Hey, we have placed the reservation for you!
By the way, why not check out events that you can attend while you pick your title.
Oops! Something went wrong.
Looks like we were not able to place the reservation. Kindly try again later.
Are you sure you want to remove the book from the shelf?
Testing group differences for confounder selection in nonrandomized studies: flawed practice
Do you wish to request the book?
Testing group differences for confounder selection in nonrandomized studies: flawed practice
Please be aware that the book you have requested cannot be checked out. If you would like to checkout this book, you can reserve another copy
We have requested the book for you!
Your request is successful and it will be processed during the Library working hours. Please check the status of your request in My Requests.
Oops! Something went wrong.
Looks like we were not able to place your request. Kindly try again later.
Journal Article
Testing group differences for confounder selection in nonrandomized studies: flawed practice
2019
Overview
Nonrandomized studies, including observational and quasi-experimental studies, are frequently used to determine the effect of a given exposure (e.g., a new practice, intervention or policy) on relevant outcomes in situations in which random assignment to the exposed or unexposed group is not feasible or ethical. A prevalent source of bias in causal inference based on nonrandomized studies is confounding. This type of bias occurs when characteristics that are causally linked to the outcome(s) of interest are imbalanced across the study groups. A known practice to identify these potential confounders is to test for group imbalances statistically based on the observed study data. Limiting the search for confounders to testing for group imbalances fails to consider possible unobserved variables that are relevant in the confounding mechanism. Variables that are strong predictors of the outcome but are only weakly associated with the exposure would be less likely to be selected for adjustment, resulting in uncontrolled confounding and biased results.
Publisher
Elsevier Inc,CMA Impact Inc,Joule Inc,CMA Impact, Inc
