Eden Hochbaum

Background: Recent advances in software systems hold great promise for transforming clinical decision-making in primary care^2,6,8. However, adoption of these technologies in the U.S. remains limited^2,6.  This gap is known as the “AI Chasm².”

Objectives: This work aims to (1) evaluate the current landscape of clinical decision-making technologies, (2) identify technical, social, and economic barriers to widespread adoption in the U.S., and (3) identify particular aspects of clinical workflows that are primed for significant near-term disruption through automation.

Search Methods: A scientific- and business-literature review was conducted with special attention paid to articles published in the past 2 years in high-impact journals.  Special consideration was devoted to (1) the application of Large Language Models (LLMs) to clinical decision-making tasks, (2) diagnostic technologies in primary care, and (3) automated Electronic Health Record (EHR) alerts for patients at risk for cerebrovascular accidents (strokes).

Results: Recent technology advances have the potential to significantly enhance clinical workflows, improving outcomes (e.g., more accurate diagnoses, reduced iatrogenic harm)^1,2,6,7,8, patient experience (e.g., efficient data collection, empathetic communication)², access², equity^4,9, and affordability^3,5.  However, significant barriers to American adoption persist.  These include regulatory constraints^2,6,8, liability concerns^2,6, patient resistance⁶, sunk costs in existing systems⁶, provider resistance⁶, and incompatible billing models⁶.

Conclusion: Automation in clinical decision-making is likely to expand gradually, driven mainly by efforts to reduce labor costs^1,6.  In the near term, technology will assist rather than replace human providers, primarily through optional decision-support tools like alerts^5,10. Fully automated, closed-loop diagnostic systems remain out of reach—not due to strict technical limitations, but because of social and regulatory obstacles².

Works Cited:

1.  Ballestero, M., de Souza, L.C., Levada, A.L.M. et al. Is artificial intelligence superior to traditional regression methods in predicting prognosis of adult traumatic brain injury?. Neurosurg Rev 48, 355 (2025). https://doi.org/10.1007/s10143-025-03506-0
2. D’Adderio, L., Bates, D.W. Transforming diagnosis through artificial intelligence. npj Digit. Med. 8, 54 (2025). https://doi.org/10.1038/s41746-025-01460-1
3. Hart, L. Gary PhD*; Deyo, Richard A. MD, MPH,†‡; Cherkin, Daniel C. PhD*§. Physician Office Visits for Low Back Pain: Frequency, Clinical Evaluation, and Treatment Patterns From a U.S. National Survey. Spine 20(1):p 11-19, January 1995.
4. Lopez L, Wilper AP, Cervantes MC, Betancourt JR, Green AR. Racial and sex differences in emergency department triage assessment and test ordering for chest pain, 1997–2006. Acad Emer Med. 2010; 17(8):801–808. [PubMed: 20670316]
5. Shafieioun, A., Ghaffari, H., Baradaran, M. et al. Predictive power of artificial intelligence for malignant cerebral edema in stroke patients: a CT-based systematic review and meta-analysis of prevalence and diagnostic performance. Neurosurg Rev 48, 318 (2025). https://doi.org/10.1007/s10143-025-03475-4
6. Shah, W.S., Elkhwesky, Z., Jasim, K.M. et al. Artificial intelligence in healthcare services: past, present and future research directions. Rev Manag Sci 18, 941–963 (2024). https://doi.org/10.1007/s11846-023-00699-w
7. Singh H , Giardina TD , Meyer AN , Forjuoh SN , Reis MD , Thomas EJ . Types and origins of diagnostic errors in primary care settings. JAMA Intern Med. 2013;173(6):418-425. doi:10.1001/jamainternmed.2013.2777
8. Tu, T., Schaekermann, M., Palepu, A. et al. Towards conversational diagnostic artificial intelligence. Nature (2025). https://doi.org/10.1038/s41586-025-08866-7
9. Van Ryn M, Burgess D, Malat J, Griffin J. Physicians’ perceptions of patients’ social and behavioral characteristics and race disparities in treatment recommendations for men with coronary artery disease. Am J Public Health. 2006; 96(2):351–357. [PubMed: 16380577]
10. Viralkumar Vaghani, Li Wei, Umair Mushtaq, Dean F Sittig, Andrea Bradford, Hardeep Singh, Validation of an electronic trigger to measure missed diagnosis of stroke in emergency departments, Journal of the American Medical Informatics Association, Volume 28, Issue 10, October 2021, Pages 2202–2211, https://doi.org/10.1093/jamia/ocab121