ѻýҕl

What are the Cost-Effective Treatment Options for Opioid Use Disorder?

<ѻýҕl class="mpt-content-deck">— Modeling study puts numbers to savings from medication-assisted treatment
MedpageToday
A photo of a bottle of Methadose Oral Concentrate

Medication-assisted treatment (MAT) -- especially when coupled with other approaches such as overdose education and naloxone distribution (OEND) and contingency management -- appeared beneficial to those with opioid use disorder (OUD) in a modeling study, while remaining cost-saving for healthcare providers in comparison to no treatment.

While MAT was only modestly effective in terms of improving quality-adjusted survival -- adding 1.02 to 1.07 quality-adjusted life years (QALYs) per person -- it was also relatively inexpensive, at some $16,000 per QALY gained with methadone and about $22,000 per QALY gained with methadone plus OEND rounded, according to Douglas Owens, MD, MS, of Stanford University in California, and colleagues .

Costs for other approaches included $42,000 per QALY gained with buprenorphine plus OEND and contingency management, and $250,000 per QALY with buprenorphine, OEND, contingency management, and psychotherapy.

All of them, even the latter, provided cost savings when compared to no treatment, in part because criminal justice costs are much higher with untreated OUD. The biggest saver was methadone plus contingency management. Psychotherapy, on the other hand, was a less cost-effective treatment option in the analyses. But, Owens and colleagues noted, they assumed that MAT included a baseline level of drug counseling.

The researchers' use of a continuous-time dynamic compartmental model, they said, helped them account for an individual's changing health status over time, such as out of treatment, receiving treatment, abstinent, dead, or not receiving treatment. The model also used a representative cohort population that was based on the makeup of individuals with OUD in the general U.S. population, and on data from the VA.

The base case analysis, with a hypothetical cohort of 100,000 patients, estimated that 42,717 overdoses (4,132 fatal, 38,585 nonfatal) and 12,660 deaths would occur over a 5-year period, with a reduction of approximately 11.58 QALYs per person. There was an estimated 10.7% reduction in overdoses when MAT with methadone was used, and a 22% reduction in overdose occurrence when MAT with buprenorphine or naltrexone was used. When MAT was combined with contingency management and psychotherapy, overdose reduction ranged from 21-31.4%. The estimated number of deaths decreased by 6% when MAT with methadone was used, and 13.9% for MAT with buprenorphine or naltrexone was used. When combined with contingency management, OEND, and psychotherapy, there was an overall 16.9% estimated reduction in overall deaths.

While it may come as no surprise to readers that MAT carries more value as a treatment option than no treatment at all, Owens told ѻýҕl that mapping out the probable costs and benefits of MAT could carry weight with policy-makers. The differences in cost-effectiveness found between the various medications used for MAT, however, are more subject to change.

"That's going to vary from place to place, because of availability or cost," Owens said. "We're not trying to say you should use this versus that [medication]. What we're saying is that, in general, these treatments work, and so getting them to people is very important."

Sean Murphy, PhD, associate professor of population health sciences at Weill Cornell Medical College, said it's important to take these results with a grain of salt. Model-based studies like this one, he told ѻýҕl, "are so heavily dependent on the inputs that are chosen as well as the structure of the model." The authors themselves identified several limitations for this study.

"The benefit of these types of models is that it really allows you to explore questions that would be very difficult in a clinical trial and over longer timeframes," Murphy added.

Catherine Maclean, PhD, associate professor of health economics at Temple University, told ѻýҕl that the authors were probably conservative in the way that they measured cost-effectiveness. Maclean pointed out that the study did not go beyond healthcare and criminal justice sectors in analyzing cost-effectiveness. Expanding the analysis to homeless populations, other social services, and labor markets, she said, could suggest even greater cost savings. Owens's group did acknowledge that as an additional limitation of the study.

Although Maclean wasn't very optimistic about the potential for policy change, she said the study's evaluations of MAT's cost-effectiveness are significant.

"This is one more piece of high-quality information suggesting that if we invest in these interventions, if we expand access, we could see cost savings," she said.

  • author['full_name']

    Kara Grant joined the Enterprise & Investigative Reporting team at ѻýҕl in February 2021. She covers psychiatry, mental health, and medical education.

Disclosures

This study was funded by the U.S. Department of Veterans Affairs. Fairley reported grant support from Stanford University's Shaper Family Graduate Fellowship. Co-author Trafton serves on the board of directors of the Institute for Brain Potential, and Asch reported that he received grants from Facebook during the course of the study.

Sean Murphy disclosed that he blind-reviewed this paper before its publication.

Catherine Maclean is a co-editor for the Journal of Policy Analysis and Management and associate editor for the Journal of Health Economics.

Primary Source

JAMA Psychiatry

Fairley M, et al "Cost-effectiveness of treatments for opioid use disorder" JAMA Psychiatry 2021; DOI: 10.1001/jamapsychiatry.2021.0247.