Peter G. Brodeur

More than 65 years ago, complex clinical diagnostic reasoning cases were introduced as the gold standard for the evaluation of expert medical computing systems, a standard that has held ever since. In this study, we report the results of a physician evaluation of a large language model (LLM) on challenging clinical cases across five experiments with a baseline of hundreds of physicians. We then report a real-world study comparing human expert and artificial intelligence (AI) second opinions in randomly selected patients in the emergency room of a major tertiary academic medical center. In all experiments, the LLM outperformed physician baselines and displayed continued improvement from prior generations of AI clinical decision support. Our study suggests that LLMs have eclipsed most benchmarks of clinical reasoning, motivating the urgent need for prospective trials.

Background: Leadless pacemakers (LPM) have been shown to be safe and effective alternatives to transvenous pacing systems. Few studies have evaluated the incidence and associated costs of post‐implant complications. The objectives of this study were to assess risk factors and causes for 1‐year mortality and all‐cause readmission, as well as characterize the total cost of care associated with index procedures and readmissions. Methods: LPM procedures, including inpatient and outpatient encounters, were captured in the Healthcare Cost and Utilization Project data in Florida, Maryland, and New York from 2016 to 2020 with 1‐year follow‐up through 2021. Cox proportional hazards regression was used to identify patient demographics, facility volume, and comorbid risk factors for 1‐year all‐cause readmission and in‐hospital mortality. Costs of inpatient cases and readmission were captured. Methods: Among 7127 patients receiving LPM, 3% died during the initial episode of care. The 1‐year all‐cause readmission rate was 45.9%, and the in‐hospital mortality rate was 8.8%. Comorbid heart failure (CHF), atrial fibrillation/flutter, chronic kidney disease, and diabetes increased the risk of 1‐year all‐cause readmission and in‐hospital mortality ( p < 0.05). CHF was the most common cause of readmission (17%). Inpatient cases resulted in a cost of $257 million, with readmissions increasing costs by 44.4%. Conclusions: The large healthcare expenditure derives from high rates of readmission and in‐hospital mortality, with readmissions potentially representing a modifiable target. CHF is a prominent cause of poor outcomes, which suggests the need to consider the overlapping roles of conduction system pacing, goal‐directed medical therapy, and close clinical follow‐up.

Background: In correlation with a growing body of evidence regarding nonoperative management for Achilles tendon rupture (ATR), studies from Europe and Canada have displayed a decreasing incidence in surgical management, which has not been noted in the United States. The primary objective of this study is to evaluate the US trend in ATR repair volume. Methods: The American Board of Orthopaedic Surgery (ABOS) Part II Oral Examination Case List Database was used. All cases using Current Procedural Terminology codes for primary ATR repair were requested from the years 2006-2020. Total submitted Achilles repair volume, the number of candidates submitting an Achilles repair case, and the overall submitted case volume per examination year was analyzed. Poisson and linear regressions were used to determine statistically significant trends. Results: The total number of Achilles repair cases submitted for the ABOS Part II Oral Examination significantly increased from 2006 to 2011 and then decreased until 2020. Taking Achilles repair cases as a proportion of total orthopaedic cases submitted, the same trend was seen. The number of candidates submitting an Achilles repair case increased from 2006 to 2009 and then decreased until 2020. Foot and Ankle fellowship-trained candidates submitted an increasing number of ATR repair cases per candidate during the time period studied. Conclusion: This is the first study to demonstrate a decline in the volume of ATR repair in the United States. The decline in ATR repair volume seen in the ABOS Part II Case Lists does not match previously published US surgeon practice patterns but is not necessarily generalizable to beyond this period. Although the overall ATR repair volume in the ABOS Part II Case Lists is decreasing, we found Foot and Ankle fellowship-trained surgeons are operating on an increasing number of ATRs during their board collection period. Level of Evidence: Level III, retrospective cohort study.

Background: Distal radius fractures account for nearly 25% of fractures in adults, with a trend toward operative fixation. The objective of this study was to assess the relationship between surgeon and hospital volume with complications following distal radius fixation. Methods: A retrospective study was performed using the New York Statewide Planning and Research Cooperative System database from 2009 to 2015. Outpatient claims were identified for distal radius fractures and surgery. The facility and surgeon’s identifier were used to calculate annual procedure volume. The risk for infection, carpal tunnel surgery, and revision/hardware removal was analyzed, and Social Deprivation Index (SDI) was linked to each patient. Patient demographics and rate of complications were compared across hospital and physician volume. Results: A total of 14 748 patients were included, finding Federal and self-pay insurance associated with low-volume (LV) facility care and private insurance with high-volume (HV) facilities. The SDI for patients treated by LV surgeons and hospitals was significantly higher compared with HV providers. Low-volume facilities and surgeons had a higher 3-month risk of infection requiring reoperation. High-volume facilities were less likely to treat Hispanic patients, those with comorbidities, higher SDI, and with Federal or self-pay insurance. Conclusions: Patients treated by LV surgeons and facilities had a higher risk of infection requiring surgery within 3 months than those treated by HV providers. Low-volume facilities were more likely to treat patients who were Hispanic, Federally insured, and with comorbidities and higher SDI than HV facilities, increasing their risk for disadvantaged care. Level of Evidence: Level III.

Objectives: The aim of this study was to characterize the case volume dependence of both facilities and surgeons on morbidity and mortality after femoral shaft fracture (FSF) fixation. Methods: Adults who had an open or closed FSF between 2011 and 2015 were identified in the New York Statewide Planning and Research Cooperative System database. Claims were identified by International Classification of Disease-9, Clinical Modification diagnostic codes for a closed or open FSF and International Classification of Disease-9, Clinical Modification procedure codes for FSF fixation. Readmission, in-hospital mortality, and other adverse events were compared across surgeon and facility volumes using multivariable Cox proportional hazards regression, controlling for patient demographic and clinical factors. Surgeon and facility volumes were compared between the lowest and highest 20% to represent low-volume and high-volume surgeons/facilities. Results: Of 4,613 FSF patients identified, 2,824 patients were treated at a high or low-volume facility or by a high or low-volume surgeon. Most of the examined complications including readmission and in-hospital mortality showed no statistically significant differences. Low-volume facilities had a higher 1-month rate of pneumonia. Low-volume surgeons had a lower 3-month rate of pulmonary embolism. Conclusion: There is minimal difference in outcomes in relation to facility or surgeon case volume for FSF fixation. As a staple of orthopaedic trauma care, FSF fixation is a procedure that may not require specialized orthopaedic traumatologists at high-volume facilities.

Introduction: This study seeks to evaluate (1) the relationship between hospital and surgeon volumes of shoulder arthroplasty and complication rates and (2) patient demographics/socioeconomic factors that may affect access to high-volume shoulder arthroplasty care. Methods: Adults older than 40 years who underwent shoulder arthroplasty between 2011 and 2015 were identified in the New York Statewide Planning and Research Cooperative System database using International Classification of Disease 9/10 and Current Procedural Terminology codes. Medical/surgical complications were compared across surgeon and facility volumes. The effects of demographic factors were analyzed to determine the relationship between such factors and surgeon/facility volume in shoulder arthroplasty. Results: Seven thousand seven hundred eighty-five patients were included. Older, Hispanic/African American, socially deprived, nonprivately insured patients were more likely to be treated by low-volume facilities. Low-volume facilities had higher rates of readmission, urinary tract infection, renal failure, pneumonia, and cellulitis than high-volume facilities. Low-volume surgeons had patients with longer hospital lengths of stay. Discussion: Important differences in patient socioeconomic factors exist in access to high-volume surgical care in shoulder arthroplasty, with older, minority, and underinsured patients markedly more likely to receive care by low-volume surgeons and facilities. This may highlight an area of potential focus to improve access to high-volume care.

Socioeconomic status, race, and insurance status are known factors affecting adult orthopaedic surgery care, but little is known about the influence of socioeconomic factors on pediatric orthopaedic care. The purpose of this study was to determine if demographic and socioeconomic related factors were associated with surgical management of pediatric supracondylar humerus fractures (SCHFs) in the inpatient versus outpatient setting. Pediatric patients (p < 0.0001), Asian (OR: 2.159, p < 0.0001) and African American (OR: 2.095, p < 0.0001) patients to have increased odds of inpatient treatment relative to White patients. Injury diagnosis on a weekend had increased odds of inpatient management (OR: 1.863, p = 0.0002). Higher social deprivation was also associated with increased odds of inpatient treatment (OR: 1.004, p < 0.0001). There are disparities among race and socioeconomic status in the surgical setting of SCHF management. Physicians and facilities should be aware of these disparities to optimize patient experience and to allow for equal access to care.