BACKGROUND/RATIONALE:
Two important missions of the VA system are to provide patient-centered care and to train new physicians. However, tensions exist between these missions as pressure increases for productivity and accountability. Patients, trainees, and faculty are likely to have different beliefs about the purpose and function of an academic clinic. These differences may create barriers to patient-centered care.
OBJECTIVE(S):
The objectives of the study are to: 1) document and analyze the values and beliefs of patients, trainees and faculty about the purpose and function of the academic medical clinic; 2) discover barriers to patient-centered care created by the three cultures identified above; and 3) evaluate the use of cultural consensus analysis (a quantitative ethnographic technique) as a tool for measuring outcomes in primary care education.
METHODS:
All three phases have been completed. Data analysis is ongoing. During phase I, trained observers collected ethnographic data in the academic medical clinics over the course of one year. Observations were made in the clinic waiting areas, workstations and exam rooms. Two analysts coded the text units into 15 recurrent themes. Breakdown was the most important theme. During phase II, the themes and observations were further explored using interviews and focus groups. A taxonomy of breakdowns and a structural model of breakdown, learning and patient care was created. During phase III, statements for cultural consensus analysis were drawn verbatim from focus group responses. These were applied to faculty, residents, and patients.
FINDINGS/RESULTS:
During phase I, we discovered fifteen themes that were applied with a Cohen’s k of 0.88 between coders. Breakdown was the most common and important theme, seen in 43 percent of all text segments. During phase II, breakdowns were found to be necessary for learning, and effective response to breakdown required six characteristics: engaging the patient, matching responsibility to authority, matching tools to tasks, matching information resources to need, matching values between co-participants, and matching expectations to capacity. During phase III, the cultural consensus analysis (CCA) showed no culture shared by the aggregate of patients, residents and faculty (eigenvalue ratio 1.433), strong cultural consensus for faculty (ratio 3.841), very strong cultural consensus in residents (ratio 6.070), and near cultural consensus in patients (ratio 2.558). Triangulation of data from all three phases was used for criterion-related validation of the CCA. These results are final and are being prepared for publication. Preliminary validation of the CCA instrument at another institution looks promising for the generalizability of this instrument.
IMPACT:
Cultural consensus analysis is a criterion-validated outcome measure at our institution, and shows promise as a generalizable tool. If true, the resource intensive ethnography done here will not need to be repeated to use the instrument at other sites. The findings from our structural model are being refined into a method for patient-centered improvements. Both of these are testable hypotheses.
PUBLICATIONS:
Journal Articles
- Smith CS, Morris M, Hill W, Francovich C, Christiano J. Developing and validating a conceptual model of recurring problems in teaching clinic. Advances in Health Sciences Education. 2006; 11(3): 279-88.
- Smith CS, Morris M, Francovich C, Hill W, Gieselman J. A qualitative study of resident learning in ambulatory clinic. The importance of exposure to 'breakdown' in settings that support effective response. Advances in Health Sciences Education. 2004; 9(2): 93-105.
- Smith CS, Morris M, Hill W, Francovich C, McMullin J, Chavez L, Rhoads C. Cultural consensus analysis as a tool for clinic improvements. Journal of General Internal Medicine. 2004; 19(5 Pt 2): 514-8.
