TaiPhone messages often brief not intended brusqueOn 29 Nov 2019, at 14:17, Anoop B <[log in to unmask]> wrote:From what I have seen, adjusting for confounders is one area that needs to be improved: Either there are number of confounders missing because they didn't measure or they didn't want to include in i the analysis. I would like to see a sort of confounder map/cloud that shows the potential confounders and mediators AND how/why they selected these as confounders. Then include the confounders that you measured. This way readers could clearly see which confounders are missing and which were could be measured in the study. Also, ideally the confounder list should be pre-registered.Dear ColleaguesWhen teaching appraisal of internal validity, it helps to give real examples with flaws (biases). It is quite easy to find flawed examples for most aspects of a study but when it comes to analysis most examples I have are limited: (lack of) Intention to Treat and failure to adjust for possible confounders. What is more, they are getting old now. Does anyone have examples they could share?Or does doing so count as “flaw shaming”? ;–)KevProf. Kev (Kevork) Hopayian,
BSc, MB BS, MD, FRCGP, DCH, DRCOG
Clinical Professor, University of Nicosia, CyprusRCGP [INT] MemberSessional GP, Suffolk
To unsubscribe from the EVIDENCE-BASED-HEALTH list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=EVIDENCE-BASED-HEALTH&A=1
To unsubscribe from the EVIDENCE-BASED-HEALTH list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=EVIDENCE-BASED-HEALTH&A=1
To unsubscribe from the EVIDENCE-BASED-HEALTH list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=EVIDENCE-BASED-HEALTH&A=1