DO DIFFERENT PHYSICIANS PANELS REACH SIMILAR CONCLUSIONS - A CASE-STUDY ON PRACTICE GUIDELINES FOR LIMITED SURGERY IN BREAST-CANCER

This study was aimed at assessing the extent to which different panels of physicians produce similar recommendations and whether their agree ment is greater when members of the panels practice in the same geogra phical area, A total of 34 physicians, organized in 3 multispeciality and one monospeciality (all surgeons) panels, working at both speciali zed and community hospitals, participated in the study, They were aske d to rate the appropriateness of an indication for limited surgery in breast cancer for 432 hypothetical patients described through clinical scenarios, The primary outcome of the study was the reliability of ap propriateness ratings between the pairs of panels, measured as the per cent absolute agreement and kappa statistic, The agreement of the rati ngs between pairs of panel of the same region was always quantitativel y good - panels I-II kappa = 0.58 (95% CI: 0.50-0.67) and panels Ill-I V kappa = 0.65 (95% CI: 0.56-0.75) - and higher than when panels from different regions were compared, We conclude that in our study an acce ptable level of agreement was reached when different panels of experts were asked to produce guidelines using a structured process which inc ludes exposure to the relevant scientific literature, The fact that lo cal (organizational, cultural, etc.) factors seemed to play a modulati ng role over scientific evidence has implications that should be consi dered when deciding on the level (local versus central) where guidelin es are produced.