Using AI to Increase Clinical Pathway Value in Oncology

Key Takeaways

• Although clinical pathways have wide adoption among oncologists, their value may be limited by data and scope restrictions, inadequate documentation, and differing performance measures from payers and health systems.
• Pathways measure drug costs and treatment adherence while payers look at health outcomes when evaluating coverage for a medication.
• Artificial intelligence (AI) tools such as ConcertAI can improve data utilization and support treatment decision-making among patients and physicians.

At the Clinical Pathways Congress (CPC) 2026, Carole Tremonti, RN, MBA and Deirdre Saulet, PhD, discussed current challenges in treatment pathways and explored the ways AI could improve their value in patient-centered care.

While adoption of clinical pathways is widespread, with 70% of oncologists using pathways and 56% of treatment decisions following a determined pathway, the proof of their value has yet to be established.

Competing Goals Between Pathways, Payers, and Health Systems

Pathways measure compliance, but payers look at health outcomes after 12 months, creating a disconnect between clinical and market performance measures. From a payer perspective, lower symptom burden, longer treatment duration, and delayed disease progression are significant markers when determining coverage for a drug regimen. Pathways, on the other hand, evaluate on-pathway rates, drug costs, and treatment adherence.

In addition, current pathways fail to answer 3 critical health system questions:

1. Before treatment: What support will the patient need?
2. During treatment: Who follows the pathway, and why not?
3. After treatment: What did the decision return?

According to Tremonti and Saulet, answering these questions requires data beyond a single organization. They also identify 5 main challenges in the current pathway landscape:

1. Inaccessible records: suboptimal utilization of patient data impedes treatment decision-making.
2. Varied rules and policies: each payer has an individual pathway which clinicians must translate by hand.
3. Narrow scope: pathways do not measure support needs, toxicity, or medical costs.
4. Inadequate documentation: reasons for off-pathway decisions are rarely recorded.
5. Limited performance measures: pathways only measure adherence, not 12-month results.

Increasing the Value of Clinical Pathways

Tremonti promoted the use of AI tools, specifically ConcertAI, in improving the value of pathways. ConcertAI is designed to build decision intelligence, which involves determining the right treatment and necessary support, creating a case for coverage, and recording proof.

According to the speakers, data matters because payers value real-world evidence, which proves a drug’s efficacy. A program like ConcertAI may improve utilization of collected data by identifying appropriate support for patients and measuring decision concordance in addition to adherence.

The goal of ConcertAI is to reduce physician workloads and administrative burdens by bridging interoperability, adoption, and governance. However, Tremonti emphasized the importance of human involvement in the model, with people prioritizing treatments and deciding what to focus on and evaluate in pathways.

Reference

Tremonti C, Saulet D. Pathway design 2.0: integrating data, AI, and real-world evidence. Presented at: Clinical Pathways Congress; November 10-11, 2026; Boston, MA.