Balancing clinical experience in outpatient residency training

Area of Science:

• Medical Education
• Health Services Research
• Clinical Informatics

Background:

• Teaching clinics require diverse patient case-mix for adequate resident training.
• Variations in case-mix across resident panels lead to inconsistent educational experiences.
• Ensuring equitable training is crucial for primary care residents.

Purpose of the Study:

• To develop and evaluate patient reassignment algorithms for balancing case-mix in resident panels.
• To improve the consistency and breadth of training experiences for residents.
• To address disparities in patient complexity and disease categories across clinical panels.

Main Methods:

• Encounter data from primary care residency clinics (Massachusetts General Hospital, 2008-2010) were analyzed.
• Patients were characterized by demographics, 44 disease categories (acute/chronic), and complexity.
• Patient reassignment algorithms sorted patients by complexity and allocated them to less complex panels, tested in 3 scenarios (within preceptor, group, or practice-wide).

Main Results:

• All 3 reassignment algorithms significantly reduced the standard deviation of disease categories/diagnoses across residents compared to baseline.
• Reassignment across the entire clinic and within groups yielded the best results.
• These improvements came with an initial reduction in patient-preceptor continuity.

Conclusions:

• Systematic patient panel reallocation can enhance the consistency and breadth of resident training.
• This approach offers a scalable method for health care system reform involving patient or clinician turnover.
• Balancing case-mix is key to optimizing medical education in residency programs.