17 August 2025 in Journal of Intensive Care Medicine
Background: The future of artificial intelligence in medicine includes the use of machine learning and large language models to improve diagnostic accuracy, as a point-of-care tool, at the time of admission to an acute care hospital. The large language model, ChatGPT-4, has been shown to diagnose complex medical conditions with accuracies comparable to experienced clinicians, however, most published studies involved curated cases or examination-like questions and are not point-of-care. To test the hypothesis that ChatGPT-4 can make an accurate medical diagnosis using real-world medical cases and a convenient cut and paste strategy, we performed a retrospective case study involving critically ill patients admitted to a community hospital medical intensive care unit. Methods: A redacted H&P was essentially cut and pasted into ChatGPT-4 with uniform instructions to make a leading diagnosis and a list of 5 possibilities as a differential diagnosis. All features that could be used to identify patients were removed to ensure privacy and HIPAA compliance. The ChatGPT-4 diagnoses were compared with critical care physician diagnoses using a blinded longitudinal chart review as the ground truth diagnosis. Results: A total of 120 randomly selected cases were included in the study. The diagnostic accuracy was 88.3% for physicians and 85.0% for ChatGPT-4, with no significant difference by McNemar testing (p-value of 0.249). The agreement between physician diagnosis and ChatGPT-4 diagnosis was moderate, 0.57 (95% CI: 0.35–0.79), based on Cohen's kappa statistic. Conclusion: These results suggest that ChatGTP-4 achieved diagnostic accuracy comparable to board certified physicians in the context of critically ill patients admitted to a community medical intensive care unit. Furthermore, the agreement was only moderate, suggesting that there may be complementary ways of combining the diagnostic acumen of physicians and ChatGPT-4 to improve overall accuracy. A prospective study would be necessary to determine if ChatGPT-4 could improve patient outcomes as a point-of-care tool at the time of admission.
29 September 2026 in JCO Oncology Practice
PURPOSE: Immune checkpoint inhibitors (ICIs) have transformed cancer care but impose a growing financial burden on Medicare. We characterized Medicare Part B spending and utilization patterns for ICIs from 2018 to 2022 to quantify expenditure trends and identify key cost drivers. METHODS: We conducted a retrospective analysis of the Centers for Medicare & Medicaid Services (CMS) Medicare Part B Spending by Drug public summary files (aggregated Healthcare Common Procedure Coding System [HCPCS] drug-year data) from 2018 through 2022. For eight ICIs, we extracted total gross program spending, total dosage units (HCPCS billing units), and unique beneficiaries and calculated average spending per dosage unit, average spending per beneficiary, and growth metrics over time. RESULTS: In 2022, total Medicare Part B spending across the eight ICIs was approximately $8.49 billion US dollars (USD). Pembrolizumab accounted for the largest spending ($4.94 billion USD) and beneficiary utilization (67,022 beneficiaries), followed by nivolumab ($1.85 billion USD; 26,957 beneficiaries). Average spending per beneficiary in 2022 (the primary cost metric, capturing dose frequency and treatment duration) ranged from $53,508 USD (durvalumab) to $80,705 USD (atezolizumab). Average spending per dosage unit varied widely as a secondary pricing measure, from $27.02 USD (cemiplimab) to $216.09 USD (dostarlimab). CONCLUSION: Medicare Part B spending on ICIs is highly concentrated among pembrolizumab and nivolumab, and is driven primarily by beneficiary volume, with substantial heterogeneity in per-beneficiary spending across agents. As indications and utilization continue to expand, these therapies' fiscal impact warrants ongoing evaluation to inform Medicare payment policy and value-based oncology care.