Need for consensus development in prehospital emergency medicine: effect of an expert panel approach

The objective of this study was to look at the need for consensus development in prehospital emergency medicine, and to determine the effect of an expert panel approach. The study took place in Euregio Scheldemond, comprising Sealand Flanders, The Netherlands, and Belgian Flanders, Belgium. Firstly, seven experts rated in organized mailings 153 (random selection out of 505) existing cases of acute or critically ill patients, situated in Dutch Sealand-Flanders. Experts were asked to decide whether assistance from neighbouring Belgian Flanders, consisting of a trauma team with or without the use of a highly equipped ambulance/helicopter, was needed or not in Dutch Sealand-Flanders, at: (1) the site of the incident, and (2) for transport to the hospital. They also had to decide on: (3) the required type of destination hospital (Belgian centre/university, versus Dutch regional). In a subsequent meeting using a modified nominal group technique 23 'worst' cases from the postal rounds with the lowest agreement were discussed and re-rated. We present a framework for the consensus measurement and development procedures. Agreement among experts was poor with multiple rater (Fleiss) kappa values for all 153 postal cases for the first, second and third decisions of 0.32, 0.08 and 0.45, respectively. After group discussions of the 23 'worst' postal cases, kappa values increased significantly and substantially; for the first, second and third decisions from 0.08 to 0.51, from -0.08 to 0.39 and from 0.16 to 0.62, respectively (all p < 0.001). Agreement increased significantly for medical cases, but not for trauma cases. It is concluded that consensus development for prehospital emergency medicine is needed. An expert panel approach seems fruitful in achieving more agreement, which forms a basis for guideline or protocol development.