Estimating the incidence of food-borne Salmonella and the effectiveness of alternative control measures using the Delphi method

The paper describes the use of the Delphi method to estimate the incidence of food-borne Salmonella in the UK and the effectiveness of alternative control measures. A panel of experts on food-borne Salmonella participated in the Delphi survey, which involved five rounds of questioning taking place in the period July 1993 to January 1994. Participants were asked to give initial estimates for a number of parameters and invited to revise these estimates through progressive rounds of the survey at which the group responses were reported back. This process resulted in a reduction in the variation between the estimates given by individual experts. The final estimated annual incidence of food-borne Salmonella in the UK was 537 000, although significant variation remained between individual estimates. The foods judged to be the most important modes of transmission were poultry and poultry products (50% of cases) and eggs and egg products (26% of cases). The panel was also requested to estimate the effectiveness of strategies available to reduce the incidence of food-borne Salmonella from all sources. The most effective methods were judged to be food irradiation and mandatory application of HACCP, although there were significant differences in the judged effectiveness of these technologies for individual respondents. The paper demonstrates the efficacy of the Delphi method as a mechanism for reconciling differences between expert judgements of the incidence of food-borne disease and the effectiveness of alternative control strategies.