Profiling medical specialties and informing aspiring physicians: a data-driven approach

Area of Science:

• Medical Education
• Health Services Research
• Data Science in Medicine

Background:

• Informed career decisions are crucial for medical students.
• Understanding inter-relations among medical fields aids specialty selection.
• A data-driven approach can objectively map medical specialty landscapes.

Purpose of the Study:

• To cluster medical specialties based on working environment similarities.
• To provide an unbiased perspective on inter-relations among medical fields.
• To guide medical students in their specialty choices.

Main Methods:

• Merged data from Canadian health information databases (2018-2019).
• Utilized Latent Profile Analysis (LPA) on 25 medical specialties and 36 variables.
• Employed Kruskal-Wallis test to identify significant differentiating variables.

Main Results:

• Identified seven distinct clusters (profiles) of medical specialties.
• Significant differences found in income, work settings, and payment styles across profiles.
• Demographics, professional satisfaction, and work-life balance satisfaction did not significantly vary between profiles.

Conclusions:

• Latent Profile Analysis successfully grouped 25 medical specialties into meaningful profiles.
• These profiles reflect common themes and indicator variables (e.g., procedural vs. clinic-based).
• The identified profiles offer valuable guidance for aspiring physicians selecting a medical specialty.