During a typical outpatient visit in today's managed care environment, the principal means by which patients can influence their care is by making requests. A request is an explicit verbal or written communication asking the physician for information or action. As the focal pont for the clinical negotiation, requests occupy a central place in the encounter. In preliminary work, the investigators have constructed a detailed Taxonomy of Requests by Patients (TORP). TORP currently consists of 11 categories of requests for information, 8 categories of requests for action, and 8 categories of physicians responses. The objectives of this application are to refine this new system for coding patients requests, to estimate the prevalence of different request types and responses, and to elucidate the antecedents and consequences of problematic requests. The investigators will accomplish these objectives by: 1) adding new codes to enhance TORP's relevance to managed care; 2) producing additional evidence for TORP's reliability; and 3) performing a series of analyses designed to assess the concurrent, construct, and predictive validity of the enhanced taxonomy. By project's end, a valid and reliable taxonomy will be available to study patients requests and physicians responses across managed healthcare settings. This project is ambitious relative to the resources requested because much of the data required for the proposed analyses has already been collected. Three-hundred tape recordings of ambulatory encounters are available from a 1994 study of patients visiting a small group of Sacramento internists; an additional 60 encounters with cardiologists will be recorded de novo. The planned research will proceed in three stages. First, using card-sorting techniques, new categories of patient requests and physician responses will be developed to increase TORP's sensitivity to problematic issues in managed care. Second, additional studies concerning the inter- and intra-rater reliability of the coding scheme will be performed. Third, the coding scheme generated in the developmental dataset (n=142 internal medicine encounters) will be assessed in a validation dataset derived from the remaining 158 internal medicine encounters plus the 60 newly recorded cardiology encounters. Validation will occur by way of five studies assessing relationships between directly-observed patient requests and 1) physician-reported requests; 2) contentious physician-patient negotiation; 3) patients demographic and health characteristics; 4) visit satisfaction; and 5) physicians speciality, once validated, TORP promises to be a robust, comprehensive system for analyzing a critical aspect of the physician- patient interaction.
