People with severe mental disorders tend to present to psychiatric services whilst those with less severe but more common disorders are more likely to be found in substance-abuse treatment services (Kessler et al., 1996; Primm et al., 2000). Currently the needs of comorbid clients are not being met by either of these services. In most western health care systems there is an artificial separation of treatments for substance abuse and mental health disorders. There is pressure to place patients in one system or the other, by determining which disorder is primary for them. This can result in no treatment for the disorders not considered primary. Distribution of funding from the government level can ensure that the separation is jealously guarded by either service, the outcome being poor or no services for people with comorbid disorders (Kessler et al., 1996).

A first step towards appropriate service provision for persons with comorbid disorders is to ensure that, wherever they are present, be it in primary care, substance abuse services or psychiatric services, careful assessment of their presenting conditions is carried out.

Assessment
Treatment
Treatment for psychotic disorders and substance abuse

Assessment

Well-documented deficiencies in assessment by treatment services are compounded in the treatment of those with comorbidity. Some common difficulties are that clinicians may fail to obtain a full history of substance use in people with a mental illness. Alternatively, people with a mental illness may deny, distort, or minimise their self-reported use of substances, particularly illicit drug use (Bryant, Rounsaville, Spitzer, & Williams, 1992; Drake & Mercer-McFadden, 1995; Mueser et al., 1997). So, despite high rates of substance misuse amongst those with mental disorders, it is under-reported in this population.

In substance treatment services the clinical picture is often unclear, because many patients seek help in a distressed condition and complain of a multiplicity of psychological symptoms. Reviewers have described the commonest of these symptoms which include anxiety, irritability, and feelings of sadness as transient, disappearing within seven days of abstinence both in young, healthy problem drinkers and in primary alcoholics (Schuckit & Monteiro, 1988). Thus assessment within this time period may produce spuriously elevated scores.

In their broad-ranging review of comorbidity of anxiety and depression with substance abuse, Scott and co-workers (1998) argue that there is a need for increased awareness by GPs as well as psychiatric and addiction service staff of the likely presence of comorbidity. They propose mandatory use of brief screening instruments for drug and alcohol abuse and for anxiety and depression as well as probes regarding self-harm. This will require increased training of staff in order that they can detect and treat comorbid disorders. In fact there is a need to incorporate into all services that deal with comorbid individuals effective assessment tools and procedures which take into account the special needs of these patients. As discussed in the following section, at the level of primary care, simple screening tools are available to assess both mental disorders and substance use disorders. Their use has been the exception rather than the rule in Australia and elsewhere (Drake, Rosenberg, & Mueser, 1996; Hickie, Koschera, Davenport, Naismith, & Scott, 2001).Top of page

The role of primary care

A major problem with service delivery for comorbid disorders is that most people do not seek help. In the recent Australian NSMHWB, only one-third of people with a mental disorder consulted any health care provider (Teesson, Hall, Lynskey, & Degenhardt, 2000). However, most people do see a GP for any disorder and this could provide an opportunity for moving comorbid patients towards treatment. However, as demonstrated below, currently GPs resist screening for either the common or psychotic disorders or drug use disorders.

In a series of articles appearing in the Medical Journal of Australia, Hickie and co-workers present findings from a survey of general practices in Australia involving 386 GPs who screened 46,515 patients in 1998 and 1999. In one of these articles they addressed the issue of comorbidity (Hickie, Koschera et al., 2001) and, of particular relevance, the likelihood that comorbidity will be identified in general practice. They found possible comorbidity in 12% of patients attending the surgeries, although diagnoses were based on a simplified classification system (ie., not DSM-IV or ICD-10), (Hickie, Davenport, Naismith, & Scott, 2001). Those classified as having comorbid mental and substance use disorders by this system were also found to be more likely to be assessed by GPs as having a psychological diagnosis and having greater health risk, and GPs were more likely to treat them or refer them on to a mental health service. However, only half of this group was actually diagnosed with any psychological disorder which reflects a missed opportunity for treatment for this vulnerable group. Although not a random sample of the population of practices and patients, the outcomes are likely to be indicative of the trend in GP practices across Australia.

Hickie argues that those with substance use disorders are more readily identified by GPs and GPs should be encouraged to proceed from such diagnoses to screen for comorbid psychiatric conditions. However, he points to the problem that GPs may be unable to treat all patients identified by screening because high prevalence rates would present too great a workload.

These sorts of concerns have also been addressed by Andrews (2001) who has argued that a strategy similar to that in the breast cancer field may be appropriate. This involves use of four concurrent strategies: identifying risk factors, using targeted population screening, producing widespread public understanding, and profession-wide acceptance of management guidelines. Such an approach should result in the number consulting being reduced by prevention and self-help strategies, and the number becoming chronic and needing continual help being reduced by effective treatment — so that overall the number needing to consult would be reduced. These suggestions tie in well with the discussion of prevention above.

It has also been argued that the workload of GPs may actually be reduced if patients with mental health problems were identified and treated; as this group tends to significantly overuse primary care services for physical health problems. Bebbington and colleagues (2000) reported that, in a recent household survey in the UK, patients identified as having a neurotic disorder were 40% more likely to consult a GP for any physical disorder than those with no neurotic disorder.

Assessment in substance abuse services

Where assessment does take place, there exists the likelihood of misdiagnosis because of common symptoms amongst the various disorders. Under-diagnosis can result by assigning symptoms to one disorder to the exclusion of the other, whilst over-diagnosis can occur when symptoms are assigned to a second disorder when they can be fully accounted for by a single disorder. For example, many more people with substance use disorders present with depressive and anxiety symptoms than would be given a specific diagnosis of anxiety or depression. These symptoms are due to over-use and withdrawal from drugs and alcohol. Thus, further probing is necessary once the possibility of a particular diagnosis is identified through screening. This is a specialised procedure requiring trained treatment staff and should be regarded as the starting point for the positive therapeutic relationship needed for successful treatment (Drake, Rosenberg et al., 1996).

Where true comorbidity exists, it is important that it is recognised and treated appropriately. This is demonstrated in the study by Westermeyer et al, (1998) described above, where the presence of dysthymia correlated with much higher substance abuse service usage. Although it is difficult to identify dysthymia in those presenting at substance use disorder facilities, it is still feasible and supported by research evidence to date, to treat depressive symptoms at the same time as treating the substance abuse. It is less clear whether comorbid anxiety disorders should be treated at the same time as treatment for substance use disorders.Top of page

Assessment in psychiatric services

A study by Drake et al. (1990) on alcohol use in schizophrenia indicated that, as a group, people with schizophrenia were particularly vulnerable to the psychiatric and social complications of drinking. The authors suggest almost any alcohol consumption at all by people with schizophrenia should be identified as problem drinking. Consequently, applying standard definitions and diagnostic criteria in assessing those with psychotic illness may substantially underestimate the problem (Smith & Hucker, 1994).

The identification of substance abuse in the psychiatric services for those with psychotic disorders has been the subject of recent reviews (Carey & Correia, 1998; Drake, Rosenberg et al., 1996). These reviews focus on psychiatric services, as it is rare that patients with psychotic disorders are accepted into treatment in substance abuse treatment facilities (Primm et al., 2000). They highlight the fact that identification of comorbidity is made more difficult for those with psychotic disorders because many of the signs and symptoms of severe drug and alcohol abuse may be masked by symptoms of psychotic disorders, e.g., social isolation or dysfunction and cognitive dysfunction such as confusion, depression, anxiety and positive psychotic symptoms due to substance abuse (Carey & Correia, 1998). Non-detection of concomitant substance abuse can lead to inappropriate treatments such as over-medication, and subsequent poor outcomes (Carey & Correia, 1998; Drake, Rosenberg et al., 1996).

As pointed out by Drake (1996), the assessment serves to inform and involve the patient in the treatment process. Diagnosis of comorbid disorders is but one aspect of the assessment process; but nevertheless essential if treatment is to proceed effectively. Assessment should reveal the severity of substance-related problems where they exist, the patient's motivation to be involved in treatment for such problems, identification of the psychosocial variables encouraging ongoing use and explication of where best to direct treatments.

Detection of drug and alcohol use can be facilitated through use of screening tools such as urine analysis, self and collateral reports and expert detection of the biological signs and symptoms of substance abuse. This would identify those who need to be subjected to more thorough-going diagnosis. When reviewed by Carey and Correia (1998), self-report measures such as the DAST, MAST and CAGE were found to be reliable in this population, although screeners which take into account the presence of a psychotic disorder are preferable. Carey and Correia refer to the DALI which was developed specially for this group and has shown promise. Also, where self-report is considered less reliable, such as during an acute psychotic episode, clinician rating scales have been developed which have been found to provide reliable information regarding substance use disorders of patients (Alcohol Use Scale — AUS and Drug Use Scale — DUS, (Drake, Rosenberg et al., 1996)).

Just as with the common mental disorders, an important factor affecting the non-detection of comorbid substance abuse in those with severe disorders is the lack of training of staff in the specialist mental health services so that they can identify comorbid substance use (Carey & Correia, 1998). Fowler et al., (1998, p 450), in a study of substance abuse by people with schizophrenia in Newcastle, Australia commented that:

"...although there was reasonable agreement between case managers' assessments and the research diagnoses, this did not reach the levels found in other studies (Drake et al., 1990; Carey et al., 1996), possibly because in the current study the case managers were not trained. Thus, efforts to train case managers and to heighten their awareness of substance use problems in their schizophrenic patients may be timely."
Thus it is of paramount importance, before the process of treating comorbid patients can begin, to identify the presence of comorbid disorders in those presenting to the specialist services. Appropriate procedures include use of valid and reliable screening instruments as well as training of staff to be able to identify likely comorbid disorders.

Treatment

Common mental disorders comorbid with substance use disorders

Top of page

(a) Anxiety

There is a common belief that the high rate of comorbidity between anxiety disorders and alcohol use is because the alcohol is used to reduce stress (the stress reduction hypothesis). But as Allan (1995) argues, evidence does not support the stress reduction hypothesis that alcohol users become more anxious with extended use; and that drinkers with many problems are realistically anxious. Thus in most patients, anxiety disappears when the alcohol use disappears. A small proportion of patients, perhaps as low as 10% (Brown, Irwin, & Schuckit, 1991), are then left with more persistent symptoms which may be an independent clinical disorder.

Recent reviews have argued that psychological interventions such as cognitive-behavioural therapy (CBT) are preferable to medication for comorbid anxiety and substance use disorders because of the potential misuse of anxiolytic medication, especially benzodiazepines (Allan, 1995; Scott et al., 1998). However a behavioural approach depends on the use of exposure as the main agent of therapeutic change, and, from a theoretical point of view, the regular use of a central nervous system suppressant such as alcohol would potentially reduce the effectiveness of this process. In fact, there is evidence that the use of alcohol acts to retard the process of desensitisation among clinically anxious patients (Cameron, Liepman, Curtis, & Thyer, 1987; Thyer & Curtis, 1984). So treatment for an anxiety disorder is likely to be ineffective if the person does not stop drinking.

This suggests that an attempt should be made to reduce or stop drinking before commencing treatment for anxiety and that simply integrating treatment might not be the best solution. This is illustrated in a recent study by Randall et al, (2001) who conducted a randomised controlled trial comparing CBT for alcohol alone with CBT for both alcohol and social phobia. While both groups improved on alcohol and social phobia measures after treatment, the group treated for both alcohol and social phobia had worse outcomes on three of the four alcohol use indices.

A further problem arises in that patients in psychiatric services tend to have a preference for having their anxiety and depression treated and are generally not interested in tackling their drug and alcohol use problems (Allan, 1995; Scott et al., 1998). This poses a difficulty for the therapist where it is clear that treatment of a substance use disorder may be sufficient to remove anxiety and depression symptoms. However such problems can be overcome with the establishment of a positive therapeutic relationship and the use of psychotherapeutic techniques such as motivational enhancement which have been found to be effective in moving patients towards recognition of the need for substance abuse treatment (Scott et al., 1998 and Kavanagh et al., Chapter 5).

(b) Depression

The confounding effects of multiple common symptoms of comorbid disorders makes decisions about treatment difficult. However, there is evidence that the comorbid disorders exacerbate each other, e.g., depression increases substance use, harm and poor treatment compliance. So the need to identify and treat a 'primary' disorder may be less important than removing the exacerbating effects of either disorder. As with anxiety disorders, many depressive symptoms are removed by abstinence from alcohol and other drugs (Grant, 1996; Hasin et al., 1996). Again, appropriate motivational counselling techniques can be used to encourage patients to have treatment for their substance use disorders (Lynskey, 1998; Scott et al., 1998).

Although it is unclear how severely the presence of depression affects outcomes in substance abuse treatment, there is a growing body of evidence which indicates that treating depression in comorbid individuals will improve outcomes for both disorders (Carroll, Nich, & Rounsaville, 1995; Lynskey, 1998). It should be noted that this is contrary to the evidence to date for treating comorbid anxiety and substance use disorders (see section on Anxiety above). In his review, Lynskey (1998) argues that the advent of SSRIs makes medical interventions for comorbid individuals more practicable. These tend to have fewer side-effects and are less toxic than the older tricyclic antidepressants. Research has found that use of the SSRIs reduces both depressive and alcohol dependence symptoms in those who are comorbid. (On the contrary, they appear not to be effective for those with alcohol dependence alone, i.e., where there are no depressive symptoms.) However, Lynskey warns that treatment with antidepressants should be accompanied by appropriate psychosocial support and that more research is needed to determine the safety of these medications when patients keep drinking. He points to the promising parallel developments in the use of cognitive-behavioural interventions for depression in alcohol dependent individuals and concludes that inclusion of efficacious treatments for depression can significantly improve the outcomes for both disorders.

Thus there are strong arguments for introducing drug abuse treatment into the treatment programs for those suffering from comorbid affective and substance use disorders in psychiatric services, and for introducing treatments for depression for those with comorbid disorders in substance abuse treatment services. In their informative review and discussion of the management of comorbidity, Scott et al., (Scott et al., 1998) conclude that staff in addiction treatment need to appropriate and implement the evidence-based skills used to treat psychiatric disorders, whilst those in psychiatric services should extend their use of such procedures to treat comorbid substance use disorders. However, in concert with most reviewers, they point to the dearth of research on treatment for comorbidity. Much research carried out to date tends to exclude people with comorbid disorders so that little is known about their specific requirements.Top of page

(c) Conclusions regarding treatment for those with comorbid substance use disorders and the common mental disorders

The evidence suggests that treating any comorbid substance use disorder prior to treatment for anxiety or depression is more likely to lead to positive outcomes for anxiety and depression. Motivational enhancement techniques, which have been demonstrated to be effective, are needed to re-orient the patient towards controlling their substance abuse, prior to management of the comorbid disorder.

For those with both an affective disorder and a substance use disorder, there is an additional benefit on outcomes for both disorders, conferred by treatment for the affective disorder alone. This provides good support for the introduction of efficacious treatments for depression into substance abuse treatment facilities where both disorders can be treated at the same time. However, attempts to treat both anxiety and substance abuse contemporaneously have to date, proven counterproductive. Further research is needed on this issue.

Treatment for psychotic disorders and substance abuse

Comorbidity of psychotic disorders and substance abuse is common and has consistently been found to be more prevalent in treatment than non-treatment samples (Helzer & Pryzbeck, 1988; Kessler et al., 1996; Ross, Glaser, & Germanson, 1988). For people with a serious mental illness, the risk of developing a substance use disorder is of particular importance as they are especially vulnerable compared to people with other psychiatric disorders (Mueser et al., 1997).

Models of service provision

Because outcomes for comorbid patients tend to be poor within the systems designed to treat single disorders, there is increasing literature discussing possible interaction between the services (Kessler et al., 1996; Ries, 1993). Three models of treatment for those with serious mental illness and substance abuse have been widely discussed in the literature, and these are described below.

Serial treatment involves treating one disorder before treating the other. The tendency has been to treat acute presentations as primary and then refer to the alternative treatment system for treatment of the other disorder. Thus acute presentations of psychotic disorders tend to be treated before referral for treatment for co-occurring substance use disorders; and severe intoxication is treated before any consideration of co-occurring mental health problems. In non-acute cases, having two independent systems treating serially in this way means that many people with comorbid disorders "fall between the cracks", being treated by neither system as neither sees it as their responsibility (Ries, 1993).

Although there are variations between international systems, there are elements of psychiatric services in most western nations which are common and which contrast with those of addiction treatment services. These include more academically qualified (especially medical) staff, use of diagnostic classification, eg, DSM-IV, and emphasis on medication to treat the core disorder. Substance abuse services differ from psychiatric services in providing more non-professional staff, often themselves with lifetime dependency problems, emphasis on confrontational interventions and self-help through 12-step programs, and an anti-drug preference. Thus the conflicting philosophies of the traditional drug and alcohol treatment services and the mental health services mean that patients receive diverse and incompatible messages from this type of serial treatment provision, with little or no opportunity to reconcile the different messages.

Parallel treatment involves being treated for one disorder at the same time as receiving treatment for another. This is likely to be less confusing for the patient as it requires some direct interaction between the services and allows more opportunity to reconcile the different messages. It also permits a better understanding of both systems by treatment service staff who have to reconcile concurrent treatments to those they are administering — which should cause better integration of treatments (Ries, 1993). An example of parallel treatment is where a patient housed in a psychiatric unit is sent for treatment to a substance abuse facility on a regular basis. This does present the risk of putting considerable stress on the patient who is already in a vulnerable state and consequently may prove counter-productive. The stress may be in the form of upsetting a routine established in in-patient care, forcing them to travel unaccompanied, or merely trying to accommodate a doubled treatment regime.Top of page

Although parallel treatment may be useful for a particular sub-sample of comorbid patients, those with psychotic disorders in particular are unlikely to be satisfactorily treated using this model because of the criticisms listed above.

The treatment response to drug and alcohol and mental disorders in many developed countries has been dominated by parallel systems. That is, drug and alcohol disorders have been treated by one co-ordinated, funded, and planned service whilst mental disorders have been treated in parallel by a separate, unconnected service. A wide range of problems have been noted with using this method to treat comorbid substance use and psychiatric disorders (Bellack & Gearon, 1998; Ridgely, Goldman, & Willenbring, 1990). There is a wealth of evidence documenting the fact that the traditional methods for treating substance use do not work for clients with psychiatric disorders (McLellan, Woody, Luborsky, O'Brien, & Druley, 1983; Rounsaville, Dolinsky, Babor, & Meyer, 1987; Woody, McLellan, & O'Brien, 1990). It is likely that this lack of success has resulted partly from the mental health and substance use services offering only separate, parallel treatment programs (Ridgely et al., 1990). However it is also likely due to traditional treatments for substance abuse not being particularly effective in themselves (Proudfoot & Teesson, 2000).

Integrated treatment has been proposed as the likely solution to some of the problems presented by the older models of parallel and serial treatment. Integrated treatment in various forms has been the subject of study and review and the definition of such treatment has been refined over time (Bellack & Gearon, 1998; Carey, 1996; Drake, Bartels, Teague, Noordsby, & Clark, 1993; Minkoff, 1989). A range of integrated treatment models has been developed which abide by the following principles (Mueser et al., 1997):
  1. The same individual, team, or service, provides both mental health and substance abuse treatments simultaneously.

  2. Behavioural strategies are utilised to help clients resist social pressures and urges to use substances.

  3. Close involvement is maintained with the patient's family.

  4. Treatment is approached in stages to ensure optimal timing of clinical interventions.

Research findings

Much of the research has been carried out in the United States where substance abuse treatment programs tend to be rigidly abstinence-oriented and there is considerable emphasis on AA-oriented self-help groups. The authoritarian and often anti-medication stance of such programs tend to clash with the regimens in place in the mental health areas dedicated to treating severe psychotic illness. As a consequence, maintaining these separate services to treat both illnesses is considered to be particularly counter-productive for comorbid patients (Bellack & Gearon, 1998).

Various research institutions and hospitals have proposed and instituted integrated models of treatment for this particularly disabled group. Research on their effectiveness is now becoming available with the completion of several randomised controlled trials. However, as the discussion below indicates, much more needs to be done to clarify best practice in service provision. In particular, broad-based implementation of 'ideal' but costly integrated programs where there is a high level of staff training and involvement and high staff-patient ratios, may not be justified if benefits are only minimal or limited to sub-groups of the patient population (Hall & Farrell, 1997).

In their review of integrated treatments, Drake et al., (1998) summarised the historical development of integrated approaches from simply adding an outpatient substance abuse treatment group to usual care, to approaches which involved 'multiple interventions daily, for several hours each day, over a period of weeks or months' (Drake et al., 1998 p 593). These could be in an outpatient, in-patient or residential setting. Currently understood best practice for interventions for patients with comorbid substance use and psychotic disorders has been summarised by Drake et al., in their review. This approach includes elements of assertive case management as well as evidence-based interventions for substance abuse treatment. Table 1, reproduced from this review (Drake et a;., 1998, Table 1, p. 591), provides a broad description of their approach. This review also summarised the research on integrated services until 1998 and concluded that although many of the studies have been poorly executed, there is some evidence that their comprehensive integrated outpatient treatment programs are effective. However, the conclusions drawn may be considered somewhat optimistic considering the quality of the studies reviewed and a later review carried out for the Cochrane Collaboration (Ley, Jeffery, McLaren, & Siegfried, 2000).Top of page

Bellack & Gearon (1998) provide a thoughtful discussion of the particular needs of those with schizophrenia and substance use disorders and conclude that there is little firm empirical support for integrated treatment programs to date. The meaning of 'comprehensive, integrated treatment' has varied across studies and Bellack and Gearon ask the important question: Which aspects of the treatments reviewed by Drake et al., really add substance to treatment? For example, they point out that the most important influence on substance abuse, found in the only study that compared specific interventions (Jerrell & Ridgely, 1995), was the behavioural program which was non-specific and not particularly intensive (one session per week).

The study by Ho et al., (1999) also raises questions about exactly which aspects of integrated programs work. This team performed a sequential analysis on consecutive intakes to a treatment facility for those with psychosis and substance use disorders.

The facility practised 'integrated' treatment which was evolving, with the quality and intensity of treatment increasing over the years of the study (1994 to 1996). They found significant improvements over time in engagement and retention rates, hospitalisation rates and level of abstinence from abused substances. There were several factors which the authors identified that may have led to these improvements. These include more case managers, addition of a special substance abuse module in relapse prevention, addition of a community re-entry module, a lunch program and a relaxation group.

Bellack and Gearon (1998) suggest that because of the cognitive deficits commonly associated with schizophrenia, treatments for this group must be designed to minimise demand on cognitive capacity. To this end they point to the promise of contingency management. They also highlight the tendency of this group to be unmotivated to change and agree that a more realistic goal is reduction in drinking rather than abstinence, and that treatment needs to be directed at raising the levels of motivation of this group. One further area that they suggest should be emphasised is social skills to assist them to develop relationships with people who do not abuse drugs.

Improving motivation to change has been investigated for this group. Addington and colleagues (1999) found that persons with schizophrenia in the later stages of Prochaska and DiClemente's (1986) stages of change, had better substance abuse and treatment outcomes. Another study suggests that the effect seems to be on adherence. A brief motivational interview at hospital discharge from an in-patient unit led to enhanced treatment attendance rates during the first three months after hospital discharge, and lower rehospitalisation rates (Swanson, Pantalon, & Cohen, 1999).

The effectiveness of integrated treatment for those with comorbid substance misuse and psychotic disorder has been the topic of an ongoing review on the Cochrane Collaboration database. Included studies had to meet minimum standards of methodological rigour with systematic cross-checking amongst reviewers. The most recent review (Ley et al., 2000) found six studies that met criteria for inclusion but concluded on the basis of various outcome criteria that there was no evidence that programs which incorporated substance abuse treatment were superior to standard psychiatric care provided for the psychotic illness. The studies were not considered of particularly high quality and the reviewer suggested that better research needs to be carried out in this area.

Although the review took some trouble to establish methodological standards, it was not made clear whether efficacious treatments for substance misuse were used (Proudfoot & Teesson, 2000). In fact it appears that no standard or manualised interventions were used across the studies. Thus it is not likely that standard care will be improved upon if interventions are added which have no evidence of effectiveness. We must agree with the reviewers who consider that simple, well-designed controlled trials are feasible and indeed necessary if we are going to progress in treating this severely ill group.

The preceding discussion emphasises the need for clear definition of treatments used and faithful implementation of treatment programs. These are significant issues when it comes to drawing conclusions from the research literature and from reviews of the literature. For example, in a recent Cochrane Collaboration report Marshall et al., (2002) conclude that there is a need to define treatment approaches in a much more rigorous fashion in order that they can be better assessed. In addition, Jerrell and Ridgely (1999) highlight the importance of implementation of treatment programs when comparing outcomes from 'robustly' and 'non-robustly' implemented interventions.

Two further studies serve to demonstrate that clearly specified (manualised) and implemented interventions based on cognitive-behavioural therapy can have positive outcomes for both substance abuse and psychological symptoms. Both could be considered as integrated programs, with one using group treatment for comorbid substance abuse and personality disorder (Fisher & Bentley, 1996), and the other integrated motivational interviewing, CBT and family intervention for comorbid substance abuse and schizophrenia (Barrowclough et al., 2001). Subject numbers in the studies were small, but both found significant improvements in outcomes for those receiving the manualised CBT-based intervention compared with usual care. Although in need of further replication, such studies increase confidence that more reliable estimates of the effectiveness of interventions are possible, and importantly that some integrated treatments for comorbid disorders are effective.
Top of page
Conclusions which can be drawn from the above review of the literature are that:
  1. persons with a dual problem of schizophrenia and substance use disorders are a particularly vulnerable subgroup with complex service needs;

  2. at present comorbid schizophrenia and substance use disorders are less than optimally recognised and managed; and

  3. the evidence for effective treatment options for this group is less than compelling. However, with improved definition and implementation of effective components, integrated treatments warrant further investigation.
Areas of future research have been highlighted by the preceding discussion. These include consideration of the effect on outcome of adding or removing aspects of integrated treatment as well as ensuring fidelity of implementation of interventions. Aspects warranting further research include: assertive community treatment, motivational enhancement, manualised interventions to assist with compliance with treatments and to ensure standard treatments for substance abuse, psychosocial support variables and contingency management.

Table 1: Drake et al., model of integrated treatment for dual disorders

  • The patient participates in one program that provides treatment for two disorders — psychotic disorder and substance use disorder.

  • The patient's mental and substance use disorders are treated by the same clinicians.

  • The clinicians are trained in psychopathology, assessment, and treatment strategies for both mental disorders and for substance use disorders.

  • The clinicians offer substance abuse treatments tailored for patients with severe mental illness. These tailored treatments differ from traditional substance abuse treatment.

    • Focus on preventing increased anxiety rather than on breaking through denial.
    • Emphasis on trust, understanding, and learning rather than on confrontation, criticism, and expression.
    • Emphasis on reduction of harm from substance use rather than on immediate abstinence.
    • Slow pace and long-term perspective rather than rapid withdrawal and short-term treatment.
    • Provision of stage-wise and motivational counselling rather than confrontation and front-loaded treatment.
    • Supportive clinicians readily available in familiar settings rather than being available only during office hours and at clinics.
    • 12-step groups available to those who choose and can benefit rather than being mandated for all patients.
    • Neuroleptics and other pharmacotherapies indicated according to patients' psychiatric and medical needs rather than being contraindicated for all patients in substance abuse treatment.

  • Some program components specifically address substance use reduction as a central focus of programming. Components focus especially on integrated treatment.

    • Substance abuse group intervention.
    • Specialised substance abuse treatment.
    • Case management.
    • Individual counselling.
    • Housing supports.
    • Medications and medication management.
    • Family psychoeducation.
    • Psychosocial rehabilitation.