Resident Autonomy in an Era of Managed Care

The widespread penetration of health care systems by managed care has posed new challenges for graduate medical education.^{1 - 3} Our institution has been no exception. For example, as a direct result of managed care, nearly all hospitalized patients at our institution currently have a private attending physician of record. This was not the case in the past. We were concerned about the effects of this change; specifically, we wondered if the influx of private generalist attending physicians had adversely affected residents' autonomy in the hospital setting. We report on a study of residents' perceptions of autonomy in the care of hospitalized patients in a managed care environment.

We conducted a prospective study at an urban, not-for-profit, academic medical center. Twenty-eight pediatric residents (16 postgraduate year 1, 12 postgraduate year 2/3) were each asked to maintain a log on 10 consecutive admissions. The survey instrument asked residents to record several pieces of data, including the admitting diagnosis, attending physician of record, and responses to questions concerning residents' level of autonomy with each case. Residents were asked to rate their level of autonomous involvement in the management decisions for each patient on a scale of 1 to 10 (with 10 being the most highly involved). Data from general pediatric diagnoses (GP) were compared with those from subspecialty/surgical diagnoses (SS). Differences in proportions were tested by a χ² test.

Results showed that 278 patients (62% GP and 38% SS) were treated by residents during the study period. Ninety-five percent of all patients had a designated private attending physician of record. Residents reported a high level of autonomy in the management of GP diagnoses compared with SS diagnoses (7.7 vs 4.4; P<.001). Residents assessed their levels of autonomy to be appropriate in 91% of GP diagnoses vs 80% of SS diagnoses (P = .02). Residents were more likely to offer suggestions for the management of GP diagnoses compared with SS diagnoses (93% vs 70%, P = .002). Patient management plans were more likely to be dictated by SS attending physicians vs GP attending physicians (97% vs 30%, P<.001). Even for the subset of patients with private GP attending physicians, SS attending physicians were still more likely to dictate a management plan than were private GP attending physicians (29% vs 9.5%, P<.001). There were no significant differences for any of the variables across years of training.

We conclude that at our institution, residents perceived that they had appropriate levels of autonomy in the hospital setting despite the advent of managed care. However, we found significant differences in resident autonomy in the care of patients hospitalized with GP vs SS diagnoses. Residents did not perceive their autonomy to be substantially affected by generalist private attending physicians. Other institutions could employ a similar method to assess residents' perceptions of the effects of changes in the health care system over time.