Residents' patient-specific clinical questions: opportunities for evidence-based learning.

PURPOSE Resident physicians generate many patient-related questions in the outpatient setting, but rarely answer them. This study examined how answering a patient-specific clinical question affected residents' patient care decisions. METHOD From October 2001 to June 2002, 43 internal medicine residents at the University of Colorado Health Sciences Center outpatient clinic formulated and attempted to answer specific clinical questions based on patients seen in the out-patient setting. Residents completed a questionnaire after finishing their clinical question exercise. RESULTS Residents completed 158 exercises (68%) out of 234 opportunities and found answers to 89% of questions (141/158). Questions most commonly addressed therapy (43%), diagnosis (15%), disease management (13%), and prevention (9%). The most frequently used methods to retrieve clinical information were Medline 73% (115/158) and UpToDate 70% (110/158). UpToDate (45%, 71/158) and journal articles (42%, 66/158) were the most helpful final information sources. Residents rated the impact of the information on a five-point scale (1 = strongly disagree; 5 = strongly agree) for the following: can be used to assist patient's care (mean 4.1), improved care (mean 4.0), improved communication (mean 4.3), improved confidence in care (mean 4.3), improved knowledge (mean 4.6), and will improve care for future patients (mean 4.4). The information affected clinical decision making in 78% (110/141) of cases. CONCLUSION Internal medicine residents practicing self-directed learning by answering patient-specific clinical questions reported improvement in knowledge and changes in patient care decisions.