Using AI-Driven Strategies to Optimize Specialty Drug Costs, Manage Polypharmacy

As health care costs rise, managed care organizations are increasingly turning to artificial intelligence (AI)-driven solutions to control specialty drug spending and manage polypharmacy. Research from the Academy of Managed Care Pharmacy 2025 conference highlights how AI approaches can help optimize billing and identify high-risk patients, delivering significant financial and clinical benefits.1,2

As health care costs rise, managed care organizations turn to artificial intelligence (AI)-driven solutions to control specialty drug spending and manage polypharmacy. | Image credit: Burlingham - stock.adobe.com

Evaluating Cost-Effective Billing for Specialty Medications

A systematic, AI-driven approach has significant potential to reduce costs and optimize specialty drug billing.1 Researchers developed an algorithm that identifies the most cost-effective billing method for specialty drugs, addressing the challenge of varying reimbursement rates between pharmacy and medical benefits. Integrated into a bot capable of ingesting any payer’s medical claims, the algorithm provides real-time recommendations to streamline decision-making.

This study, presented in a poster, analyzed medical claims from March 1 to August 31, 2024, within a regional health plan. Claims included Healthcare Common Procedure Coding System (HCPCS) codes, HCPCS unit quantities, and ingredient cost paid (ICP). The research team simulated pharmacy benefit claims for specialty drugs traditionally billed under the medical benefit. By leveraging average wholesale price (AWP) data from Medi-Span and integrating specialty pharmacy AWP discounts, the algorithm compared expected ICP under pharmacy benefits with ICP under medical benefits to determine the optimal billing strategy.

Findings are still in progress, but the researchers believe that the approach will both enhance financial efficiency and offer a replicable framework for reducing drug costs across the health care system.

AI-Driven Identification of High-Risk Polypharmacy Patients

AI-driven medication management can significantly reduce health care utilization and costs by proactively identifying high-risk hyper-polypharmacy patients, as highlighted in another study.2

With over 40% of older adults classified as polypharmacy (≥ 5 medications) or hyper-polypharmacy (≥ 10 medications) patients, adverse drug reactions (ADRs) contribute to up to 30% of hospitalizations and impose a $50 billion annual burden on the US health care system. By leveraging AI-driven insights, this study underscores the potential for more efficient and cost-effective medication management strategies to mitigate the clinical and financial impact of polypharmacy.

The study analyzed deidentified patient data from the Optum Market Clarity database, focusing on 26,277 patients aged 65 and older who had been prescribed at least 5 medications for 120 days or more between January 1, 2023, and December 31, 2023. Researchers compared the AI-assisted platform to conventional methods in identifying patients at risk for ADRs, emergency department (ED) visits, and hospitalizations. Treatment costs were then evaluated based on intervention programs targeting identified populations.

The AI-assisted platform detected 80% of high-risk patients missed by conventional methods, leading to a 6% reduction in hospitalization rates and a 2.6-fold decrease in ADR-related ED visits. The findings suggest that integrating AI into medication management could help payers anticipate substantial savings while improving patient outcomes.

Both studies highlight the importance of AI-driven solutions in managed care and how new tools can help reinforce the value of data-driven strategies in controlling specialty drug expenses and managing polypharmacy risks.

References

1. Zhu J, Stapley M, Wilson A. Artificial intelligence-driven optimization of medical versus pharmacy benefit strategies for cost efficiency. Presented at: AMCP 2025; March 31-April 3, 2025; Houston, TX. Poster U31.

2. Roy A, Verma V, Mishra R, Nayyar A, Goyal R, Marken R. AI-assisted medication management platform for hyper-polypharmacy patients to reduce healthcare expenditure in the US. Presented at: AMCP 2025; March 31-April 3, 2025; Houston, TX. Poster U21.