UMich deploys an AI-based system to combat fraud, waste, and abuse in healthcare, reducing costs and safeguarding patients. Dr. Mohammed Saeed highlights AI tools’ potential to improve provider practices, preventing inappropriate care. Health at Scale’s specialized algorithms deeply analyzes transactions, aiming to avoid provider abrasion and identify missed expenditures. The system’s rapid response time and high accuracy offer promising results, showing potential for immediate ROI and a long-term impact on patient safety. Seamless integration of AI tools into existing workflows is recommended for effective implementation.
UMich is implementing an AI-based system to combat fraud, waste, and abuse in healthcare, aiming to reduce costs and safeguard patients. Dr. Mohammed Saeed from the University of Michigan Medical School emphasizes the potential of AI tools to enhance provider practices, protecting patients from unnecessary or inappropriate care.
The United States faces soaring healthcare costs, with up to one-third of expenditures attributed to wasteful or fraudulent services. Underserved communities are especially vulnerable, receiving unnecessary care that burdens them financially and can harm their well-being.
Dr. Mohammed Saeed, a cardiologist, and researcher, points out that existing utilization management and payment integrity systems have limitations. They lack precision and a nuanced understanding of patients, providers, and healthcare settings. Consequently, treatments appropriate for one patient may be wasteful for another, leading to errors in identifying fraud and abuse.
To address these challenges, IT vendor Health at Scale has developed specialized AI algorithms. These algorithms deeply analyze each transaction, considering historical patient data and provider practice patterns while incorporating the latest evidence and care guidelines. By focusing on a small fraction of providers responsible for the most egregious inappropriate care, the system aims to minimize provider abrasion.
The AI algorithms are intended for use in prior authorization and claim adjudication workflows. By scanning administrative claims and transactions, they aim to prevent fraud, waste, and abuse before care is delivered or payments are made. The system can integrate with existing platforms through APIs or standalone interfaces.
While the University of Michigan Medical School has not yet published results, Health at Scale reports promising outcomes with other healthcare providers. The response time of their system is under 200 milliseconds, and it identifies 3% to 7% of missed spending by existing payment integrity tools. Approximately 95% of flagged activities were correctly classified as fraud, waste, or abuse due to multiple factors considered in the decision-making process.
Dr. Saeed advises healthcare organizations to opt for simple and scalable fraud, waste, and abuse systems. Seamless integration of AI-based tools into existing workflows can lead to an immediate return on investment. He encourages organizations to prioritize the prevention of inappropriate care services and embrace innovative AI-driven systems to improve provider practice patterns and protect patients from unnecessary harm.
Overall, the implementation of AI technology holds significant potential to curb healthcare costs and enhance patient safety by identifying and preventing fraud, waste, and abuse in the healthcare system.
