"More Than A Quick Fix" BMJ 11/01/08 Article on Prescibing Heroin to Addicts

By Jatelka · Jan 11, 2008 · ·
  1. Jatelka
    Tony Sheldon, freelance journalist

    1 Utrecht
    [email protected]

    doi: 10.1136/bmj.39422.503241.ADdoi: 10.1136/bmj.39421.593692.94

    Despite its long use in the UK, prescribing heroin to misusers remains controversial. Tony Sheldon looks at the history and the evidence behind the increased prescribing in Europe

    "The past is a foreign country: they do things differently there." The famous line from L P Hartley’s The Go-Between seems to sum up the medical prescribing of heroin to addicts. British medicine has a history of prescribing heroin, and the practice is now also largely accepted in Switzerland and the Netherlands. But use of the British system, as it was known abroad, in the United Kingdom has declined in favour of methadone maintenance—although not vanished completely.

    It was 1926 when a government committee chaired by Humphry Rolleston, president of the Royal College of Physicians, advised it was legitimate medical practice to supply heroin to addicts for their maintenance. Only later was the practice restricted to doctors licensed by the Home Office. Past UK examples include the drug dependency clinic of London’s University College Hospital, which prescribed injectable heroin during the 1970s.1 Later, a team led by psychiatrist John Marks offered heroin on prescription in Widnes, Merseyside, in an attempt to restrict the spread of HIV.

    However, by 1992 researchers estimated there were little more than 100 addicts prescribed heroin in the UK, while 17 000 were prescribed oral methadone.2 Today the number of addicts in the UK regularly prescribed heroin is around 300, although the practice is enjoying a revival.3 A trial of supervised injecting, the randomised injecting opioid treatment trial (RIOTT), among 150 addicts at clinics in Brighton, London, and Darlington is set to run until 2008.

    Similar relatively small scale trials of heroin assisted treatment have taken place in Canada, Spain, and Germany, where a multicentre trial with over 1000 participants took place between 2003 and 2005. However, only Switzerland and the Netherlands have had the political will to build a long term policy.

    Wider acceptance

    Switzerland was the first country to launch a large scale evaluation of heroin prescribing with a national cohort study running from 1994-6. It included more than 1000 people who were in poor health and who had repeatedly failed to benefit from conventional drug treatments during years of heroin addiction. They were offered controlled medical prescribing of heroin as part of a comprehensive programme of social and medical care.

    By 1997, the Swiss federal authorities claimed a "substantial improvement" in addicts’ physical and mental health and social situation as well as a 60% reduction in criminal behaviour. Clients’ reported use of non-prescribed heroin fell significantly, and there were no deaths from overdose or complaints from local neighbourhoods. The study concluded that it is possible and clinically effective to provide injectable heroin at a clinic three times a day, seven days a week.4 Meanwhile public opinion swung behind the project. A referendum in September 1997 returned 71% in favour of heroin maintenance.
    A later cohort study in 21 community outpatient treatment centres assessing more than 1600 heroin users between 1994 and 2000 concluded: "Heroin-assisted substitution treatment might be an effective option for chronically addicted patients for whom other treatments have failed." It showed 70% of users remained in treatment for more than a year with positive health and social outcomes.5

    Currently there are about 1300 addicts enrolled in 23 clinics across 16 Swiss cities. Criteria for inclusion include age 18 or older, being dependent on heroin for at least two years, having had two unsuccessful attempts at other treatment, and existence of severe medical, psychological, or social problems associated with heroin use.

    Typically, people prescribed heroin are in their late 30s and have been addicted to heroin for 10 years; a quarter are women and more than half will stay in treatment for two years or more. Of the up to 200 addicts who stop heroin prescribing each year, about 40% will transfer to methadone maintenance and a quarter will move to treatment based on abstinence.
    Sandra Wuethrich, head of the heroin assisted treatment programme, said: "In many cases patients’ physical and mental health has improved, their housing situation has become considerably more stable, and they have gradually managed to find employment." In addition, there is a substantial decline in consumption of non-prescribed substances and earnings from illegal activities. Heroin assisted treatment, she argues, is nationally and internationally acknowledged as an "established treatment for severely dependent heroin users."

    However, she adds, the programme only represents 8% of heroin replacement treatment. It was never intended to replace methadone maintenance, prescribed for 15 000 Swiss heroin users. It is not seen as a first line treatment but an option for a small minority.

    But the scheme has critics too. A review of the trials published in 1999 by Swiss and US researchers claimed they "did not withstand scientific scrutiny," in particular because they did not "randomly assign patients to comparison groups."6 Eric Voth, chair of the US Institute on Global Drug Policy, and Ernst Aeschbach, a Swiss doctor argued: "As seen in Switzerland, heroin handouts simply further the addiction and enslavement of suffering addicts."7 However, the Swiss trials encouraged the Dutch to act. The Netherlands Health Council, a government scientific advisory body, was charged in spring 1994 to report on prescribing heroin.

    Just as in Switzerland, the context was concern for a section of heroin users that current treatment failed to reach. The council estimated that methadone maintenance programmes had proved inadequate for 8000 out of 25 000 Dutch addicts. It believed heroin prescribing could establish contact with difficult to reach addicts, leading to benefits such as limiting the spread of infectious diseases.

    In 1995 it advised conducting a randomised clinical trial into the "benefits or harmful consequences of prescribing heroin to severely addicted heroin users who fail to respond to current treatment."

    The council stressed that it was "good clinical practice" for a doctor to prescribe medicines to an addict that will bring about "an improvement of the patient’s medical situation and will not in principle harm the patient." However, it distanced itself from any concept of "free supply," saying it did not take a position on any form of legalisation of heroin.
    Wim van den Brink, scientific director of the Central Committee on the Treatment of Heroin Addicts, said in 1997 that prescribing heroin would be one small piece of the puzzle of treatment—a "pharmacological intervention to stop a destructive pattern of behaviour."

    A three month pilot project was launched in 1998 and, after parliamentary approval, a larger experiment with 300 heroin users in six cities, followed. Addicts had to attend clinics three times a day and use heroin under medical supervision, thus encouraging a daily routine that enabled close contact with medical services. Also the relatively safer method of inhaling rather than injecting heroin was encouraged.

    Between 1998 and 2001, 549 patients participated in two randomised controlled trials comparing heroin and methadone with methadone alone over 6-12 months. The researchers claimed that the proportion of patients with a favourable response—defined as clinically relevant improvements in physical and psychological health as well as improvements in social functioning and criminality—was 20-25% higher among the group receiving heroin.8 In addition, despite their long term addiction 13% of addicts stopped using heroin in the second year, choosing either to return to methadone or total abstinence.

    As a result the Dutch cabinet extended the scheme in 2004. Criteria for inclusion are being addicted for five years, unsuccessfully treated in methadone programmes, in poor physical and mental health, and aged over 35.

    Today up to 450 patients are prescribed heroin in six cities. This is set to grow to 850 patients in 15 cities during the next year. Last year the Medicines Evaluation Board registered diacetylmorphine (heroin) as an effective treatment for a specific group of addicts, accepting it offers a positive balance between effectiveness and risk.

    Ten years on, Professor Van den Brink believes prescribing heroin has gained scientific and ideological acceptance as a normal treatment, albeit with a specific indication. He hopes, although he accepts this is harder to judge, that it also "shows that we can deal in another way with addicts by not regarding them as criminals but as patients who are treatable."

    He believes that in the future, as can be expected in any chronic disease, treatment will diversify. Heroin may not be a first line treatment but can play a small part in the range of care offered.

    The latest figures from the UK’s National Treatment Agency said 58% of addicts who attended drug clinics up to March 2007 failed to complete treatment. In the light of such data the successes claimed by the Dutch and Swiss with addicts who were all but beyond help, will continue to demand attention.

    doi: 10.1136/bmj.39422.503241.ADdoi: 10.1136/bmj.39421.593692.94
    Competing interests: None declared.


    1. Hartnoll R, Mitcheson MC, Battersby A, Brown G, Ellis M, Fleming P, et al. Evaluation of heroin maintenance in controlled trial. Arch Gen Psychiatry 1980;37:877-84.[Abstract]
    2. Gossop M. Verschreibung von Heroin und anderen injizierbaren Drogen an Abhängige aus britischer Sicht. Sucht 1994;5:325-3.
    3. Fischer B, Oviedo-Joekes E, Blanken P, Haasen C, Rehm J, Schechter MT, et al. Heroin-assisted treatment (HAT) a decade later. J Urban Health 2007;84:552-62.[CrossRef][ISI][Medline]
    4. Uchtenhagen A, Gutzwiller F, Dobler-Mikola A, eds. Programme for a medical prescription of narcotics: final report of the research representatives.Zurich: Zurich University, 1997.
    5. Rehm J, Gschwend P, Steffen T, Gutzwiller F, Dobler-Mikola A, Uchtenhagen A. Feasibility, safety, and efficacy of injectable heroin prescription for refractory opioid addicts: a follow-up study. Lancet 2001;358:1417-23.[CrossRef][ISI][Medline]
    6. Satel SL, Aeschbach E.The Swiss heroin trials. Scientifically sound? J Subst Abuse Treat 1999;17:331-5.[CrossRef][ISI][Medline]
    7. Voth EA, Aeschbach E. Heroin handouts are flawed policy. BMJ 2004;328:229.[Free Full Text]
    8. Van den Brink W, Hendriks VM, Blanken P, Koeter MWJ, van Zwieten BJ, van Ree JM. Medical prescription of heroin to treatment resistant heroin addicts: two randomised controlled trials. BMJ 2003;327:310.[Abstract/Free Full Text]

    Share This Article


  1. Jatelka
    Also in the same issue is a "Head to Head"

    BMJ 2008;336:70 (12 January), doi:10.1136/bmj.39421.593692.94

    Should heroin be prescribed to heroin misusers?


    Jürgen Rehm, chair, Benedikt Fischer, professor

    1 Centre for Addiction and Mental Health, Toronto, ON M5S 2S1, Canada, 2 Centre for Addictions Research of British Columbia, University of Victoria, Canada
    Correspondence to: J Rehm [email protected]

    doi: 10.1136/bmj.39422.503241.ADdoi: 10.1136/bmj.39434.460694.AD

    Some heroin addicts are very difficult to treat. Jürgen Rehm and Benedikt Fischer believe that maintenance with heroin is the way forward for this group, but Neil McKeganey argues that it is treating the effects of misuse not the addiction

    Recently, a public hearing of a Danish parliament subcommittee discussed whether heroin assisted maintenance treatment should be offered experimentally to reduce health and social harm related to use of heroin (www.tekno.dk/ordineret-heroin). This is just one in a series of similar—existing or proposed—programmes in Europe, North America, and Australia.1 We believe that such treatment is appropriate for heroin misusers under certain circumstances.

    Supporting evidence

    Increasing heroin misuse in the United States in the early 1970s led to a public debate about prescribing heroin as a last resort form of opioid maintenance therapy for people with chronic heroin dependence. In 1973 Lorrin Koran advocated in the New England Journal of Medicine that "carefully designed clinical research on the safety and efficacy of heroin maintenance should be undertaken, particularly with addicts not helped in current treatments."2

    Some 35 years later, three important research studies have been completed. In Switzerland, a small randomised trial3 and a study using natural cohort designs4 found heroin assisted maintenance treatment to be feasible and effective for a group of heroin misusers who were refractory to treatment, as characterised by long term heroin dependence; physical, psychological, or social deficits; and unsuccessful previous treatment.4 5 Effectiveness was observed in treatment retention; somatic health outcomes such as epileptic episodes, abscesses, or cachexia; mental health outcomes such as affective or anxiety disorders; heroin and cocaine misuse; and criminal outcomes such as property offences or drug trafficking (on the basis of self report and objective measures).5

    Large randomised controlled clinical trials in the Netherlands and Germany, which compared different modes of heroin assisted maintenance treatment with methadone maintenance treatment, obtained positive results on similar outcomes.6 7 Moreover, heroin assisted maintenance treatment was found to be cost beneficial in Switzerland8 and cost effective in the Netherlands compared with methadone maintenance treatment.9

    Since these results were obtained, this treatment option has been extended beyond the trial periods, and heroin has been approved by the regulatory bodies for treating opioid dependence. In all three countries, the intake of medical heroin is supervised and occurs a maximum of three times a day, and patients recover from acute intoxication before leaving the treatment clinic. Notably, heroin has been a treatment option for heroin misusers in the United Kingdom for several decades, albeit on a relatively small scale and under different conditions—with lower average dosing and less supervised intake.10

    Use of maintenance

    The above summary makes the recent use of heroin assisted maintenance treatment look like a straightforward scientific success story, and not like a topic for debate in the BMJ. However, since the original heroin assisted maintenance treatment programme was proposed in Switzerland in the early 1990s, there has been scientific, and perhaps more importantly, larger public debate on the ethics, safety, and clinical value of prescribing heroin, and to a lesser degree, on maintenance treatment in general. Overall, maintenance with buprenorphine and, to a larger degree, methadone is more successful than treatment focusing on abstinence or using placebos.11 Given the nature of opioid dependence as a chronic relapsing disease,12 these results are not surprising.

    Opioid maintenance treatment generally seems to be well justified for treating this disease. And if maintenance is generally justifiable as a form of treatment, why should heroin not be used as one such pharmacological agent? One reason that has been cited is safety, both for the patient13 and for the general public (for example, through diversion or the risk of trivialising the dangers of heroin, leading to an increase in use). Results from the Swiss studies, however, show that mortality among patients in heroin assisted maintenance programmes is low, and lower than for patients in other maintenance programmes.14 In addition, the wider safety concerns could not be empirically confirmed in Switzerland or the Netherlands.15 Finally, the incidence of heroin dependence has decreased greatly in Switzerland since the start of the trials, and currently heroin has a more negative image than it did 15 years ago.16

    Overall, we see no convincing reason why heroin assisted maintenance treatment should not be part of a comprehensive treatment system for opioid dependence. However, the development of an overall integrated treatment system is crucial. All studies to date have been conducted in samples of refractory addicts with severely compromised health and several previous failed attempts of methadone maintenance treatment. Our current knowledge about the effectiveness of heroin assisted maintenance treatment is restricted to these groups and to the context of countries where there is already an established and effective comprehensive system for treating opioid dependence. Although we currently do not have the necessary empirical evidence for establishing heroin assisted maintenance treatment in other circumstances, addition of heroin assisted maintenance treatment would be likely to improve the overall treatment system, especially with respect to so called treatment resistant and refractory opioid addicts.

    doi: 10.1136/bmj.39422.503241.ADdoi: 10.1136/bmj.39434.460694.AD
    Competing interests: None declared.


    1. Fischer B, Oviedo-Joekes E, Blanken P, Haasen C, Rehm J, Schechter M, et al. Heroin-assisted treatment (HAT) a decade later: a brief update on science and politics. J Urban Health 2007;84:552-62.[CrossRef][ISI][Medline]
    2. Koran L. Heroin maintenance for heroin addicts: Issues and evidence. N Engl J Med 1973;288:654-60.[ISI][Medline]
    3. Perneger T, Giner F, del Rio M, Mino A. Randomized trial of heroin maintenance programme for addicts who fail in conventional drug treatments. BMJ 1998;317:13-8.[Abstract/Free Full Text]
    4. Rehm J, Gschwend P, Steffen T, Gutzwiller F, Dobler-Mikola A, Uchtenhagen A. Feasibility, safety, and efficacy of injectable heroin prescription for refractory opioid addicts: a follow-up study. Lancet 2001;358:1417-20.[CrossRef][ISI][Medline]
    5. Uchtenhagen A, Dobler-Mikola A, Steffen T, Gutzwiller F, Blattler R, Pfeifer S, eds. Prescription of narcotics for heroin addicts: main results of the Swiss national cohort study. Basel: Karger, 1999.
    6. Van den Brink W, Hendriks V, Blanken P, Koeter M, van Zwieten B, van Ree J. Medical prescription of heroin to treatment resistant heroin addicts: two randomised controlled trials. BMJ 2003;327:310-5.[Abstract/Free Full Text]
    7. Haasen C, Verthein U, Degkwitz P, Berger J, Krausz M, Naber D. Heroin assisted treatment for opioid dependence: a randomised, controlled trial. Br J Psychiatry 2007;191:55-62.[Abstract/Free Full Text]
    8. Frei A. Economic evaluation of the Swiss project on medically prescribed heroin substitution treatment. Psychiatrische Praxis 2001;28:S41-4.[CrossRef][ISI][Medline]
    9. Dijkgraaf M, van der Zanden B, de Borgie C, Blanken P, Van Ree J, Van den Brink W. Cost utility analysis of co-prescribed heroin compared with methadone maintenance treatment in heroin addicts in two randomised trials. BMJ 2005;330:1297.[Abstract/Free Full Text]
    10. Metrebian N, Carnwath Z, Mott J, Carnwath T, Stimson G, Sell L. Patients receiving a prescription for diamorphine (heroin) in the United Kingdom. Drug Alcohol Rev 2006;25:115-21.[CrossRef][ISI][Medline]
    11. Amato L, Davoli M, Perucci C, Ferri M, Faggiano F, Mattick R. An overview of systematic reviews of the effectiveness of opiate maintenance therapies: available evidence to inform clinical practice and research. J Subst Abuse Treat 2005;28:321-9.[CrossRef][ISI][Medline]
    12. Leshner A. Addiction is a brain disease, and it matters. Science 1997;278:45-7.[Abstract/Free Full Text]
    13. Stoermer R, Drewe J, Dursteler-Mac Farland K, Hock C, Mueller-Spahn F, Ladewig D, et al. Safety of injectable opioid maintenance treatment for heroin dependence. Biol Psychiatry 2003;54:854-61.[CrossRef][ISI][Medline]
    14. Rehm J, Frick U, Hartwig C, Gutzwiller F, Gschwend P, Uchtenhagen A. Mortality in heroin-assisted treatment in Switzerland 1994-2000. Drug Alcohol Depend 2005;79:137-43.[CrossRef][ISI][Medline]
    15. Bammer G, van den Brink W, Gschwend P, Hendriks V, Rehm J. What can the Swiss and Dutch trials tell us about the potential risks associated with heroin prescribing? Drug Alcohol Rev 2003;22:363-71.[CrossRef][ISI][Medline]
    16. Nordt C, Stohler R. Incidence of heroin use in Zurich, Switzerland: a treatment case register analysis. Lancet 2006;367:1830-4.[CrossRef][ISI][Medline]

    Neil McKeganey, professor of drug misuse research

    1 University of Glasgow, Glasgow G11 6PW
    [email protected]

    doi: 10.1136/bmj.39421.593692.94doi: 10.1136/bmj.39434.460694.AD

    Some heroin addicts are very difficult to treat. Jürgen Rehm and Benedikt Fischer believe that maintenance with heroin is the way forward for this group, but Neil McKeganey argues that it is treating the effects of misuse not the addiction

    Prescribing heroin to heroin addicts is a strategy beloved by top police officers1 and successive home secretaries.2 It is a strategy, though, borne of utter frustration at our seeming inability to tackle an escalating drug problem. If you cannot stop addicts committing crimes to fund their drug habit then, so the argument goes, the next best thing is to provide them with the drugs that are the reason they are committing the crimes in the first place. The logic may seem faultless, but at the back of your mind is the nagging question, "Is it treatment or is it social problem prescribing?"

    The evidence in relation to heroin prescribing is far from conclusive. On the positive side Nordt and Stohler have suggested that heroin prescribing led to a large reduction in incidence of heroin addiction in Switzerland, although the authors also point out that such prescribing may have reduced individual’s inclinations to cease their heroin use.3 A London study found no health benefits associated with heroin prescribing,4 whereas various Dutch and Swiss heroin trials have identified a range of benefits including improved social functioning and psychological and physical health.5 6 What is often quite difficult to identify from these studies is the degree to which the improved outcomes are the result of the heroin prescribed or other elements of the therapeutic programme provided. The cost of treating an addict with heroin is estimated to be three to four times that of treating an addict with methadone.7

    Risks of prescribing

    In the face of the additional costs and inconclusive evidence, many clinicians are wary of prescribing heroin. Their anxieties are understandable, given the high profile cases of doctors who have prescribed heroin to addicts and then subsequently found themselves facing a General Medical Council inquiry into their prescribing practices.8

    At a clinical level prescribing heroin to heroin addicts is a risky strategy. Once you start, it is difficult not to feel that you have ceded authority for your prescribing to your patient. What, for example, do you say to patients who threaten to resume their previous life of crime if you reduce their heroin prescription? What do you say to the cocaine addict who asks why he cannot have cocaine provided in the same way as the heroin addict? Opening up heroin prescribing to addicts could lead to massive pressure on doctors to prescribe increasing amounts of the drug.

    It was in part as a result of that pressure that the Interdepartmental Committee on Drug Addiction advised the UK government in 1965 that only appropriately certified doctors should prescribe heroin to addicts. The committee’s decision was influenced by the case of Lady Frankau, a noted London psychiatrist who in 1962 prescribed more than 600 000 heroin tablets to her addict patients.9

    What are we treating?

    Prescribing heroin to heroin addicts, however, makes sense only if your primary concern is to treat not their drug dependency but the consequences of their drug use. You may want to reduce their use of street drugs, the risks to health from HIV or hepatitis C virus, the risks of overdose, or their criminality. With all of these aims in mind you may conclude that it makes sense to provide addicts with a prescription for the drug that they have become dependent on. And yet the reason they are committing those crimes, and taking such enormous and persistent risks with their health, is because the drugs have become more important than life itself—that is the nature of drug addiction. And that is the problem that drug treatment services need to tackle.

    Research has shown that with the right services in place it is possible to do more than simply stabilise addicts’ continued drug use through the prescribing route. For example, the Australian treatment outcome study, which followed up 429 heroin users recruited from a random sample of drug treatment agencies 36 months after starting treatment, found that 40% of drug users had been abstinent for the preceding 12 months and 25% had been abstinent for the preceding 24 months.10 In a similar Scottish study of 695 addicts, re-interviewed 33 months after they had started treatment for drug misuse, 29.4% of those who had been provided with residential rehabilitation had been abstinent for at least 90 days before being interviewed compared with only 3.4% of those receiving methadone maintenance.11 All of the residential rehabilitation services included in this study followed an abstinence based programme.

    But do addicts coming forward for treatment actually want heroin to be prescribed to them? A study of over 1033 drug users starting treatment in 2001 asked participants what they wanted to get from the drug treatment services they were contacting.12 Most of those questioned said that they wanted the services to help them become drug free. Health services need to ensure that they are supporting addicts’ attempts to become drug free, and they need to be extremely cautious about any extension of a policy that could be seen as a route to maintaining rather than reducing an individual’s drug dependency.

    doi: 10.1136/bmj.39421.593692.94doi: 10.1136/bmj.39434.460694.AD
    Competing interests: None declared,


    1. Bright M. Police urge major rethink on heroin. Observer 2001 Dec 9.
    2. Blunkett D. David Blunkett’s speech on cannabis. Guardian 2002 Jul 10.
    3. Nordt C, Stohler R. Incidence of heroin use in Zurich, Switzerland: a treatment case register analysis. Lancet 2006;367:1930-4.
    4. Hartnoll R, Mitcheson MC, Battersby A, Brown G, Ellis M, Flemming P, et al. Evaluation of heroin maintenance in controlled trial. Arch Gen Psychiatry 1980;37:877-84.[Abstract]
    5. Van den Brink W, Hendricks V, Blanken P, Koeter M, van Zweiten B, van Ree J. Medical prescription of heroin to treatment resistant heroin addicts: two randomised controlled trials. BMJ 2003;327:310.[Abstract/Free Full Text]
    6. Uchtenhagen A, Dobler-Mikola A, Steffen T, Gutzwiller F, Blatter R, Pfeifer S. Prescription of narcotics for heroin addicts: main results of Swiss national cohort study. Basel: Karger, 1999.
    7. Stimson G, Metrebian N. Prescribing heroin what is the evidence?. York: Joseph Rowntree Foundation, 2003.
    8. Dyer O. Seven doctors accused of over-prescribing heroin. BMJ 2007;328:483.
    9. Interdepartmental Committee on Drug Addiction. Drug addiction in the United Kingdom; the second report of the interdepartmental committee on drug addiction. London: HMSO, 1965.
    10. Darke S, Ross J, Mills K, Williamson A, Harvard A, Teesson M. Patterns of sustained heroin abstinence among long term dependent heroin users: 36 months findings from the Australian treatment outcome study. Addict Behav 2007;32:1897-1906.[CrossRef][ISI][Medline]
    11. McKeganey N, Bloor M, Robertson M, Neale J, MacDougal J. Abstinence and drug abuse treatment: results for the drug outcome research in Scotland study. Drug Educ Prev Policy 2006;13:537-50.[CrossRef]
    12. McKeganey N, Morris Z, Neale J, Robertson M. What are drug users looking for when they contact drug services: abstinence or harm reduction? Drugs Educ Prev Policy 2004;11:423-35.[CrossRef]
To make a comment simply sign up and become a member!