Peter L. Elkin and Kenneth E. Leonard.

Peter L. Elkin, MD, left, and Kenneth E. Leonard, PhD, have been awarded a grant to assess risks for people with AUD who use prescription painkillers.

Grant Helps Assess Risks of Combined AUD, Opioid Use

Published December 11, 2018

Kenneth E. Leonard, PhD, professor of psychiatry and director of the UB Clinical and Research Institute on Addictions (CRIA), and Peter L. Elkin, MD, professor and chair of biomedical informatics, have been awarded a two-year, $386,000 grant from the National Institute on Alcohol Abuse and Alcoholism to assess what risks may occur for people with alcohol use disorder (AUD) who use prescription painkillers.

“People who are dependent on alcohol or who have alcohol use disorder are at greater risk for chronic pain for a variety of reasons.”
Professor of psychiatry and director of the UB Clinical and Research Institute on Addictions

Integrated Database Being Created

The grant will help the CRIA create an integrated database with information from the New York State Office of Alcoholism and Substance Abuse Services (OASAS) and the New York State Medicaid office. Leonard and Elkin are co-principal investigators on the grant.

“People who are dependent on alcohol or who have alcohol use disorder are at greater risk for chronic pain for a variety of reasons,” Leonard says. “For example, their alcohol use can lead to physical conditions that produce significant pain or keep them from adhering to medical regimens for certain diseases, leading to increased pain.” In addition, people with AUD are more prone to accidents.

Cross-Referencing Records Could Curtail Abuse

The number of people with AUD who also have and receive treatment for chronic pain — often with opioid-type painkillers — raises the risk of overdose or death from opioids, especially if combined with excessive alcohol use.

“By cross-referencing these records, we should be able to determine the risk of painkiller use and misuse, such as opioids and benzodiazepine, for patients with alcohol use disorder and look at the medical and treatment factors that increase risk in this population,” Leonard says.

Predictive Models Can Link People to Treatment

Leonard and Elkin will use machine learning to create predictive models to know which patients are at high risk for withdrawal or overdose. They will use these models to identify patients at risk for opioid overdoses and to link them to treatment before opioids become a problem.

“These findings will have important clinical implications for the management of patients in primary care with an unrecognized history of an alcohol use disorder as well as policy implications with respect to medical access to alcohol treatment records,” Elkin adds.