Production-Ready Retrieval-Augmented Generation and Agentic AI Systems for Healthcare Claims and Prior Authorization
Keywords:
Retrieval-Augmented Generation (RAG), Agentic AI Systems, Healthcare Claims Adjudication, Prior Authorization, Gradient Boosting Classifier, Support Vector Machine (SVM), Machine Learning in Healthcare, Intelligent Decision Support SystemsAbstract
Fragmented data systems, rule-based automation, and poor contextual reasoning capabilities continue to make healthcare claims adjudication and prior authorization complex and resource intensive processes. The work gives a production-ready system based on Retrieval-Augmented Generation (RAG) and agentic Artificial Intelligence systems that could improve the efficiency, accuracy, and scalability of healthcare processes. Real-world claims processing is simulated by using a secondary Kaggle dataset of 10,000 records and 15 features. This system is written in Python and tested on Gradient Boosting and Support Vector Machine (SVM) models, a RAG-augmented contextual retrieval layer and a multi-agent decision workflow. Accuracy, precision, recall, F1-score, and processing time are used as metrics to measure performance. Findings show that Gradient Boosting is more effective than SVM and that RAG and agentic integration lead to high predictive ability and efficiency. The architecture has lower processing time, high interpretability, and scalability, and presents an actual route to self-made production-scale intelligent healthcare claims automation systems in the real world.
Downloads
References
Ni, Y., Lingren, T., Hall, E.S., Leonard, M., Melton, K. and Kirkendall, E.S., 2018. Designing and evaluating an automated system for real-time medication administration error detection in a neonatal intensive care unit. Journal of the American Medical Informatics Association, 25(5), pp.555-563.
Psotka, M.A., Singletary, E.A., Bleser, W.K., Roiland, R.A., Hamilton Lopez, M., Saunders, R.S., Wang, T.Y., McClellan, M.B., Brown, N. and American Heart Association Prior Authorization Learning Collaborative*, 2020. Streamlining and reimagining prior authorization under value-based contracts: a call to action from the value in healthcare initiative’s prior authorization learning collaborative. Circulation: Cardiovascular Quality and Outcomes, 13(7), p.e006564.
Gaskin, C.M., Ellenbogen, A.L., Parkhurst, K.L. and Matsumoto, A.H., 2021. Use of a commercially available clinical decision support tool to expedite prior authorization in partnership with a private payer. Journal of the American College of Radiology, 18(6), pp.857-863.
Machireddy, J.R., 2022. Integrating predictive modeling with policy interventions to address fraud, waste, and abuse (fwa) in us healthcare systems. Advances in Computational Systems, Algorithms, and Emerging Technologies, 7(1), pp.35-65.
Sinsky, C.A., Biddison, L.D., Mallick, A., Dopp, A.L., Perlo, J., Lynn, L. and Smith, C.D., 2020. Organizational evidence-based and promising practices for improving clinician well-being. NAM perspectives, 2020, pp.10-31478.
Maliha, G., Gerke, S., Cohen, I.G. and Parikh, R.B., 2021. Artificial intelligence and liability in medicine: balancing safety and innovation. The Milbank Quarterly, 99(3), p.629.
Caplan, A.L., Teagarden, J.R., Kearns, L., Bateman-House, A.S., Mitchell, E., Arawi, T., Upshur, R., Singh, I., Rozynska, J., Cwik, V. and Gardner, S.L., 2018. Fair, just and compassionate: A pilot for making allocation decisions for patients requesting experimental drugs outside of clinical trials. Journal of medical ethics, 44(11), pp.761-767.
Caplan, A.L., Teagarden, J.R., Kearns, L., Bateman-House, A.S., Mitchell, E., Arawi, T., Upshur, R., Singh, I., Rozynska, J., Cwik, V. and Gardner, S.L., 2018. Fair, just and compassionate: A pilot for making allocation decisions for patients requesting experimental drugs outside of clinical trials. Journal of medical ethics, 44(11), pp.761-767.
McCullough, R., 2018. Merit-based Claim Adjudication for Cancer Treatment Toxicities–Policy Trends that Lower Downstream Costs. Journal of Insurance Medicine, 47(4), pp.236-248.
Machireddy, J.R., 2022. Revolutionizing claims processing in the healthcare industry: The expanding role of automation and AI. Hong Kong Journal of AI and Medicine, 2(1), pp.10-36.
Mandl, K.D., Gottlieb, D., Mandel, J.C., Ignatov, V., Sayeed, R., Grieve, G., Jones, J., Ellis, A. and Culbertson, A., 2020. Push button population health: the SMART/HL7 FHIR bulk data access application programming interface. NPJ digital medicine, 3(1), p.151.
Ju, J.H., Sim, B., Lee, J. and Lee, J.Y., 2022. Reimbursement of digital therapeutics: future perspectives in Korea. Korean circulation journal, 52(4), pp.265-279.
Egilman, A.C., Kapczynski, A., McCarthy, M.E., Luxkaranayagam, A.T., Morten, C.J., Herder, M., Wallach, J.D. and Ross, J.S., 2021. Transparency of regulatory data across the European medicines agency, health Canada, and US food and Drug Administration. Journal of Law, Medicine & Ethics, 49(3), pp.456-485.
Kaplan, S., Lang, N., Gasparic, M., Rainville, C., Haralabopoulos, G. and Borroni, E., 2019. Prescribing patterns from medical chart abstraction of patients administered lipegfilgrastim: a pilot study in Europe. Journal of Drug Assessment, 8(1), pp.70-76.
Ball, R., Toh, S., Nolan, J., Haynes, K., Forshee, R. and Botsis, T., 2018. Evaluating automated approaches to anaphylaxis case classification using unstructured data from the FDA Sentinel System. Pharmacoepidemiology and drug safety, 27(10), pp.1077-1084.
Goyal, N., Tsivgoulis, G., Malhotra, K., Katsanos, A.H., Pandhi, A., Alsherbini, K.A., Chang, J.J., Hoit, D., Alexandrov, A.V., Elijovich, L. and Fiorella, D., 2019. Minimally invasive endoscopic hematoma evacuation vs best medical management for spontaneous basal-ganglia intracerebral hemorrhage. Journal of NeuroInterventional Surgery, 11(6), pp.579-583.
Herzberg, S., Hansen, M., Schoonover, A., Skarica, B., McNulty, J., Harrod, T., Snowden, J.M., Lambert, W. and Guise, J.M., 2019. Association between measured teamwork and medical errors: an observational study of prehospital care in the USA. BMJ open, 9(10), p.e025314.
Webb, J., Shah, L.D. and Lynch, H.F., 2020. Ethically allocating COVID-19 drugs via pre-approval access and emergency use authorization. The American Journal of Bioethics, 20(9), pp.4-17.
Wiese, A.D., Griffin, M.R., Stein, C.M., Schaffner, W., Greevy, R.A., Mitchel Jr, E.F. and Grijalva, C.G., 2018. Validation of discharge diagnosis codes to identify serious infections among middle age and older adults. BMJ open, 8(6), p.e020857.
Jorstad, K.T., 2020. Intersection of artificial intelligence and medicine: tort liability in the technological age. Journal of Medical Artificial Intelligence, 3.
Eloranta, S. and Boman, M., 2022. Predictive models for clinical decision making: Deep dives in practical machine learning. Journal of internal medicine, 292(2), pp.278-295.
Szalados, J.E., 2021. Statutory Controls and Regulation of Pharmaceuticals and Medical Devices. In The Medical-Legal Aspects of Acute Care Medicine: A Resource for Clinicians, Administrators, and Risk Managers (pp. 547-571). Cham: Springer International Publishing.
Mouchantaf, R., Auth, D., Moride, Y., Raine, J., Han, S.Y. and Smith, M.Y., 2021. Risk management for the 21st century: current status and future needs. Drug Safety, 44(4), p.409.
Holbrook, A.M., Vidug, K., Yoo, L., Troyan, S., Schulman, S., Douketis, J., Thabane, L., Giilck, S., Koubaesh, Y., Hyland, S. and Keshavjee, K., 2022. Coordination of Oral Anticoagulant Care at Hospital Discharge (COACHeD): protocol for a pilot randomised controlled trial. Pilot and Feasibility Studies, 8(1), p.166.
Bajandas, F.F. and Ray, G.K., 2018. Implementation and Use of Electronic Case Management Systems in Federal Agency Adjudication. In Administrative Conference Recommendation (Vol. 3, pp. 44-49).
Bommasani, R., Hudson, D.A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M.S., Bohg, J., Bosselut, A., Brunskill, E. and Brynjolfsson, E., 2021. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258.
Sharma, A., Mahaffey, K.W., Gibson, C.M., Hicks, K.A., Alexander, K.P., Ali, M., Chaitman, B.R., Held, C., Hlatky, M., Jones, W.S. and Mehran, R., 2022. Clinical events classification (CEC) in clinical trials: report on the current landscape and future directions—proceedings from the CEC Summit 2018. American Heart Journal, 246, pp.93-104.
Coravos, A., Goldsack, J.C., Karlin, D.R., Nebeker, C., Perakslis, E., Zimmerman, N. and Erb, M.K., 2019. Digital medicine: a primer on measurement. Digital biomarkers, 3(2), pp.31-71.
Coravos, A., Goldsack, J.C., Karlin, D.R., Nebeker, C., Perakslis, E., Zimmerman, N. and Erb, M.K., 2019. Digital medicine: a primer on measurement. Digital biomarkers, 3(2), pp.31-71.
Herbst, R.B., Margolis, K.L., McClellan, B.B., Herndon, J.L., Millar, A.M. and Talmi, A., 2018. Sustaining integrated behavioral health practice without sacrificing the continuum of care. Clinical Practice in Pediatric Psychology, 6(2), pp.117-128.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
