The American Journal of Bioethics

We applaud Louis C. Charland (2002) for raising for bioethics debate the provocative and underdiscussed subject of heroin-prescription trials. He does an excellent job of discussing informed consent and subject competency. Charland does not, however, adequately contextualize his discussion within the world of heroin addiction and thereby overlooks certain practical aspects of research participation. This ultimately leaves many of his concerns ill-placed.

Charland argues that, in the absence of empirical evidence, it cannot be presumed that heroin addicts are competent to consent to a trial that offers heroin at no cost, because they "have an impaired decisional capacity to make choices about heroin." Their addiction, he asserts, compromises their values such that they are likely to exaggerate the benefits of participating while undervaluing the risks so that they "can hardly be considered mentally capable of rationally weighing the risks and benefits." Charland specifies only one benefit of participation (free heroin) and never specifies any risks. So, ironically, like his imagined subjects, he does not exhibit full appreciation of the risks and benefits of participation.

In light of the absence of empirical evidence to support Charland's claim, we cannot agree with his presumption that addicts or, more specifically, the subject population of heroin-prescription trials, would likely exaggerate the benefits of participation. Addicts are extremely knowledgeable about heroin, its availability, its illegality, its social stigma, and its composition—addicts frequently have a sophisticated understanding of chemistry. Within the context of single-arm heroin-prescription trials, addicts may very well weigh the risks and benefits of participation very rationally.

About the Subject Population

Heroin prescription is being evaluated as a treatment for severely and chronically dependent treatment-refractory abusers who are socially disintegrated. They need to "fix" three to four times a day and find their way into "the system" through social, legal, or economic crisis (Bammer et al. 1999). To begin, it is important to note that the main harms associated with heroin addiction stem first from its illegality and second from its impurity, both of which are minimized in the medical research model.

The illegality of heroin defines the culture of its use and treatment. Direct harms of use come in many forms:

1. reluctance to seek emergency medical attention out of fear of police intervention;

2. exacerbated economic hardship as a result of arrests, incarceration, and, for some, imposed, state-sponsored treatment programs that are inconsistently effective; and

3. increased health risk.

Numerous studies confirm that heroin addicts are at increased risk to experience violence and/or contract HIV/AIDS and hepatitis (Darke and Zador 1996). Addicts also tend to concomitantly use central nervous system depressants such as alcohol and benzodiazepines (prescription tranquilizers such as Valium, Halcyon, and Xanax), which some argue are the leading cause of respiratory failure and death among heroin users (Darke and Zador 1996). [Begin Page 56] Finally, the contaminants found in street heroin, such as starch, quinine, and strychnine, make it inherently more dangerous than prescription heroin, both because of impurities (Cherubin et al. 1972; Ruttenber and Luke 1984) and because variations in strength coupled with individual pharmacokinetics make overdose by even seasoned users a constant risk.

Benefits of Participation

The medical environment reduces many of the physical and environmental risks inherent in a "hard-core" addict's real, everyday life. We do not mean to incorrectly conflate treatment and research in this context. While concerns regarding therapeutic misconception may seem applicable, we argue that for the population enrolled in a reasonably well-designed heroin-prescription study the benefits of participation need not be exaggerated to compel participation, because participation is inherently less risky than not.

The two primary benefits of participation speak directly to the two primary risks of use "on the street." The user's context is shifted from criminal to medical, and the user receives pure, carefully dosed heroin. Most studies administer methadone, which is long acting, along with heroin, which is short acting, in order to minimize withdrawal symptoms and reduce reliance on heroin. Some research studies, such as the Swiss Trials to which Charland refers, offer subjects counseling, psychiatric treatment, and social assistance (such as welfare benefits, housing, and medical care). While it is argued that such services undermine heroin-prescription trial design, it cannot be argued that their provision presents a serious risk to the subject population. In fact, these services may be a direct benefit.

What Risks?

Charland does not identify any risks of participation in heroin-prescription trials. Bammer et al. offer two, neither of which can be substantiated by data. The first is that study participation may be a disincentive to striving for abstinence. It must be remembered, however, that entry criteria require that subjects have experienced failure in prior treatment attempts. While a harm-reduction strategy may be more suited to this population, it in no way precludes them from striving for abstinence. Further, data have shown that heroin-prescription does not reduce the rate of achievement of abstinence (Bammer et al. 1999).

A second, related risk is that heroin prescription may undermine the attractiveness and effectiveness of other treatments. There is little evidence on which to assess this risk (Bammer et al. 1999). One cannot become "more addicted" after participation in a heroin-prescription trial. At worst, one goes back to one's state prior to the research, no better or worse off for the experience. Other perceived risks of heroin-prescription trials are not directed at subjects, but are larger social concerns such as:

1. trial cities receiving an influx of addicts;

2. gitimacy of the trial leading to increased permissiveness of illegal drug use; and

3. concerns that if heroin prescription is deemed therapeutically effective, tax payers would have to foot the bill for sustaining the heroin addict's habit (Bammer et al. 1999).

These concerns, however, are largely preventable and/or unfounded and are superfluous to the issue of competency and consent.

Conclusion

Major obstacles to effective heroin-addiction treatment include the barriers imposed on the access to and the institution of novel, often stigmatized approaches. When we begin to assess competencies, we may again be imposing a system on potentially valuable research or treatment that may doom the research or treatment and restrict its acceptance among those it is meant to study or assist.

Charland questions the competency of heroin addicts to consent to a heroin-prescription trial. We argue that one is not incompetent because one is sick, and, in the absence of any empirical evidence, Charland's concern regarding subject protection, while interesting and well meaning, may be unfounded in this context because the research presents no new dangers. Charland's assertion that new mechanisms for informed consent ought to be explored may simply chase away those for whom the research is intended, which has happened in American methadone treatment.

Participation in heroin-prescription trials for treatment-refractory addicts would reduce many of the risks associated with heroin abuse, and we can think of no risk that would be increased due to such participation. As to the question of competence, one would be incompetent, in a sense, not to participate.