Low-Value Services and Longitudinal Knowledge Assessment Performance

Importance: Low-value services (LVSs) remain common and account for a substantial amount of wasteful US health care spending. Objective: To assess whether LVS use is associated with physicians' general clinical knowledge as measured by performance on the American Board of Internal Medicine's Longitudinal Knowledge Assessment (LKA). Design, Setting, and Participants: This cross-sectional study included general internists enrolled in the LKA in 2022 or 2023 caring for Medicare fee-for-service beneficiaries aged 67 years or older in an outpatient setting during their first year of participation. Exposure: Physician first-year LKA scores. Main Outcomes and Measures: Receiving any of 25 LVSs designed to measure low-value primary care. Patient level regressions were adjusted for patient demographics, comorbidities, physician practice characteristics, and patient location. Results: Of 898 365 included patients, 59% were female with a mean (SE) age of 76.8 (0.02) years, and 30% received at least 1 LVS. Of 7089 physicians, 3568 (50%) were male with a mean (SE) age of 52.5 (0.1) years. Having a physician with a top vs bottom quartile LKA score was associated with a logistic-regression adjusted 7.9% lower (95% CI, -10.5% to -5.3%, P < .001) chance of receiving any LVS during the year (bottom: 31.0% [95% CI, 30.4% to 31.7%]; top: 28.6% [95% CI, 28.1% to 29.1%]); absolute difference: -2.5 percentage points (95% CI, -3.3 to -1.6, P < .001). Significant differences were observed by LVS subtype including diagnostic or preventive testing (bottom: 10.4% [95% CI, 9.8% to 11.0%], top: 8.7% [95% CI, 8.3% to 9.1%]; absolute difference: -1.7 percentage points [95% CI, -2.5 to -0.9]; P < .001), cancer screening (bottom: 13.8% [95% CI, 13.4% to 14.3%], top: 12.3% [95% CI, 11.9% to 12.7%]; absolute difference: -1.5 [-2.1 to -0.9], P < .001), and imaging (bottom: 13.8% [95% CI, 13.4% to 14.2%]; top: 13.2% [95% CI, 12.9% to 13.5%]; absolute difference: -0.6 [95% CI, -1.1 to -0.1]; P = .02). LKA score quartile differences for individual LVSs were most prominent for total and free triiodothyronine testing in patients with hypothyroidism and prostate-specific antigen testing in males 75 aged years and older. Conclusions and Relevance: The findings of this cross-sectional study suggest patients cared for by physicians with higher first-year LKA scores were less likely to receive LVSs.