Interpreting and reading pharmacy benefit manager contracts and maintaining 100% oversight of pharmacy claims can be some of the biggest challenges in the managed care pharmacy space.
But, according to CareSource Vice President of Pharmacy Jessica Hutton, PharmD, BCACP, and Nick Trego, PharmD, Senior Vice President, Pharmacy, managing pharmacies are equipped with evolving computer system tools such as artificial intelligence and machine learning. This has led to improved efficiency in dealing with these issues. President, Clinical Analytics and Client Services, HealthPlan Data Solutions, Inc.
Hatton and Trego presented on these tools and their benefits for the PBM space at this year’s AMCP Nexus conference in Orlando.
While AI is still being developed every day, one of its few subsets, machine learning (ML), helps machines extract knowledge from data and learn autonomously. Together, these tools allow users to significantly improve productivity and performance accuracy.
As mentioned earlier, AI and ML are addressing the challenges associated with reading PBM contracts.
For example, Hutton shared that the volume and complexity of contracts, which often include key “single-sentence sentencing concepts,” can be efficiently managed through these technologies.
Currently, these contracts are done manually and disorganization is widespread. These factors lead to avoidance by plans that often limit interaction throughout the contract.
Through their experience, they shared that the integration of AI and ML in contract reading can address these challenges by:
- Rapid contract review: AI and ML enable full contract and amendment reviews within minutes. This is a big difference compared to the days or weeks it usually takes.
- Indexing key terms or elements: These tools can be trained to identify specific elements within a contract and can find important terms within a document.
- Digital contract repository: AI and ML create a hub for all contracts and amendments, organized by effective date or type.
The next challenge these tools address is 100% monitoring of pharmacy claims. This task traditionally requires a large amount of processing each month and requires complex and variable logic and custom plan design. Currently, PBM practices include cursory reviews, partial monitoring, or reliance on annual audits.
Hatton and Trego emphasized that AI and ML tools can efficiently manage PBM claims through:
- High-throughput technology platform: Quickly find solutions that guarantee 100% accuracy of pharmacy claims.
- Customized plan design model: Monitor your claims and train your software engine to identify issues from your plan settings.
- Near real-time and retrospective approaches: Identify errors at the time they occur and at the end of the year. You can also avoid repeating unintended costs by making changes mid-year.
- Fraud, waste and abuse: tool IIdentify errors specific to FWA using pre-programmed decision tree logic. This trains AI models to identify emerging trends at the provider, patient, and pharmacy levels.
Hutton also shared real-world situations that CareSource has experienced when AI and ML solve problems in the pharmacy space.
For example, in our insulin copay cap case study, implementation of state laws regarding insulin copay caps created challenges that required custom coding changes. AI and ML systems identify errors in out-of-pocket awarding, ensuring quick correction, compliance with state laws, and prevention of future errors.
Another case study regarding COVID-19 testing mandates required last-minute notice of coding changes to quickly integrate CMS coverage of commercially available COVID-19 tests. AI/ML models effectively monitor test logic, identify discrepancies and ensure compliance with initiatives, minimizing unnecessary planning costs.
They shared that beyond contract reading and billing monitoring, AI and ML are also being used for predictive modeling for member engagement, customer service, and targeted outreach. Additionally, these tools enhance member interaction, connect with patients at highest risk based on social determinants of health, assist with prescription management, and support base modeling based on a plan’s unique demographics. To do.