Computational markers have revealed that patients in treatment for opioid addiction are more likely to relapse when they develop greater tolerance to risks, according to results published in JAMA Psychiatry.
“Although it is well known that people addicted to opioids cycle through periods of abstinence and use, we lack the tools needed to prospectively identify when these transitions are more likely to occur,” Anna Konova, PhD, assistant professor at Rutgers University Behavioral Health Care, said in a press release. “Here, given that opioid use during treatment is quite risky, we wanted to examine whether a patient’s tolerance for risky decisions is informative about their vulnerability to relapse.”
Konova and colleagues longitudinally studied a cohort of 70 patients with opioid use disorder at a community-based treatment setting for up to 7 months. Across up to 15 sessions, patients played a computer game created for the study wherein they made decisions that assessed two types of risk tolerance — known-risk tolerance, defined as “a willingness to accept courses of action for which the precise odds of a given outcome are known”; and ambiguity tolerance, defined as “a willingness to accept actions for which these odds are not fully known and cannot be estimated.” The researchers used time-lagged mixed-effects logistic regression analyses to determine the likelihood of opioid use between sessions by collecting data at each session. They compared results for this group of patients with those of 55 controls.
Konova and colleagues found that 252 (45.7%) of 552 sessions completed with patients directly preceded opioid use events. Ambiguity tolerance was the only task parameter significantly associated with increased odds of prospective opioid use (adjusted OR = 1.37; 95% CI, 1.07-1.76), which indicated that patients were more tolerant specifically of ambiguous risks prior to use events, the researchers noted. The association of prospective use with ambiguity tolerance was independent of established clinical factors (aOR = 1.29; 95% CI, 1.01-1.65), and a model that combined these factors explained more variance in reuse risk. When patients were compared with controls, the researchers found no significant differences in ambiguity tolerance, but patients were more tolerant of known risks.
“Used in conjunction with clinical assessments, the computer model can be an important risk calculator, allowing clinicians in large but short-staffed treatment centers to allocate appropriate attention to those at greater risk for relapse and treatment failure,” Konova said. “The goal is to eventually create a mobile app based on the game that people can play remotely, which could convey information about relapse risk in real time to the patient, clinician or caretaker.” – by Joe Gramigna
Disclosures: Konova reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.