Cocaine Use: Recommendations in Treatment and Rehabilitation

5. Treatment Approaches

Two types of treatment interventions for cocaine-related problems are described in the following section: pharmacotherapy and behavioural treatment. These interventions pursue a variety of goals. The focus of pharmacotherapy research has been on finding a medication that will block or substantially reduce the effects of cocaine, and also block the severe craving experienced by users. In addition, some studies have focussed on pharmacological treatment for the depression that is associated with withdrawal from cocaine. Such medications may address both the management of withdrawal, as well as longer-term maintenance. Behavioural interventions address the reduction or elimination of attitudes, feelings or behaviours that support or contribute to substance use, and help the individual to develop healthier behaviours and a corresponding healthier lifestyle.

For the evaluation of medications, all articles that used double-blind randomized control procedures were selected as the basis for making best practice statements. In addition, some information is provided on drugs that show potential, but have not yet demonstrated effectiveness in rigorous scientific trials. For the evaluation of behavioural treatment, studies that used random assignment and a credible control or comparison treatment procedure were included.

For each study that was evaluated, the following information (when available) was obtained: a) retention in the treatment program, b) reduction of cocaine usage, c) reduction of cocaine cravings, and d) improvement in medical or psychological status. These categories are consistent with those recommended by the Treatment Protocol Effectiveness Study (Treatment Outcome Working Group, 1996). Most of the pharmacotherapy studies provided information for some or all of these variables. However, the studies evaluating behavioural treatment procedures provided less information.

Some studies involved interventions with subjects who were dependent on other drugs besides cocaine, particularly opioids and alcohol. There is some evidence that people with addictions to more than one drug may not respond as well to treatment compared to those abusing only one substance (e.g. Brown, Seraganian and Tremblay, 1994). This is consistent with the treatment of other disorders.

It is important to note that most of the drug studies also used psychosocial treatment interventions in addition to medication. The purpose of these studies was to determine if the addition of medication improved treatment effectiveness over psycho-social treatment only. In most of these studies, participants in both groups showed improvement, and the addition of medication generally did not improve treatment efficacy.

It should also be noted that, although most studies included both genders (the proportion of women was usually about 25-30%), study results did not usually include analysis of outcome by gender.

5.1 Pharmacotherapy

Cocaine acts by blocking the re-uptake of three neurotransmitters (dopamine, norepinephrine and serotonin) which produces cocaine's acute reinforcing or pleasurable effects. Studies have examined drugs that either block the reinforcing effects of cocaine (antagonists) or share some of the reinforcing effects of cocaine, but have a longer duration of action (agonists or analogs). Other studies have examined the usefulness of drugs that address some of cocaine's side effects such as depression or seizures. Based on their pharmacological properties, four primary classes of drugs have been used to treat cocaine-related problems: antidepressants, dopamine regulators, anti-seizure medications and drugs that are used to maintain opioid users.

5.1.1 Antidepressants

Table 1 reviews studies that used antidepressants to treat cocaine addiction. All of the drugs described in this Table have effects on the serotonergic and norepinephrine neurotransmitter systems. The first six studies used tricyclic antidepressants. The final four used fluoxetine, a selective serotonin re-uptake inhibitor. Antidepressants are thought to be useful for treating cocaine-related problems for two reasons. First, many cocaine users experience features of depression when withdrawing from cocaine. Second, repeated exposure to cocaine can cause prolonged deficits in serotonergic function (Levy et al., 1993). All antidepressants work partly by regulating serotonergic systems.

Table 1 - Comparison of Antidepressants and Placebo for Treating Cocaine Addiction

Study	Medicine	Subj- ects	Ret- en. 1	Red. Coc. 2	Red. Urg-es 3	Med/ Psy-ch Improv. 4	Comments
Nunes et al., 1995	Imipramine	Cocaine	-	?			Poor response for intravenous, freebase users & non-depressed.
Gallo- way et al., 1994	Imipramine	Cocaine		-	-	-	Study conducted in community setting.
Camp- bell et al., 1994	Desipramine	Cocaine	-	-			Also evaluated carbamazepine. No evidence for effectiveness.
Carroll et al, 1994 (a & b)	Desipramine	Cocaine		?			Group differences did not persist formedication beyond 6 weeks, improvementsdue to cognitive behavioural therapy,notmedication
Gawin et al., 1989	Desipramine	Cocaine					Included weekly psychotherapy.
Oliveto et al., 1995	Desipramine	Cocaine /Opioid					Results are compared to a group receiving fluoxetine.
Covi et al., 1995	Fluoxetive	Cocaine	-	-	-	-	Highest dose (60 mg) of fluoxetine produced poorest outcome.
Batki et al., 1994	Fluoxetive	Cocaine /Opioid					Reductions found in methadone maintained cocaine dependent subjects only.
Batki et al., 1996	Fluoxetive	Cocaine		-	-	-	Could only compare usage and craving for first 6 weeks.
Wash-burn et al., 1994	Fluoxetive	Cocaine
Note: 1. Increased Retention in Program 2. Reduced Cocaine Usage 3. Reduced Cocaine Craving 4. Medical/Psychological Improvement = treatment group improved compared to control group - = no differences between groups ? = could not determine if differences occurred

Tricyclic Antidepressants

Several studies have examined the efficacy of tricyclic antidepressants for the treatment of cocaine dependence and associated problems. Galloway et al. (1994) found that imipramine was effective in retaining people in treatment when compared to a placebo, but had no effect on cocaine craving or cocaine use. In contrast, Nunes et al. (1995) found that imipramine significantly reduced cravings among nasal cocaine users, but those who injected or freebased showed a very poor response to imipramine. This same study also found that cocaine users who were depressed showed a better response to imipramine than those who were not, indicating the need for careful client assessment when considering the suitability of this type of pharmacotherapy.

Oliveto et al. (1995) found that, similarly to imipramine, desipramine was effective in retaining people in treatment, but also had no effect on craving or cocaine use. In this study (Oliveto et al., 1995), subjects were addicted to both heroin and cocaine, and all were being maintained on buprenorphine. Desipramine was more effective than either amantadine or fluoxetine in retaining patients in treatment over a 12-week period. In addition, the desipramine and amantadine groups had a greater number of drug-free days than the fluoxetine group.

Gawin et al. (1989) in a double-blind study compared desipramine with lithium and a placebo. In addition, subjects in all three conditions attended weekly individual outpatient psychotherapy sessions. Over the six-week period of the study, subjects in the desipramine group were significantly more likely to achieve continuous periods of abstinence, to show reductions in cocaine craving and to have longer retention in treatment than the other two conditions. Carroll et al. (1994) compared behavioural relapse prevention to clinical management, paired with either desipramine or a placebo over a 12-week period. The study found that: subjects in all four groups showed significant improvement over the 12-week period; desipramine was more effective than a placebo in reducing cocaine use over the first six weeks, but these differences did not persist beyond six weeks; and subjects with low-severity cocaine use had significantly longer periods of consecutive abstinence in the desipramine condition than in the placebo condition. This study also involved a 12-month follow-up in which the authors conclude that the efficacy of desipramine did not persist after short-term treatment and that its effects were most apparent in the early stages of treatment.

Another study (Campbell et al., 1994), compared desipramine and carbamazine (an anticonvulsant), and found that neither was more effective than a placebo in reducing cocaine usage. Finally, a study by Bystritsky et al. (1991) found some evidence that desipramine can be effective in reducing cocaine-induced panic attacks.

The effectiveness of desipramine (and possibly other antidepressants) may be reduced by use of other drugs. Kosten, et al. (1990) compared plasma levels of desipramine in patients who were being treated for depression or cocaine use. Plasma levels were found to be much lower in a subset of cocaine users being maintained on methadone for heroin addiction. The authors suggest that methadone may have affected the metabolism of desipramine, reducing its effectiveness.

Selective Sertonin Re-uptake Inhibitors (SSRIs)

Fluoxetine is a cocaine antagonist that has been one of the more extensively studied drugs because of its role in serotonergic regulation (McCance, 1997). Several studies indicate fluoxetine's utility in reducing the effects of cocaine. Some studies have found that fluoxetine is effective in retaining people in treatment, (Batki et al., 1994; Batki et al., 1996; Washburn et al., 1994). The study by Washburn et al. (1994) also found longer periods of abstinence for subjects using fluoxetine versus a placebo, and Batki et al. (1994) found reduced cocaine use and craving in methadone-maintained cocaine-dependent individuals, but not those with a primary cocaine dependence. However, other studies have found fluoxetine to be less effective in treatment retention than desipramine or amantadine (Oliveto et al., 1995). A study by Covi et al. (1995) found that fluoxetine could be a detriment when used in conjunction with interpersonal counselling. The study found that fluoxetine at 20 mg., 40 mg. or 60 mg. did not add to the improvement produced by counselling, and the 60 mg. doses may have interfered with the effects of counselling.

5.1.2 Dopamine Regulators

There have been very few studies using antidepressants other than the tricyclics or SSRIs. However, some of the "second-generation antidepressants" such as bupropion, which inhibit norepinepherine and dopamine, may be effective in reducing cocaine craving.

The addictive and euphorogenic effects of cocaine result primarily from inhibition of dopamine re-uptake (Rothman, 1990). This inhibition of uptake leads to a rapid accumulation in the synapse resulting in activation of dopamine receptors. Spealman et al. (1992), in their review of non-human research related to cocaine and the dopamine system, have shown that dopamine agonists produce cocaine-like effects in animals and that dopamine antagonists reduce the response to self-injected cocaine.

Fortunately, some potent dopamine re-uptake blockers have not been reported to produce euphoria or addiction in humans. Based on these observations, dopamine re-uptake inhibitors have been classified into two groups: type 1, which produce euphoria and type 2, which do not (Rothman, 1990). Given that the two types act at the same site (dopamine transport), it has been hypothesized that type 2 blockers may be useful in reducing the euphoric effects of cocaine and, as a result, may decrease cocaine usage. This has lead to the evaluation of several dopamine-regulating drugs in the treatment of cocaine usage.

It should be noted that there is some evidence that the effects of dopamine agonists vary depending on the phase of the cocaine-abuse cycle. Phases of the cocaine-abuse cycle include euphoria, crash and craving. Hollander et al. (1990) treated cocaine addicts during different phases of the cycle with apomorphine, a dopamine agonist. They found that subjects reported less craving during the craving phase than in the crash phase.

Table 2 presents six studies that evaluated medications that operate primarily on the dopamine neurotransmitter system. Amantadine, an indirect dopamine agonist, has been evaluated in a series of randomized, double-blind studies. The results suggest that amantadine might be useful, especially in the early stages of treatment. Alterman et al. (1992) compared amantadine with a placebo. The subjects received medications for a 10-day period. At the end of the drug trial, subjects receiving amantadine were significantly more likely to be free of cocaine than were those on the placebo. Similar findings were obtained at a one-month follow-up. A more recent study by Handelsman et al. (1995), which evaluated amantadine and a placebo over a longer period of time, did not find differences in cocaine consumption or craving between the two groups. Two possible reasons for the differences in these two studies may be related to subject characteristics and/or longevity of amanadine's effect. For example, Alterman et al. (1992) specifically excluded subjects who were addicted to other substances, whereas the subjects in the Handelsman et al. (1995) study were methadone-maintenance patients. Alternatively, in a brief review of the efficacy of amantadine for treating cocaine withdrawal, Thompson (1992) speculated that "amantadine's effectiveness in maintaining short-term abstinence may decrease with time" (p. 934).

Table 2 Comparison of Dopamine Regulators and Placebo

Study	Medicine	Subj- ects	Ret- en. 1	Red. Coc. 2	Red. Urg-es 3	Med/ Psy-ch Imp-rov. 4	Comments
Eiler et al., 1995	Bromocriptine	Cocaine	-			-	Very high dropout rates.
Stine et al., 1995	Mazindol	Cocaine	-	-	-	-	All participants received group therapy.
Hande- lsman et al., 1995	Amantadine	Cocaine/ Opioid	?	-	-	-	Participants with high SCL-90 scores improved more on medication.
Alterman et al., 1992	Amantadine	Cocaine	-		-	-	Differenes in positive urine samples developed over two weeks.
Giannine et al., 1993	Buspirone	Cocaine		?			Buspirone is a nonbenzodiazepine tranquilzer affecting the dopamine system. The improvement was for withdrawal symptoms.
Note: 1. Increased Retention in Program 2. Reduced Cocaine Usage 3. Reduced Cocaine Craving 4. Medical/Psychological Improvement = treatment group improved compared to control group - = no differences between groups ? = could not determine if differences occurred

Bromocriptine is another dopamine agonist that acts through stimulation of postsynaptic dopamine receptors, and may reduce craving and withdrawal by reversing the depletion of dopamine resulting from cocaine use. Eiler et al.(1995) compared bromocriptine to a placebo with patients that were only dependent on cocaine. Their results showed that the two treatments did not differ except for a possible advantage for bromocriptine during the first three weeks.

Mazindol, is a dopamine re-uptake inhibitor (cocaine antagonist). One study (Stine et al., 1995) compared mazindol to a placebo over a six-week period . The results showed that the two groups did not differ on any of the primary measures.

5.1.3 Anticonvulsants

Animal research has shown that animals that are given large doses of cocaine experience seizures. With repeated cocaine experiences, there is an increased probability of seizure activity occurring. This has been referred to as "kindling" (Cornish et al., 1995).

Carbamazepine, an anticonvulsant medication, has been found to reduce both seizure activity and cocaine use, in open clinical trials. As a result, several studies have evaluated the use of carbamazepine in cocaine-dependent populations. The major studies using double-blind randomization procedures have not found carbamazepine to be effective in reducing cocaine use or craving, or for retaining patients in treatment programs (Cornish et al., 1995; Kranzler et al., 1995; Montoya et al., 1995). These results are presented in Table 3.

Table 3 Comparison of Anticonvulsants and Placebo

Study	Medicine	Subj- ects	Ret- en. 1	Red. Coc. 2	Red. Urg-es 3	Med/ Psy-ch Imp-rov. 4	Comments
Kranzler et al., 1995	Carbama- zepine	Cocaine Males only	-	-	-	-	The majority of participants smoked cocaine.
Montoya et al., 1995	Carbama- zepine therapy	Cocaine	-	-	-	-	Participants also had cognitive behaviour.
Cornish et al., 1995	Carbama- zepine	Cocaine	-	-	-		Carb. group showed better retention during early phase of study.
Note: 1. Increased Retention in Program 2. Reduced Cocaine Usage 3. Reduced Cocaine Craving 4. Medical/Psychological Improvement = treatment group improved compared to control group - = no differences between groups ? = could not determine if differences occurred

5.1.4 Buprenorphine

Because of the co-morbidity of opioid and cocaine use, several studies have evaluated the relative efficacy of methadone and buprenorphine for retaining subjects in treatment programs, and in reducing opioid and cocaine use. Research has shown that buprenorphine attenuates the effects of cocaine on adrenocorticotropin (ACTH) in cocaine-dependent men (Mendelson et al., 1992). Plasma levels of ACTH parallel plasma cocaine levels and self-reported mood states. Strain et al. (1994) compared methadone to buprenorphine for reducing cocaine use in patients that were addicted to both cocaine and heroin. Their results showed that the two drugs were equally effective in retaining subjects and in reducing cocaine-positive urines. Both groups showed about a one-third reduction in cocaine-positive urine samples. Finally, in a very recent report, Eissenberg et al. (1997) have shown that daily injections of buprenorphine are not necessary to maintain cocaine abstinence. These results are shown in Table 4.

Table 4 Comparison of Buprenorphine and Methadone

Study	Medicine	Subj- ects	Ret- en. 1	Red. Coc. 2	Red. Urg-es 3	Med/ Psy-ch Imp-rov. 4	Comments
Strain et al., 1994	Buprenorphine  Methadone	Opioid/Coc.	-	-			No differences between drugs for cocaine measure.
Schottenfeld et al., 1997	Buprenorphine  Methadone	Opioid/Coc.		-	-	-	The larger the dose of both drugs the greater the retention.
Note: 1. Increased Retention in Program 2. Reduced Cocaine Usage 3. Reduced Cocaine Craving 4. Medical/Psychological Improvement = treatment group improved compared to control group - = no differences between groups ? = could not determine if differences occurred

5.1.5 Disulfiram

McCance (1997) notes that many cocaine abusers are also dependent on alcohol, and that alcohol may precipitate cocaine use, because it is used to enhance the euphoric effects of cocaine and to alleviate some of the dysphoric effects. Although McCance (1997) reports that a number of open trials with disulfiram resulted in decreases in both cocaine and alcohol use, its efficacy needs to be confirmed in large, well-controlled trials (McCance, 1997). However, a recent randomized clinical trial (Carroll et al., 1998) concluded that disulfiram in conjunction with outpatient psychotherapy (cognitive behavioral therapy (CBT) or 12-step facilitation (TSF)) was effective in reducing both alcohol and cocaine use, and retaining clients in treatment.

5.16 Emerging Pharmacotherapies

There are several new drugs that show potential for treating cocaine addiction. The most promising of these compounds is GBR 12909. Studies have shown that both cocaine and GBR 12909 inhibit the activation of a protein called the dopamine transporter, increasing the levels of dopamine outside the nerve cells. This prolongs dopamine's pleasurable effects. GBR 12909 produces a much smaller dopamine "spike" but maintains levels for a longer period of time. In studies using monkeys, it was found that the injection of GBR 12909 greatly reduced cocaine self-administration (Stocker, 1997).

Best Practice Guideline #1: The literature does not yet provide sufficient evidence for the efficacy of specific drugs in the treatment of cocaine dependence. However, several antidepressant drugs have shown promise in retaining users in the initial stages of treatment, particularly depressed patients and those who "snort' cocaine. There appears to be some evidence that drugs used in the treatment of opiate or alcohol dependence may be useful in reducing cocaine use in patients addicted to cocaine and heroin or cocaine and alcohol.

5.2 Behavioural Treatments

As with treatment for other substance-abuse disorders, behavioural treatments have been found to be an effective approach to the treatment of cocaine dependency. A number of behavioural interventions have been found to be particularly effective: Contingency Management, Cognitive Behavioural Therapy (CBT) and broad-based behavioural therapy.

5.2.1. Contingency Management

To date, one of the most effective approaches to treating cocaine dependency has been developed by Stephen Higgins and his colleagues (Higgins et al., 1993; Higgins et al., 1994; Higgins et al., 1995) combining a community reinforcement approach (CRA) with a contingency management component (vouchers). The base CRA program is designed to enhance the client's family relations, and vocational, recreational and social activities, and is provided over a six-month period. Typically, it involves a functional analysis of the client's substance use to identify antecedents and consequences of substance use, so that strategies can be developed to avoid high-risk antecedents and to put in place alternatives to the consequences of cocaine use. As well, the base program includes social, recreational and employment counselling, drug-refusal training, social-skills training and reciprocal relationship counselling.

To this base, a voucher system has been added. Clients receive vouchers (that can be exchanged for various merchandise) when they remain drug-free and in treatment. The vouchers increase in value for cocaine-free urine samples. It should be noted that the therapists for this program were very experienced doctoral-level students. The three studies by Higgins and his colleagues reported

in Table 5, all show that the contingent use of vouchers when combined with other cognitive behavioural interventions, markedly reduces cocaine use and increases the psychological well-being of the participants. In addition, these studies show that the use of these procedures, especially the contingent vouchers, increases retention over alternative treatment conditions, including CRA alone. The literature consistently suggests that the longer people remain in a treatment program the greater is the likelihood that they will decrease cocaine usage. In a recent review of the literature, Higgins (1996) reports on 13 studies using contingency management to reduce cocaine usage. In just over one half of the studies, the participants were on methadone maintenance for opioid addiction in addition to their cocaine dependency. He reports that, in 11 of the 13 studies, there was a reduction in cocaine usage. These studies have been carried out both in inner-city and rural locations.

In some studies, Higgins and his colleagues (Higgins et al., 1994a) incorporated family members or friends into the contingency management component of their behavioural treatment program. Family members were informed of the results of urinalysis, and they provided social reinforcement for a negative-urine screen. Although preliminary results indicated that this was an effective intervention, randomized trials did not support its efficacy (NIDA, 1998).

The National Institute on Drug Abuse in the United States has recently produced a manual incorporating a contingency management approach, which is available in hard copy and on-line: A Community Reinforcement Plus Vouchers Approach: Treating Cocaine Addiction (1998).

Table 5 Behavioural Treatment

Study	Medicine	Subj- ects	Ret- en. 1	Red. Coc. 2	Red. Urg-es 3	Med/ Psy-ch Imp-rov. 4	Comments
Higgins et al., 1993	Contingency Man.	Cocaine					Incentives were contingent on urine free samples.
Higgins et al., 1994	Contingency Man.	Cocaine					Both groups in behavioural program; one group received vouchers.
Higgins et al., 1995	Contingency  Man.	Cocaine					One year follow-up of above participants.
Carroll et al., 1991	CBT	Cocaine					More severely dependent did better in CBT.
Carroll et al., 1994 (a & b)	CBT	Cocaine					More severely dependent did better in CBT.
Carroll et al., 1998	CBT & TSF	Cocaine &Alcohol					CBT and TSF more effective than clinical management.
Wells et al.,	CBT and 12 Step	Cocaine		-			CBT and 12 step equally effective in reducing drug use.
Azrin et al., 1994	Behaviour Mod.	Cocaine					Youth showed better results compared to adults.
Note: 1. Increased Retention in Program 2. Reduced Cocaine Usage 3. Reduced Cocaine Craving 4. Medical/Psychological Improvement = treatment group improved compared to control group - = no differences between groups ? = could not determine if differences occurred

5.2.2. Cognitive Behavioural Therapy

Another approach that has shown strong evidence of effectiveness for the treatment of substance abuse in general, but also with clients who are cocaine-dependent is Cognitive Behavioural Therapy (CBT). This approach is a short-term (usually 12-16 sessions over 12 weeks) focussed intervention involving somewhat similar components to CRA. As with CRA, it includes a functional analysis of antecedents and consequences to develop strategies to avoid high-risk situations and to identify alternatives to cocaine's reinforcing effects. It also includes a heavy emphasis on the development of coping skills. It may also be delivered as part of a broader range of interventions that could include pharmacotherapy, counselling for adjunctive areas such as family counselling or vocational counselling, or attending a mutual-aid group. Unlike CRA, it does not usually include a voucher component or intervention with the client outside the treatment setting.

Studies of the effectiveness of CBT with cocaine dependent clients have been carried out by Carroll and her colleagues (Carroll, 1991, Carroll et al., 1994a; 1994b; Carroll et al., 1996) and compare CBT to other interventions for clients with different profiles. As Table 5 indicates, CBT (Relapse Prevention) has been found to be more effective in terms of both retention in treatment and reduction in cocaine use when compared to Interpersonal Psychotherapy (Carroll et al., 1991) and to Clinical Management (CM) (Carroll et al., 1994 a, 1994b). In both these studies, there was an interaction effect. Subjects who were more severely dependent on cocaine did better in the CBT condition, while there were no differences in outcome for those less severely dependent. In the CBT/CM study, at the one year follow-up, continued gains in reducing cocaine use were found in the CBT group, but not the CM group (Carroll et al., 1994b). These studies also found that CBT was more effective than CM in retaining depressed subjects in treatment and was somewhat more effective in reducing cocaine use (Carroll et al., 1994a).

One interesting finding from the series of studies by Carroll and colleagues and also by Wells et al. (1994) is that other "active therapies" using a different theoretical approach may be as effective as cognitive behavioural therapy. Carroll (1998) evaluated the efficacy of CBT in comparison to CM and TSF (12- Steps Facilitation) in a group of subjects that met criteria for both cocaine and alcohol dependence. This study also included the use of disulfiram. The study found that both CBT and TSF were more effective than CM in retaining clients in treatment and reducing cocaine use (as well as alcohol use). Wells et al. (1994), compared skills-training and relapse-prevention techniques based on Marlatt and Gordon (1995) to a recovery support group based on the Twelve Steps of AA, and recovery support group as described in a study by Well et al.(1994). Subjects in both treatment conditions reduced their use of cocaine and other substances (alcohol and marihuana), and there were no significant differences in cocaine outcomes for the two interventions.

It may be, as Carroll et al. (1998) point out, that the more active and directive therapeutic approaches, which also require clients to carry out assignments outside of scheduled sessions, are more powerful than the less demanding and less directive clinical management approach, particularly for more severely dependent clients.

CBT has also been published in manual form by the National Institute on Drug Abuse: A Cognitive Behavioural Approach: Treating Cocaine Addiction (1998).

5.2.3. Broad-based Behavioural Therapy

Azrin and his colleagues (1994) have reported similar success using a broad-based behaviour-therapy program. In this study, the three primary treatment procedures were a) stimulus control/competing response training, b) urge control and c) social control/contracting. The control condition produced minimal change in drug use, whereas over 60% of the patients in the behavioural group discontinued drug use. In addition, they were more improved on measures related to work, school and alcohol use.

Best Practice Guideline #2: The literature shows good evidence that behavioural treatment procedures (particularly contingency management and cognitive behavioural therapy) are effective in reducing cocaine use and retaining clients in treatment. Further, other active, directive therapeutic approaches using different theoretical approaches may be as effective as CBT.

5.2.4 Treatment for Women and Pregnant Women

The recent Health Canada report: Best Practices - Substance Abuse Treatment and Rehabilitation (Health Canada, 1999) concludes that there is insufficient evidence to support the provision of specific types of interventions for women. However, it also notes that it is important to consider the barriers to treatment and to provide a range of modifications and support services.

Hughes et al. (1994) examined treatment retention among women who were in an 18 month therapeutic community program and were randomly assigned to two conditions. Although both groups received the same therapeutic community program, the experimental group was permitted to bring their children to live with them during treatment. Preliminary results from this study indicate significantly longer retention in treatment for the experimental group in comparison with the control group.

Pregnant women who use cocaine have benefited from involvement in a therapeutic community. Using a rigorous outcome measure of cocaine-positive urinalysis, Egelko and her colleagues (1996) found that 87.2% of the final three urine samples (prior to discharge) were negative in a perinatal cohort attending a modified day-program therapeutic community. Female participants averaged two months in treatment. Those women who began the program while pregnant did substantially better than those who joined the program postpartum, illustrating the need for early intervention.

Studies have also shown that intervention programs directed at improving parenting skills can have positive effects on infant development, and can help "make up ground" for children born to cocaine-using mothers (Kane et al., 1997; Howard et al., 1995; Zuckerman and Frank, 1994).