Of the 52,404 drug overdose deaths in the U.S. in 2015, roughly 63 percent involved an opioid. Prescription painkillers, the most widely used opioids, still accounted for the largest share of opioid overdose deaths — nearly half — in 2015. But over the last few years, the rise in overdose deaths has been driven primarily by a spike in deaths related to heroin and synthetic opioids. From 2014 to 2015, there was a 20.6 percent increase in deaths involving heroin, and a 72.2 percent increase in deaths from synthetic opioids other than methadone, particularly fentanyl and its analogues. If current trends continue, we will see an increase in the share of the heroin supply that is not heroin at all but much more powerful opioids like fentanyl. This poses a life-threatening risk to users who, unable to determine the content of drugs they purchase off the street or the internet, are more likely to consume a lethal dose, incorrectly assuming that they are taking an appropriate amount.
In response to the growing epidemic of opioid dependence and the alarming death rates, there is increased support for medication-assisted treatment (MAT). MAT uses opioid replacement drugs such as methadone or buprenorphine to stabilize patients’ physical dependency on opioids.
MAT is critical to treating opioid use disorders. It can reduce risky behaviors and criminal activity associated with illicit drugs, improve the lives of patients and help many patients achieve abstinence. And while MAT can certainly prevent overdose deaths, it also has some limitations. Several studies of methadone maintenance programs have found that some patients are resistant to opioid replacement therapies, either due to physical aversions to methadone or buprenorphine or issues with the dosing and delivery characteristics of these drugs. Thus, MAT may not keep people from using illicit heroin; some studies have found that while heroin use declines among methadone maintenance patients, up to 60 percent continue to use heroin while in treatment.
These findings should be troubling to policymakers seeking to reduce overdose deaths. The increasing prevalence of fentanyl and other synthetic opioids in the heroin supply likely means that deaths among those who use from this supply will also go up. MAT is viewed as a viable treatment option to help people dependent on opioids transition away from illicit heroin. But if a sizable portion of the MAT population continues to use opioids illicitly, it is likely that they will use heroin or counterfeit pills that are laced with fentanyl or other stronger opioids and therefore will still be at a high risk for overdose.
What if, instead of replacement drugs, this heroin-dependent population was offered their drug of choice — that is, heroin? This may sound counterintuitive, but heroin-assisted treatment, or HAT, is a well-established treatment method that was available in the U.S. until the early 1920s and is currently used in several countries. HAT programs provide severely addicted individuals with access to pharmaceutical-grade heroin (diacetylmorphine) on the grounds that doing so will decrease demand for illicit heroin, reduce criminal activity associated with obtaining heroin, improve patients’ lives, reduce the spread of communicable diseases and reduce overdose deaths by providing unadulterated heroin in a supervised setting.
Germany, Belgium, Denmark, the Netherlands and the United Kingdom all have HAT programs. The most well-known is the Swiss model, which began in 1994 in response to a persistent open-air heroin market and the spread of HIV through needle sharing. The Swiss program has demonstrated success in several areas. Criminal activity and new incidences of hepatitis C and HIV among HAT patients have declined significantly. No patient has died of an overdose from heroin received through HAT; this is because of the high quality of the heroin administered and because patients use under supervision, ensuring that someone is available to revive them if they do have an adverse reaction. Rather than encouraging heroin use, the heroin-dependent population declined following Switzerland’s implementation of HAT from approximately 30,000 in 1992 to 26,000 in 2002.
Comparisons of HAT with more traditional MAT programs that rely on methadone have consistently found HAT to be superior on a number of measures. Compared to methadone maintenance patients, HAT participants have exhibited greater declines in criminal behavior, greater physical and mental health improvements and greater reductions in behaviors associated with HIV transmission. And, most relevant to U.S. overdose trends, HAT patients are more likely to remain in treatment and are less likely to use street heroin and other illicit drugs compared to methadone patients.
Not everyone with an opioid use disorder should be treated with HAT. General rules of thumb are that HAT should be used only when other substitution therapies have failed, and it should not be used for younger people or for people with a short history of heroin dependence. It is also important to offer some form of cognitive therapy in conjunction with drug maintenance, and to approach HAT as a long-term treatment option, providing it to patients for at least 12 months but sometimes several years or more. This drives home another point, which is that HAT — and all other opioid maintenance therapies — will not lead to abstinence for all patients. While many do achieve abstinence, this is not a realistic goal for everyone.
The effectiveness of HAT has been demonstrated through numerous studies across multiple measures. If the federal government would allow states and localities to operate HAT programs, it would be an opportunity to see how HAT works in the U.S. on a small scale. HAT could initially be made available to people in communities suffering from the highest rates of heroin overdose. It should not be available to everyone with an opioid use disorder, but only to those individuals who have persistent problems with heroin dependence and who have demonstrated resistance to other MAT therapies. For appropriate candidates, administrative barriers to access should be kept low, and communities that want to experiment with HAT should offer it in conjunction with cognitive-behavioral therapy and integrate it with other harm reduction programs.
Addressing the underlying psychological, social and economic causes of the opioid epidemic is a serious challenge, one which we do not fully know how to meet. But on the problem of people already dependent on opioids, there is a wealth of evidence about what works. The overall approach should be one of reducing harms associated with opioid use. This means providing access to overdose reversal drugs, clean needles for people who inject drugs and safe spaces to use drugs. It means providing access to medication-assisted treatment and cognitive-behavioral therapy. And in some cases it means providing users with pharmaceutical-grade heroin to prevent them from shooting up with more dangerous, unknown substances purchased off the street or online.
Political barriers to the adoption of HAT programs are high. Despite growing understanding of the opioid epidemic, heroin remains one of the most heavily stigmatized drugs in our culture. Such negative perceptions make the suggestion of providing heroin to users seem contradictory to the very idea of drug treatment. But scientific evidence should drive policy decisions, not misguided perceptions. And the evidence available clearly indicates that HAT programs could make a significant dent in deaths from opioid use. This possibility alone should be enough motivation for the U.S. to experiment with HAT programs.
Katharine Neill Harris, Ph.D., is the Alfred C. Glassell, III, Fellow in Drug Policy at the Baker Institute. Her current research focuses on state sentencing policies for drug offenders and the legalization of medical and recreational marijuana.