Penn medicine researchers have developed a model to predict stimulant-related overdose risk among individuals insured by Medicaid.
The model — published in a study in the JAMA Health Forum — was conducted by several senior fellows at the Leonard Davis Institute of Health Economics. Researchers used Medicaid claims data from 2016 to 2020 to identify high-risk patients for emergency department treatment or hospitalization due to overdose, analyzing both individual and area-level sociodemographic factors such as race, income, and local inequality indexes.
Penn Perelman School of Medicine professor Rebecca Arden Harris, who co-authored the project, explained in an interview with LDI that “prior substance use diagnoses, combined with area-level indicators” enabled researchers to locate populations and communities at risk of stimulant-induced overdoses with great accuracy.
“This study breaks new ground as the first internally validated model for stimulant-involved overdose among Medicaid beneficiaries, establishing a transparent, reproducible framework for public health use,” Harris added.
The prediction model effectively scored above nine out of 10 on statistical accuracy tests.
In recent years, the number of overdoses involving stimulants rose drastically across the United States.
Researchers conducted the project after noticing that that enrollees in the Medicaid program have been disproportionately impacted.
The prediction from the model not only provides further insight into this heightened risk for stimulant use disorder among Medicaid patients, but also improves treatments prescribed for individuals suffering from substance abuse.
Tuhina Srivastava, the lead author of the study and a former epidemiology PhD student at Penn, discussed the model’s potential to change the current state of substance use disorder treatment in an interview with Penn Today.
RELATED:
Penn Medicine breaks ground on $401M Princeton Cancer Center set to open in May 2028
Medical School graduate, immunologist to receive John Scott Award by American Philosophical Society
“Too often, the response to people with substance use disorder is reactive or even punitive,” Srivastava said. “ We believe this [model] provides a potential step toward minimizing or eliminating that. It’s classified as a chronic disease and should be treated as such.”
Looking ahead, the research team is exploring opportunities to integrate the model into public health surveillance systems, allowing policymakers to identify high-risk geographic regions for targeted interventions.
