Skip to main content Skip to footer
Case study

The challenge

Drug addiction interferes with positive health outcomes for patients being treated for other conditions. Additionally, caregivers providing drug-addiction treatment must divert much-needed resources from other patients.

Treating addiction is very expensive. U.S. healthcare organizations spend more than $500 billion annually caring for patients suffering from opioid addiction alone. Across a large healthcare organization it’s challenging to consistently identify patients at risk of becoming addicted and alert physicians to that risk. Our client asked us to explore ways to identify potential drug-seeking behavior to lessen the incidence of addiction and lower healthcare costs.

Our approach

People looking to secure opioids or other addictive drugs tend to behave in predictable ways and have common characteristics. We proposed an artificial intelligence-based solution that links text analytics performed on physicians’ notes from patient visits—including their impressions of a patient’s behavior, appearance and diagnoses—with data in our client’s confidential third-party electronic medical records (EMR) system.

Our solution, which uses text analytics and advanced machine learning, generates system alerts for doctors during patients’ visits when a pattern of at-risk behavior is identified. This enables caregivers to intercede with patients in real time and take corrective actions.

Intuitive AI-based solution identifies potential drug-seeking behavior

We’re helping one of the nation’s largest integrated healthcare services companies implement an intuitive AI-based solution to identify potential drug-seeking behaviors to alert caregivers about patients at risk, improve health outcomes and lower treatment costs.


drug-seeking patients identified

$60 million

targeted organizational savings for this healthcare provider


behavior and symptoms in real time as patients interact with a physician