Background
It is very common for people with alcohol or substance dependence disorders to also have other co-occurring mental illnesses. This is particularly the case for people attending residential alcohol and other substance abuse treatment services, where it is estimated that between 64% and 71% of participants have co-occurring mental illness [
1]. When compared to people with a substance abuse problem unaccompanied by mental illness, people with co-occurring mental illnesses tend to have much poorer treatment outcomes. For example, they are significantly more likely to have a poor treatment response, higher rate of relapse, more hospital visits, increased incidence of Hepatitis C and HIV infection, higher incarceration rate, family difficulties and homelessness [see 2].
Traditionally, alcohol and other substance abuse services have delivered drug and alcohol interventions and relied on mental health services, working in either a parallel or sequential fashion, to address the mental health of their patients [
3,
4]. This type of approach has proven largely ineffective, often resulting in increased fragmentation between the services, higher treatment dropout and exclusion of participants from substance abuse services [
5]. In response, integrated approaches have been developed that concurrently target both the substance abuse and mental illness [see 3 review]. There is evidence that integrated residential treatment facilities are more effective in treating complex mental illness and substance abuse disorders, than less integrated facilities [
2].
Cognitive Behavioural Therapy (CBT) is an approach that lends its self very successfully to being delivered in an integrated fashion. There is strong support for its effectiveness in the treatment of alcohol and substance abuse disorders [
6], as well as across a wide range of psychiatric conditions [
7]. Whilst CBT has demonstrated comparative effectiveness to other therapeutic approaches [
8], it tends to have more durable long-term effects when compared to other treatment conditions [
9]. For example, in a randomized trial comparing integrated CBT with a12-Step facilitation therapy for people diagnosed with substance dependence and depression, both interventions produced similar reductions in substance use and depression at the end of treatment [
10]. However, participants in the integrated CBT condition continued to demonstrate improvements in their depression at 6-months follow-up, while depression for participants in the 12-Step condition deteriorated. More stable reductions in substance use were also demonstrated for people in the integrated CBT condition at follow-up.
Whilst there is increasing empirical support for integrated treatment approaches, the majority of substance abuse services do not address co-occurring mental illness in a systematic way. For example, a review of residential substance abuse services across three Australian states indicated that 88% of services were not equipped to provide integrated mental health treatment [
11]. This appears to be largely the result of organisations employing people with limited or no mental health qualifications and failing to prioritise co-morbidity treatment [
12]. Moreover, high staff turnover rates in the substance abuse field [
13,
14] makes attempts to train staff in new therapeutic approaches, such as integrated CBT, problematic and difficult to sustain in the longer term.
One approach to improve the utilization of evidence-based approaches in health settings is to use computer based interventions [
15]. Benefits of computer based approaches are that they do not require additional staff, do not rely on staff having specialist mental health or substance abuse training, and are relatively cost-efficient to deliver. Emerging research has demonstrated that computer delivered interventions can produce clinically significant improvements across a range of outcome domains [
16]. Computer delivered interventions are increasingly being recommended for use in the substance abuse field [
17‐
21]. Whilst it is still a developing area of clinical research, the use of computer based CBT interventions for drug and alcohol abuse appears to be efficacious [
22]. For example, CBT4CBT is a 6-session, computer-delivered program that is based on cognitive behavioural principles. It is a self-directed program that is used as an adjunct to treatment as usual. Evaluation of the program indicated that it is superior to treatment as usual, both at the end of treatment, and at 8-weeks and 6-month follow-up [
23,
24]. Whilst CBT4CBT appears to be quite a promising intervention, it does not specifically target co-occurring mental illnesses.
An integrated computer delivered CBT intervention for co-occurring depression and alcohol or substance use disorders has been developed. The Self-Help for Alcohol/other drug use and Depression program [SHADE; 25] is a 10-session multimedia psychotherapy treatment program incorporating motivational, behavioural and cognitive components. The SHADE program delivers its therapeutic content via a number of interactive components including video demonstrations, voiceovers, and in session exercises. Results from an initial randomised clinical trial indicated that the SHADE Program outcomes were equivalent when a therapist delivered the same intervention to people attending outpatient substance abuse treatment. It also produced significantly better treatment outcomes than a brief intervention alone [
25]. To date research has not examined the use of SHADE within a residential treatment setting or the extent that it might enhance standard care [
26].
The current project will be conducted at The Salvation Army Sydney Recovery Service Centres located in the Australian states of Sydney and Brisbane. These centres provide a 12-step based residential alcohol and other substance abuse program. The purpose of the current study is to examine the effectiveness of 'adding' the SHADE program to an already established treatment facility. The study will be conducted as a randomized trial, in which participants allocated to the treatment condition will complete the SHADE Program. A typing training program will be used as an active control condition. As mental illness is not specifically targeted as part of The Salvation Army program, and CBT is not systematically available for substance use disorders, it is hypothesized that individuals in the Treatment Condition will report greater improvements in their mental health and greater reductions in their substance use at follow-up, than individuals in the Control Condition. Recruitment for the study is currently underway. The study is funded by a competitive research grant from the Australian Rotary Health Fund. The University of Wollongong Human Research Ethics Committee (HE11/091) has approved the research trial, which is registered with the Australian New Zealand Clinical Trials Registry (ACTRN12611000618954).
Discussion
The present study is examining the effectiveness of 'adding' a computer delivered co-morbidity intervention to residential alcohol and other substance abuse treatment. A large proportion of people accessing residential drug and alcohol treatment screen positive for mental illness, with depression being the most common [
1]. It is expected that long-term mental health outcomes for participants who complete the SHADE program will be significantly better than for those allocated to the control condition. Additionally, as the residential treatment facility is based on a 12-step approach, rather than a CBT approach, it is expected that the CBT based SHADE intervention will also improve alcohol and substance abuse outcomes for participants relative to the control condition.
Strengths and limitations
A significant strength of the current research is that it will be conducted as an effectiveness study. Unlike efficacy studies, where clinical trials are typically conducted in highly controlled research environments, effectiveness studies are conducted in 'real world' treatment settings. The advantage of using this approach is that the results are more representative of 'actual' clinical practice and provide evidence regarding the feasibility of using the intervention as part of ongoing routine care. The research design also includes additional attempts to increase the generalizability of the results by using very inclusive eligibility criteria.
Previous trials of the SHADE program have used therapists to support the delivery of the program [
25,
26]. This has included the delivery of a one-session intervention at the commencement of treatment and 10-minute 'check in' sessions at the conclusion of each computer session. Possible advantage of such an approach are that therapists can check participants understanding of the intervention, confirm homework assignment, and therapists can address any motivational issues. It has been suggested that therapist support is likely to improve client utilisation of computer interventions and subsequently improve client outcomes [
18]. However, a decision has been made in this study to not provide therapist support in the delivery of the program. This is in line with previous computer delivered interventions for substance abuse clients [
44] and likely reflects more closely how interventions would ultimately be implemented within routine care.
A further strength is that the research design includes an active attention control condition. A computer delivered typing tutor will be used as the control, with participants being provided with the rationale that improving their typing skills will help their future employment opportunities. This active control has previously been used in computer based trials in residential alcohol and other substance abuse treatment settings [
31], and addresses limitations with previous clinical trials in which comparisons have only been made to no-treatment conditions [
22].
A significant challenge for this project will be participant dropout due to participants leaving the residential treatment facility prior to completing the intervention. Dropout rates from substance abuse treatments are extremely high. Fifty-seven percent of participants prematurely leave within the first 3-months of treatment [
50]. Similar percentages have also been reported in the broader alcohol and other substance abuse treatment literature [see [
51] for review]. To help address this concern, participants will complete the SHADE or computer delivered typing interventions twice a week (i.e. over a 5-week period). This represents a shift from the protocol used by Kay-Lambkin et al. [
25], where SHADE was delivered weekly for 10-weeks.
A further challenge for the study will be retaining participants at follow-up. People with alcohol and other substance abuse disorders are traditionally very difficult to follow-up. This is further complicated with residential facilities, as participants often move outside of their local area to attend treatment. Attempts to improve follow-up rates in the current study will include using telephone follow-up, obtaining contact details of significant others to help with locating participants, reinforcing to participants the importance of conducting follow-up and financially compensating participants for the time required to complete the assessments (AUD$20).
Competing interests
Kelly and Deane both hold research consultancies with The Salvation Army. The SHADE computer based program has been licensed for use in the United States of America by Cobalt Therapeutics. The authors (FKL, AB) receive no financial benefit as a result of this licensing agreement.
Authors' contributions
All authors have made an intellectual contribution to this research trial. The study chief investigators PJK, FKL, AB, FPD, ACB and GAD were responsible for identifying the research questions, design of the study and overseeing the implementation of the study. Associate investigator AM was responsible for the development of additional research questions, selection of process measures and development of inter-rater reliability procedures. Research assistants SM and MW contributed to the development of support materials, recruitment of participants and study implementation. All authors were responsible for drafting of this manuscript and have read and approved the final version.