Hospitalization records as a tool for evaluating performance of food- and water-borne disease surveillance systems: a Massachusetts case study.

We outline a framework for evaluating food- and water-borne surveillance systems using hospitalization records, and demonstrate the approach using data on salmonellosis, campylobacteriosis and giardiasis in persons aged ≥65 years in Massachusetts. For each infection, and for each reporting jurisdiction, we generated smoothed standardized morbidity ratios (SMR) and surveillance to hospitalization ratios (SHR) by comparing observed surveillance counts with expected values or the number of hospitalized cases, respectively. We examined the spatial distribution of SHR and related this to the mean for the entire state. Through this approach municipalities that deviated from the typical experience were identified and suspected of under-reporting. Regression analysis revealed that SHR was a significant predictor of SMR, after adjusting for population age-structure. This confirms that the spatial "signal" depicted by surveillance is in part influenced by inconsistent testing and reporting practices since municipalities that reported fewer cases relative to the number of hospitalizations had a lower relative risk (as estimated by SMR). Periodic assessment of SHR has potential in assessing the performance of surveillance systems.