OBJECTIVES: To develop an economic model of the cost effectiveness of using molecular (polymerase chain reaction, PCR) compared to conventional microbiological methods to diagnose pathogens contributing to acute intestinal infectious diseases.

METHODS: Agent-based models of population interaction, exposure to sources of infection and transmission were developed for major pathogens. Model parameters were identified from a literature review and from the Integrate project. A genetic algorithm was used to determine unknown parameters relating to infectiousness on encounter. The models considered 50,000 agents over a period of one year. Each agent could be infected by a source and infect other agents. A subset of infected agents in the community visit a general practitioner (GP), and a subset of these provide a stool sample for testing. We simulated a PCR method by representing a greater number of agents having their pathogen correctly identified. A threshold of correctly identified pathogens triggers outbreaks. Simulations determine the effect a PCR method could have in determining outbreaks and closing of infected sources. A UK NHS perspective was adopted.

RESULTS: The combined incidence rate for Campylobacter, Clostridium difficile, E. coli O157, Salmonella and Norovirus is 35 per 1000 person years in the community. Simulations of PCR systems show a 25% drop to 26 per 1000 person years. The number of GP visit falls 33% from 3.8 to 2.6 per 1000 person years. Multiplex PCR costs £33.33 per sample including staff time but conventional methods cost £6.58. The PCR method results in a gain of 0.3 QALYs per 1000 person years, at a cost which is within acceptable limits of cost-effectiveness

CONCLUSIONS: The model demonstrates that a multiplex PCR method is a cost-effective method of screening for acute intestinal infectious diseases.