This article appeared in the March 19, 2012 issue of Alcoholism & Drug Abuse Weekly, received by electronic subscribers March 15.

A controversial study from Canada comparing the costs of heroin — as a treatment drug – to methadone has found that heroin is more cost-effective for some patients. Reports on the study in the press in Canada and the United States just reported the findings, without noting the flaws. ADAW spoke to the top federal regulator of pharmacologic therapy to get insights into the methodology of the Canadian study and implications of the results for treatment providers.

For the study, “Cost-effectiveness of diacetylmorphine versus methadone for chronic opioid dependence refractory to treatment,” the researchers used mathematical modeling to derive a cost comparison between methadone maintenance treatment and medical heroin (diacetylmorphine) in which the dose is carefully controlled. The study, published March 12 in the online version of the Canadian Medical Association Journal (CMAJ), is based on data from the North American Opiate Medication Initiative (NAOMI) released in 2008 and published in 2009 in the New England Journal of Medicine (see ADAW Nov. 3, 2008 and August 31, 2009).

The study was based on a hypothetical cohort of patients, assigned the baseline characteristics of the patients in the NAOMI study. It was not based on any of the patients in the NAOMI trial, but constructed out of the data and extrapolated to project possible future savings.

The article, by Bohdan Nosyk, Ph.D., and colleagues, starts with the premise that medicinal heroin is better at retaining patients than methadone maintenance at treating chronic opioid addiction for the subgroup of patients who have relapsed at least twice from methadone treatment, as the NAOMI data indicated.

However, the direct costs of treatment with heroin are higher, because the drug must be injected on a daily basis at a clinic. For this study, the researchers calculated incremental cost-effectiveness over 1-year, 5-year, 10-year, and lifetime periods. They found that over a lifetime, people on methadone gained 7.46 quality-adjusted life-years (QALYs) and generated a social cost of $1.14 million, while those on heroin gained 7.92 QALYs and generated a social cost of $1.10 million. The cost savings was primarily due to reduced criminal activity, the researchers said.

The cost of heroin maintenance is much higher than the cost of methadone maintenance — $14,891 a year compared to $3,192.

CSAT’s response

We asked Robert Lubran, director of the division of pharmacologic therapies at the Center for Substance Abuse Treatment at the Substance Abuse and Mental Health Services Administration, to comment on the recent study.

He said that the cost savings are not huge — $1.14 million compared to $1.10 million — and that the savings all come from the criminal justice system.

“And you have to keep in mind that the study population is very narrow,” he said. “They failed at least two episodes of methadone maintenance, but we don’t know why methadone treatment didn’t work.”

The cost benefits in the heroin patients were due to better retention in that group than in the methadone group.

The treatment system is different in Canada, noted Lubran. “It’s not regulated the way it is here,” he said. Methadone maintenance treatment must include counseling and comprehensive services in the United States, with steps taken to help patients who are in jeopardy of relapse. That is not the case in Canada. For that reason, it’s “really inappropriate” to draw any conclusions about whether heroin treatment would be effective here based on this study, he said.

Only looking at costs from the criminal justice system creates a severe limitation, said Lubran. “They didn’t look at the whole range of social and medical costs,” he said. Not counting the medical costs is a major limitation as well, he said. In Vermont, for example, patients in methadone maintenance have Medicaid costs that are three times the costs of other patients not in OTPs, he said.

About three quarters of the heroin patients were unemployed, said Lubran, noting that that is not the case in opioid treatment programs in the United States. Adding the costs of welfare — not factored into the CMAJ study – would create a different equation altogether, said Lubran.

“It doesn’t seem as if someone on heroin maintenance is functional, if they’re not working and not going to school,” he said.

“The other thing we don’t know is whether the dose of methadone was adequate” in the Canadian program, said Lubran. The average dose was 96 milligrams, which “may not be adequate for these severely dependent people,” he said. This could have contributed to criminal activity. If that is the case, then giving them more methadone might have been a better option than giving them heroin.

“There are a lot of unknowns here,” said Lubran. “This seems more of a palliative rather than a therapeutic approach,” said Lubran.

Lubran also said that underlying psychiatric comorbidity should be treated. It’s not just a matter of giving a patient medication. “We’re trying to promote a wholistic approach, integrated with primary care and behavioral health care so that you’re treating the whole person – the psychiatric problems, and the medical problems.”

The National Institute on Drug Abuse (NIDA) refused to comment on the study.

NAOMI is funded by the Canadian Institutes of Health Research.

For the study, go to