Statistics from Altmetric.com
We read the excellent paper by Shawihdi et al with interest.1 The authors demonstrated in a retrospective study, using hospital episode statistics, that outcomes from gastro-oesophageal malignancy were worse among patients from practices with a low referral rate for upper gastrointestinal (GI) endoscopy.
While the authors address some weaknesses in their data, we believe they have not highlighted all problems with their approach. As with all observational data, there is the problem of confounding, which often cannot be completely corrected in analyses. However, the major concern is that they looked at aggregate data, not individual patient data. This is open to the ecological fallacy; what is seen in populations …
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.