This study had two primary aims: (1) to determine the usefulness of a telephone-administered health status questionnaire as an epidemiological survey instrument in groups at high risk for morbidity and disability; and (2) to evaluate the value of a telephone interview for making judgments about individual patient management. A structured telephone interview was compared with a face-to-face interview in 366 individuals with a stroke or an orthopedic condition up to 5 years after discharge from a rehabilitation hospital. Standardized assessments including the Barthel Index, the Zung Scale, the Reintegration to Normal Living (RNL) Index, and the Pfeiffer Short Portable Mental Status Questionnaire (SPMSQ), along with questions on health, medication use, and health events were administered using both modes. Comparability between the modes was also assessed according to the type of respondent, self, or proxy. On the Barthel Index, the Zung Scale, the RNL Index, and the SPMSQ the percentage of patients indicated as having no disability was similar between the modes. Proxies' responses were, in general, as consistent between the modes as patients' responses. The reliability coefficients indicated moderate to substantial agreement between the modes on the majority of indices and health-related questions. Discord between modes, when present, was greatest for individuals with moderate and severe disability, with less frequent reporting of disability on the telephone. This study supports the use of telephone interviews in determining the prevalence of disability in the community and supports the use of the telephone interview in the case finding process.