International Journal for Quality in Health Care Advance Access originally published online on April 7, 2006
International Journal for Quality in Health Care 2006 18(3):177-182; doi:10.1093/intqhc/mzl005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Appropriateness of colorectal cancer screening: appraisal of evidence by experts
1 Health Care Evaluation Unit, 2 Health Services Research Unit, Institute of Social and Preventive Medicine, Hospices-CHUV, 3 Division of Gastroenterology & Medical Outpatient Department, Hospices-CHUV, University of Lausanne, Switzerland, and 4 Gastroenterology Department, University of Basle, Switzerland
Objectives. To evaluate how the level of evidence perceived by an international panel of experts was concordant with the level of evidence found in the literature, to compare experts perceived level of evidence to their appropriateness scores, and to compare appropriateness criteria for colonoscopy between experts and an evidence-based approach.
Design. Comparison of expert panel opinions and systematic literature review regarding the level of evidence and appropriateness of colonoscopy indications.
Participants. European Panel on the Appropriateness of Gastrointestinal Endoscopy multidisciplinary experts from 14 European countries.
Main outcome measures. Concordance and weighted kappa coefficient between level of evidence as perceived by the experts’ and that found in the literature, and between panel- and literature-based appropriateness categories.
Results. Experts overestimated the level of published evidence of 57 indications. Concordance between the level of evidence perceived by the experts and the actual level of evidence found in the literature was 36% (weighted kappa 0.18). Indications for colonoscopy were reported to be appropriate, uncertain, and inappropriate by the experts in 54, 19, and 27% of the cases, and by the literature in 37, 46, and 17% of the cases. A 46% agreement (weighted kappa 0.29) was found between literature-based and experts’ appropriateness criteria.
Conclusions. Experts often overestimated the level of evidence on which they based their decisions. However, rarely did the experts’ judgement completely disagree with the literature, although concordance between panel- and literature-based appropriateness was only fair. A more explicit discussion of existing evidence should be undertaken with the experts before they evaluate appropriateness criteria.
Keywords: colonoscopy, colorectal cancer, concordance, level of evidence, RAND appropriateness method, weighted kappa
Address reprint requests to Isabelle Peytremann Bridevaux. Institute of Social and Preventive Medicine, 17 Bugnon, CH-1005 Lausanne, Switzerland. E-mail: isabelle.peytremann-bridevaux{at}hospvd.ch
Accepted for publication March 6, 2006.