Validation of guidelines for investigating foodborne disease outbreaks: the experience of the Lazio region, Italy.

Information about risk ratios, exposures, and vehicles for foodborne diseases tends to be more reliable when it is obtained from outbreak surveillance than when it is obtained from disease notifications. In 1997, guidelines for methods of investigating foodborne disease outbreaks (FBDOs) were implemented in the Lazio region. To evaluate the impact of these guidelines, we analyzed information about 410 FBDOs from 1996 to 2000. Under these guidelines, the delay in the reporting of outbreaks decreased from 10 to 2 days. An analysis of 82 large FBDOs (in which > 30 people were exposed) showed increases in the calculation of attack rates with a cohort approach (up to 83%), correctly drawn epidemic curves (up to 79%), and the calculation of food-specific relative risk (up to 60%). On the other hand, the level of the determination of etiology remained low: tests were performed on patients in 57% of the cases considered, resulting in an agent identification rate of 38%; for 22 outbreaks, tests were performed on food, resulting in three positive identifications. Analysis of the contamination route with the use of hazard analysis critical control point criteria a posteriori was carried out for 15 outbreaks, and nine of these analyses were successful. The implementation of the 1997 guidelines was successful with regard to epidemiological and statistical methods but did not improve etiological diagnosis for FBDOs. These guidelines improved surveillance for outbreaks in which > 30 people were exposed in well-defined exposure situations; however, the guidelines did not significantly improve epidemiological investigations of small household outbreaks.