Important demographic variables impact the musculoskeletal knowledge and confidence of academic primary care physicians

Background: Although most musculoskeletal illness is managed by primary care providers, and not by surgeons, evidence suggests that primary care physicians may receive inadequate training in musculoskeletal medicine. We evaluated the musculoskeletal knowledge and self-perceived confidence of fully trained, practicing academic primary care physicians and tested the following hypotheses: (1) a relationship exists between a provider's musculoskeletal knowledge and self-perceived confidence, (2) demographic variables are associated with differences in the knowledge-confidence relationship, and (3) specific education or training affects a provider's musculoskeletal knowledge and clinical confidence. Methods: An examination of basic musculoskeletal knowledge and a 10-point Likert scale assessing self-perceived confidence were administered to family practice, internal medicine, and pediatric faculty at a large, regional, academic primary care institution serving both rural and urban populations across a five-state region. Subspecialty physicians were excluded. Individual examination scores and self-reported confidence scores were correlated and compared with demographic variables. Results: One hundred and five physicians participated. Ninety-two physicians adequately completed the musculoskeletal knowledge examination. Fifty-nine (64%) of the ninety-two physicians scored < 70%. Higher examination scores were associated with male gender (p = 0.01) and participation in a musculoskeletal course (p = 0.009). Practitioners who took elective courses demonstrated higher scores compared with those who took required courses (p = 0.014). Greater musculoskeletal confidence was associated with the number of years in clinical practice (p = 0.045), male gender (p = 0.01), residency training in family practice (p < 0.00001), and prior participation in a musculoskeletal course (p = 0.0004). Physicians demonstrated greater confidence with medical issues than with musculoskeletal issues (mean confidence scores, 8.3 and 5.1, respectively; p < 0.00001). Higher scores for musculoskeletal knowledge correlated significantly with increasing levels of musculoskeletal confidence (r = 0.416, p < 0.0001). Conclusions: Although a large proportion of primary care visits are for musculoskeletal symptoms, the majority of primary care providers tested at a large, regional, academic primary care institution failed to demonstrate adequate musculoskeletal knowledge and confidence. Further characterization of the relationship between knowledge and confidence and its association with demographic variables might benefit the education of musculoskeletal providers in the future.