I was wondering if I need to provide a regression table including the estimation of all main varibales + confounders? Can I just present the final result (coefficeint and pvalues) of the main variables after adjustment?
$\begingroup$
$\endgroup$
4
-
$\begingroup$ Include where? In a study report? In an abstract? In a full peer-reviewed manuscript? On a slide? $\endgroup$– AlexCommented Oct 2, 2022 at 0:56
-
$\begingroup$ @ Alex, Thanks for the reply. For the manuscript. Can I just report the pvalue and the coeffient of the main variables? Or I need to provide the full table including all info regarding main and confounders> $\endgroup$– statsCommented Oct 2, 2022 at 1:12
-
$\begingroup$ In articles, these tables are normally never provided. But the correspondent author is expected to provide them if requested by the journal reviewers or readers. $\endgroup$– AlexCommented Oct 2, 2022 at 10:09
-
1$\begingroup$ This paper might be relevant to read: Daniel Westreich, Sander Greenland, The Table 2 Fallacy: Presenting and Interpreting Confounder and Modifier Coefficients, American Journal of Epidemiology, Volume 177, Issue 4, 15 February 2013, Pages 292–298, doi.org/10.1093/aje/kws412 $\endgroup$– dipetkovCommented Oct 2, 2022 at 16:22
Add a comment
|