Abstract
OBJECTIVE: To develop a research methodology to apply a single case-finding tool to multiple related diseases and to evaluate the ability of a single tool to detect two or more related chronic diseases. METHODS: A case-finding study to detect two related respiratory diseases is used to demonstrate and explain the proposed methodology. Adults in the community with no prior history of physician-diagnosed lung disease who self-reported respiratory symptoms were contacted via random-digit dialing. Multiple risk scores, one for asthma and one for COPD, were developed using data from a single case-finding questionnaire administered to the study population. Each score was statistically optimized for targeted detection of cases having one disease in the class. External validation of tandem risk scores was prospectively conducted in an independent sample and predictive performance re-evaluated. RESULTS: Sensitivity for detection of asthma improved from 87% using single risk scores to 96% using tandem risk scores, and sensitivity for detection of COPD similarly improved from 87% to 99%. In the independent validation cohort, case-finding sensitivities increased from 64% and 59% using single risk scores to 95% and 96% using tandem risk scores for asthma and for COPD, respectively. CONCLUSIONS: Use of a single questionnaire which incorporates risk scores for multiple diseases considered in tandem, rather than individually, enhances the yield of cases detected when compared with one-at-a-time application of risk scores for case discovery. Benefits include greater efficiency in case-finding and improved sensitivities for detection of each disease.