Co-prescribing stimulants and opioids may fuel the twin epidemic of overdose

The combination of prescribed central nervous system stimulants, such as drugs that relieve ADHD symptoms, with prescribed opioid medications is associated with a pattern of escalating opioid intake, a new study has found.

The analysis of health insurance claims data from almost 3 million U.S. patients investigated prescribed stimulants' impact on prescription opioid use over 10 years, looking for origins of the so-called "twin epidemic" of combining the two classes of drugs, which can increase the risk for overdose deaths.

Combining the two drugs is associated with an increase in overdose deaths. This is something we know. But we didn't know whether stimulant use has a causal role in high use of opioids, so we conducted a big data analysis of how these two patterns interacted over a long period of time," said

Ping Zhang, senior study author, associate professor of computer science and engineering and biomedical informatics at The Ohio State University

"What we found is that if someone is taking a stimulant and an opioid at the same time, they're generally taking a high dose of the opioid," he said. "And if the patient in this study population takes the stimulant before beginning opioid use, they are more likely to have higher doses of subsequent opioids."

The study was published Feb. 17 in The Lancet Regional Health – Americas.

The research team obtained data on 22 million patients with 96 million opioid prescriptions from MarketScan Commercial Claims and Encounters, a large U.S. health insurance database. Researchers established a cohort for this study of 2.9 million patients with an average age of 44 who had at least two independent opioid prescriptions between 2012 and 2021.

Because these prescriptions included a range of oral formulas – codeine, hydrocodone, methadone, oxycodone, morphine and others – researchers standardized every prescription to morphine milligram equivalents (MME) and calculated each patient's monthly intake of opioids. The MME computation from electronic health records was previously co-developed by co-senior author Wenyu Song, an instructor at Harvard Medical School.

First author Seungyeon Lee, a PhD student in Zhang's lab, used statistical modeling and classified patients into five baseline groups of opioid dosage trajectory over the 10-year study period: very low-dose, low-dose decreasing, low-dose increasing, moderate-dose increasing and high-dose sustained use.

"Some patients had stable low-dose opioid use, while others had increasing or high dose patterns over time," Lee said.

Of the total cohort, 160,243 patients (5.5%) also were prescribed stimulants. The addition of a monthly calculated cumulative number of stimulant prescriptions to the model and statistical analysis showed a shift in the trajectory groups. Characteristics that could serve as risk factors for increasing opioid use also emerged in the data, Lee said.

Moderate-dose increasing and high-dose groups had an overall higher average MME and a higher proportion of patients with diagnoses of depression, anxiety and attention-deficit/hyperactivity disorder compared to other groups. The low-dose increasing group also had a higher proportion of patients with ADHD compared to the low-dose decreasing group.

The most common diagnoses linked to co-prescription of stimulants and opioids were depression and ADHD or ADHD and chronic pain.

"This was an important finding, that many patients with ADHD and depression, also experiencing chronic pain, have an opioid prescription," said Zhang, also a core faculty member in the Translational Data Analytics Institute at Ohio State. "This cohort represents a very realistic health care problem."

Even taking those factors into account, the model showed that stimulant use was key to driving up the odds that patients who took both stimulants and opioids would belong to a group of people who increased their doses of opioids.

"Stimulant use before initiating opioids and stimulant co-prescription with opioids are both positively associated with escalating opioid doses compared to other factors," Lee said.

Analysis of geographic and gender data also offered some clues to opioid use patterns in the United States. Patients in the South and West regions had higher total opioid intakes over the 10-year study period compared to the Northeast and North Central regions, with the highest frequency of opioid prescriptions in the South and higher MMEs per prescription in the West. Males also had higher average daily opioid intakes than females.

The results linking high opioid doses and stimulant use suggest stimulants may be a driving force behind the emergence of the twin epidemic and offer evidence that regulation of stimulant prescribing may be needed for patients already taking prescription opioids, the researchers said. In addition to the increased risk of overdose death, co-using prescription stimulants and opioids can increase the risk for cardiovascular events and mental health problems, previous research has shown.

Zhang's Artificial Intelligence in Medicine Lab focuses primarily on using AI to aid in clinician decision making, and these findings are part of a larger project aimed at development of safer personalized treatment recommendations for people who are prescribed both opioids and stimulants.

"We want to reduce the risk of opioid- or stimulant-related adverse drug events in real-world practice," Zhang said.

This work was funded by the National Institute of General Medical Sciences, the National Institute on Drug Abuse and the National Science Foundation.

Additional co-authors were David Bates of Harvard Medical School and Richard Urman, chair of anesthesiology in Ohio State's College of Medicine.