The ACT heroin trial proposal: an overview

The question to be asked is: If maintenance treatment for opioid dependence is expanded, so that both injectable diacetylmorphine (heroin) and oral methadone are available, is this more effective than current maintenance treatment with oral methadone alone?

Measures of effectiveness are:

• Ability to attract dependent heroin users into treatment;
• Ability to prevent premature drop-out from treatment;
• Ability to improve health and well-being, including reducing drug use and criminal behaviour and improving social functioning; and
• Cost-effectiveness.

First pilot study: This would involve 40 participants who meet the following eligibility criteria:

• Either currently or formerly on the ACT methadone program; and
• Able to prove ACT residence since 1993.

To warrant moving to the second pilot study, the first would have to show that:

• A stable maintenance dose of injectable diacetylmorphine or injectable diacetylmorphine plus oral methadone could be found for more than half the participants;
• Participants could safely and easily move between the three treatment options; and
• There was improvement in at least half of the outcome measures for health and well-being.

Stability is a key issue for the first pilot study. Stabilised consumption within a defined therapeutic range was identified as a criterion for effective maintenance treatment by a meeting of experts on drug substitution organised by the World Health Organization in May 1995. Participants in the proposed trial would be able to attend the clinic to inject heroin up to three times a day. Current Swiss experience is that this works very well;⁵ those who cannot be stabilised on heroin under these conditions are prescribed a low dose of methadone as well.

Second pilot study: This would be a small randomised controlled trial with 250 participants and the same eligibility criteria as the first pilot study. In contrast to the first pilot study, half the participants would be allocated to the choice of treatment options and half to oral methadone alone.

The second pilot study would:

• Further investigate the questions addressed in the first pilot study;
• Begin to examine attraction into, and retention in, treatment; and
• Assess if randomisation is practicable for this group.

• 1000 participants in three cities; and
• Volunteers drawn evenly from three groups: dependent heroin users who have never been in treatment; those who have dropped out of treatment; and those currently in methadone treatment.

The trial would run for two years. In the first year it would be a randomised controlled trial, but in the second all participants would be given choice of treatment.

At the end of the trial there could be evidence-based assessment of the role of diacetylmorphine in maintenance treatment, the subgroups in whom this treatment is most likely to be useful, and other indications and contraindications. This would allow a more balanced perspective on medical prescription of this drug.

Some of the controversy arises from uncertainties about whether prescribing diacetylmorphine can have positive outcomes, whether it can be cost-effective and whether stability is achievable on this short-acting opioid. These questions can be resolved only through empirical research and are the focus of the trial.

Moral arguments about the value of maintenance treatment, about providing treatment for self-inflicted problems and about making a currently illicit substance available under carefully controlled conditions are not resolvable but are open to ethical debate (some issues are covered by Ostini et al.⁶).

Finally, controversy arises because a trial has risks. These include that dependent heroin users may move to the ACT; a trial may lead to more permissive attitudes to illicit drug use; the trial drugs may cause road accidents or be diverted onto the black market; participants may congregate at the trial site; women in the trial may give birth to diacetylmorphine-affected babies; and a trial might further institutionalise or marginalise dependent heroin users.

Much of the feasibility research involved working with critics of a trial, firstly to identify these risks and then to develop ways to minimise them. In summary, these include using well-defined eligibility criteria, not providing take-away doses of heroin, strictly supervising injection at the clinic, carefully monitoring participants before they leave and setting the trial within the current context of law enforcement and preventive activities. Potential risks would also be carefully monitored.

The feasibility research was conducted in collaboration with the Australian Institute of Criminology. Well over 100 people have been involved -- as collaborators, assistants and advisers -- and many hundreds have provided feedback through workshops, seminars and discussions. Opinions have also been elicited from around 5000 members of the general community through ACT and national surveys.

A wide range of options was initially explored. The development of a proposal that was clinically workable, able to be rigorously evaluated and minimised risks was an iterative process -- any one change to the protocol could have multiple ramifications. The process involved integrating both the findings of many disciplines (anthropology, clinical science and health care, criminology, demography, economics, epidemiology, law, pharmacology, philosophy, political science, policy analysis, psychology, sociology and statistics) and the insights of the key interest groups (people who are or have been dependent on heroin, police, people involved in providing treatment and other services to illicit drug users, the general community and policy makers).

Heroin has long been prescribed for dependent users in the United Kingdom and, while there is evidence that this can be a useful option, it is contested. The same is true of historical evidence from the United States. No evaluation to date has been as rigorous as would now be required before introducing a new treatment. Other countries, most notably Switzerland and the Netherlands, are either undertaking, or about to undertake, "heroin trials".^5,7 Much will be learnt from them, but many questions will remain unanswered and these are the focus of the ACT proposal.

In the meantime, we welcome public and private critiques and debate on the proposal. If a trial does eventuate it must be as well conceived as possible. Opportunities for clinical trials are rare and justified only when there is a real research question, with doubt about the outcome. In addition, trials are expensive. We estimate the cost of the two pilot studies alone at $2.3 million. A trial cannot be paid for from funding currently allocated to drug treatment or research; there are too many other urgent priorities. New money will have to be allocated --ultimately, this is the real test of political will.

Of necessity this overview must be brief. A more detailed proposal can be found in the 1995 report on feasibility of the heroin trial.⁹ The results of the feasibility research are presented in four reports, thirteen working papers and, to date, 16 peer-reviewed papers, which are available from the authors (for a selection see references 10-14). A detailed response to the critique by Dr Matt Gaughwin is also available from the authors.

1. Gerstein DR, Johnson RA, Harwood HJ, et al. Evaluating recovery services: The California drug and alcohol treatment assessment (CALDATA). General report submitted to the State of California, Department of Alcohol and Drug Programs by National Opinion Research Center at the University of Chicago and Lewin-VHI Inc, Fairfax Virginia. Sacramento: California Department of Alcohol and Drug Programs, 1994.
2. National Institute in Drug Abuse. Drug abuse treatment. An economical approach to addressing the drug problem in America. Rockville MD: US Department of Health and Human Services, Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration, 1991.
3. Odyssey House. Drugs in our community. Unpublished report. Melbourne: Odyssey House.
4. Rydell CP, Everingham SS. Controlling cocaine. Supply versus demand programs. Santa Monica: RAND Drug Policy Research Centre, 1994.
5. Uchtenhagen A, Dobler-Mikola A, Gutzwiller F. Medically controlled prescription of narcotics: fundamentals, research plan, first experiences. In: Rihs-Middel M, Lewis DC, Clerc J, et al., editors. The medical prescription of narcotics: scientific foundations and practical experiences. Freiburg: Huber verlag. In press.
6. Ostini R, Bammer G, Dance P, Goodin R. The ethics of experimental heroin maintenance. J Med Ethics 1993; 19: 175-182.
7. Health Council of the Netherlands: Committee on Pharmacological Interventions in Heroin Addicts. The prescription of heroin to heroin addicts. The Hague: Health Council of the Netherlands, 1995. (Publication no. 1995/12E.)
8. Legislative Assembly for the Australian Capital Territory. Select Committee on HIV, Illegal drugs and Prostitution. Second interim report. A feasibility study on the controlled availability of opioids. Canberra: Legislative Assembly for the Australian Capital Territory, 1991.
9. Bammer, G. Report and recommendations of stage 2 feasibility research into the controlled availability of opioids. Canberra: National Centre for Epidemiology and Population Health, Australian National University and the Australian Institute of Criminology, 1995.
10. Hartland N, McDonald D, Dance P, Bammer G. Australian reports into drug use and the possibility of heroin maintenance. Drug Alcohol Rev 1992; 11: 175-182.
11. Bammer G. Should the controlled provision of heroin be a treatment option? Australian feasibility considerations. Addict 1993; 88: 467-475.
12. Bammer G, Weekes S. Becoming an ex-user: insights into the process and implications for treatment and policy. Drug Alcohol Rev 1994; 13: 285-292.
13. Bammer G, Stevens A, Dance P, et al. Controlled heroin availability in Australia? How and to what end? Int J Addict 1995; 30: 991-1007.
14. Stevens A, Ostini R, Dance P, et al. Police opinions of a proposal for controlled availability of heroin in Australia. Policing Soc 1995; 5: 303-312.

No reprints will be available.
Correspondence: Dr G Bammer, National Centre for Epidemiology and Population Health, The Australian National University, Canberra, ACT 0200.
Email: Gabriele.BammerATanu.edu.au