A recent Cochrane review of psychosocial treatments for co-occurring severe mental illness (predominantly psychotic spectrum disorders) and AOD use concluded that there is no clear evidence supporting the use of one treatment approach over another . As noted by Lubman and colleagues , however, it is difficult to draw any firm conclusions from the current evidence base due to issues with study design (e.g., inconsistent or absent measures of key outcome variables, significant variation within ‘treatment as usual’ control groups). Integrated psychosocial treatments have shown some promise - in particular, programs in which clients receive treatments addressing both disorders, in combination with case management, vocational rehabilitation, family counselling and housing, as well as medications [525- 527].
The majority of studies examining the efficacy of psychological treatments for people with comorbid psychotic spectrum disorders and AOD use have examined MI, either alone or in conjunction with another therapy. Although study findings have been mixed, there is some support for MI in improving AOD use and, when used in conjunction with CBT, improved mental state . One study which added MI, CBT and a family intervention to usual care for clients with schizophrenia comorbid with AOD use found significant improvements in outcomes for both disorders over care as usual . An Australian study which used a 10-session intervention comprising both MI and CBT for this comorbid group also found modest improvements in outcomes . In contrast to these positive findings, two studies have reported opposing findings in regards to MI. The first examined CBT plus MI, and found no significant differences between the treatment and treatment as usual comparison groups on some key outcome measures (e.g., AOD use, positive symptoms of schizophrenia) . In the second study, conducted among young people with psychosis and cannabis use, the use of MI did not lead to improved outcomes compared to treatment as usual, for AOD use or symptoms of psychosis .
Barrowclough and colleagues  suggest that MI techniques may need to be adapted for clients with psychotic disorders because disorganised thoughts and speech may make it difficult for AOD workers to understand what the client is trying to say, and psychotic symptoms (combined with AOD use and heavy medication regimes) may impair clients’ cognitive abilities. For this reason it is recommended that therapists:
- Make use of more frequent and shorter reflections to clarify meaning.
- Use frequent and concise summaries to draw together information.
- Avoid emotionally salient material that is likely to increase thought disorder.
- Provide sufficient time for the client to respond to reflections and summaries.
- Ask simple open questions and avoid multiple choices or complicated language.
Several studies have examined the efficacy of CBT on symptoms of psychosis and AOD use [532, 533]; again evidence regarding the efficacy of CBT in treating co-occurring psychotic disorders and AOD problems is mixed. Naeem and colleagues  found that although CBT led to better outcomes for symptoms of psychopathology, there were no differences between CBT and treatment as usual groups on AOD use outcomes. Similarly, Edwards and colleagues  found no significant differences between the CBT and psychoeducation groups for the key outcomes of cannabis use or psychopathology.
A small number of studies have examined contingency management as a means of treating clients with comorbid psychotic spectrum disorders and AOD use. As discussed in Chapter B5, contingency management involves the use of reinforcement to encourage particular behaviours (and discourage undesired behaviours). In a systematic review of psychosocial interventions for people with comorbid severe mental health (i.e., schizophrenia, schizoaffective disorder, bipolar disorder, or severe depression) and AOD use disorders, Drake and colleagues  found that the use of contingency management led to improved outcomes for AOD use. These findings indicate that contingency management may be a useful adjunct to other treatments for psychotic spectrum disorders and AOD use.
Reviews of the literature have also highlighted that residential, ‘dual diagnosis’ treatment programs may lead to positive outcomes, particularly for people with severe psychosis and AOD use [524, 534]. Long-term residential programs (at least one year) are more likely to be associated with positive outcomes than short- term programs, in terms of increased abstinence from substances, and decreased risk of homelessness .