of University of Florida (UF) announced an official update on a new study that uses machine learning to predict discontinuation of opioid treatment. Below is his official release from UF, written by Tyler Francischine.
Gainesville, Florida — Researchers at the University of Florida have developed a system designed to identify patients at high risk of discontinuing buprenorphine treatment due to opioid use disorder.
Buprenorphine, an FDA-approved prescription drug, is one of three over-the-counter treatments for opioid use disorder that have proven effective in treating both pain and addiction.
In a study published in the journal Computers in Biology and Medicine, Dr. Md Mahmudul Hasan and his research team found that approximately 15% of patients did not complete the clinically recommended one year of buprenorphine treatment, and approximately We found that 46% discontinued it. Get treatment within the first 3 months. With the help of artificial intelligence (AI), the team also identified several factors associated with high-risk patients and treatment discontinuation.
Hasan, who is an assistant professor in the Department of Pharmaceutical Outcomes and Policy at the University of Florida College of Pharmacy and with the Department of Information Systems and Operations Management in the Warrington College of Business, said the retrospective study included subjects aged 18 to 64 who: He said the insurer was included. Buprenorphine prescribed to treat opioid use disorder Researchers provide new insights to help fight national public health epidemic that claimed more than 80,000 lives in the U.S. in 2021 .
The study measured a gap of 30 days or more when a buprenorphine prescription was not filled within the first year of treatment. By building predictive models that focused on different treatment stages (starting treatment, 1 month, and 3 months after starting treatment), Hasan's team found that nearly 15% of patients stopped treatment prematurely. I discovered that The research team noted that this is a conservative estimate as some patient exclusion criteria may have lowered the discontinuation rate.
“We know that continuing the buprenorphine treatment regimen is beneficial. Early discontinuation can increase the risk of hospitalization, drug overdose, and most importantly, death.” said Hasan. “If AI can be used to predict which patients are at high risk for this behavior, clinicians can get to the root cause, make more informed decisions, and be more targeted for those patients. This will allow us to design targeted interventions.”
Hasan's team used machine learning prediction and risk stratification frameworks to identify high-risk patients and determine which factors contributed to lack of buprenorphine treatment compliance. Risk factors identified in this study include age, gender, early treatment adherence, use of stimulants or antipsychotics, and days of supply related to the first buprenorphine prescription a patient receives. The study also found that living in a rural area and other treatment access barriers contributed to an increased risk of treatment discontinuation.
“Younger patients are at increased risk of discontinuing treatment prematurely, as are patients with a history of stimulant use, including nicotine,” Dr. Hasan said. “We also found that patients with poor buprenorphine adherence early in treatment were at higher risk of discontinuing treatment early.”
Hasan said that if the technology developed in this study is made available to medical centers across the country, patients will have greater access to buprenorphine treatment, while saving valuable time for front-line clinicians.
“Primary care physicians are already overburdened, overworked, and have limited resources. Tools like this that can reliably predict which patients are at high risk could be helpful.” Hasan said. “Healthcare providers can quickly identify the interventions needed for each patient without increasing their workload, allowing them to optimally allocate limited resources.”
Javed Al Faisal, a graduate student at the University of California, was the lead author of the study.
