Coexisting severe mental illness and substance misuse: community health and social care services

Current practice

Committee members were aware from their experience that people with coexisting severe mental illness and substance misuse may present in crisis (for example, at A&E). But they may also be found opportunistically in other settings (for example, homeless shelters) and identified as needing immediate assistance with a range of needs. This includes their mental or physical health, substance misuse or social care needs.

They noted that the physical health and social care needs of this group are often overlooked because of the challenging nature of dealing with both mental health and substance misuse issues. They also noted that this group is often excluded from services because no one wants to take responsibility for them and they need help to access a wide range of services.

In addition, members noted that a policy guide in 2002 (Department of Health's Dual diagnosis good practice guide) has advised that care for people with coexisting severe mental illness and substance misuse should be delivered within mental health services.

Evidence

The committee noted from the evidence and members' experience that people with coexisting severe mental illness and substance misuse are a vulnerable group, who often have poor physical health, are unemployed, homeless or are at risk of other people taking advantage of them. The latter includes being subjected to sexual exploitation or being taken advantage of in relation to their housing or financial situation.

It noted there was strong evidence from a meta-analysis of 3 cohort and case–control UK studies (2 high quality [++] and 1 low quality [−]) that people with coexisting severe mental illness and substance misuse (compared with those with severe mental illness only) were more likely to have a history of homelessness or housing problems. There was also evidence from 1 high-quality UK case–control study that this group of people are more likely to live in the most deprived areas. There was moderate evidence from 3 high-quality UK cohort studies that showed a greater number of people with coexisting severe mental illness and substance misuse are unemployed than those with severe mental illness only [ES1.1.9].

The committee noted that a meta-analysis of 2 UK case–control studies (1 high and 1 moderate quality [+]) showed no difference in social functioning between this group and people with a severe mental illness only. However, 1 high-quality UK cohort study showed poorer social functioning in people with coexisting severe mental illness and substance misuse than in those with substance misuse [ES1.1.9].

The committee also noted that this evidence was mainly from people in contact with secondary care mental health services and may not reflect the needs of the wider population of people with coexisting severe mental illness and substance misuse [ES1.1.9].

The committee noted inconsistent evidence for educational outcomes [ES1.1.9]. But members also noted from their experience that the point at which a person is diagnosed would have an effect on their educational attainment.

The committee was aware, from the evidence and its experience, that this group is often stigmatised by staff or because of the type of services they are using. For example, this may be a negative attitude towards substance misuse within mental health settings or vice versa. This is based on evidence from 7 qualitative studies (2 high, 3 moderate and 2 low quality) reporting on barriers related to stigma and attitudes towards this group [ES2.1.3].

Six qualitative studies showed that people with coexisting severe mental illness and substance misuse face a number of barriers or facilitators when accessing social care services, particularly housing support. Of the 4 studies that identified barriers to accessing housing support, 1 high-quality qualitative study reported that people with coexisting severe mental illness and substance misuse often feel there is a social stigma associated with seeking help [ES2.2.1]. Also, services are often not easy to access.

The committee felt that it is important for all services to address these issues from an inequalities perspective and to prevent further deterioration in the person's mental and physical health, social care and substance misuse needs. It was also aware from 7 qualitative studies (2 high, 3 moderate and 2 low quality) reporting on fragmented care, that a consequence of fragmented care is a negative impact on a person's experience of care and willingness to engage with services [ES2.2.4].

So it made a strong recommendation that all staff coming into contact with this group should be able to understand their needs and help them access services.

Committee members were aware from their practice and the evidence from 1 high-, 3 moderate- and 2 low-quality qualitative studies (3 set in the UK) that mental health and substance misuse services often fail to take responsibility for people with coexisting severe mental illness and substance misuse [ES2.1.10].

The committee also noted the evidence from 1 low-quality UK qualitative study that highlighted commissioners' views that the health and wellbeing of this group need to be addressed [ES2.1.2]. The committee noted that wherever people with coexisting severe mental illness and substance misuse present, a similar approach to helping them access care is needed.

The committee advised that secondary care mental health services need to be the lead organisation responsible for delivery of services and therefore made a recommendation to refer people with coexisting severe mental illness and substance misuse to secondary care mental health services.

The committee heard from an expert about the physical health issues that can affect people with coexisting severe mental illness and substance misuse [EP4]. It noted that although the expertise was from a perspective of primary care services for homeless people, the range of health needs identified could be transferable to the wider population of people with coexisting severe mental illness and substance misuse. So the committee made a weak recommendation on the range of physical health conditions (for example, cardiovascular, cancer or communicable diseases) that staff need to be aware of. However, it noted that this is not an exhaustive list.

It also reflected on the lack of evidence on the prevalence of coexisting physical health problems [ES1.1.8] and agreed further research is needed (see the recommendation for research on needs assessment).

The committee noted that because of the complexity of their needs, people with coexisting severe mental illness and substance misuse are at increased risk of poor self-care, losing contact with family and friends, social isolation or living in poor housing or having their homes abused by others as venues for substance misuse or drug dealing.

Based on moderate to strong evidence from 4 cohort and 6 case–control studies, committee members were aware of the range of social care needs of people with coexisting severe mental illness and substance misuse in the UK [ES1.1.9]. They were also aware from expert testimony [EP2], and their own experience of working with this group, of the detrimental effects that unmet needs (such as social isolation or poor housing) can have on a person's health and recovery, which could lead to relapse [ES2.2.1, EP2]. This was based on 2 high-quality and 1 moderate-quality qualitative studies reporting on barriers when seeking housing support.

The committee was aware that these unmet needs may lead to physical health problems, offending behaviour or disengagement from services. It was also aware that a person may have issues with both poor housing and physical health and that this may not always be a 'cause–effect' relationship.

Cost effectiveness

There was no evidence for cost effectiveness for this set of recommendations.

Additional factors taken into account

Committee members agreed that recommendation 1.1.1 is for staff working in all general services. But they also noted that it would be applicable to other services, such as criminal justice system and urgent care.

Committee members were aware that the criminal justice system was not included in the scope and that the evidence reviews did not specifically search for studies on the transition between criminal justice systems and healthcare services. They were also aware that NICE is developing guidance on the mental health of adults in contact with the criminal justice system. However, they felt it was important to include because it is a potential route for people with coexisting severe mental illness and substance misuse to come into contact with healthcare services. This was also reflected in the expert testimony on primary care services for homeless people [EP4].

The committee was aware from its experience of the importance of highlighting safeguarding issues for this vulnerable population. It felt that this point needs to be for general services. The committee acknowledged that safeguarding has been made a statutory duty under the Care Act 2014. It was also aware of statutory safeguarding arrangements specific to children (see the Department of Education's guidance on working together to safeguard children) and statutory guidance to the 1989 and 2004 Children Acts (see Ofsted's report What about the children? ). The committee was also aware of the safeguarding needs of dependents and carers.

Current practice

The committee was advised by the topic experts that secondary care mental health services are usually the lead agency that supports people with coexisting severe mental illness and substance misuse.

Evidence

Although this guideline focuses on people with diagnosed coexisting severe mental illness and substance misuse, the committee felt it was important to address the general issue of ensuring people are properly assessed so they can be offered an effective care plan.

The committee noted from 1 moderate-quality study, 1 low-quality UK study and members' experience that timely assessments can help people to access services and stay involved with their care plan [ES2.1.1].

The committee agreed with the recommendations on the principles of recognition and assessment in NICE's guideline on coexisting severe mental illness (psychosis) and substance misuse: assessment and management in healthcare settings, even though it has a narrower focus than this guideline. The committee also agreed that the recommendations on identification and diagnosis were useful (identification and diagnosis was outside the scope of this guideline). Although the psychosis with substance misuse guideline was specific to psychosis and not the range of severe mental illnesses covered in this guideline, members agreed it would be useful for readers to refer to both recommendations.

The committee agreed to develop a recommendation on what needs to happen once a person is referred to and accepted into secondary care mental health services based on the evidence, expert testimony and members' own experience.

Cost effectiveness

There was no evidence for cost effectiveness for this recommendation.

Additional factors taken into account

The committee agreed that substance misuse should not be a reason to exclude people from secondary care mental health services. Based on the evidence and from members' experience this is a common problem [EP2]. The committee also noted from members' experience that the person's wider needs are often not recognised, or they are not given a routine assessment of their mental health or substance misuse needs to develop a care plan.

From their experience, committee members were aware of the importance of a person-centred approach. This was reinforced by review 2. The committee was also aware of NICE's guidelines on coexisting severe mental illness (psychosis) and substance misuse and service user experience in adult mental health. Both outline the need for a non‑judgemental and empathetic approach built on trust and respect. The committee felt it was important to take a person-centred approach when developing and reviewing the care plan and made a strong recommendation on involving people in their care planning. This was based on evidence from:

Current practice

The committee agreed that secondary care mental health services take the lead in coordinating services and developing a care plan. The committee noted that care planning is usually led by a care coordinator because this is part of the Department of Health's Care Programme Approach.

Evidence

The committee was aware of the importance of continuing care. It was also aware that the continuity provided by a key contact encourages people to keep in touch with services (evidence review 2). This was based on the evidence from:

Cost effectiveness

There was no evidence for cost effectiveness for this set of recommendations.

Additional factors taken into account

Based on their expertise and the responsibilities outlined in the Care Programme Approach, committee members made a strong recommendation that a care coordinator from community mental health services is assigned once a person has been referred to secondary care mental health services.

They agreed that the care coordinator should take the lead in developing and reviewing the care plan and should take responsibility for organising delivery of a range of services, with the support of a wider team.

Resource implications and implementation issues

Committee members advised that the role of care coordinator already exists within secondary care mental health services. They noted that care coordinators are part of a multidisciplinary team. But they also noted that overall responsibility (for example, for discharging a person) would lie with a consultant psychiatrist.

Evidence

The committee agreed that it is important to take a person-centred approach, by focusing on actions that are agreed with the person and by offering, not imposing, services on them. So it developed a set of recommendations on 'involving people' in care planning. These recommendations are deliberately separate from the recommendations on the actual content of the care plan (see the section on the care plan: multi-agency approach to address physical health, social care, housing and other support needs).

The committee took into account qualitative evidence from 3 studies reporting on the barriers or facilitators that face people with severe mental illness and substance misuse face when trying to make decisions about their care [ES2.2.9]:

It felt that these factors can help a person adhere to their care plan.

The committee was also aware from the evidence (5 qualitative studies: 2 high quality, 2 moderate quality and 1 low quality) and their experience that a good relationship between the health or social care professional and the person with coexisting severe mental illness and substance misuse is key to effective delivery of health and social care services [ES2.1.4]. Members noted that a good relationship can affect a person's willingness to engage with and respond to care, and can also affect their recovery.

Bearing in mind all these factors, it made a strong recommendation on the need to take them all into account when developing a care plan.

Additional factors taken into account

The committee noted from members' experience that providers need to understand what is having an effect on the person each time they see them, so that they can provide the right level of support, including information, each time. It noted that the frequency of contact can vary depending on the person's circumstances. It also noted the importance of sharing the care plan between services.

The committee noted that people can recover. But it also noted that for this group of people, 'recovery' may not necessarily only be about reducing their substance use but about leading a productive life. The members felt that although recovery may take time, providers need to always convey a sense of optimism whenever possible.

The committee was aware that changing behaviour may be a lengthy process and that NICE's guideline on behaviour change: individual approaches may provide useful strategies on personalising messages.

Cost effectiveness

There was no evidence for cost effectiveness for this set of recommendations.

Current practice

Social care needs should be assessed in line with the Care Act 2014.

Evidence

The committee noted from the evidence from 1 high-, 4 moderate- and 3 low-quality qualitative studies (including 4 studies in the UK) that the lack of a shared approach between services could act as a barrier to providing health and social care services [ES2.1.7]. The committee heard from an expert on local partnership working and experts working with people with coexisting severe mental illness and substance misuse who are homeless [EP1, EP2]. The experts highlighted factors that could help with a coordinated approach.

Based on the evidence, expert testimonies and their own experience, members agreed that important factors in providing a coordinated approach included a shared vision, joint responsibilities and regular communication [ES2.1.7, EP1, EP2].

The committee highlighted the range of agencies or providers the care coordinator in secondary mental health services would need to work with to ensure people receive care for their wider health or social care, housing or support needs. The committee highlighted the physical health, social care, housing and other support needs that need to be considered when developing and reviewing a care plan.

Members reflected on the evidence, expert testimony and their own experience to inform their recommendations on social care, housing and other support needs [ES1.1.9, ES2.2.1, ES2.2.2, EP2]. The committee referred to evidence previously noted in the discussion for section 1.1 [ES1.1.9, ES2.2.1]. It also considered the evidence from 2 qualitative studies (1 high quality and 1 moderate) which described the barriers faced by this group in relation to employment support [ES2.2.2].

Members reflected on expert testimony and existing NICE guidelines on a range of health behaviours [EP4]. Based on this and their knowledge and experience they made a weak recommendation to decide on how a person's physical health could be improved and provided examples of how this may be achieved. This included addressing health behaviours (such as improving diet, quitting smoking or increasing physical activity) and minimising risky behaviours (such as unprotected sex, sharing needles). They realised this is not an exhaustive list and that the care plan may need to address other behaviours.

The committee also noted that care coordinators may need to help people with practical tasks so that the person can look after their own physical health. The examples were based on the committee's expertise on the type of tasks undertaken by care coordinators. So the committee made a weak recommendation on approaches to keep people involved in their care plan.

In addition, the committee noted the importance of encouraging activities to improve physical wellbeing (for example football or walking groups). But it was aware of the risk of widening inequalities if this only reaches people who already use services. The committee agreed that potential inequalities could be addressed by recommending providing inclusive services and strategies to improve engagement.

The committee made a weak recommendation on practical strategies that may help improve uptake of services and prevent relapse. This was based on evidence, expert testimony and the committee's expertise [ES2.2.3, EP2]. The committee was aware from 1 moderate and 1 low-quality study (set in the UK) of barriers or facilitators associated with providing information or training [ES2.2.3]. One moderate-quality study showed supporting people to develop self-care skills helped with daily living. The committee used this evidence combined with their experience to give other examples of practical skills to include in the recommendation.

Additional factors taken into account

Recommendations on how to encourage use of services and the suitability of different types of support were based on evidence review 2, expert testimony and the committee's expertise. The committee noted that people with coexisting severe mental illness and substance misuse are particularly at risk of being taken advantage of, so it is important to ensure the type of support they are offered is suitable for them [EP2].

The committee was aware from members' experience and expert testimony that communication between services is often poor [EP2]. The committee was also aware, from members' experience, that people are often discharged early or denied access to services because of missed appointments. There is often a good reason why the appointment was missed – for example, because the person was having side effects from their medication – but this has not been shared among the agencies involved. This was also highlighted in 2 qualitative studies (1 high, 1 moderate quality) in review 2. Members made a strong recommendation, noting that this can be addressed by making sure practitioners communicate and share information with each other, particularly in relation to non-attendance, so that it does not lead to an automatic discharge.

Cost effectiveness

There was no evidence for cost effectiveness for this set of recommendations.

Evidence

Committee members were aware from their experience and from the evidence of the barriers or facilitators associated with an integrated approach to care from 1 high-quality, 6 moderate- quality, and 2 low-quality qualitative studies (2 set in the UK) [ES2.2.6]. They noted from members' experience, expert testimony and the evidence that this could increase engagement and result in positive improvements in health, functioning and wellbeing. Although the UK studies were low quality, the committee felt the findings were relevant because they reflected the views of providers and users in voluntary sector services.

They also noted the importance of different disciplines working collaboratively, and taking part in case review meetings. This was based on the evidence from 8 qualitative studies reporting on barriers or facilitators associated with the management of cases with members of the same team and across different health and social care agencies [ES2.1.7].

The committee noted that the frequency of case review meetings would vary and would involve multidisciplinary team members and several different agencies. This is important to make sure a person's care plan is up to date and relevant. The strong recommendation to review the plan annually was based on the Care Programme Approach. But the committee recognised that this would depend on the person's level of need and circumstances and so recommended review meetings could be more frequent, if needed.

The committee noted the importance of regular monitoring of physical health, including for adverse effects of medications [EP4]. It was aware of strong evidence from 3 UK studies (2 case control and 1 cohort) that people with coexisting severe mental illness and substance misuse are less likely to adhere to medications than those with severe mental illness only [ES1.1.8].

The committee heard expert testimony about the side effects of medication and was aware from members' experience that this includes weight gain and other adverse effects [EP4]. Members felt this could be a barrier to adhering to treatment and could have a negative impact on a person's mental or physical health.

Committee members acknowledged that the evidence on working collaboratively and the views expressed in the expert testimony reflected their own experiences of working with people with coexisting severe mental illness and substance misuse [ES2.1.7, EP4]. They noted that changes in circumstances need to be taken into account in a person's care plan and physical health or social care, support or housing needs revised accordingly.

Cost effectiveness

There was no evidence for cost effectiveness for this recommendation.

Current practice

Committee members noted from their experience that transfer between services and discharge from the Care Programme Approach are key points when a person can lose touch with services.

Evidence

The committee felt that a robust relapse prevention plan and re-entry into the system would help to mitigate the risk of suicide or death from unintentional overdose. The committee agreed that housing needs are a priority before discharge and referred to evidence previously noted in the discussion for section 1.1 and discussion for section 1.3 [ES1.1.9, ES2.2.1].

The committee noted that the discharge plan should also include information on managing risky situations because of the challenging nature of working with people who may be intoxicated or in withdrawal. This was based on members' experience and evidence from 2 moderate-quality and 1 low-quality studies [ES2.1.8]. The committee was aware of NICE's guideline on violence and aggression and agreed it was a useful source for providers.

Members noted the evidence on challenges people can face when moving between services and felt this was applicable to other key points in a person's life [ES2.2.1]. The committee acknowledged the need to take a 'life course' approach. So it strongly recommended that provision for continuity of care needs to be in place when transition between services occurs and at key points in a person's life. This was based on members' experience and evidence from 4 qualitative studies (2 high and 2 moderate quality) [ES2.2.4]. The members highlighted particular groups who may need additional support based on their expertise and existing NICE guidelines.

Cost effectiveness

There was no evidence for cost effectiveness for this set of recommendations.

Additional factors taken into account

The committee heard expert testimony on the importance of making sure the guideline included the referral of young people to adult services [EP3]. It also reflected on members' experience and noted that groups such as looked after children and older people may need additional help. So handover of care on discharge, or when a person transfers to another service (in consultation with other providers), was included in the recommendation.

The committee agreed that encouraging practitioners to meet at multidisciplinary and multi-agency meetings is likely to improve physical health, social care and support outcomes and potentially reduce admissions for crisis care. But it also noted that this may be a new approach for the non-mental health sectors and that releasing staff for these meetings could be problematic without additional resources.

Current practice

The committee noted that although a policy guide in 2002 (the Department of Health's dual diagnosis good practice guide) had set out the vision for how services and care could be delivered, it was not being implemented. The committee was aware of Public Health England's guidance on co-existing alcohol and drug misuse with mental health issues: guidance to support local commissioning and delivery of care [to be published December 2016] which sets out the importance of joint working.

The committee discussed the fact that since April 2013 there have been separate funding streams for mental health and substance misuse services, with mental health services funded by clinical commissioning groups and substance misuse services by local authorities. The committee felt this exacerbated a longstanding division between the mental health and substance misuse sectors. It has also led to 2 different sets of organising paradigms for commissioners, which does not serve people with coexisting severe mental illness and substance misuse.

Members also noted that funding for addiction services comes from local authority budgets and is subject to commissioning contracts (which may exclude provision of mental health assessment or prescribing) and competitive tendering.

Evidence

The committee decided to recommend partnership working because there is a lack of provision of health and social care services for people with coexisting severe mental illness and substance misuse. Where it exists, it is often fragmented and inconsistent and this can affect continuity of care. This was based on members' experience and the evidence previously noted in the discussion for section 1.2 [ES2.2.4].

The committee noted from the qualitative evidence previously described in the discussion for section 1.3 that different disciplines working together to support people with coexisting severe mental illness and substance misuse could help with coordinating care. If they work together and share responsibility for this group, the evidence showed it could improve the quality of health and social care services offered [ES2.1.7]. The evidence showed this could be done by joint management of cases and regular communication.

The committee also heard from an expert in local partnership working who described a framework designed to help local areas design and deliver flexible and coordinated services for people with multiple needs [EP1].

The committee noted that there needs to be a strategic framework for services that work with people with coexisting severe mental illness and substance misuse. And that commitment from providers and commissioners is essential for services to collaborate locally.

Based on the evidence, the expert testimony and their own experience, committee members agreed that a cross-sector partnership, with a shared understanding of the problem (based on assessment of local needs) and a shared vision for the future were important factors [ES2.1.7, EP1]. Based on their expertise and expert testimony, they developed a recommendation on how services need to work together. They also noted the lack of evidence from review question 1.2 on existing care pathways and agreed further research is needed (see the recommendation for research on care pathway).

The committee was aware of evidence from review 2 that a lack of policy on referrals has an effect on the organisation and continuity of care. Evidence from 4 qualitative studies conducted in different settings (including 1 UK study set in the voluntary sector) noted that uncertainty on who should make referrals can also have an impact [ES2.1.11].

Committee members noted that the evidence from qualitative studies (previously noted in sections 1.1 and 1.2) was consistent with their experience [ES2.1.10, ES2.1.11]. This showed that pathways were inadequately planned and supported and that movement across a care pathway was often restricted because none of the specialist services took responsibility for this group. They also noted that continuity of care can be interrupted because of changes in the commissioning process or cycle. For example, re-tendering for services can lead to disruption and the need to build new care pathways.

One UK low-quality qualitative study exploring the views of commissioners provided evidence of a facilitator associated with organisation and continuity of care. The study noted that good links between the statutory and voluntary sectors improved outcomes, such as reduced waiting times and delivery of care [ES2.1.11]. This could also help with organisation and continuity of care.

The same study also highlighted that existing resources were stretched and that investment in the non-statutory sector could lead to provision of services not available in the statutory sector [ES2.1.5; ES2.1.6]. But the committee noted that this study was published in 2006. It also noted that commissioning and service provision for addiction services, the demography of people who use the services, treatment and the types of substances used have all changed markedly since 2002.

The committee noted from the evidence that there is no national service configuration in place (review question 1.2). Members acknowledged the importance of including the needs of people with coexisting severe mental illness and substance misuse in the joint strategic needs assessment. They agreed the needs of this group could be included in local strategies (for example, housing, alcohol or drug services and crime prevention). The committee noted that referral processes and pathways need to be in place to ensure this happens – and that a joined-up approach would help because this group often falls through the gaps in services.

Committee members also highlighted the importance of prompt access to services, based on their own experience and evidence. This was based on 1 high-, 2 moderate- and 3 low- quality qualitative studies (3 set in the UK) reporting on barriers and facilitators when seeking access to health advice. Barriers included long waiting lists, and 1 low-quality UK study indicated that direct referrals by alcohol and addictions teams could act as a facilitator [ES2.2.8]. Members agreed that direct referrals may be useful. They noted that direct access to services may be beneficial (compared with, for example, open access drop-in clinics) because this would give the person a sense of continuity of care. In turn, this may also enhance feelings of trust [ES2.2.4].

Cost effectiveness

There was no evidence for cost effectiveness for this recommendation.

Evidence

The committee made recommendations to highlight the importance of information sharing. The committee noted an expert testimony that highlighted that confidentiality is a barrier often faced by voluntary sector as an excuse not to share information [EP2]. The committee also noted the importance of services knowing about other local services and being able to tell people with coexisting severe mental illness and substance misuse or their families or carers about them [ES2.2.3, ES2.2.10]. For example, 1 UK low-quality study set in the voluntary sector noted that GPs were unaware of local community groups that people with coexisting severe mental illness and substance misuse could use [ES2.2.3].

Cost effectiveness

There was no evidence for cost effectiveness for this recommendation.

Current practice

The committee observed what appears to be an inequity in the way that people with coexisting severe mental illness and substance misuse are treated by services compared with other groups. It noted that the needs of this group are often not taken into account and they risk being excluded from mainstream services. Therefore the committee made a strong recommendation on improving delivery of existing services to make them more inclusive.

Evidence

Committee members were aware, from their own experience, the evidence and expert testimonies of the benefits of supporting people to participate in improving services [review 1, EP1, EP2]. The committee also noted from the evidence (previously described in the discussion for section 1.2) the importance of involving people with coexisting severe mental illness and substance misuse (and their family or carers), and providing them with information and support [ES2.2.9, ES2.2.10]. The ways in which people with coexisting severe mental illness and substance misuse, and their family or carers, could be involved in design and delivery of services were based on the findings from the review on epidemiology and current configuration [review 1].

The committee noted from its expertise and evidence (previously noted in the discussion for section 1.1) that people are often passed between services without being provided with appropriate care and support and that this may be because of negative attitudes or stereotyping by staff or services [ES2.1.3, ES2.2.5].

The committee also noted from its experience and the evidence that these factors can lead to a mistrust of professionals, resulting in poor engagement with services [ES2.2.5]. This was based on evidence from 3 (1 high and 2 moderate quality) of the 9 qualitative studies (1 high, 4 moderate and 4 low quality) reporting on barriers associated with access to effective care by trusted professionals. In addition, members agreed that a pessimistic attitude among professionals, about the likelihood of the person staying in the service may also be a contributing factor to the poor service.

The committee was aware from the evidence review on epidemiology that the prevalence of coexisting severe mental illness and substance misuse varied across regions. The evidence showed that semi-rural areas seem to have the highest need [ES1.1.2]. This was based on moderate evidence from 9 cohort studies (4 high, 1 moderate and 4 low quality) and 7 case–control studies (2 high, 2 moderate and 3 low quality) reporting on the prevalence of coexisting severe mental illness and substance misuse among those in contact with secondary mental health services.

Expert testimony suggested there is a high incidence of early psychosis in rural areas, but the committee noted from the evidence that specialist services are mostly in urban areas [review 1, EP3]. The committee agreed not to make a recommendation specifying content or configuration of service delivery by geographical settings. Instead it felt that the most important message was to ensure that any services needed (as identified by the joint strategic needs assessment) are delivered locally.

The committee made a strong recommendation on locating services in places that are safe and where there is minimal stigma attached to attending. It acknowledged the evidence (1 moderate- and 1 low-quality study) on co‑location of services (for example, services based in the same facility) was mixed but recognised that there may be stigma in accessing certain services [ES2.1.12]. Committee members were aware from their experience and from expert testimony that people with coexisting severe mental illness and substance misuse are particularly vulnerable. They may be at risk of exploitation (for example, being forced to become sex workers or being taken advantage of in relation to their housing or financial situation). Or they may have experienced trauma (for example, women may have experienced rape) [EP1, EP2]. It agreed that a 'trauma-informed' approach would provide the best support for this group.

Members were also aware from the evidence that even if people knew about services, barriers to access included difficulty in contacting or gaining admission to services outside hours, long waiting lists and services not being local [review 2]. The committee considered the evidence review (review question 1.2) on current configuration of services and developed a recommendation highlighting the importance of safety of location, low stigma and flexibility in opening times as factors that can help make services more accessible.

Cost effectiveness

See the end of this section for details on cost effectiveness.

Current practice

The committee was aware of moderate evidence from 13 UK studies (2 high, 9 moderate and 2 low quality) that there were inconsistencies in the current configuration of 'dual diagnosis' services in NHS trusts across the UK [ES1.2.1]. These inconsistencies lie in several areas, including sources of funding, structure of services, type of staff members, services delivered and coordination of care. The committee considered the evidence on configuration of services and observed there were few specialist services for adults [ES1.2.1].

The committee agreed that the recommendations for specialist services (secondary care mental health services and 'dual diagnosis' services) need to focus on improving existing services using the expertise that is available instead of creating a specialist 'dual diagnosis' service. It felt that the standard care delivered in the UK could be improved by increasing the level of engagement people with severe mental illness and substance misuse have with existing services and that existing capacity and resources could be used to deliver this.

Evidence

The committee made recommendations about the design, delivery and content of the service model, based on the evidence, economic model, expert testimony and members' expertise.

The committee considered the evidence for the effectiveness and efficiency of service delivery models, which included randomised controlled trials (RCTs) and observational studies [ES3.1, ES3.2, ES3.3, ES3.4, ES3.5, ES3.6, ES3.7, ES3.8, ES3.9, ES3.10]. The evidence covered a range of service delivery interventions, showing some positive outcomes and that there was value in what the models were aiming to achieve. However, the members agreed that there was no overwhelming evidence of benefit to indicate a particular model should be recommended.

The committee agreed that there was limited evidence of effect for assertive community treatment and integrated treatment interventions in relation to mental health and substance misuse outcomes [ES3.1, ES3.2, ES3.3]. The committee noted that fidelity to delivery of interventions (whether the intervention was delivered as designed) in the service models was reported for only 5 studies. Where reported, the fidelity was considered to be good.

There was weak evidence for assertive community treatment based on 5 US RCTs [ES3.1]. The committee noted that the assertive community treatment intervention model is no longer used in the US and is rare in the UK. There was moderate evidence from 6 RCTs and 1 observational study (3 studies based in the UK) for integrated treatment interventions compared with treatment as usual [ES3.2]. There was weak evidence from 1 RCT for integrated treatment intervention compared with enhanced assessment and monitoring. The RCTs did not all show a clear evidence of benefit [ES3.3].

There was some improvement in service use outcomes (increase in physical and telephone contact) but members noted that it was debatable whether this was necessarily an evidence of benefit, because the reasons for contacts were not reported [ES3.1]. There was some evidence of effect on social care outcomes such as housing, employment and social functioning [ES3.1].

The committee felt that although the follow-up in the studies ranged from 24 weeks to 3 years, the length of time needed to observe small improvements can sometimes be 5 to 10 years [ES3.1, ES3.2].

There was moderate to weak evidence from 8 RCTs and 1 non-randomised controlled trial evaluating a range of interventions. The intervention included:

The comparator arms were no intervention, treatment as usual or an active comparator.

The committee noted that there was mainly weak evidence from small studies, with short follow-up (ranging from 16 to 78 weeks). Three studies were based in UK and Ireland but most of the evidence was from US. It noted that fidelity to delivery of the intervention was reported in only 2 studies (1 reported as low and 1 as high fidelity). Members discussed the potential value of service models incorporating contingency management, peer support (delivered as part of a care coordination intervention in 1 US study) or text messaging, and considered these further under research recommendations [ES3.5, ES3.6, ES3.10] (see the recommendation for research on what works).

The committee agreed that there was weak evidence for a staff training intervention considered in the review of effectiveness of service delivery models [ES3.8]. It noted that the 2 UK studies were of low quality, the evidence was inconsistent and did not appear to show an overall benefit. In addition, a committee member reflected on their own involvement in delivery of the intervention in 1 of the studies. The committee member noted that there were a number of challenges: staff often moved between services, there was a high turnover of staff, and low fidelity to delivery of the intervention.

The committee agreed not make a recommendation on training because the evidence did not show an overall benefit.

The committee agreed there were several gaps in the evidence from review 3 including:

Cost effectiveness

See the end of this section for details on cost effectiveness.

Additional factors taken into account

The committee was aware of evidence from 4 qualitative studies (1 moderate- and 3 low-quality studies) of barriers or facilitators associated with integrated services. One low-quality UK study, for example, described mixed views among staff in a specialist 'dual diagnosis' service on whether services should be separate or integrated with mental health or substance misuse services [ES2.1.13]. It noted that there was evidence from the same study indicating that most commissioners felt that integrating services is essential for the effective and efficient delivery of care for people with complex needs. Some commissioners also noted that relationships between different services could be expected to improve if they were required to share budgets and resources.

Committee members felt this finding (published in 2006) should be treated with caution because the funding landscape has changed considerably since 2002. Based on their experience they noted that:

The committee noted that there was limited description of the comparator arms (often described as 'treatment as usual') in the studies included in review 3 and that most of the studies were conducted in the US. The committee's view was that 'usual care' in the US is likely to differ from that in the UK and the level of 'usual care' in the UK was considered to be of a better standard.

The committee used members' expert knowledge and the evidence to develop a recommendation on aspects that could be included in a service. This includes interventions that have shown to be effective in NICE guidelines for either severe mental illness or substance misuse. The committee was aware of the study by Wenze (2015; adjunctive psychosocial intervention following Hospital discharge for Patients with bipolar disorder and comorbid substance use: a pilot randomized controlled trial) included in the economic model. It reflected on the components of the 'treatment–engagement' sessions in the Wenze (2015) study as well as members' own experience to develop a recommendation on ways to improve engagement.

The committee noted that any recommendation on improving service delivery needs to take into account the needs of those who reach crisis and those who experience a relapse after discharge. This recommendation was based on members' expertise. Members were aware from the evidence and their experience that people's care is often fragmented and that plans need to be in place to allow people to return for additional support after being discharged or losing touch with the system. They noted the evidence on facilitators for consistent care, including from 1 low-quality UK study that highlighted that good aftercare is an important means of preventing relapse [ES2.2.4]. They also noted that the Department of Health's Mental Health Crisis Care Concordat has information on developing an action plan for people in a crisis.

Cost effectiveness

The committee discussed the evidence from the cost effectiveness studies and the economic model when developing the recommendations on improving service delivery.

An economic analysis was undertaken. This comprised a review of existing cost effectiveness studies and a bespoke economic model.

The findings from the review of evidence (from 1 UK and 7 US studies) were inconsistent [ES4.1, ES4.2, ES4.3, ES4.4, ES4.5]. The US studies found that integrated treatment leads to minor cost savings but the UK study found that the intervention resulted in an increase in public sector costs.

In all studies, integrated treatment appears to result in improvement in some outcomes. But economic analyses used different outcome measures, reported as changes on various scales, making comparisons challenging. Three studies adopted before-and-after design, studies used different perspectives and time horizons, only 1 included economic study was judged to be directly applicable, 3 studies were judged to be characterised by minor limitations [++], 4 by potentially serious limitations [+], and 1 by very serious limitations [−]. Overall, there is little evidence to support one service delivery model over another, based on existing economic evidence.

The model was based on 3 studies. The first study, conducted in the US, comprised a treatment–engagement intervention (using resources more intensively than in standard care) for people with bipolar disorder and substance misuse. It was a small study whose health outcome was inconclusive, but yielded resource use data. The remaining 2 studies, both from the UK, were used to estimate baseline admissions rates for people with dual diagnosis.

The model's time-horizon was 1 year only. So increases in life expectancy that might have occurred as a result of an intervention were not included as benefits in the model. Because of the lack of data, a further conservative assumption was that wider costs, particularly those falling on the criminal justice system, were not included. Further, the model's measured outcome might not have measured all of the health outcome benefits.

The model showed that an intervention that combined enhanced engagement with standard care would need to reduce relapses by about 12% for the intervention to become cost saving.

The committee members had differing views about whether UK standard care is better than that reported in the US studies. It was felt that standard care in the UK may be more similar to the enhanced intervention modelled.

Assuming standard care in the UK is equivalent to the enhanced intervention modelled, it would be offering better outcomes at the same cost. By definition, that would be a cost-effective approach. However, assuming standard care in the UK would need to be enhanced and therefore need additional resources, at a cost of £226 per person and assuming an effect size of 10% the intervention would need to result in a small quality-adjusted life year (QALY) gain of 0.002 (equivalent to 0.73 days in full health) to be considered cost effective at an incremental cost effectiveness ratio threshold of £20,000 per QALY [ES4.6].

Given the results that were obtained even though a number of potential benefits were not considered because of the lack of data (for example on a person's life expectancy, improvement in the substance misuse problem, improvement in the mental health of service users the reduction in health and social care and the criminal justice system costs) the treatment–engagement intervention is very likely to be a cost-effective option.

Additional factors taken into account

The committee noted that maintaining engagement can lead to improved outcomes and may place less burden on crisis care or inpatient admissions.