Drug use can be considered to exist along a continuum (Epstein, 2001), with experimental use at one end and regular (hazardous, harmful and dependent) use at the other. Other common types of use include instrumental or situational use and heavy although infrequent use (Wickes, 1992). Accordingly, interventions should be tailored to the client's point on the continuum (Wickes, 1992).

Approaches applicable to all psychostimulant users
Polydrug use
Approaches to experimental psychostimulant use
Approaches to infrequent, heavy use of psychostimulants
Approaches to instrumental use of psychostimulants
Approaches to ecstasy use
Assessment of regular amphetamine use
Psychosocial approaches to regular psychostimulant use (hazardous, harmful or dependent users)
Cognitive behavioural interventions
Characteristics of amphetamine users in outpatient treatment and retention
Residential rehabilitation
Self-help groups

Approaches applicable to all psychostimulant users

Given the risks associated with psychostimulant use (detailed in Chapter 4: Risks associated with psychostimulant use), Hando and Hall (1993) recommended that all users be encouraged to practise safer sexual behaviours and use sterile injecting equipment if injecting. They further recommended that all users be informed about the adverse consequences of heavy use so that they can moderate or cease their use if adverse consequences are experienced (see Chapter 4: Risks associated with psychostimulant use) and if resources allow, be provided with a self-help guide (e.g. Lintzeris, Dunlop & Thornton, 1999; Topp, McKetin, Hando & Dillon, 2001).

Polydrug use

The majority of amphetamine users are polydrug users (Darke & Hall, 1995). Benzodiazepine use among amphetamine users is common (Darke, Ross & Cohen, 1994) and may be used to assist with amphetamine-related problems (Hando, Topp et al., 1997). Heroin has also been used to self-medicate or as a substitute for amphetamines (Hando, O'Brien, Darke, Maher & Hall, 1997). Furr, Delva and Anthony (2000) have reported a significant association between daily alcohol intoxication and methamphetamine ('ice') smoking, independent of potentially confounding factors such as other recent drug use, age and sex. The authors hypothesised that heavy drinkers may use ice to counteract the performance deficits arising from the CNS depressant effects of alcohol.

In their review of the physical and mental health problems experienced by amphetamine users, Vincent and colleagues (Vincent et al., 1998) recommended that an appropriately tailored management program should be negotiated with each client, that polydrug use needs to be considered and that the client may be placed on withdrawal or maintenance programs for other drugs while being treated for amphetamine use. O'Connor and Bradley (1990) have reported a case study successfully employing cognitive therapy for the treatment of amphetamine and benzodiazepine abuse. Top of page

Approaches to experimental psychostimulant use

Recommendations regarding approaches to experimental psychostimulant users have primarily focused on reducing transition to injecting. Hall, Darke, Ross and Wodak (1993) recommended that for people at risk of experimenting with amphetamines, clinicians should discuss the hazards of injection, without exaggerating the risks of occasional low dose oral use. For current users, advice to avoid injection and daily use has been recommended (Hando & Hall, 1993). Presently, there are no recommended safe limits for amphetamine use, but Hall and Hando (1994) have offered the following suggestions to reduce the risk of experiencing adverse effects of amphetamine use: to use less than twice a week and to use small amounts.

Darke, Cohen, Ross, Hando and Hall (1994) reported survey data from 301 regular amphetamine users regarding transitions between routes of administration of amphetamines. The main reasons given for the transition to injecting were enjoying the 'rush' from injecting and viewing it as a more economical and healthier way to use. Only 9% reported a transition away from injection, the main reason being concern over vascular damage. Darke et al. (1994) recommended that interventions to encourage safer use of amphetamines needed to address misconceptions that injecting is more economical and healthy and to emphasise the vascular problems associated with injecting.

Des Jarlais, Casriel, Friedman and Rosenblum (1992) conducted a randomised controlled trial (RCT) in order to evaluate the effectiveness of CBT in preventing transition to injecting among 104 intranasal heroin users in four sessions conducted across two weeks. The four-session small group prevention program has been described in detail by Casriel and colleagues (1990). At nine-month follow-up interviews there was a reduction in injecting in the intervention group with only 15% injecting during the follow-up period, compared to 33% of the control group. Thus it would appear that the intervention had a modest effect in reducing IDU. The authors suggested that intranasal heroin users needed to develop skills to manage social pressures to inject and resources to cope with a reduction in or elimination of their intranasal use. The study by Des Jarlais and colleagues represents a progressive utilisation of CBT among people at an early stage of change (Prochaska, DiClemente & Norcross, 1992) for injecting. Replication of the study with a larger sample of primary amphetamine users is necessary to determine its appropriateness for that group, although the intensity and extent of the intervention needed for sustained change is not yet known.

Hunt and colleagues (1998) reported three-month follow-up data from an uncontrolled study of a brief intervention (less than one hour) among current IDUs. Subjects reported increased disapproval of initiating non-injectors into injecting; reduced requests from non-injectors for subjects to assist with initiation into injecting; and reduced rates of injecting in front of non-injectors. Results suggested that brief interventions with the aim of preventing initiation of non-injectors into injecting are feasible, acceptable and potentially effective. However, only 27% of the sample of 73 subjects reported amphetamines as the main drug injected. Further RCTs of such interventions among amphetamine users are recommended.

Approaches to infrequent, heavy use of psychostimulants

Ten years ago, simple suggestions for interventions with infrequent heavy users were provided by Hando and Hall (1993). These included encouraging awareness of the purity of the drug; adverse consequences of heavy use; the need for moderation or cessation of use if adverse consequences were experienced; a false sense of psychomotor competence that may be produced when used in combination with alcohol; the need to avoid driving when using; and the need to take precautions to reduce harmful side-effects (e.g., obtaining the drug from reliable sources and using smaller amounts per occasion). However, there has been no published research since this time providing an evidence base for such simple interventions.

Approaches to instrumental use of psychostimulants

Instrumental users are those who use amphetamines for specific (non-recreational) purposes. They include, for example, long distance truck drivers, chefs, shift workers and students. No studies have been identified that offered specific harm reduction measures for this group. The advice for experimental and infrequent, heavy users (above) may be appropriate. CBT interventions such as those described below may be indicated. There may be additional opportunities for peer education among different occupational groups, but again these have not been systematically studied. Top of page

Approaches to ecstasy use

In general, ecstasy users do not present for treatment, except in instances of adverse effects serious enough to require medical assessment, or in instances of significant concomitant use of alcohol or other drugs. This is likely to largely be a reflection of typical patterns of ecstasy use. It also determines the type of interventions that can be considered for ecstasy users.

Ecstasy is generally used infrequently, in small amounts (1 to 2 tablets a time, taken orally), in association with social events. This pattern of intermittent use, that is usually self-limited, does not suggest the need for treatment specifically directed at ecstasy use. The occasional occurrence of significant adverse effects, particularly the highly publicised deaths of young people in Australia and the UK subsequent to ecstasy use, have negated the benign image of ecstasy to some extent. Such events have triggered primary prevention initiatives directed at the youth dance party culture.

Given the low numbers of ecstasy users seeking treatment, interventions need to be largely opportunistic. An approach that is well suited to these purposes is that of brief interventions (Barry, 1999). Brief interventions aim to investigate a potential problem and motivate an individual to begin to do something about their substance use. The primary goal of a brief intervention is to reduce the risk of harm that could result from continued substance use. Brief interventions on their own can promote behaviour change, or can act as the first stage of more intense treatment. Furthermore, brief interventions are applicable to individuals from a wide range of cultures and backgrounds and they can be used in a variety of settings, both opportunistic or within specialised substance abuse treatment.

Potential settings for opportunistic use of brief interventions to address ecstasy use include emergency departments of hospitals, subsequent to attendance for acute adverse effects, support services at major events such as dance parties, primary health care (doctors and dentists may detect ecstasy use in the context of other consultations), law enforcement settings (subsequent to being found in possession of an illicit drug) and computer-based applications (the target group is likely to be frequent internet users).

These strengths identify the potential value of brief interventions in addressing ecstasy use, but brief interventions need to be structured and much of the evidence of their effectiveness relates to tobacco and alcohol abuse. The development and evaluation, through structured research, of brief interventions appropriate to ecstasy users and the various contexts for delivery of the interventions is required.

More intense forms of psychological interventions are appropriate to those with problematic ecstasy use. However, as discussed previously, this group is likely to constitute a minority of ecstasy users who are likely to be polydrug users and hence may require additional interventions appropriate to other drugs that are being used. In general, the psychosocial intervention modalities appropriate for cocaine and amphetamine users would also be appropriate for ecstasy users. This is particularly relevant as most ecstasy sold in Australia is actually methamphetamine as detailed in Chapter 2: Prevalence and patterns of psychostimulant use.

Assessment of regular amphetamine use

Teesson, Degenhardt and Hall (2002) have reviewed a number of self-report questionnaires for psychostimulant users, all of which have good reliability and validity. They emphasise the importance of conducting an assessment within the context of a non-confrontational, empathic and mutually respectful therapeutic relationship. The instruments they recommended were: Topp and colleagues (Topp & Darke, 1997;Topp & Mattick, 1997b) have validated the Severity of Amphetamine Dependence Questionnaire (SamDQ) (Churchill, Burgess, Pead & Gill, 1993) as an instrument measuring dependence and capable of discriminating between individuals with different use patterns. Proudfoot and Teesson (2000) also suggest that broad instruments such as the Opiate Treatment Index (OTI) (Darke, Hall, Heather, Wodak & Ward, 1992) and the Addiction Severity Index (ASI) (McLellan, Luborsky, Cacciola, Griffiths et al., 1985), which have good psychometric properties, are useful to assess drug use. Top of page

Psychosocial approaches to regular psychostimulant use (hazardous, harmful or dependent users)

Pharmacological approaches to psychostimulant dependence are reviewed in Chapter 8: Pharmacological interventions. Most of the psychosocial approaches described below are compatible with pharmacotherapy and many people are likely to benefit from a combination of both types of intervention.

Motivational interviewing (MI)

Following their survey of treatment preferences among regular amphetamine users, Hando et al. (1997) have suggested that MI may be appropriate for users who have difficulty perceiving amphetamine-related problems or who are not motivated to attend treatment. Vincent et al. (1998) have suggested that emphasising the associations between severity of dependence on amphetamines and poor mental and physical health may help improve motivation. They suggest that this information could be most effective if provided within the context of the damage to social functioning with which such problems may be associated. Hando et al. (Hando, Topp et al., 1997) have suggested that key factors in defining amphetamine use as a problem, such as dependence and financial difficulties, should be emphasised, increasing awareness of dependence symptoms and the possible adverse consequences of dependence.Treatment seeking amphetamine users have reported that they are especially interested in interventions that are amphetamine-specific, non-judgemental and allow a variety of goals, including abstinence and controlled use (Hando, Topp et al., 1997).

Behaviour therapy and cognitive behaviour therapy

There have been very few studies of non-pharmacological approaches for the treatment of amphetamine use (Baker, Boggs & Lewin, 2001a; Baker, Boggs & Lewin, 2001b) and the effectiveness of different types of psychological therapy for cocaine use has been found to be variable (Gowing et al., 2001). The American Psychiatric Association (APA) (1995) emphasises that the different findings may be due more to intensity of treatment than type of therapy. However, outcomes of the Collaborative Cocaine Treatment Study (Crits-Christoph, Siqueland, Blaine, Frank et al., 1999) suggest that differences may be due to the quality of treatments provided.

Psychosocial therapy for cocaine dependence has traditionally been based on the 12-step approach and much of the controlled research in this area has concentrated on comparing newer therapies with this approach. The APA (1995) concluded that attendance at self-help groups (which are generally based on the 12-step model) might improve long-term outcomes.They also noted that psychodynamic approaches have been the subject of little research to date, but that two psychotherapeutic approaches based on behavioural and cognitive behavioural theory have shown promise. These are discussed below, following consideration of assessment strategies.

Behavioural reinforcement

There is some evidence of the effectiveness of behavioural reinforcement and CBT from the cocaine literature that may be extrapolated for use with amphetamine users. However, caution is warranted because of both the differences between cocaine and amphetamine use and the fact that much of the research on the treatment of cocaine use has come from the USA, which has a strong abstinence orientation and may influence the treatment goals and outcomes measured.

As reviewed by Proudfoot and Teesson (2000), Higgins and colleagues, in research on non-drug reinforcers, used vouchers that were exchangeable for retail items or housing and job opportunities as positive reinforcers for cocaine abstinence (Higgins, Budney, Bickel & Badger, 1994; Higgins, Budney, Bickel, Foerg et al., 1994). Vouchers were employed in combination with a community reinforcement approach (CRA).This intervention produced substantial reductions in rates of cocaine use. CRA involves individual therapy directed at relationships and other living skills in order to increase non-cocaine reinforcers in the individual's environment. The researchers found this approach to be superior to standard outpatient drug abuse counselling. In addition, there were significant improvements in outcomes for the voucher plus CRA compared with CRA condition (Higgins & Wong, 1998). Higgins et al. (1998) also found significantly greater abstinence rates for a group given contingent vouchers compared with another group given non-contingent vouchers. These researchers also incorporated monitored disulfiram therapy in their program for those cocaine users also abusing alcohol and found promising reductions in cocaine as well as alcohol use. Considering that it is estimated that some 60% of cocaine abusers are also alcohol dependent, this finding is important.

In his general review of literature on cocaine addiction, Platt (1997) commented that research had indicated that the magnitude of reinforcement and immediacy of reinforcement might be critical in determining efficacy of a voucher system. He also pointed to some research that has not supported the use of vouchers to encourage abstinence from cocaine, especially on a longer-term basis. In attempting to explain the disparities in the literature, he suggests that the study samples were from widely divergent social settings — those that obtained best results were from a rural environment, whilst those with negative findings were from an inner-city environment.
Top of page
Subsequent to publication of Platt's review, there have been a number of published studies investigating the effectiveness of voucher systems. In two studies involving 90 severely socio-economically disadvantaged cocaine users (88% crack cocaine), Kirby et al. (1998) investigated the effect of adding voucher payments for cocaine-free urine screens to a comprehensive treatment package. The treatment package consisted of 26 sessions of CBT plus 10 one-hour sessions of interpersonal problem solving carried out over the 12 weeks of the study. In the first study, voucher delivery was on a weekly basis with initial values low, increasing with production of consecutive negative urine results and reset to zero on production of positive screens. In this study the use of vouchers was found to have no effect. This is consistent with Platt's view that negative results tend to be associated with an inner-city environment.

The second study involved 23 subjects. Half the group received vouchers on a weekly basis while the other half received vouchers immediately upon producing the cocaine-free urine. The values of the vouchers started high ($30 for the first nine cocaine-free specimens) with no punishment for positive screens. Repayments became more intermittent after this, but overall maximum earnings were greater. There was a trend for this system of voucher delivery to improve retention and attendance outcomes, but low numbers are likely to have prevented these differences from being significant. There were also significant improvements on measures of abstinence for immediate compared with weekly voucher delivery. About half the participants on immediate voucher delivery completed treatment and showed continuous abstinence at one month following treatment, whereas no participant on weekly voucher delivery achieved one month of continuous abstinence. This finding provides some support to Platt's conclusion that immediacy of reinforcement may be an important determinant of efficacy.

Further support for this is provided by an RCT comparing behavioural day treatment (DT) only with DT plus abstinent-contingent housing (available immediately on achievement of four consecutive urine samples over two weeks) and DT plus work therapy during aftercare in a sample of homeless persons with substance use disorders (primarily crack cocaine) and non-psychotic mental disorders. DT was associated with greater abstinence at two and six months and more days of treatment attendance (Milby, Schumacher, McNamara, Wallace et al., 2000). The odds of being cocaine-abstinent increased with days of treatment attendance (Schumacher, Usdan, Milby, Wallace & McNamara, 2000).

Cognitive behavioural interventions

CBT for cocaine use is aimed at helping individuals to recognise that they have a problem with their cocaine use, to understand their problem and to assist users to modify the dysfunctional cognitions underlying this problem behaviour. Therapy typically involves skills training and practise to deal with craving, monitoring thoughts about drugs and monitoring high-risk situations associated with relapse (Carroll, 1998). Cognitive behavioural interventions have not generally been demonstrated to be superior to other psychotherapies in initiating abstinence, but research suggested that its effects may be more durable and thus protective against relapse. Furthermore, CBT may be more effective with more severely dependent users (Carroll, 1998). This was also the conclusion of the APA (1995).

Baker and colleagues (2001b), in an RCT, compared a brief cognitive behavioural intervention (either two or four sessions duration) with a self-help booklet (control condition). Participants were regular (at least monthly) users of amphetamines. Moderate reductions of amphetamine use were reported by both groups, but significantly more people in the CBT condition abstained from amphetamines at six-month follow-up compared to the control condition. This study demonstrated the feasibility of brief CBT for the treatment of regular amphetamine use.

Over a decade ago, Hawkins, Catalano, Gillmore and Wells (1989) reported 12-month follow-up data for a randomised trial of CBT (Hawkins, Catalano & Wells, 1986) among people in the re-entry phase of residential therapeutic communities (TCs). The intervention consisted of drug refusal and avoidance skills, problem-solving, social and stress coping skills, how to deal with depression or with being treated unfairly, coping with a slip into drug use and coping with personal high-risk situations. Community volunteers also became involved in sessions and attended bimonthly support groups for six months. Subjects were expected to remain in treatment during the 10-week period when the CBT group received their intervention. At 12-month follow-up subjects who completed CBT had significantly higher skill scores than did controls. The CBT intervention did not significantly affect subjects' drug use except for a marginal effect on amphetamine use (p<.05) at 12 months for the entire sample and for fully treated subjects at 6 months compared to controls. Urinalysis results corroborated self-reported drug use.

Hawkins et al. (1989) stated that the generalisability of their findings is questionable given that the subjects were a highly select sample of volunteers who had completed a lengthy and demanding drug treatment program. They also noted that the design, in which CBT was in addition to an already intensive and lengthy program, does not address the effectiveness of CBT as the primary treatment modality or in combination with other treatments (e.g. methadone maintenance treatment). In addition, these studies were not conducted among primary amphetamine users. However, this study provides some initial evidence that adjunctive CBT may be effective within the context of residential programs in reducing amphetamine use.
Top of page
Maude-Griffin and colleagues (1998) compared CBT with 12-step facilitation in a randomised study involving 128 crack cocaine smokers. This was a socio-economically disadvantaged group with 75% homeless or marginally housed, 84% unemployed, 82% with comorbid psychiatric disorders (and almost half with two other psychiatric disorders) and a mean length of cocaine use of 19 years. Participants attended three groups and one individual therapy session per week over 12 weeks.Treatments were manualised and administered by counsellors with extensive experience, with the same counsellors administering both therapies. The 12-step facilitation group was encouraged to attend Cocaine Anonymous, while the CBT group was encouraged to attend Rational Recovery, a cognitively based self-help group.

Attendance at treatment groups was low — only 17 participants (13%) attended at least 75% of both group and individual sessions. Overall, 44% of the cognitive behavioural group and 32% of the 12-step facilitated group achieved four consecutive weeks of abstinence from cocaine (p<0.05). However, the outcomes varied for different subgroups of participants. For those assessed as having high levels of abstract reasoning, 50% in the cognitive behavioural group achieved four weeks of abstinence compared to 25% in the 12-step facilitated group. This result was virtually reversed (18% compared to 48%) for those assessed as having low levels of abstract reasoning.

For those assessed as having a low degree of religious belief, 48% in the cognitive behavioural group achieved four weeks of abstinence, compared to 12% in the 12-step facilitated group. For those assessed as having a high degree of religious belief there was little difference between the two groups: 35% in the cognitive behavioural group and 40% in the 12-step facilitated group achieved four weeks of abstinence. This variability indicates the importance of providing treatment that is relevant to the individual.

Monti et al. (1997) compared the effects of adding brief coping skills training or 'attention placebo' to a comprehensive treatment package incorporating both 12-step and social learning principles. The coping skills training was directed towards high-risk situations while the 'attention placebo' involved the same number of hours in manualised meditation and relaxation training, which the researchers regarded as a credible but ineffective treatment. Both approaches were administered on an individual basis in eight one-hour sessions. Self-reported cocaine use at six months pre-treatment and one-month and three-month follow-up assessments was confirmed with urine tests as well as collateral reports. Demographic information and indices of psychosocial wellbeing were also obtained at pre-treatment and 3-month follow-up.

Monti et al. found that there were no differential effects of the two additional interventions in terms of total abstinence during the 3-month follow-up period, or on longest continuous abstinence. However, there were significant reductions in days of use as well as length of bingeing for participants in the coping skills treatment condition compared with placebo, variables that are considered to be more sensitive than the categorical abstinence measure.

Overall, the authors concluded that the brief coping skills intervention led to shorter and less severe relapses. These results fit with prior findings that interventions based on cognitive behavioural principles may have more impact on longer-term relapse prevention than on more immediate broad measures of drug use or abstinence.

In a multicentre collaborative cocaine treatment study, supported by the National Institute on Drug Abuse in the US (NIDA) (Crits-Christoph et al., 1999), 487 participants were randomised to four treatment conditions:
  1. individual drug counselling plus group drug counselling;
  2. cognitive therapy plus group drug counselling;
  3. supportive-expressive therapy plus group drug counselling; and
  4. group drug counselling alone.
All treatments were manualised with a six-month active phase and a three-month booster phase. The individual drug counselling and group drug counselling were based on the disease model of substance use, with strong encouragement to participate in 12-step programs, and they taught participants how to progress through stages of recovery from addiction. Cognitive therapy followed a program for substance abuse based on social learning theory. Supportive-expressive therapy was based on the psychoanalytic approach to treatment for substance abuse.

The Crits-Christoph et al. study employed a composite outcome measure of cocaine use, which ascribed the rating 'abstinent' or 'not abstinent' for each month. Any indication of drug use from either urine tests, Addiction Severity Index responses or a weekly cocaine use inventory led to a 'not abstinent' rating. Where no measures were available (which occurred on 19% of possible occasions) participants were rated as 'not abstinent'. However, as only 42.6% of all potential urine specimens were collected, this global abstinence rating may have been unreliable.
Top of page
Participants in the study conducted by Crits-Christoph et al. (1999) were obtained from a total of 2,197 persons screened by phone, of whom 1,777 met inclusion criteria and 870 were considered to have begun what was termed the orientation phase of treatment. During this phase participants were required to attend three clinic visits within 14 days to demonstrate their motivation. At this time the participants were encouraged by group counsellors to attend self-help groups based on 12-step principles. Housing, employment and financial needs were also addressed during the orientation phase. Only 487 (56%) proceeded to randomisation and the active therapy stage.

It was found that participants in the three groups which received individual therapy had significantly better outcomes than those who received only group drug counselling. Despite poorer retention, it was also found that individual counselling plus group drug counselling was more effective than cognitive therapy plus group drug counselling or supportive-expressive therapy plus group drug counselling in promoting abstinence (in the past 12 months).

However, as the authors point out, the superiority of individual counselling plus group drug counselling in this study may be due to the additive effect of the single focus (on 12-step principles). Further, as Carroll (1999) comments, a focus on the 12-step principles in the orientation phase may have proven selective for those who were more amenable towards this approach. This, along with possible differential attendance at AA-type self-help meetings, would also help explain the need for less treatment in this group and thus lower retention rates.These factors are yet to be examined by the researchers.

Crits-Christoph et al. (1999) suggested that one reason for the effectiveness of individual counselling, when it had not been found to be effective in previous studies, was the use of high quality manualised counselling with highly selected and experienced counsellors. Thus, the greater intensity of treatment provided by individual counselling plus group drug counselling compared with group drug counselling alone may be interpreted as a response to a higher dose of treatment. On the other hand, the interaction of two approaches based on different models (as with the psychotherapies plus 12-step orientated group drug counselling) may be counterproductive.

It could be argued that this study demonstrates that a singular concerted approach may be more effective than the more eclectic approach often found in drug counselling in community settings. This point was also raised by Carroll (1999) in relation to the transfer from orientation to active phase. The Crits-Christoph et al. study demonstrates that manualised individual therapy in addition to group counselling leads to significant improvements in outcome. However, because of the correlation of selection (orientation), group and individual counselling procedures offered, it is difficult to draw definitive conclusions from this study regarding the relative merits of individual counselling versus cognitive therapy and supportive-expressive therapy.

Characteristics of amphetamine users in outpatient treatment and retention

Copeland and Sorensen (2001) investigated differences between primary methamphetamine and cocaine-dependent outpatients in a retrospective chart review of 345 admissions to the Stimulant Treatment Outcome Program (STOP) in San Francisco during 1995–1997. Methamphetamine users were found to engage in higher rates of injecting risk-taking behaviour, were more likely to be HIV positive, have a psychiatric diagnosis and be prescribed psychiatric medications. Only 18% of all clients completed the six-month treatment program and there were no differences in retention rates between methamphetamine and cocaine patients.

The authors suggested that the findings highlighted the need for more effective treatments for psychostimulant abuse and dependence, although not necessarily the development of novel treatments for amphetamine users. They suggested it might be more productive to provide ancillary services in order to address amphetamine users' more severe medical and psychiatric problems.

Maglione, Chao and Anglin (2000) examined retention among 2,337 methamphetamine users entering public outpatient treatment programs from the California Alcohol and Drug Database System (CADDS) between January 1994 and September 1997. Dropout was defined as receiving less than 180 days of treatment. Overall, 23% completed treatment and the average stay in treatment was 112 days. Men were 1.35 times more likely to drop out of treatment than women and people 40 years of age and older were significantly less likely to drop out. Referral from the criminal justice system was a strong predictor of treatment retention. Those who reported injecting drug use (IDU) were 1.5 times more likely to drop out compared to those who smoked or snorted the drug. In addition, daily users were more likely to drop out. Thus, it appears that completion of lengthy outpatient treatments is low and strategies to improve treatment completion rates of men, younger people and IDUs are needed.

As part of an ongoing study to describe use ecology and drug use motivation among amphetamine users, Von Mayrhauser, Brecht and Anglin (2002) have interviewed 260 participants from the CADDS study. Thus far, the most commonly stated reasons for amphetamine use are as a substitute for other psychostimulants (28%); to cope with mental illness, mental distress or trauma (28%); to stay awake (23%); to enhance sexual experience (11%); and to lose weight (10%).Von Mayrhauser and colleagues expect that developing a profile of amphetamine users will help the development of locally relevant treatment protocols for amphetamine users and identify areas worthy of further research. Top of page

Matrix Model program

The outpatient Matrix Model program for psychostimulant users was designed to integrate several interventions into a structured approach (e.g. Huber, Ling, Shoptaw, Gulati et al., 1997). Elements of the treatment include individual therapy, family education groups and relapse prevention groups, conjoint sessions and 12-step involvement.

Specific goals are to stop drug use, learn about issues critical to addiction and relapse, educate family members regarding addiction and relapse, become familiar with self-help programs, and receive weekly urine screening and breath alcohol testing. Treatment materials are manualised. The recommended treatment duration was 26 weeks (52 individual sessions, two stabilisation groups, 24 relapse prevention groups, 12 family education groups and numerous 12-step groups) from 1987 to 1990 and 16 weeks from 1991 to the present (Huber et al., 1997; Shoptaw, Rawson, McCann & Obert, 1994).

The program has been employed extensively in Southern California for over 15 years. Currently, a seven-site randomised controlled trial is being conducted among methamphetamine users in the USA, with subjects being randomly assigned to either the standardised Matrix 8- and 16-week protocols or usual treatment (Freese, Obert, Dickow, Cohen & Lord, 2000; Galloway, Marinelli-Casey, Stalcup, Lord et al., 2000; Herrell, Taylor, Gallagher & Dawud-Noursi, 2000; Huber, Lord, Gulati, Marinelli-Casey et al., 2000; Obert, McCann, Marinelli-Casey, Weiner et al., 2000; Rawson, McCann, Huber, Marinelli-Casey & Williams, 2000; Reiber, Galloway, Cohen, Hsu & Lord, 2000).

A number of studies describing outcomes of the program have been published (Rawson, Huber, Brethen, Obert et al., 2000; Simon, Richardson, Dacey, Glynn et al., 2002). Rawson et al. (2000) compared the characteristics and treatment retention among 500 methamphetamine and 224 cocaine users between 1989 and 1995 in California. Cocaine users reported more episodic use patterns, spent more money on purchasing their drugs and used alcohol more heavily. Methamphetamine users included a higher proportion of women, individuals who tended to use on a daily basis, used cannabis more often and experienced more severe medical and psychiatric consequences. Despite these differences in sample characteristics, there were no differences in treatment retention between the samples. Mean retention was 118 days for methamphetamine users and 125 days for cocaine users.

Recently, Rawson et al. (2002) described the outcome status at 2–5 years of a convenience sample of 114 of the 500 methamphetamine users recruited in the original sample. Methamphetamine and other drug use were significantly reduced from pre-treatment levels and the follow-up status of the sample was much improved. However, the authors note that this type of follow-up data does not allow conclusions regarding the specific impact of the Matrix program. Also, as the follow-up group stayed longer in treatment than that which was not followed up, it should be assumed that this outcome is better than would be reported for the sample as a whole. Nevertheless, these results are promising and we await the results of the seven-site study with great interest, particularly whether there are differences in effectiveness between the 8- and 16-week programs.

However, this sort of therapy is resource intensive, even in the eight-week form, and it may be that shorter interventions may be suitable for some people. For example, there is evidence from a randomised controlled trial (see CBT section above) that briefer outpatient CBT can be effective among regular amphetamine users (Baker et al., 2001b).

Residential rehabilitation

Psychosocial approaches vary considerably in their setting (outpatient, residential, self-help group) and treatment orientation (Swindle, Peterson, Paradise & Moos, 1995). Research evidence in this area is limited. The evidence that exists comes mainly from observational studies (Gowing, Cooke, Biven & Watts, 2002) such as the Drug Abuse Reporting Program (DARP), the Treatment Outcome Prospective Study (TOPS), the Drug Abuse Treatment Outcome Study (DATOS), all undertaken in the USA, and the National Treatment Outcome Research Study (NTORS), undertaken in the UK. In part the limited research evidence reflects ethical and procedural difficulties in conducting randomised controlled trials with clients of residential rehabilitation facilities (Toumbourou & Hamilton, 1993).

Residential rehabilitation is based on the principle that a structured residential setting provides an appropriate context to address the underlying causes of addictive behaviour. These programs assist the client to develop appropriate skills and attitudes to make positive changes towards a dependence-free lifestyle. Wickes (1993) has noted the comparative efficacy and cost-effectiveness of outpatient versus in-patient treatments among cocaine users. She cited recommendations by Taylor and Gold (1990) that in-patient treatment may be considered when there is polysubstance dependence; severe withdrawal is a possibility; medical complications may require close observation or treatment; there may be psychiatric complications; living conditions are undesirable; outpatient treatment has repeatedly failed; or social supports are absent. The duration of stay should be tailored to the individual and their goal to be achieved and in all cases be long enough for resolution of withdrawal symptoms (Wickes, 1993).

Several studies of the effectiveness of residential treatment among drug users have reported results separately for amphetamine users. Thirty years ago, Melin and Gotestam (1973) reported on a residential contingency management program for injecting amphetamine users. Significantly more people who had received the program were drug free at 6 and 12 months compared to a comparison group who received residential treatment without the contingency management program. They suggested further research was required to establish appropriate schedules and the most appropriate contexts for these. The effectiveness of these interventions for amphetamine users has not been investigated. In addition, Australian trials of these interventions were needed, as the methods used in US research may not easily translate to Australian treatment services.
Top of page
The study by Hawkins, Catalano, Gillmore and Wells (1989) reported above (CBT section) provided some initial evidence that adjunctive CBT may be effective within the context of residential programs in reducing amphetamine use. However, these results need to be replicated among primary amphetamine users.

Evidence of the effectiveness of a CBT program adjunctive to residential treatment among primary amphetamine users was reported in a non-randomised comparative study by Smith, Volpe, Hashima and Schuckit (1999). Data from two groups of consecutive admissions of male veterans with alcohol dependence, dependence on amphetamines or cocaine, or both, were reported for the 383 subjects who completed at least 21 days of the 28 day in-patient treatment. All patients were assigned to aftercare groups for up to six months and after discharge approximately 71% went to abstinence-oriented recovery houses for two months. The enriched program consisted of all the elements of the standard program, including the aftercare and recovery housing, to which psychostimulant-focused elements were added for a total of 10 additional hours per week. Two one hour relapse prevention group sessions per week were conducted and two hours of related homework was required per week. Two one and a half hour sessions of interpersonal counselling groups per week were held, along with two hours per week of related homework. Both groups utilised a therapist manual. The remaining one hour per week came through additional educational material focused on psychostimulants added to the weekly meeting of the family and friends group. Follow-up occurred at 3 and 12 months. At 3 months, abstinence from substances were 63% and 49% for the standard and enhanced groups respectively and 43% and 24% at 12 months respectively.

Although the enhanced program showed a lower percentage of subjects returning to psychostimulant use than for the entire group, the results were difficult to interpret for the small numbers of subjects remaining who were dependent on psychostimulants only. Smith and colleagues (1999) concluded that despite the enhanced treatment focus on psychostimulants, both alcohol and stimulant-dependent participants appeared to benefit from the enhanced treatment, suggesting that different substance problems do not require different treatment interventions and that more intense interventions produce better outcomes.

The absence of randomisation of subjects in this study is a serious flaw and the study should be replicated with RCT methodology. Although different treatment interventions may not be required for different drug classes, it is important to note that the intervention in this study was manual driven and delivered by therapists who received six weeks of training. Furthermore, the results are only generalisable to those subjects who had remained in treatment for at least 21 days.

Overall, there is support from three trials that enhancing residential treatment with behavioural (contingency) management or CBT is associated with better outcomes for amphetamine users. However, further randomised controlled trials among primary amphetamine users are required. In addition, longer-term residential treatment may only be suitable for a small proportion of psychostimulant users. Hando, Topp and Hall (1997) reported that as most of their sample of 200 regular users of amphetamines in Sydney was employed, home detoxification or short-term residential treatments may be more appropriate for this population.

Therapeutic communities

Therapeutic communities (TCs) represent a subset of residential rehabilitation where residents participate in the management and operation of the community. The community is the principal means for promoting behavioural change and there is a focus on social, psychological and behavioural dimensions of substance use (Gowing, Cooke et al., 2002). The philosophies of TCs and 12-step groups are such that they tend to be available to AOD users in general. Hence research evidence is generally not related to specific illicit drugs, although psychostimulant users are often included in studies of these approaches.

Residential rehabilitation originally was based around lengthy periods of stay. However, in the last two decades, short-term residential rehabilitation programs have emerged. There is also a developing trend for both therapeutic community and 12-step approaches to be used in conjunction with other treatment approaches (both pharmacological and psychosocial). This diversity of intervention approach complicates the task of assessing the effectiveness of general drug-free approaches.

Gowing et al. (2002) recently reviewed the research literature on the effectiveness of TCs. They noted that there have been very few comparative studies of the effectiveness of TC treatment with good control of bias and confounding factors, making it difficult to form an accurate view of the effectiveness of this approach relative to other treatment modalities. Furthermore, the major longitudinal studies, such as DARP, DATOS and NTORS, combine TCs with other residential rehabilitation approaches, further limiting the data available specific to the effectiveness of TCs. Consequently Gowing et al. made their assessment of effectiveness based on the consistency of outcomes to the multiple follow-up studies that are available.
Top of page
Concern has been expressed over a period of many years regarding high rates of dropout from TCs, particularly early in treatment. At the same time, there is a long-standing view among residential treatment services that three months or more in treatment is necessary for enduring behavioural change. The studies reviewed by Gowing et al. (2002) indicate that between 30% and 50% of those entering TCs remain in treatment at around the three month mark. Reported median or mean lengths of stay ranged from 54 to 100 days. Hence the majority of those entering TCs do not remain in treatment for the length of time considered necessary for enduring change.

Some strategies, such as preparatory interventions prior to entry, have the potential to improve retention rates, as do approaches such as providing additional services to meet individual needs. Perhaps the strongest message from the reported retention rates is that the TC approach does not suit all people and individuals are likely to vary in their receptiveness to the approach at different stages of substance abuse and recovery. This emphasises the importance of linking TCs to other treatment approaches to ensure there are alternatives available for those who find themselves unable to complete treatment.

As with other forms of treatment, relapse to substance use is common following TC treatment. Nonetheless, overall levels and frequency of drug use are significantly reduced by TC treatment, with the reduction still apparent one to two years after exit. The degree of reduction is at least similar to and possibly more enduring than the changes achieved with methadone maintenance treatment. Findings in relation to levels of criminal behaviour are similar. Other aspects of health, particularly psychological symptoms, are also significantly improved with TC treatment and there is a trend of increasing participation in employment and education or training. These reported areas of significant improvement indicate the benefits that can be gained by those who respond positively to the TC approach and justify the continued availability of this approach as part of a treatment system.

There is a strong indication provided by the studies reviewed that time in treatment is a significant determinant of treatment outcome, but this is a complex issue with time being something of a proxy indicator for engagement, participation and progress in treatment. Nonetheless the evidence from the studies reviewed here is consistent with the accepted benchmark of at least three months in treatment before enduring behaviour change is likely to be seen. Given that time is a proxy for other factors, it would be useful to give greater attention to issues of participation and motivation during treatment, with a view to increasing the average length of stay in TCs and therefore potentially improving outcomes on average. Other factors worthy of consideration include involvement of the family, childcare, comorbidity (particularly psychiatric conditions) and cultural issues. Information on the cost effectiveness of the TC approach is particularly lacking.

Self-help groups

Self-help or mutual support groups are most commonly based on the principles of Alcoholics Anonymous (AA) or Narcotics Anonymous (NA), which espouse a disease concept of drug and alcohol dependence with the potential for recovery, but not cure, for those who adhere to it. The '12 steps' of AA/NA contain a strong spiritual component. They emphasise the importance of reconstructing relationships with other people, including confession, restitution and an injunction to help other alcoholics or addicts. They contain an implication that a decision to change is within the power of the individual, even if the power to effect that change is not (Cook, 1988). One of the perceived benefits of self-help or mutual support groups is that they provide a mechanism to promote alternative social networks that do not support drug use. It has been found that abstinence is more likely in individuals who have formed new social networks (Powell & Taylor, 1989).

The efficacy of self-help groups based on the 12-step approach of AA to support the maintenance of abstinence has been briefly reviewed by Fiorentine (1999).

Fiorentine (1999) noted that claims by AA of efficacy are often based on testimonies of long-term, abstinent participants, which may exaggerate the effectiveness of AA if those who drop out are more likely to continue or resume alcohol or drug use. Fiorentine (1999) identified more rigorous studies; both observational after-treatment studies and some controlled studies and noted that both groups of studies offer mixed results as to the effectiveness of the 12-step approach. One possible explanation given for the inconsistent findings is that some 12-step groups are more effective than others, but it remains unclear what comprises an effective or ineffective 12-step approach. It is also probable that some individuals will respond better to the 12-step approach than others (Maude-Griffin et al., 1998).
Top of page
Fiorentine and colleagues (1999; 2000) have used a longitudinal study of more than 400 adult clients entering 25 outpatient treatment facilities in Los Angeles to investigate a number of aspects of 12-step programs, with attempts to control for the confounders of motivation and simultaneous activities. In this group the primary drugs most commonly used in the year prior to treatment were crack cocaine (56%), cannabis (46%), methamphetamine (24%) and cocaine (22%), with around half the cohort being polydrug users. Key conclusions were:
  • weekly or more frequent 12-step participation may be an effective step in maintaining relatively long-term abstinence;

  • less than weekly participation does not seem to be any more effective than non-participation;

  • formal drug treatment and 12-step programs were seen as integrated recovery activities, rather than alternatives;

  • individuals with pre-treatment involvement in 12-step programs stayed in treatment longer and were more likely to complete a formal 24-week treatment program; and

  • individuals who participated in both formal drug treatment and a 12-step program had higher rates of abstinence than those who participated only in formal treatment (consistent with findings that intensity and duration of treatment is important for a successful outcome).
Weiss et al. (1996) made the point that attending self-help groups in itself is not sufficient — it is participation in self-help group meetings that is critical. They support this view with data from a survey of 519 cocaine-dependent people entering a psychotherapy study. In the week prior to study entry 34% had attended a self-help group. Of those who attended and actively participated in a self-help group meeting, 55% initiated abstinence within the next month, compared with 40% of non-attenders and 38% of non-participating attenders. The majority of self-help attendees may not continue with the program for long enough to accrue significant benefit, with more than half dropping out by three months (Fiorentine, 1999). Many find the heavy emphasis on spirituality difficult to reconcile and the 12-step method has been criticised for emphasising reliance on external forces for recovery (Li, Feifer & Strohm, 2000) with some professionals expressing concern about potential disempowerment of users in treatment.

As a result of some of these limitations other self-help groups have been developed to provide an alternative to the 12-step model. SMART Recovery (Self Management And Recovery Training) is an abstinence-based self-help group based on a cognitive behavioural model designed to provide similar support mechanisms and be more compatible with mainstream drug treatment, which is also based on a cognitive behavioural model. Unlike AA, SMART recovery has group facilitators, some of whom are professionals, and professional volunteer advisers, and the groups are more focused on developing skills and education (Fletcher, 2001). However, because it is relatively new, like AA, little research has been conducted into its efficacy, although the advisers point out that it is based on an evidence-based treatment intervention.