Care models for coexisting serious mental health and alcohol/drug conditions: the RECO realist evidence synthesis and case study evaluation

Summary of literature mapping

A total of 23 papers met the inclusion criteria, which described 19 UK COSMHAD service models. Service models described in multiple publications were combined into a single entry on the data extraction form. Nine of the studies were delivered by mental health trusts in either community or inpatient settings, four by substance use services (three in the community and one specialist ward), and six studies involved collaboration between multiple service settings either through networks or liaison with specialist workers.

Models of service provision fell into three broad categories: (1) multidisciplinary COSMHAD groups or networks to establish provider relationships and co-ordinate care pathways; (2) placement of specialist COSMHAD link workers within existing mental health or substance use teams; and (3) specialist COSMHAD liaison workers working with mental health and substance use teams from a separate specialist COSMHAD team. These three categories are indicated in the first column of the Table 1, with the allocated colours indicating elements specific to that model.

Summary of evidence

The concept of ‘everyone’s business’ comes from the position that a significant number of people who use mental health and substance use services have co-occurring disorders and therefore need to be equipped to engage people in treatment and attend to their multiple needs. This PT (Figure 2) describes how this concept can actually be operationalised. A key contextual factor facilitating successful engagement (especially on initial contact) is that staff should recognise that working with COSMHAD is a part of their job. Adams3 describes how ‘professional ambivalence towards comorbidity [context] … may influence the assessment process and subsequent interactions [mechanism–resource]’ (p. 102) and several studies have highlighted the importance of using assessment protocols and screening tools to help the clinician formulate a thorough picture of the client’s life circumstances. 40–44 This in turn allows the clinician to develop a richer understanding of the person’s situation, which promotes compassion. This was also supported by the case study data with staff emphasising the importance of a flexible and non-judgemental approach, that it’s ‘everyone’s business’ and a needs-led holistic approach can be effective in identifying the right care and treatment.

FIGURE 2.

Refined PT 1: first contact and assessment (‘It’s everyone’s business’).

Across all sites, staff demonstrated awareness of their responsibility to assess and identify treatment for people with co-occurring disorders. Staff emphasised the importance of ‘responsiveness and flexibility’ (SP22, Case Study C) and described their approach as being ‘friendly, non-judgemental, just listening to them’ (SP92, Case Study B) and having this responsible and flexible context for first contact would lead to people with a co-occurring disorder feeling less judged and more likely to engage with services (mechanism – response) (Figure 3).

FIGURE 3.

Programme theory 1 context to mechanism.

I would like them to be met at where they’re at and for us not to focus on what is wrong with you, but what’s happened to you as a first encounter … They’re not going to feel judged, are they? For a start. So, you’re optimising engagement. And I think it’s important for us not to come with our own personal, professional, organizational agendas.

SP10, Case Study C

Staff also demonstrated their awareness of national dual diagnosis guidance and policy: ‘You want to implement the principles of everybody’s business and no wrong door’ (SP52, Case Study F). There was a sense from some services that the high prevalence of co-occurring disorders had led to acceptance of these principles. However, this context varied between different mental health services, with Assertive Outreach, Early Intervention (EI) and Improving Access to Psychological Therapies (IAPT) (now known as NHS Talking Therapies) services viewed as being more proactive when compared with community and inpatient mental health services. When ‘no wrong door’ context was present within services, staff participants described health professionals feeling empowered to address clients’ needs (mechanism – resource) leading to reduced exclusion and stigmatisation of people with COSMHAD who feel reassured that they will have ongoing support for their mental health and substance use from knowledgeable staff (mechanism – response).

… another way of thinking about it is that the existing pathway, the treatment as usual pathway is for dual presentation because that’s the norm … And then it’s almost like unusual if you come into an EI team without substance use … the EI model of care has to work for people with this presentation, ’cause it’s most people.

SP37, Case Study A

However, service user and carer data suggested that challenges in accessing help still remained. Many service users discussed being unable to access mental health support until they had halted their drug or alcohol use, or they were unable to access drug or alcohol rehabilitation programmes while taking medication for their mental health conditions (context). Participants felt there was little awareness or appreciation of COSMHAD among services making their initial assessments (mechanism – resource) and described this scenario like ‘chicken and egg’ [P29, (Carer) Case Study B; P38, (Carer) Case Study A; P6, Case Study C], where services had stipulated that ‘we can’t fix both things. We have to do one or the other’ (P6, Case Study C). However, participants reported that in their experience, their mental health conditions and substance use were strongly related to each other. For example, substance use was described as a coping mechanism:

to self-medicate. You know to try and block out thoughts and feelings that you know that I was having.

P8, Case Study E

Service users and carers suggested that first response services (e.g. A&E or the police) were often the only way to receive any kind of support in a crisis (mechanism – resource). A number of carers discussed how they had often called the police as a last resort because they feared for their loved one’s safety and were at a loss as to where to receive support. In some cases, carers reported that they had been advised to call 999 by their local mental health crisis team (mechanism – response). Service users described being caught in a cycle of representing at times of crisis in an attempt to get support for their mental health with discharge often resulting in further suicide attempts or relapse of their drug/alcohol problems (outcome).

For service users, this separation of their mental health and substance use by services was frustrating. Some felt despondent about not being able to access mental health support until they had stopped their substance use (mechanism – response). These participants described themselves as having ‘fell through the cracks for many years’ (P8, Case Study E), ‘been in limbo’ [P28, (Carer) Case Study A] or ‘all the doors closed for him and it was blamed on … his drug using’ [P28, (Carer) Case Study A], which reduced their motivation to work towards their goals and could lead to relapse (outcomes).

The prevailing context for assessment was the importance of a relationship between the health professional and the person with a co-occurring disorder (mechanism – response). Key to this relationship was the client feeling ‘comfortable enough to just sit and be assessed and be happy to meet again’ (SP61, Case Study F) and ‘get to the point where they’re actually able to engage in therapy’ (SP51, Case Study D). However, it was also recognised that it was not always possible to build this relationship within a single, initial assessment and that

[I]t depends on their motivation at that specific time. It might be that when you initially see them that they don’t want any support, they’re not ready to make any kind of change. But then three months down the line, it might be a different kettle of fish.

SP91, Case Study B

This demonstrates the importance of a positive experience which could yield a positive outcome (engagement) in the future. Where a positive relationship exists (context), people with co-occurring disorders feel supported and hopeful and so are more willing and able to engage with services (mechanism – response).

Better for them to build a relationship rather than being ping-ponged around and inevitably falling through the gaps and having to restart again and fail and not … have any faith in services … this is part of why we’re trying to transform our community services so that people aren’t being pinged around between different services. So, they, you know, they stay put and you, you’re with them through the whole journey …

SP50, Case Study D

A trauma-informed approach was also emphasised as a way of engendering hope and engagement (mechanism – response). Experience of trauma can impact on engagement with statutory services and when a positive relationship is lacking (context), existing service structures can replicate people’s experiences (and in some ways be re-traumatising) and reinforce feelings of hopelessness and disengagement.

[P]eople start to get passed around … it’s that idea. Oh, here’s some risk that I can’t deal with. I don’t know what to do with that or I can’t deal with that. So, what I’m gonna do is send you to someone else now, I can’t help thinking that in some way, what we start to replicate is … that they were removed from their families of origin for a reason. They would possibly be removed from other services who said they’d keep themselves and help them because they were unmanageable and then here, we are again potentially replicating those, those relationships […], and so we’re undermining their basic … view of themselves … We’re replicating it and reinforcing a negative scheme.

SP38, Case Study A

Additional context

Additional contextual factors that negatively impacted first contact with services included operational processes that hindered engagement including telephone-only first assessments, limited follow-up before discharge, service remit which explicitly excluded clients due to their co-occurring substance use or severity of their mental health condition, overly long and repetitive assessment pro formas (due to services not being able to share care records) and a lack of clear point of contact following assessment.

Staff expressed frustration at ‘having to work in a system that maybe isn’t best placed to offer the service that you’re hoping to offer’ (SP5, Case Study A). They felt that this led to a fragmented experience of services for clients, disengagement and a lack of hope (mechanism – response).

… everybody has to cover their personal, professional, organizational, litigious liability thing with endless risk assessments … That assurance work would probably be much more limited if there was a much more joined up IT governance system across GP Services primary acute drug mental health services … if you could read it and say is there anything changed to this? As opposed to could you talk all about your childhood again or you know, blah, blah blah? I think that that would be quite a big difference.

SP38, Case Study A

Following assessment, people with COSMHAD were often referred to services that did not meet all their needs and were expected to engage with a ‘one-size-fits-all approach’ to treatment. As a result, clients could become despondent and disengage or fall into the gaps between services, often resulting in exacerbation of both their substance use and mental health problems (and leading to further crisis-led service use).

It’s very much a one size fits all, isn’t it? … you’re gonna get filtered out … if you don’t engage or you don’t fit into any mental health … So you get despondent. Drop out and then go back again because … mental health won’t pick him up until they stop drinking. And it’s like, well, they can’t stop drinking because … it’s such a vicious circle […] we’re seeing people die, aren’t we because nothing ever gets really addressed?

SP30, Case Study A

By that time, he is 18 … it’s really, really hard to actually get any help. The only way we could get help was to continuously ring the police to come and take him out of the house because he smashed the house. Until, finally, there was a psychiatric trauma analysis … and [son’s name] found himself in the [hospital name] at the [psychiatric intensive care unit] and from then a more supportive network was built around him. But from the word go it’s been horrendous just trying to access help in the 1st place. That’s the biggest, biggest problem.

P36, (Carer) Case Study B

Staff participants emphasised the importance of engagement with service users at first contact (mechanism – resource) and, if this was achieved, it could result in improved access to services (outcome) (Figure 4). However, this outcome required staff to support people with a co-occurring disorder to reach a level of stability where they felt able to access and attend services (outcome) (see Figure 4).

FIGURE 4.

Programme theory 1 mechanism and outcome.

It makes a massive difference to our service users because they are being followed up in both camps and if we don’t do that in partnership working, it’s really difficult to get good outcomes for our service users … Yes. I think we have better outcomes if that happens because we see often if those two don’t go hand in hand, people become like frequent flyers, they come back through the same services, but not always jointly – separately. So we see more admissions, we see more, on the inpatient side, more mental health admissions. We also see on the physical health side, people are turning up more in A&E they are attending more with their general practitioner (GP). So if we can do that joint working together, those frequent admissions and attending places reduces as well.

SP41, Case Study D

Carers described feeling fearful, angry and frustrated that they were unable to access mental health support for their loved ones (mechanism – response). They felt that substance use services were ‘not equipped to deal with someone who has a mental health problem’ [P28, (Carer) Case Study A] and that more education was needed for staff around COSMHAD (mechanism – resource).

Obtaining a diagnosis could mean access to specific services; however, diagnostic labels could also represent a barrier (e.g. if that diagnosis was an exclusion criterion for a service). Many individuals had received multiple diagnoses over time, and this could lead to confusion over which service should be providing care, whereas it was suggested that taking a needs-led approach could lead clients to receiving a more integrated and trauma-informed approach to care.

Does diagnostic uncertainty broaden our thinking and help us integrate under that trauma-informed kind of framework rather than thinking you’ve got this disorder and that disorder and diagnostic labels are everything … then they’re problematic use is a behaviour alongside a bunch of other behaviours that we can think about. I think that’s what I was trying to say at the beginning. That sometimes that notion of dual disorder, I don’t know it I can’t quite get my head to think about it properly. But anyway, yeah, I think that the diagnostic uncertainty incomes could be quite helpful in integrating.

SP37, Case Study A

Overall, the literature from the realist synthesis and data from the staff interviews and client and carer focus groups suggested that an open door and flexible policy was essential for successful COSMHAD services. COSMHAD needed to be on the agenda for all staff wherever they worked to enhance first contact success and improve the engagement of service users.

Summary of evidence

In the literature, there was a range of attitudes identified towards COSMHAD among healthcare professionals, and this also varied across professional group and work experience. 3 For staff in mental health services, attitudes were influenced by exposure within undergraduate and continuing professional development training. 45–47 Staff with positive attitudes were often interested in working with COSMHAD clients, expressed non-punitive beliefs about substance use, had a commitment to generate and maintain therapeutic relationships, and demonstrated pragmatic, flexible and individually tailored approaches3,47–55 (Figure 5).

FIGURE 5.

Refined PT 2: attitudes.

In the staff data, a wide range of staff attitudes were noted (context) which impacted on the way in which training and policy around staff attitudes (mechanism – resource) was implemented (Figure 6). Staff working within Case Study B (which had been operating for 16 years) reported an open approach towards clients with co-occurring disorders that focused on harm reduction and staff ‘not frowning on people, we’re trying to reduce the stress and triggers for use really and keep somebody well and out of hospital’ (SP92, Case Study B) (context). Staff in Assertive Outreach noted that the increasing numbers of people with COSMHAD within their service had been challenging (context) but that the support they received from the COSMHAD specialist team which included training and consultation-liaison (mechanism – resource) had helped them to develop the skills required to assist these clients in a non-judgemental manner (mechanism – response).

FIGURE 6.

Programme theory 2 context and mechanism.

[W]e’ve had great support from the [name removed] team throughout … We’ve been really lucky to have had a worker working alongside us with our guys to try and look at the issues that come up and trying to build people to have the skills to deal with the issues around in the substances and not be judgmental or telling people ‘no … don’t do that’….

SP92, Case Study B

Service users and carers described several facets of positive staff attitudes, namely, trust, advocating for their clients’ needs, feeling listened to, and their level of experience. Service users and carers acknowledged that with individual staff members it ‘takes some time to trust somebody’ (P14, Case Study E) but that it was a key aspect of a positive relationship ‘for mental health patients not to feel threatened by mental health professionals. Because they don’t trust … And … when they trust, then half the job done’ [P21, (Carer) Case Study A]. A trusting relationship consisted of ‘openness … there’s no stigma, no judgement’ (P18, Case Study C) and showing ‘compassion, empathy and a bit of understanding’ (P25, Case Study A).

Where positive attitudes exist (context), service users felt that they were able to be open about their problems. There were more varied staff attitudes in other sites which were influenced by individual staff interest and organisational philosophy (context). This sometimes stemmed from philosophical approaches towards substances in mental health services (context):

[T]here’s a lot of emphasis, sometimes on medication, to the extent of almost omitting the psychological and the social approaches … I think that’s what feeds into sometimes attitude because there’s historically this thing was you couldn’t access psychological therapies if you were drinking [alcohol] or whatever.

SP65, Case Study F

Engagement with COSMHAD issues was often dependent on ‘an individual being interested’ and the existence of ‘little flagship teams’, and a broader cultural shift was required within organisations to ensure that all staff were supported and encouraged to work with these clients (mechanism – resource) which would lead to staff feeling enthusiastic, confident and supported to use the skills they have been trained in (mechanism – response). This was also confirmed by a participant from Case Study E, who described ‘group think’ within certain teams which impacted upon both current and new staff attitudes (context) and thus the effectiveness of training on co-occurring disorders (mechanism – resource).

There were examples where positive attitudes were not present and the reasons for this were connected to both a lack of organisational commitment and pressure on resources (context).

[S]taff are so busy. It’s also very hard to … do something consistently … I find that in our ward there’s always something that needs you know to be actioned or something that needs to be addressed … so it’s very hard to keep focus on it, and to keep the interest in it.

SP72, Case Study E

There was inconsistency in training coverage:

[I]t’s slightly pot luck about who will be in the room with that person on a given day … there are sometimes issues with staff recruitment and retention and some staff might not be as well informed or motivated to work with people with more dual diagnosis issues … whether it’s values or beliefs – can get in the way with engagement of someone … a sort of non-MI stance or just a lack of knowledge.

SP61, Case Study F

The impact of these positive attitudes being absent (context) led to a lack of joint working between mental health and substance use services leading to clients being left without access to services (mechanism – response).

[M]any people we support get the knock back from mental health services … and our clients have told us this … so, when we make referrals, we have to say how long somebody’s been abstinent for … and really, really labour that point and it’s hard sometimes what I want to write on a referral is this person is likely to be consuming less alcohol units than the person reading this referral. Because the judgment is felt so keenly from other professionals … the automatic assumption that any of their distress is automatically going to be put down to their substance use … and it needs to be addressed there and then otherwise they’re gonna relapse again.

SP16, Case Study C

… I joined the Community mental health teams as a CPN [community psychiatric nurse] and a team lead and I was kind of quite shocked and horrified, kind of the attitude was opposed towards people with drug and alcohol issues … They’re really, really busy, but there was great assessments going on by really qualified, really experienced members of staff. But the second a substance or alcohol misuse issue was kind of suggested or proposed or suspected all the other information would be lost in it would just be they need to refer themselves to [name of substance use service] … It was kind of like we can’t do anything for you until you’ve addressed this … that is something that’s just became maybe embedded, but it’s the attitude sort from the bottom up … I think is a real kind of issue because it’s almost like at anything else falls on deaf ears as soon as we hear drug or alcohol use is involved.

SP44, Case Study D

Four of the case study areas described offering formal training to address staff attitudes (Case Studies B, C, D, F) as well as boost skills. However, training in isolation of other initiatives is not sufficient. Staff participants in Case Study F proposed that if training was to have any longer-term impact on practice it must be accompanied with embedding positive attitudes into the structure and policy of teams as well as the wider organisation (mechanism – resource) (Figure 7). When this policy/procedural support was in place (mechanism – resource), staff ‘start feeling confident to do it and feeling they have permission to do it from the service or other disciplines sometimes’ (SP73, Case Study E) by ‘empowering other professionals to feel like there is something they can do and also to understand what success looks like’ (SP52, Case Study F). This is known as ‘role legitimacy’. The key to this is taking a team-wide approach as opposed to individual workers (mechanism – resource).

FIGURE 7.

Programme theory 2 mechanism and outcome.

So you start teaching people on the course and … the language changes … that begins to change right there in the … seeing that change in how people are communicating about these issues and thinking about being flexible with some of the challenges, they’re also telling you that ‘oh, I feel much more able to talk to people about their substance use and no longer feel pressured to looking for solutions and now just have a chat with people and listen’. To them it’s easier … it’s not just about changing attitudes, but it’s about it creates a sense of confidence across the organisation about what you know, how to proceed and what needs to happen within and … ensuring that that is, you know we are embedding that and it’s ongoing, isn’t it? So, building this into the structure.

SP60, Case Study F

I think if individual practitioners are interested and feel confident and skilled and it’s on their radar, then they are much more likely to be more proactive with this. And I also think if they’ve got, if they’ve had good positive relation, working relationships with say staffing, turning point, that’s much more they’re much more than inclined to. Be more proactive with the people that they’re working with because they’ve already got those links.

SP52, Case Study F

Service user participants also emphasised the critical role of empathy, positive attitudes and better therapeutic relationships in creating a safe space where they feel able to open up about their feelings and issues:

[S]omebody that’s gonna be empathetic … when they come in … they took it kind of a step back … tell us what’s been going on. Tell us how you’re feeling … they let you get off your chest what you need to get off your chest … they then use that information to then open a dialogue with you. And I think that’s what you need is just someone who’s empathetic because, you know, when you are going through mental health issues or when you’re going through addictions, you know, your emotions are gonna be all over the place … Feel like they’re listening to what you’re saying. And that’s not to say that they’re gonna agree everything that you said because you know … You need to want to challenge you or to want to give you a bit of advice and to look at your reasoning and the way of thinking. You know, but you want someone who’s got a kind of way to be collaboratively make you feel safe … if you don’t feel safe, then you’re not really, really good at opening up.

P19, Case Study C

In addition, one person wanted to talk about the positive experience they had with the expert lead for COSMHAD, who they acknowledged had gone above and beyond their role to help, and in doing so had instilled hope in what was a bleak situation:

[H]e’s the one that is leading the dual diagnosis team from, you know […] and he is prodding and pushing in all of the right places behind the scenes and … he is doing that to prompt people from not just saying something in that meeting, but actually turning what they’re saying and what we talk about into reality. And that’s why I’m his biggest fan, by the way. I think he’s been fantastic … what the NHS really needs is about 100 of him … because you know my constant thing has been, it’s just there’s no joined up working … [dual diagnosis lead]’s come in and you know, join that working together for us, you know. And the unfortunate thing, [dual diagnosis lead]’s team isn’t a patient facing team, unfortunately, but they should be for me. And we’ve been very fortunate in so much that [he] has … been more customer facing than potentially he should be and he’s done that because he’s seen, you know, what desperate situation we’re in and you know, we power to his arm for that and credit to him for doing it … because I think then that just gives that assurances when things look so bleak.

P3, (Carer) Case Study C

Where training and policy was in place that promoted therapeutic optimism (mechanism – resource), staff described people with co-occurring disorder as being more engaged with treatment (outcome): ‘they’re not going to feel judged, are they? For a start. So, you’re optimising engagement. And I think it’s important for us not to come with our own personal, professional, organizational agendas’ (SP10, Case Study C). Staff participants described increased empathy among staff (outcome), this was achieved not simply by acquiring knowledge on co-occurring disorders but recognising that progress with co-occurring disorders is often small and incremental (mechanism – response) and having the confidence to listen to what the individual is telling them and respond reflexively to their needs.

Sometimes progress can be really slow contact with some service users and through no fault people can go, you know, back a few steps and that that you can take that personally as much as you don’t want to because you see as a failure in yourself to actually help somebody.

SP51, Case Study F

[I]t’s not coming with any, any preconceived ideas about what that person is going through? Because everybody is going to be having a different relationship with their drugs and their mental health. And I think sometimes if you go with too much, you can have the knowledge, but you have to go there with the premise that you haven’t got any knowledge almost, do you know what I mean? And well, you have got knowledge and that you understand what they’re going through, but I’m not that you’re going to make assumptions about what they’re going through because I think. I don’t know. I think I think from my point of view, I think that would be unhelpful if I had to be in a service where someone made assumptions about me and it wasn’t right.

SP11, Case Study C

Political, structural and philosophical differences were highlighted as obstructive to providing successful COSMHAD services throughout the realist synthesis and evaluation. These more macro contexts impacted on staff attitudes and subsequently therapeutic relationships with service users.

Summary of evidence

Evidence from the realist synthesis demonstrated that when leaders implement (context) care protocols that clearly describe the co-ordination of care from initiation of care through to referral/discharge44,56–58 (mechanism – resource), this helped staff feel supported in their roles47 and enabled them to use their skills and knowledge (Figure 8). Furthermore, it provided permission for staff to take a more pre-emptive, preventative whole person approach to clients with COSMHAD59 (mechanism – response).

FIGURE 8.

Refined PT 3: collaborative interdisciplinary/professional case management.

Staff data concurred with evidence from the realist synthesis which suggested that collaborative case management required both formal co-ordination and informal collaboration (context). They described several aspects of successful joint working including effective communication between teams, joint information technology (IT) systems (shared data) and joint care planning meetings (mechanism – resource) (Figure 9). Where these resources were in place staff felt that they had a good overview of service users’ care and were promptly alerted of any significant changes that could impact progress. Joint working was more successful when the staff already had an existing professional relationship with the other health professionals involved (context). This highlights the importance of having time for building those external relationships (mechanism – resource). When collaborative multiagency and interdisciplinary relationships were absent, it became harder to facilitate a service user accessing the right care. According to a substance use worker collaborative, they had to ‘advocate quite sort of strenuously for people to access psychological therapies’ (SP52, Case Study F).

FIGURE 9.

Programme theory 3 context and mechanism.

I think relationships with staff within other areas is key with this because you can actually do that joined up working much better if you’ve got those connections and you can actually sort of just pick up the phone and speak to somebody and say I’m not sure what to do with this person, you know? Can you help?

SP51, Case Study D

There is a need for organisational structures to support collaboration such as ‘just carving out some time to do that joint working, multi-agency working’ (SP66, Case Study F) (context). If collaboration wasn’t formalised through policy and protected time (mechanism – resource), there was a danger that all the responsibility for collaboration could fall to individual staff who were already overburdened by large caseloads (mechanism – response). In contrast, having structures for joint working (mechanism – resource) would allow staff to focus on building a trusting relationship with their clients that enabled them to successfully link them into other services (mechanism – response). Participants felt this collaborative way of working required a ‘cultural shift in an organisation that looks at a patient from the point of entry and says … we will work with substance use agencies from day one to manage this individual’ (context), which ‘demystifies it and takes if from being a separate thing to something that’s part of how people cope and manage’ (SP40, Case Study D).

… the structure of the service is very, very important … it’s about a whole model. It’s not just about picking out some people and saying they know stuff. You know I think the complete tragedy really that we have is we put everything on CMHT, any needs that we have in society around mental health or risk goes the doors of CMHTs and you know … that one care co-ordinator is meant to be an expert in, you know, care package commissioning, any type of mental health issue, anything based on, you know, anxiety and depression that’s affecting people’s functioning right up to extreme, you know, ends of mental illness and functioning. Now that is, that is not right to expect one person to know all that … We’re [homelessness service] able to have the contacts … that we’ve built up over time. We have contacts in substance misuse … So, we are able to react, and we know where to go to and we know who we need to get involved … whereas a CMHT cannot be like that at all, they cannot be really flexible and …all have a different approach for every single person because … People have caseloads of 40 like they have to just get on with their job.

SP2, Case Study C

The importance of the COSMHAD lead role in terms of fostering interprofessional relationships is supported by the case study data. Five of the six sites had some version of a COSMHAD lead role, and the remaining case study (A) had a vacant post. In addition, Case Study E and Case Study F had addictions psychologists employed within certain areas of their trusts, which made ‘joint working … probably on average easier’ (SP66, Case Study F). COSMHAD leads were described as leading on policy, training and strategy as well as supporting staff through joint meetings and case study discussion (mechanism – resource). One of the key things about the presence of such a role was that it ‘held [COSMHAD] in people’s minds and also possibly the organisations mind’ (SP37, Case Study A) and ‘keep it on the agenda and keep people thinking about issues within multidisciplinary team (MDT) meetings’ (SP61, Case Study F) (mechanism – response).

As well as working across their own organisation, the COSMHAD leads were key players in facilitating relationships with external organisations and this was helped by their oversight of the different service structures (mechanism – resource). If this lead role ‘was a senior position and … seemed to have some sort of organisational influence’ (SP47, Case Study A), it resulted in easier facilitation of collaboration (mechanism – response). However, several participants at Case Study C also felt that the personal qualities of the COSMHAD lead impacted upon their success in collaboration (context): ‘people who work with [COSMHAD lead] really get on with him, and maybe he brings the skills that he uses with service users to colleagues as well’ (SP4, Case Study C).

Finally, the participants thought that clarity of role as COSMHAD lead as well as for the link worker role was important especially in terms of protected time to focus specifically on COSMHAD issues otherwise other clinical priorities tended to take over: ‘whether that’s 50% of their role, whatever percentage that is, that has to be protected’ (SP51, Case Study D). Otherwise ‘the dual diagnosis practitioner has to sort of take it all on’ (SP52, Case Study F) or

they will get pulled into other stuff, … the link workers that we’ve had in place don’t get time to attend the meetings, that they don’t necessarily have the time to feed back that they try, you know, but that, falls at that hurdle.

SP51, Case Study D

I don’t like closing people, don’t like discharging them without going to every single effort. And if [dual diagnosis lead] can tell me who the Care coordinator is, and they might then ignore me. Probably not, because they’re being malicious because they’re stretched or whatever the reason, I don’t know. I can’t say what the reason is, but just ccing [dual diagnosis lead], just that alone has a really powerful effect and tends to generate a response.

SP13, Case Study C

Working collaboratively with other health professionals both informally and formally (context) brought a number of benefits including increased understanding of how other services worked, and increased effectiveness of communication between services regarding service user progress. Participants described how working together can reduce ‘duality … of thinking’ by allowing staff to bring together their differing expertise and pick up skills (mechanism – response) leading to ‘having really good conversations either for assessment or at different stages’ (SP37, Case Study A) (outcome) (Figure 10). Staff felt that working with those who had expertise in both mental health and substance use had the potential to reduce stigma and increase engagement (outcome). One participant describes a kind of ‘inreach’ approach where they were able to capitalise on the engagement in one service to gently introduce themselves and engage with the person rather than expect someone to attend another service to speak to an unknown professional.

FIGURE 10.

Programme theory 3 mechanism and outcome.

I can think back to times when I’ve not been able to engage with somebody, but the [substance use service] has been actively seeing them because there’s a reason for the person to engage with them and I’ve tagged along on, on, on a sort of like an appointment and said, you know, do you mind me being here? And that’s started that relationship which has been really useful. And just I think it’s also sort of really useful for staff because if you’ve got that sharing of information. And you develop that relationship with [substance use service] workers. You’re more likely to contact them, more likely to engage with them. And that actually gives you that level of confidence that you know what’s going on with somebody because there’s nothing worse than feeling that there’s a whole host of stuff going on with somebody and you’re not really privy to it and you don’t know what’s happening … and that feels risky and itself.

SP51, Case Study D

When staff were able to work collaboratively (mechanism – response) (mechanism – resource), it prevented service users from being passed between services (or falling through the gaps) and ensured a holistic approach to the treatment of their co-occurring disorders (outcome). Staff also agreed that a collaborative approach to care management (mechanism – resource) provided service users with consistency, which allowed them to develop trusting therapeutic relationships and engage with other services that they may require (outcome):

that one worker who has is in some ways been like one of the only secure attachments that they’ve had … they build up that trust with a service or a worker, to the point that they then trusting them to engage with the other services which would be specialist, a lot of our clients are really suspicious and not trusting.

SP6, Case Study C

I mean there have been times when I’ve had clients on my caseload and with certain mental health practitioners, we’ve done sort of joint work and that’s really worked well. But few and far between to be honest and that’s more about … having the time within my job to do all that networking with other people … And I think what happened is clients just get battered between drug services and adult mental health. It’s like chicken and egg, isn’t it? … And I think part of my frustration is even when I say it out loud it doesn’t make sense. But that’s the reality of the services … clients just get batted between one and the other … I know it comes down to resources but doesn’t even save any money. Never mind about what it costs the client, which … should be more important, but it doesn’t even save services money. You end up putting more money in and getting less positive outcomes for clients and it should be about the client.

SP24, Case Study A

Service users and carer participants reported a lack of perceived collaboration in services (mechanism – resource); communication between services often felt fragmented and inconsistent. Service users were often seen by multiple different staff members who did not seem to be fully aware of their case history. This meant that service users and carers had to repeat information to multiple health professionals. As a result, service users and carers felt they were not informed about their care or received conflicting and outdated information which left them feeling frustrated, confused and lacking in confidence in the care providers (mechanism – response). This led to a poor therapeutic relationship with health professionals and disengagement with services (outcome).

A bit too fragmented. They don’t have seemed to speak to each other. You know you hear so many times the communication breaks down between people and between services, between the mental health people between the police between the drug worker … you seem to find yourself saying the same thing over and over and over again to different members of the team. Because you lose your confidence that that that’s actually going to get passed on ….

P36, (Carer) Case Study B

The realist synthesis identified the importance of both top-down and bottom-up collaboration, to help staff feel supported to build collaborative relationships across services. This improved consistency of care for service users. However, service user and carer data highlighted that this is not currently the case in practice, with examples of disjoined working and miscommunications, leading to frustration, confusion and lack of confidence in the system. A lack of time and resources is cited as one of the reasons for this, but participants also recognise that resources are being used in trying to help but not being used in the most effective ways which leads to poor outcomes.

Summary of evidence

Staff working in mental health services could be influenced by how much exposure they have to people with COSMHAD during their undergraduate training. 45–47 This includes being exposed to knowledge and skills in the theory aspect of the courses as well as by exposure to people with COSMHAD as part of clinical placements. Drugs and alcohol may be perceived as a ‘choice’ and not as a health and social problem deserving of help and compassion. Mental health service staff typically saw continued substance use that exacerbated mental health symptomatology as a sign of poor compliance and engagement. Often the focus for mental health service staff is abstinence (which is a requirement for inpatient settings), with limited attention given to the variety of available harm reduction strategies, possibly due to lack of skills and knowledge in this area. In addition, there are differences in the use of pharmacotherapies, ontological understandings of health, understandings of aetiology for COSMHAD, differing symptom classification frameworks and views on client autonomy that manifest themselves in how substance use and mental health services structure service delivery and set outcome expectations for treatment51,54,60 (Figure 11).

FIGURE 11.

Programme theory 4: exposure to COSMHAD pre-registration education.

The realist synthesis identified that it was important that the workforce was exposed to issues related to COSMHAD during pre-registration training. However, in the case study data, there was little coverage of undergraduate and pre-registration training in the focus group discussions and, even after prompting by researchers, not much was elicited in relation to this PT. Thus, this PT was considered theoretically not rigorous.

Case Studies B, E and F briefly discussed training and shadowing arrangements for students and new intakes of staff which were felt to increase awareness of substance use and the harm reduction approach [‘so if we can [teach] our student nurses what the impacts of substances is in our guys. Then they need to be aware of how vulnerable they are as people in the community’ (SP88, Case Study B)] and procedural knowledge [‘training new intakes of students trying to skill them up to do assessments with patients on the wards and then knowing what to do with, you know, whether they need brief interventions, whether need more than that, linking in with specialist services’ (SP67, Case Study F)].

One participant argued against the traditional classroom-based training and suggested that working directly with people with co-occurring disorders could have a have a greater impact:

They’re just new shiny nurses [band 5] just finished their training and then they come to the service, they’ll go into a ward. They’re gonna learn more in the world than I’m going to be teaching them. So if the attitude on one ward is very much about, you know that’s not our problem … or you have a very energised, interested group of staff and dual diagnosis. And yeah, their mindset. So, it definitely comes down to the staff centres, their own experiences. I suppose if there was any training and it’s not just about the training as how does the ward manager incorporate that training into what’s going on in the ward.

SP83, Case Study E

The content and delivery of training for new trainees were very much controlled by external factors. For example, participants in Case Study F spoke about how a national project to increase addictions competencies in trainee psychiatrists (with two work-based assessments in addictions) had led to the development of a network of trainers, university-funded extra places in addictions for senior trainees, and the development of new training materials. In contrast, staff at Case Study B had been asked to deliver training to medical students at a local university but had been asked to halt the work by management within their organisation because there was no funding to support this work, not enough staff time to deliver it and no direct benefit from this work to the NHS trust.

Service users and carers were keenly aware of health professionals’ level of experiences, particularly key workers in substance use settings and care co-ordinators in mental health services. They felt that newly qualified staff were assigned their own caseloads too soon after registration and instead should be given time to shadow existing staff and gain experience of working with clients before working alone. They felt that inexperienced staff could not handle complex or crisis situations, had insufficient knowledge and did not build empathetic relationships with clients due to their lack of experience (mechanism – response).

The problem is with the mental health services … it’s the recruitment of the people who can actually do the job … so for getting a graduate in psychology degree … they need to have six months before they actually get into community … And very often they’re a lot more about the mental illness than the people.

P21, (Carer) Case Study A

… the intensive team came out to see her … who really were totally out of their depth, not their fault, but they were out of their depth, in that she was saying that people were trying to get her and … she’s gonna build a big fire … She was gonna throw herself on to. Something she wanted to do, and they had, they just stopped and looked at her like a rabbit in the headlights. They have no idea what to do … I think they didn’t have the experience or knowledge how to deal with the situation.

P21, (Carer) Case Study A

We have been allocated a key worker and I know that you should give everybody a chance, and what I’m going to say now I don’t mean it in a disrespectful way to them. However, they have allocated a very young person who is new to the service to [husband’s name] and I’m not alone in my thoughts that it should not have been the case with where [husband’s name] is and what’s gone on. It should have been someone with more experience and not somebody who’s literally three weeks ago started working in the service. And that’s horrible thing to say because people only get experienced by experience. However, you do have to take into account where people have been, what journey they’ve been on and where they are going and everything that’s happened. And you shouldn’t be putting … a fresh-faced graduate the job. That, you know, find no objections if that person is there to shadow – shouldn’t be the lead.

P3, (Carer) Case Study 3

Service users and carers described that when they encountered inexperienced staff who lacked basic knowledge, they also seemed to have a lack of empathy (mechanism – response) which made the service user feel uncomfortable and misunderstood, potentially leading to disengagement.

I was accessing their community care for the first time, actually … and the counsellor was there a bit, I’m not sure, that I was referred to didn’t have any … training on drug abuse whatsoever and they like, I was talking about some of the substances that I been using and they were pretty common […] but [they] didn’t know what I was talking about … didn’t have any knowledge of it, didn’t have any understanding of it. And instantly at that point when I realized that the person I was talking to didn’t understand my experience even from like an external perspective, yeah, I like just shut down then. I didn’t feel comfortable talking to you anymore. And I just thought blank it for the rest of the time until I could get discharged. Yeah. So I think like it’s really important for healthcare professionals to be tolerant and in their understanding not to necessarily have an encyclopaedic knowledge, but to like know a little bit about different classes of drugs like different effects that they might have, what kind of effect they might have on people’s mental health.

P17, Case Study C

In summary, while this theory was low in terms of rigour, partly as the focus of the study was not on pre-registration and we did not interview education providers, it was clear from a service user and carer perspective that a lack of confidence and knowledge on the part of a clinician could impact on how they express empathy and ability to engage.

Summary of evidence

The realist synthesis61–67 indicated that when leaders valued the need for continuous workforce development (CPD) for COSMHAD (context) it was supported, and staff could be released from their routine responsibilities and supported to undertake such training. Traditional ‘classroom-based’ learning is important but insufficient to enact change unless accompanied by supervision and practice-based learning (mechanism resource). This can produce lasting changes in staff skills, values and confidence. 61–67 However, attitudinal shifts are less affected by training but can be improved when staff see that working in a different way (such as using motivational interviewing techniques) has a positive effect on service users⁵⁹ (mechanism response) (Figure 12).

FIGURE 12.

Programme theory 5: continuous professional development.

Some form of in-house ‘dual diagnosis’ training was offered in five of the six case studies and typically included a combination of mandatory training for all staff, supplemented by more flexible training and supervision opportunities which were guided by staff need, demand and local and national priorities (mechanism – resource) (Figure 13). This was typically delivered by a COSMHAD lead with support from staff in additional link worker or lead roles. In the one case study where training was not currently offered, this was due to a vacancy in the COSMHAD lead role.

FIGURE 13.

Programme theory 5 context and mechanism.

As in the realist synthesis, it was evident that the success of a training programme depended on organisational leadership recognising the need and ensuring that staff could attend (context). A participant from Case Study F highlighted that training programmes required a combination of ‘bottom-up and top-down support’ (SP60, Case Study F).

[O]ur leaders are very supportive of the development. You know, as much as they can be really and it’s crucial, you know. And cultural change is really quick when managers and more senior people are involved, isn’t it, because the opportunities are created for that to happen? … having the COSMHAD lead within services and at the base level, that still requires support of the team managers, you know giving them you know maybe time to come out for the development days and you know, maybe some ring-fencing some time once every week for them to do any work that they need to be doing to support their teams.

SP60, Case Study F

So having the space, having the time to think together as a team. So, it’s maybe about managers creating that space and that time and that ring-fenced, protected time where people, staff are supported to think collaboratively in terms of developing cultures, of compassionate understanding and skills.

SP20, Case Study C

Staff noted that where organisational commitment leadership context was absent, there was often over-reliance on individual interest: ‘if individual practitioners are interested and feel confident and skilled and it’s on their radar, then they are much more likely to be more proactive’ (SP50, Case Study D) and ‘start to train people and you get proper supervision in place and proper support, it’s a lot easier than to keep it on the agenda and you start to change the culture within the organization and within teams’ (SP50, Case Study D) but this will need wider organisation support.

Staff participants concurred with the synthesis that formal training was important (for increasing staff knowledge) but also needed to be combined with more experiential work-based development opportunities including regular supervision and team discussions to enable staff to embed new skills and knowledge into practice (mechanism – resource). As one participant summarised ‘your basic training is gonna provide some level of knowledge but if you don’t use it you lose it’ (SP15, Case Study C). When they had this opportunity to implement their knowledge on working with co-occurring disorders in a supported, supervised team environment then they would increase their confidence and skills to address the needs of people with co-occurring disorders and feel reassured and motivated to continue the work that they were doing (mechanism – response). If this was lacking, ‘it kind of goes off your radar if you don’t feel like you have your skill set to be able to deal with it’ (SP50, Case Study D).

[I]f I could have one of these guys [COSMHAD specialists] in my team at all time to come and join me lots of times with lots of people and having really good conversations either for assessment or at different stages to kind of check out or motivation or think about interventions or whatever, I think that would be really great. And I suspect that over time if we co-work enough, I would pick up some of your skills … You’d need a really sort of skilled practitioner in your team to really skill you up I think.

SP37, Case Study A

I think as well like just getting a bit of assurance. So for me, I’ve been running groups now since I started in November, never done anything like this before … and I’ve still not had anyone sitting in watching me in a group … and you know in the group lots of things happen and come up that I’ve not experienced before and it’s really difficult to know how to navigate … And just having someone be there, maybe even once a month and just say, yeah, I know you’re doing right. You handled that well, that kind of thing would be really great.

SP27, Case Study A

Being supported and supervised to practise their skills (mechanism – resource) was increasingly important in the current NHS climate, where factors including finances, time pressures, staff turnover and COVID-19 had reduced both the availability and uptake of formal training sessions (context). For example, a staff member working in mental health in Case Study C described how ‘we just have like a weekly CPD slot that we’ve, our consultant set up himself just because training is often lacking these days ’cause of funding and whatever else’ (SP15, Case Study C). As illustrated in the quotes from Case Studies D and E below, these time pressures were particularly felt in inpatient and community mental health where participation in training was limited due to large caseloads and competing priorities. The consequence of these pressures in Case Study D was that the training offered to staff had gradually been reduced by management.

I think also I mean obviously staff are so busy … It’s also very hard to and also to do something consistently is really hard because there’s always something else … I find that in our ward there’s always something that needs you know to be actioned or something that needs to be addressed … even in terms of quality improvement, there’s multiple things going on, so it’s very hard to keep the focus on it, and to keep the interest in … whenever you’ve got a high turnover or patients who all have different needs and different priorities I think that that makes it difficult to keep to keep running with it as well.

SP72, Case Study E

I think over the years it was longer. It was a whole day, then the trust said we didn’t have enough time. In the beginning we had linked workers and we did 10 days of training with them, and we’d have them out of the service. But what happened to us? We recently took on two band 6s that were going to work on dual diagnosis and they weren’t allowed to have the training … So then people don’t want to do the role because they’re not trained to do that. So, if you if you’re kind of given the label and then nobody supports you … here’s the training we’ll deliver and ‘Oh no. Can’t release. You’ve got too many patients. It’s not gonna happen’. And then it starts … why am I doing this job? … It needs to be funded to make it work. That’s the only thing that will make this work. I mean if I leave tomorrow, SP41 leaves tomorrow, this will collapse. There will be no service that will be here.

SP40, Case Study D

While staff recognised the contribution of austerity and national policy to this context, they still felt leadership played an important role (context): ‘they kind of prioritise other training before the dual diagnosis and that’s the realistic answer, isn’t it really’ (SP45, Case Study D) and without this prioritisation from management a culture shift could not be achieved. This is described by participant SP38 (Case Study A) as a barrier to staff developing compassionate values and confidence in their skills (mechanism – response):

[T]his hub and spoke thing became kind of the in vogue way of managing austerity … instead of having a team of people that hold expertise and an ability to respond to people, you have a hub of maybe 1 specialist, if you’re lucky, who kind of, you know, points at all the other people … so this issue of resources, investing in staff and investing in expertise that can be kind of available I think is a potential gap … and it’s a question about how you’re going to shift the culture into operating differently … takes a lot of tenacity, takes a lot of resilience. You’ve got to want to do it … you need people in motivated task force that’s valued.

SP38, Case Study A

Staff concurred that training and experience (mechanism – resource) led to increased confidence in their skills, increased compassion and empathy, and they were better able respond to the needs of individuals with co-occurring disorders (mechanism – response) leading to better therapeutic relationships and engagement (outcomes) (Figure 14). Communication and use of language was seen as really important. Participants described learning ‘the correct language, you’re having a deeper understanding of what they’re going through’ (SP14, Case Study C), ‘a problem-based learning approach … your learning is applicable to the people that that you work with’ (SP11, Case Study C). A COSMHAD lead who delivered training summarised this as:

FIGURE 14.

Programme theory 5 mechanism and outcome.

seeing that change in how people are communicating … and thinking about being flexible with some of the challenges, they’re also telling you … I feel much more able to talk to people about their substance use and no longer feel pressured to looking for solutions and now just have a chat with people and listen.

SP60, Case Study F

When staff were confident in their skills and knowledge (mechanism – response), they felt more able to have conversations with clients and be flexible in developing interventions for people with co-occurring disorders which in turn created better therapeutic relationships (outcome), whereas staff who lacked this confidence (mechanism – response) could shy away from these conversations with their clients (outcome).

Then you get into a situation where people are just scared to ask and you’re not able to start having a conversation with the service users because they always know more than you do. So if you haven’t got the vocabulary, if you don’t know what it is that you’re talking about, service users will pick that up straight away, and you’ve got no credibility … if you’ve got somebody who hasn’t got a clue what they’re talking about or comes with a particular set of opinions or preconceived opinions that aren’t particularly based on anything, I mean, you’re not, it’s not a respectful conversation, really.

SP12, Case Study C

To me, we all need to be confident to start that lower-level conversations … a lot of nurses are very scared of starting that conversation, ’cause they’re not sure where it’s going to go.

SP73, Case Study E

Key to this outcome is the combination of training along with supervised practice to help build skills and experience (mechanism – resource). With experience and knowledge regarding the nature of COSMHAD, staff learnt not to expect quick results and therefore become despondent when service users did not achieve this. Instead, staff become more empathetic towards people’s recovery and thus develop more flexible, tailored responses and better therapeutic relationships (outcomes).

[Y]ou get a lot of the staff thinking that they’ve engaged with somebody they’ve worked with somebody … the staff will say I gave him a lot of leaflets. He said he wanted to stop and he was off for three weeks and we were doing really well doing really well … And the next thing he’s back on the drink again and it’s almost as if it’s like a personal affront to them because he promised me he wasn’t … You’re getting staff concerned, but you’re getting staff frustrated and stuck as well … one of the ways when I start my training is to start and say give me the issues that you’re having with this cohort. Give me the problems. Give me the things that if you had a magic wand, you would fix. You never have to deal with again. And then you line it up against the symptoms of a substance use disorder under the DSM 5. And you’re saying that, oh, gosh … we’ve got the same thing. So, what you’re actually frustrated about as the symptoms of somebody not being well … They get that. But then what do they do? What do they do with that information? Now they have that. So yes, training is an issue there … And having the confidence to do … would come from the training support at work, good supervision, these things being picked up, understanding how this plays out, being able to talk to your colleagues about it, then everybody’s in the one path, you know, having that relationship with the carers … In terms of the staff, I think they do convey that, but … because our staff aren’t trained in them is so they wouldn’t be using that language.

SP83, Case Study E

Supervised practice is exemplified in the COSMHAD model delivered in Case Study B. In this model, staff receive training to deliver the interventions and then the consultant nurse from the specialist team works alongside them to implement into practice. This encourages staff ‘to try and do the work as well alongside her, which is right because we’ve all been trained in in it … So you know, although she’s a bit of the guide clinically, we try and look at our people’ (SP92, Case Study B). The intention of this model is ‘just to give to people that bit of confidence … it’s everyone’s responsibility … not just say oh, just refer to [specialist service name]. And so, people need that confidence to, you know, to be able to do some of the interventions themselves, really’ (SP84, Case Study B).

This balance between knowledge-based training and learning through experience (mechanism – resource) is summarised by SP11 (Case Study C) as becoming the ‘anti-expert’ – having the confidence, skills, knowledge of resources and flexibility to be able to respond (mechanism – response) to the needs of the person with co-occurring disorders in front of them in a way that best suits their needs (outcome).

I think the whole thing about learning so many facts about drugs, I don’t think it really helps my head because I can’t remember facts in a rote manner, but I think interventions in motivational interviewing and the work that [SP15’s] been doing, it’s been really useful because it’s a way of, of probably being an anti-expert and getting people to thinking a different way about their substances and to try and recognize the things as to why they use and why they don’t use rather than being an expert in you know the substances … And being the expert, I think it’s motivation interviewing I think is a bit like the anti-expert. You know the service user’s the expert you know you just ask the questions and it’s more of a formula rather than facts.

SP11, Case Study C

Service users’ and carers’ experiences echoed the PT in recognising that staff trained staff (mechanism – resource) who had confidence in their skills, experience of working with co-occurring disorders and compassion and empathy towards clients (mechanism – response) which lead to better therapeutic relationships with service users and carers (outcome). Service users recognised that lack of confidence in skills and lack of experience of working with people with COSMHAD wasn’t just an issue with new staff, but also in more experienced staff, hence the need for CPD (mechanism – resource).

… some people have worked in the Trust too long and … they kind of forget their role … sometimes how long they’ve been working there, I think they got a bit complacent …even, you know, the nursing staff can become just a bit hardened, really … people who [start] working at the support role level and then like work themselves up to maybe nursing and management or ward manager, when you get to that level you completely lose what you had when you was at the lower end because things are different – you got restraints, budgets and all that kind of thing.

P6, male, Case Study C

Service users and carers described how a lack of confidence and experience in providing care for people with co-occurring disorders (mechanism – response) could lead to poor communication between services; therefore health professionals were not passing on information to ensure that they got effective care (outcomes). Staff who had not been sufficiently trained (mechanism – resource) to be confident in their roles (mechanism – response) often delivered very ‘formulaic’ (P17, Case Study C) services where many of the conversations felt like ‘a box ticking exercise … it’s all very formal and paperwork’ (P13, Case Study C) rather than being more ‘holistic’ (P17, Case Study C), ‘build[ing] a rapport straight away’ (P13, Case Study C) and ‘being mindful because not everybody’s gonna be at the same stage’ (P6, Case Study C). Service users and carers observed that staff who were more confident in using a more tailored approach (mechanism – response) meant that service users were better informed about their care and involved in making choices about care for themselves (outcome). Such staff were infrequently encountered and often described as those who were seen to ‘go beyond the scope of their role’ (P17, Case Study C) when this should be ‘part of the job’ (mechanism – resource).

[I]f we get to the point where more people understand it’s functionality and where the care services are more holistic … I think that would make a big difference, at least in my experience. The good ones have been the ones that sort of go beyond the scope of their role, which they shouldn’t have to do, but the really, really good ones want to, and they do that. But ….

P17, Case Study C

Once service users and carers came into contact with staff who ‘go above and beyond’ [P3, (Carer) Case Study C], they described receiving more joined-up care, resolution of crisis situations and signposting to additional external support that they needed (outcomes). One carer described how a COSMHAD lead (in a non-patient facing role) had taken on their husband’s case (mechanism – response) and ensured that they received collaborative care (outcome) which they had not received from the community mental health team (CMHT). Another carer described how a newly appointed care co-ordinator intervened in an issue which had put her son in contact with the criminal justice system. They had previously sought support, but it was only because this individual staff member ‘really just came on side’ (mechanism – response) that they felt they got support from services (outcome).

… that one worker [trust name] … I think care co-ordinator … I don’t know who that man is, as far as I’m concerned … he wrote a letter … If he hadn’t written a letter and said that [son’s name] had been misdiagnosed, I believe that this lady’s son has been suffered with psychosis for quite a while and he has been misdiagnosed and has been battered backward and forward and I feel that this is a culmination of what happened. He is just as vulnerable … has not had the services … don’t know why he did that because, you know what, nobody else [at trust name] was prepared to do that … He looked through his notes … We had a couple of months with the early intervention, and he was [son’s] care co-ordinator. And he really just came on side, tried to look after [son], tried to look after the family.

P38, (Carer) Case Study A

Service users and carers felt that skilled staff (mechanism – resource) were those who had knowledge of mental health and substances, showed commitment to assisting those with co-occurring disorders and who communicated assertively and effectively with their service users (mechanism response). Service users often felt that they could not assertively communicate what they needed in terms of contacting services (e.g. in times of crisis) and engaging with treatment. They described how staff who communicated with them proactively were ‘like a safety net’ (P8, Case Study E) and supported them to remain engaged with treatment and work towards their goals even when they experienced setbacks (outcomes).

It’s like a safety net … if you’re failing and you reach out, you know, to somebody in the community mental health team and they say they’ll ring you back and you wait and you wait. And you don’t get that full call back. You’re unlikely to reach out again, you know, to try and access that help … but whenever you make that phone call and you do get speaking to somebody and, you know, you tell them that you need help … they get back to you … You’re more likely to reach out again … for me. Impulse is a big thing … but just being able to kind of reach out and say, you know, listen, I’m solo, this is what I feel like. Just knowing that there’s somebody professional there and that they’re going to get back to me can be enough to kinda take that impulse away.

P8, Case Study E

In line with the initial PT, service users and carers felt that when trained staff (mechanism – resource) had confidence in their skills, empathy towards clients and were sufficiently experienced (mechanism – response), then they developed better therapeutic relationships with clients (outcome). However, service users and carers suggested that staff like this were relatively rare in their own experiences, which highlights the need for continuous training and supervision to be integrated into services at a leadership level to ensure effective treatment for co-occurring disorders (context).

… but if it just makes them just treat somebody else. The child, lover, husband, daughter, son, Auntie, uncle, differently. In my heart it will be worth it to me because I can’t do anything else. Yeah, but if I can help somebody else, would they just have the milk of human kindness? Follow the policies and procedures they got because I don’t think that much really needs to change. They just need to follow the policies and procedures. Just follow them.

P38, (Carer) Case Study A

Summary of evidence

The SELFIE (Sustainable intEgrated chronic care modeLs for multi-morbidity: delivery, FInancing, and performancE) framework proposes that supportive leaders with clear accountability, visions and ambitions for integrating care are key to successful integration for multimorbidities. 68 The realist synthesis identified that leaders supporting the development of effective COSMHAD services (context) who took action to develop the listed policies, processes and procedures (mechanism – resources) led staff to feeling supported in taking a whole person approach (Figure 15).

FIGURE 15.

Refined PT6: leadership.

Staff participants in all six case study sites emphasised the importance of having visible leadership for co-occurring disorders (context) to advance the implementation of integrated treatment across their organisation. They identified that these leaders needed to exist at different levels of their organisation including at a senior, strategic level and operating at a more clinical management level.

Supportive and engaged senior leaders (context) were needed to keep co-occurring disorders on the organisational agenda (mechanism – resource) (Figure 16). Participants in Case Study F made several references to a supportive director who was active in promoting co-occurring disorders, but they were aware that this often depended upon the interests and priorities of the individual in a leadership role (context). A participant from Case Study E describes how senior managers need ‘clout to get things done’ (SP84, Case Study E), with this authority allowing them to get co-occurring disorders higher up on the organisational agenda and formalised through measures such as pathways and mandatory quality awards (mechanism – resource). These measures implemented by leaders were felt to emphasise the importance of co-occurring disorders to staff (mechanism – response), leading to better engagement with patients (outcome). In Case Study D, staff described support from leaders as ‘patchy’ and ‘not lead from the top down’, and without senior level leadership ‘it relies on clinicians doing it, if they choose to do so … unless you have a lead in every service at a senior level that says we must do this, it doesn’t happen’ (SP40, Case Study D).

FIGURE 16.

Programme theory 6 context and mechanism.

I have been in my post for about 6 years now and I’ve had different managers and not one of them has had any clout to get things done. I don’t have clout and I think there needs to be. It’s not that people are disinterested, it’s just that it’s not high enough in the agendas … and nobody is taking the lead or pushing at medical level, at psychology level, at clinical level … there’s nobody taking the lead and pushing for this. And yet, if you go and ask staff in the wards what’s your biggest worry, it’s these patients. If you ask staff in the emergency departments or the cardiac ward and the General Hospital, it’s these patients … but it’s not given priority at strategic level. It gets a couple of paragraphs in documents. … But in terms of dual diagnosis, something like that might be good where you have quality awards for it […] there’s that sense that this is important and therefore staff they’re forced to see how important is, they also see the benefits of that that we engage with patients that they stay with us, that they work with us better.

SP84, Case Study E

Senior leaders (context) were needed to achieve a

cultural shift in an organisation [mechanism – response] that looks at the patient from the point of entry and says ‘right, you have mental health and substance use issues’, we will work with substance use agencies from day one to manage this individual

and where ‘staff are using those skills that they … have been trained in with that client … because that’s the only way you’ll get change with individuals. If they’re constantly supported’ (SP40, Case Study E).

In contrast, staff felt that when senior managers did not have a sufficient understanding of people with co-occurring needs (context), they would create strategies, design services, and set outcome targets that were not feasible for staff and people with co-occurring disorders (mechanism – resource). This was described as creating additional pressures on both staff and service users leading to reduced staff motivation in their roles (mechanism – response) and reduced engagement with services (outcome).

I think that if the people at the top don’t know what their population group looks like and what their needs are … I think people need to have all of that knowledge in order for them to be able to actually strategise and put the right services in place … [for] people to be able to do their job, they need to be motivated. They need to have an interest in in what they do, because otherwise why would they turn up for it?

SP48, Case Study D

… the most difficult battles I’ve had have been with management who don’t understand co-occurring needs. It’s well, they do understand, but maybe not to the level or the acuity […] some managers you can come across and maybe they don’t have that understanding of trauma as well as you know co-occurring needs … [they’re] putting pressure on the caseworkers to maybe fit things into boxes or tick boxes … have you not got an outcome for that person yet? … But then when you look at the actual caseload of people, that’s never gonna be achieved, […] if you don’t have the right kind of goals and they’re unachievable then no one does well in that.

SP11, Case Study C

The other level of leadership that staff discussed frequently was the existence of (and resources for) dedicated clinical leaders for COSMHAD. These were typically positioned at a clinical management level and generally took responsibility for training, local strategy and provided specialist supervision (context). Four of the case studies had such COSMHAD lead roles who have an organisation-wide remit and these posts had been in existence for many years. Often these posts were held or created by people who had a long-standing interest in co-occurring disorders and had been ‘a sort of a pioneer for a long time’ (SP61, Case Study F). Staff described the key role played by the clinical leads were all ‘about relationships’ (SP10, Case Study C) which were developed by these individuals (mechanism – resource) which encouraged joint working from staff in other teams (mechanism – response) and increased accessibility for people with COSMHAD. Participants also described how these leads ‘let us know what was being discussed at a higher level … because it just doesn’t get communicated’ and responded to wider policy changes. As the quote below demonstrates, key to the success of these leads was someone who ‘leads by example’ (SP16, Case Study C) by putting the skills they taught to staff into action, and this helped staff feel motivated and supported to continue in their work and ‘stand a chance of making sure … support is in place’ (SP16, Case Study C) for service users.

However, it was clear that staff in these lead posts often felt it was challenging to maintain organisational commitment to co-occurring disorders without support from senior level leaders. There was also acknowledgement across all staff that because many of these posts had grown up from individual passion and interest to drive things forward (context), if these roles were taken away or individuals left the organisation then there would be no one to keep co-occurring disorders on the organisational agenda and ‘the whole system would fall apart’ (SP10, Case Study C) and ‘we will lose the drive for change’ (SP13, Case Study C). This was confirmed by staff in Case Study A, where the trust COSMHAD lead post was currently vacant and had been for some time (context), and staff described the support they were now lacking including someone to lead on co-occurring policy, strategy and training, assistance with case discussion and upskilling for team leads and the accessibility of ‘someone who people can just pick up the phone and call’ (SP28, Case Study A) (mechanism – resource). Participants described how when the post had been in place it ‘helped people to hold the work in mind in a more co-ordinated way … it held it in people’s minds and also possibly the organisation’s mind’ (SP37, Case Study A) (mechanism – response).

… what I reflect on a lot of the time and feel really concerned about is that I feel at times I’m an invisible voice in an organization. And although I’m quite vocal, but I actually represent these service users so I can see why they don’t get a service … you look at all these other services and will say, oh, we need … a hub service, we need an assertive outreach team – here’s your money. So, you’ve got the most complex need group of people with substance issue, some mental health issues, but actually you can’t have a penny. We just expect you to do it. So what happens is they’re difficult to engage, so they get brushed off.

SP40, Case Study D

[T]here was a time when, you know when all the services went out of the NHS and everyone’s budget was being so squeezed and [former and current trust dual diagnosis leads] basically between them … kept dual diagnosis alive and they were passionate advocates for it and … because I think it’s well part both of their own personal commitment and their knowledge and skills and their networks with dual diagnosis. But I think they just negotiated for their time and budget.

SP66, Case Study F

[W]here it comes from the ground up, that’s brilliant and it works well, and managers don’t obstruct it, maybe don’t support it, but they don’t obstruct it. And it works brilliantly. But if one key member of staff leaves, that folds, it just collapses.

SP84, Case Study E

Staff described how, when they were working alongside their NHS organisation’s COSMHAD lead or specialist team (mechanism – resource), they felt reassured that they were taking the right approach towards working with service users and that they also became more confident to listen to people with COSMHAD and work in a way that was responsive to their needs (mechanism – response). This was described as ‘empowering other professionals to understand what success looks like … it’s just about having a conversation and forming that relationship and seeing where patient’s are at’ (SP52, Case Study F) and for service users, interacting with more empowered staff could end the ‘vicious cycle that they weren’t engaged with services because they don’t think they’re being listened to or particularly relevant’ (SP3, Case Study C) (Figure 17).

FIGURE 17.

Programme theory 6 mechanism and outcome.

[W]e have got [dual diagnosis service name] worker that … will tend to be the person that we are drawn to, to discuss things. And we see her as a bit of our expert, although I know she likes us to try and do the work as well alongside her – which is right because we’ve all been trained in it. And I think we naturally just looked to her as our expert in that field, really … And I’m just getting the person involved, you know, because again, they’re the experts in what they’re doing and how they’re feeling and. So, you know, although she’s a bit of the guide clinically, we try and look at our people that are using really.

SP92, Case Study B

[A]nd [trust dual diagnosis lead] for example … is sort of my point of contact from the service to come along to sort safeguarding meetings and complex MDTs etc. And having that even if you know we’ve not got that specialist in that area engaging directly with the client, but to offer their advice and I suppose trying to as much as we can to manage that risk and when, when it happens for very complex cases, to be honest. So, I have had that recently, which was really, really helpful. And I think to offer reassurance that you are doing or you can or if there’s any other ideas to come forward about harm reduction that I wouldn’t usually think of in terms of, you know, substance use in particular. And so yeah, just having that sort of end-point to contact, it’s been really useful.

SP1, Case Study C

However, support was also needed from senior leaders in the organisation, and this support was required to help embed integrated care into routine practice. When this support was absent, there was too great a focus on key performance indicators (KPIs) over the needs of service users, and unclear structures to enable them to access the support needed from other teams. Under these circumstances, staff often had to advocate hard to get the support and resources service users needed and this often served to demotivate staff and make them despondent in their work (mechanism – response) which in turn demotivated service users (outcome).

That then becomes a little bit of a disjoint and the different services sort of battle against each other, and that doesn’t help because actually it’s not my client/your client. The client needs a service. We all look at them and through a different lens. So, there’s just work with the client. It’s not my statistic and I want to take that for my commissioner, but that’s what it comes down to. It’s like we all need to hit certain targets … the client has less, but we expect more from them and from a practitioner’s point of view you have to advocate quite hard for your client to get what they you shouldn’t need to advocate for. They should just get it and that can get you despondent let’s say that.

SP24, Case Study A

It’s quite difficult working in a system where that system is broken and where resources are less and where you can see the clients have less. You have to be really mindful of your motivation around the work you do. I’m really lucky I still really enjoy the client work and that’s what keeps me here. But to be honest it’s not an easy job and it’s not getting easier, and I don’t want to be negative too much about it. I’m just being realistic. The thing that keeps me here is I still like that connection with the clients. Well, my colleagues as well.

SP24, Case Study A

The service users and carers did not specifically mention leadership, not surprisingly as this aspect of their care would be invisible to them. However, they did mention that people in dedicated COSMHAD roles had gone the extra mile for them, and they appreciated this advocacy, compassion and commitment. The data indicated that the COSMHAD role was a lynchpin of good service, but they were stretched and given little resource to sustain their role. This led to service users having inappropriate support that was disjointed and not person-centred (e.g. KPI focused).

Summary of evidence

A critical aspect of integrated care is communication and effective relationships, both between colleagues in the same organisation as well as colleagues in external agencies. One way of creating and sustaining this is by providing formal opportunities to meet. By doing this, it allows for communication and relationships (context) that lead to networks that allow staff from different teams and services to work collaboratively (mechanism – resource) (Figure 18). The evidence from the realist synthesis suggests that these networks work best when they are formal, structured, sustained and responsive to the complexity and variety of needs experienced by people with COSMHAD. 69 There are numerous examples in the literature including steering committees,56,70 staff learning groups,71 communities of practice,72 collaborative case conferences73–75 and large multidisciplinary networks such as those in Leeds44 and Manchester. 76 Studies from both Europe and the USA found that these formalised networking opportunities for COSMHAD (context) led to opportunities for multidisciplinary peer support and ethos. 73,77–79

FIGURE 18.

Refined PT 7: formalised networking opportunities.

Co-occurring disorders networks existed across four (Case Studies C, D, E and F) of the six case study sites. In three sites (Case Studies D, E and F), there were networks which brought together designated COSMHAD champions or link workers who were positioned in different teams within the trust. Two case study areas (Case Studies C and F) also had more local level networks (at borough or local authority level) with multidisciplinary membership for case discussion (with Case Study D currently in the process of setting this up). Case Study C had a trust-wide network which was open to any professional within the geographical area who was working with people with co-occurring disorders. These networks provided multiple different functions including planning, training, information and resource sharing, dissemination of policy, case discussion, supervision and reflective practice. In line with the literature, the participants felt that these networks were most successful when they were formalised and sustained with support from management within each trust (context). Where this formal support was not present, it was challenging to maintain staff attendance and the networks had limited power to push forward the agenda of integrating care for COSMHAD (mechanism – response) (Figure 19).

FIGURE 19.

Programme theory 7 context and mechanism.

[O]ne of our difficulty is for them to find time to attend … the coordination of the ward managers to ask the staff to attend those training and so we are working with the ward managers as well to allow staff to attend those and encourage their staff and themselves as well to attend the said opportunities for any training or any conversations or forum that they can attend too.

SP57, Case Study F

Staff discussed several benefits of having these formal network opportunities (context) including bringing mental health and substance misuse services together (mechanism – resource) (see Figure 19): ‘and all those are different agencies and they’ve come together and that also helped it to be in peoples’ minds … reminding people what they do, how we work. So that would get filtered down the system’ (SP28, Case Study A) (mechanism – response). A key aspect of these networks was described as ‘forming an alliance’ (mechanism – resource) which led to ‘collaboration’ and ‘knowing how to navigate the system’ which was then viewed as spreading beyond those attending the network to colleagues within their teams: ‘so you can see how the collaboration and that knowledge and skills set and even the resources is spread so it can have a bigger impact, so you don’t necessarily need a bigger dual diagnosis team’ (SP10, Case Study C). Staff also felt that networks created the opportunities to raise awareness of co-occurring disorder policy (mechanism – resource) which could lead to more consistent policy implementation and thus more consistent care (mechanism – response).

Finally, staff also discussed how these networks gave them the opportunity for specialised supervision, case discussion with peers and reflective practice (mechanism – resource). This was often something they were lacking in relation to co-occurring disorders in their everyday roles, and that this provided them with emotional support, reassurance and confidence to continue pursuing effective care for people with co-occurring disorders (mechanism – response).

[T]he dual diagnosis network … that I go to, I find that that is more almost like reflective practice. For me anyway, I find that that really helps me emotionally with some you know dealing with challenging the system ’cause it is quite difficult to constantly challenge the system for your clients, so … as a way of cleansing myself.

SP2, Case Study C

I kind of see the Champions meetings as being a little bit like supervision and so although we might not discuss cases necessarily, it is about talking about … how things are on the ward or how are you managing with these new things … so I think that probably is the height of the sort of specific dual diagnosis supervision.

SP72, Case Study E

Staff described how the relationships built in these networks could lead to more collaborative working (mechanism – response) which would ensure staff had ‘a good overview of their [service user’s] care’ allowing them to ‘provide a good level of care’ (SP28, Case Study A) (outcome) (Figure 20). There was a sense that building a network amongst staff in different teams would mean staff were more likely to encounter a familiar face when co-ordinating care and have the support they needed in complex situations (mechanism – resource). As shown in the quotes below, there was a general sense that over time this could lead to more collaborative and consistent care for individuals with co-occurring disorders (outcome), but the challenges they experienced in developing and sustaining these networks in the short term (context) meant that they were uncertain if these long-term outcomes would be realised.

FIGURE 20.

Programme theory 7 mechanism and outcome.

I suppose what they get out of it is the sense that they’re not on their own, there’s lots of them if they happen to be on a ward and they see one of their other champions that they’ve met at the meetings, there’s a connection there.

SP83, Case Study E

The networks largely functioned at a staff level offering opportunities for peer support and improved care co-ordination for individuals with co-occurring disorders, and so for this reason, it is perhaps unsurprising that there was little discussion of these networks within the service user and carer interviews. Only one carer had come into direct contact with these networks. As previously discussed, the COSMHAD lead at Case Study C had become involved in their husband’s case in order to facilitate a multidisciplinary working group to respond to their needs, but the carer acknowledged that this was outside of the staff member and group’s official role. P3 described how they felt the multiagency group had given them a voice and brought disjointed services together. This reduced the pressure they felt to be ‘constantly firefighting’ to co-ordinate care (outcome); however, they did also express some frustration at some members of the group ‘playing lip service to it rather than making a difference’ (context) and that it was the commitment of the COSMHAD lead that was making a difference to their husband’s care.

… we’ve had multi-agency meetings which have involved CMHT, [drug and alcohol service] and [recovery housing] … I think it’s giving me more of a voice … we’ve obviously got one of the Commissioners for mental health … her being on the call has meant that I can vent my frustrations … the biggest problem is that … the different services, don’t talk at the moment and it’s very, very disjointed. And so this has given an opportunity for the relevant services to all be in one place at one time, which … doesn’t happen otherwise … before it was constantly me, firefighting, picking up the phone, trying to speak to the right people, trying to, you know, getting one piece of information from one team, then getting a piece of information from one team and it was me that was acting as the conduit … […] I think the one person that’s made the biggest difference to us, to me, to our family, to [husband], has been [dual diagnosis lead].

P3, (Carer) Case Study C

However, what was not captured in the initial PT was the important role that formal peer networks and groups played in supporting service users and carers. The majority of service users accessed some sort of support group external to the NHS and they described how these groups were valuable in their recovery as they reassured them that others were experiencing the same things, reduced their isolation (mechanism – response), presented new strategies for approaching problems related to their mental health and substance use, and held them accountable to work towards their goals (outcomes).

… because you’re not on your own and people are experiencing what you’re doing to yourself … that is like reassuring for you.

P11, Case Study B

[F]rom the mental health side of things, it was absolutely beneficial in terms of reintegrating, feeling safe, feeling, trusted, feeling, able to talk openly, really openly and really honestly and feeling welcome, feeling valued, feeling like other people understand what you’re going through from the substance … you can get feedback and cross sharing and people talking about their experience or how they’ve been in a similar situation and how have they managed to navigate through it, whether it be positive or negative.

P6, Case Study C

However, participants were also clear that these kinds of external groups should not be seen as a replacement for one-to-one support with a therapist or key worker (context) and that these groups would not necessarily suit everyone depending on a range of factors including the stage in their recovery, treatment preferences and desired level of anonymity. For example, one participant described how being held accountable by an external group for their behaviours was not helpful at their current stage of recovery and instead created additional stress and pressure. Service users also acknowledged that it could take time ‘to find a group you feel comfortable in’ (P20, Case Study C). For example, a number of LGBT participants had sought out LGBT-specific groups because ‘you don’t have to like explain things or challenge things or explaining that sort of additional like mental effort and emotional labour to try and like challenge those things’ (P17, Case Study C) and ‘within the LGBT [community], the different drugs of choice like there’s a lot of people on crystal meth … if you’ve got a common understanding of where we’re coming from this community in terms of how society sees that, it just helps a little bit more’ (P20, Case Study C). Similarly, one participant described how she found the groups in a residential rehabilitation programme ‘a man heavy environment’ and so she set up a women’s group to allow women to ‘be vulnerable’ and discuss with other women.

And I think one-to-one support is important for couple of reasons. So for example the issue with a group setting is … they are useful because you’re getting more than one person’s opinion and you know there’s social aspects to it … but with one-to-one you can make me feel more comfortable talking about things that you couldn’t talk, and … to sort of focus on you and, you know, what’s going on with you and work through that … And … the lack of anonymity it is a problem, especially [with]… specific substances.

P16 Case Study C

Carers also described the benefits of having these kinds of external groups with other carers as they provided them with opportunities to learn more about their loved one’s addictions, to get time and support for themselves, and to talk openly about their situation and hear from others in this situation (outcome). In addition, some of these carer groups had taken on advocacy roles and were using them as a platform to feedback their service needs to the NHS and local authority.

[Y]ou go to these groups and you see other people who are just normal people and the same things happen to them and you hear their stories. And that’s exactly what’s happened to you … what they’re going through, and you realize, you know, you’re not alone. You can share experiences and, you know, you get affirmation that it’s not your fault … You also get practical advice and we can just support each other as well … and you become more empathetic … And yeah, it’s just mutual support and help.

P24, Case Study A

The synthesis and evaluation findings concur that formalised and structured opportunities for collaboration (such as COSMHAD networks) led to increased awareness of other networks, and peer support. However, the practicalities of attending these groups (e.g. cover for the ward) were often difficult and required senior level management prioritisation and buy-in. Carers and service users also highlighted the benefits they got from group peer support but were also clear in outlining that this should not be a replacement for one-to-one clinical support. Since the research focused on clinical services, additional programme theories to explain peer support groups, such as those that exist outside statutory services for service users and carers, were seen as out of scope, and therefore additional PTs were not developed.

Summary of evidence

The evidence from the realist synthesis suggests that when committed and accountable leaders implement (context) care protocols that clearly describe the co-ordination of care from initiation of care through to referral/discharge44,56–58 (mechanism – resource), this helped staff feel supported in their roles47 and enabled them to use their skills and knowledge (Figure 21).

FIGURE 21.

Programme theory 8: care pathways.

All six case study areas provided their COSMHAD policies as part of the research study, all of which included a diagram and description of their pathway for people with COSMHAD. However, except for the trust COSMHAD leads, awareness of these policies and pathways by the staff was quite low (context). This was because there was not a sufficient drive to promote their policies, and an acknowledgement that ‘it’s how those policies are actually then put into practice and made relevant is a difficult thing’ (SP3, Case Study C) (mechanism – resource) (Figure 22).

FIGURE 22.

Programme theory 8 context to mechanism.

It wasn’t widely advertised, or published and the chap that wrote it left a couple of months after he did it. And what it lacks is a strategy, to follow it up, so at the moment it just feels like a document that sits there going this is what we should do. But there’s no kind of thought or plan of how it’s going to get done and what’s required to fulfil the policy so make of that what you will.

SP15, Case Study C

I mean, I think the pathways are quite clear and we’ve got policies and protocols in place and we’ve got a lot of staff if compared to other trusts, in terms of dual diagnosis. So I would say from my perspective that they’re they are clear but actually getting that message in across is the challenge.

SP67, Case Study F

The extent to which COSMHAD pathways were implemented within a trust (mechanism – resource) was felt to be dependent on funding, commissioning priorities and senior level support (context) (see Figure 22). As one participant described: ‘it depends on what the flavour of the month is as to what the push is’ (SP28, Case Study A) and ‘drugs and alcohol is a poor sister, mental health is another poor sister and dual diagnosis has been locked up in a shed at the back of the garden … dual diagnosis is definitely not going to the ball’ (SP38, Case Study A).

In contrast, ‘cultural change is really quick when managers and more senior people are involved’ (SP60, Case Study F). SP40 from Case Study D described how they had created ‘very clear pathways from start to finish’ and ‘every policy going’ but ‘none of it is followed … it just doesn’t happen’ ‘except where we force our way in’. SP40 felt the ‘biggest issue is that this is not a commissioned service’, ‘they don’t have anybody in every part of the organisation … committed to it’. This was linked to a lack of knowledge of co-occurring disorders among senior leadership: ‘I think people need to have all that knowledge in order for them to be able to actually strategize and put the right services in place’ (SP48, Case Study D). Participants in Case Study F similarly had developed multiple policies around COSMHAD and were in the process of ‘trying to put this all into one framework … the dual diagnosis pathway … it’s really us coming together and pulling all this thinking to one place so that we see where the gaps are’ (SP60, Case Study F). Much like Case Study E, staff were clear that ‘we don’t want this to be something sat on a shelf or in a file in the system and we try to build this into how we work naturally’, but they acknowledged that ‘there’s always hardly any new money in the system to create new posts and services. It’s quite often really about, you know, bending what we have to reduce this gap’ (SP60, Case Study F).

Researcher: But if the care pathway wasn’t there at all, what difference would that make at the moment?

SP40: I think it would lose the voice completely. So at least I’ve got something to battle people with. So I can say when it doesn’t work, I say, well, you were told to do this, this and this. So it is there to provide some safeguard. And it gives us something when we have inquiries or investigations to go to and say. But why didn’t you read this? And so I think it is necessary and I think it’s gaining more ownership of that and I think a lot of it will be when you have new starters that have or integrated into better practice.

SP40, Case Study D

In addition to these challenges in developing specialist pathways, staff felt that on some occasions existing pathways (mechanism resource) could hinder them from providing the support people with COSMHAD needed. Staff often felt constrained by the remit of their commissioned service meaning there was uncertainty about where responsibility to provide care (mechanism – response), thus limiting the care provided (outcome).

… because my contract ended with them and I came back to this team and then the manager was like we’re not commissioned for this anymore. We’re stopping it. And then it all stopped. And then they went back to being like, what do we do? We’ve got this knowledge that we can’t use anymore, but we’re not allowed to assess these people.

SP27, Case Study A

Staff described how accessing both mental health services and substance use services could be challenging for people with COSMHAD because often services present ‘one-size-fits-all’ which could lead to those who were not eligible becoming despondent and dropping out (mechanism – response). Clear pathways were therefore seen as beneficial in facilitating referrals between services (mechanism – resource), relieving the anxiety of having to access a new service (mechanism – response) and ensuring that they could move between services (as needs dictated) in a more co-ordinated manner (outcome) (Figure 23). Staff at Case Study D describe a project of active referral to ensure support for people who were not ready to access therapy or who had not followed through after assessment due to anxiety about taking the first step to access a service.

FIGURE 23.

Programme theory 8 mechanism and outcome.

It’s very much a one size fits all, isn’t it? It’s like we go through the spot [substance use service] bit. Then you’re gonna get filtered out and various things. If you don’t engage or you don’t fit into any mental health that wanna know ’cause, he’s still drinking. So you get despondent. Drop out and then go back again.

SP30, Case Study A

Staff also discussed some barriers which could hinder the functioning of pathways (mechanism – resource) and thus prevent service users receiving comprehensive care which met their needs (outcome). This included not aligning with pre-existing KPIs within their services, risk assessment requirements and multiple IT systems which prevented the sharing of patient data. As each service required their own, data this meant that people had to repeatedly describe their often traumatic experiences to a range of health professionals. This could be avoided if there was better data-sharing infrastructure within a local area. Participants also discussed accountability and responsibility, highlighting that it was not always clear who was responsible for ensuring that service users received care at both a practitioner and an organisational level.

[W]ho almost takes the responsibility for what happens. For me, putting them forward for a detox. The feedback I get back is, well, who supports them during this and who supports him afterwards and who supports them in the meantime leading up to it. So, who takes that responsibility of detoxing them? But it’s a mental health issue because … that’s why he’s using or whatever it is, just an ongoing circle and it is just so frustrating and disheartening. And that’s for me, as the professional never mind the client who I’m trying to go back through and let him keep them updated as to what’s happening, particularly for the one I’ve got in mind. I’ve got his daughter, bless her who’s always trying to figure out what’s going on and try and take that responsibility herself, you know, driving to appointments, do everything they can, but it just doesn’t work. We both for the mental health services and for me, one of us has to say that we can take that responsibility, but we’re not actually allowed to because of the policy and procedure of it. It’s just really difficult, obviously.

SP27, Case Study A

Service user and carer participants had limited perceptions that local care was integrated and co-ordinated. Service user participants described the services they received as ‘definitely not integrated … I think they’ve been entirely separate’ (P17, Case Study C). There were exceptions to this. Two service users from Case Study F were being treated by an integrated alcohol care team based at one of the city’s hospitals (mechanism – resource) and they described this as the first time their alcohol problems, physical and mental health had been considered in an integrated way. They reported that this had increased awareness of the comorbid nature of their conditions, improved mental health support, and a sense of relief that they had consistent, continuous support available (outcomes). Similarly, two carers (P24, Case Study A and P29, Case Study B) described how their adult children had recently been given places in residential programmes run by local charities (mechanism – resource) after being made homeless and being released from prison respectively. While both carers appreciated that this was not a long-term solution, there was a sense of relief in knowing that their loved ones had consistent and integrated support for all their needs (outcome). P29 (Case Study B) presented this in contrast to the ‘up and down and up and down’ of accessing support in the community.

[T]he difference between going to the integrated clinic, [name hospital removed]. I don’t know what will happen in the future, but … that ongoing support … eventually takes the pressure off … Because of the nature of it, it’s the first consultant ever that I’ll probably be going back there at least every six months or so. That’s the rest of my life. So that does take some of the pressure away … You know, it’s not all physical. It’s not all mental. I’ve got a combination of the two and the integrated clinic.

P3, Case Study F

Service users generally reported poor communication between services. They felt that they often had to repeat themselves and received information from different services that was incorrect or outdated (mechanism – resource). As a result, service users and carers reported losing confidence that information about care and referrals would be passed between services and thus lost confidence in the services themselves (mechanism – response). Service users and carers described having to frequently collate information and follow up care with various services themselves. This placed a significant burden on them (particularly when service users were struggling with their mental health symptoms) and they felt this would be reduced by having integrated and co-ordinated care (outcome).

A bit too fragmented. They don’t seem to speak to each other. You know, you hear so many times the communication breaks down between people and between services: between the mental health people, between the police between the drug worker and … even within the same team you seem to find yourself saying the same thing over and over and over again to different members of the team because you lose your confidence that that’s actually going to get passed on ….

P36, (Carer) Case Study B

Sometimes, I’m on the phone all day … because they’ll say you, you need to speak to this person. You speak to that person or you need to go to A&E if you feel suicidal … I don’t feel suicidal, I said, I just don’t feel great, I said this medicatio isn’t working. Oh well, I’m afraid I can’t deal with that. Shouldn't be done by psychology, try phoning psychiatry. The psychiatrist never called.

P15, Case Study C

… when I’m not feeling well, everything is exhausting … simple things like send an email or the phone or things that I guess that for most people seem simple … Those things are really hard and like if you’re struggling … The thought of picking up the phone is like people are gonna dread it. We’re not gonna do it. Like managing those kinds of things … get very difficult very quickly … it’s definitely something that I’ve struggled with especially like because there tends to be a lot of like admin … like the amount of forms to fill in … there are then many obstacles in your way to being able to access these services, not just so within forms, but while completing online application or making phone calls or a bunch of stuff that they’re very challenging problem to solve … if it’s challenging systemically, imagine how challenging it is for the individual.

P17, Case Study C

Where this communication was lacking (context), service users and carers described experiencing inappropriate referrals, immediate discharge from services and service users often falling in the gaps between services due to a lack of follow-up on their mental health from services (mechanism – response). Service users often felt that when they were discharged from mental health inpatient settings or reached the end of their allocated number of therapy sessions, they were left without any support. Some participants described losing faith in NHS mental health services and pursued private health care as an alternative. Among others, for whom private health care was not an option, this led to relapse and worsening of mental health symptoms (including suicide attempts) (outcomes). Participants attributed this failed communication (mechanism – resource) to a lack of individual staff members ‘to be the central point who co-ordinated care’ [P32, (Carer) Case Study B] and who were ‘willing to, you know, join the dots’ (P10, Case Study E) and ‘if these services were a little bit more holistic and focused on, like, what kind of health problems does someone have, what kind of identity does someone have, like in what ways are they marginalised’ (P17, Case Study C).

I’ve been referred to like [name of mental health service] and they said it was too complex. Didn’t really understand that. I went to psychologist, and they said there was no behavioural problems with me and he’s discharged me … I don’t understand why I didn’t have the psychologist do a bit more work with me and she literally did about four assessments and it was about like history and she said, oh no, there’s no behavioural issues. But we will be seeing you again, we will be referring you back. So GP … I had to go and refer back to the access team to go back to some you know psychology services to go to psychiatry to wait on the waiting list. So it’s just been … like … four years just to get this kind of level of service [for] dual diagnosis or, you know, and the trust and the drugs team and the doctors actually working together.

P15, Case Study C

It just made me kind of distrust the NHS really and think that this is this isn’t something that I can use that have to go private again, but luckily, I’m in a position to go private. Not everyone is, so I think it was more just a disappointment in the process of going through the NHS.

P20, Case Study C

Participants held almost universally favourable views around greater co-ordination of care pathways and services (mechanism – resource) and this was frequently described in terms of a ‘safety net’. The key characteristics of this safety net were having a key staff member within services who they knew they could contact in times of need and who would provide referral and access to services they needed (outcome). Another key defining characteristic of the ‘safety net’ was a service that would regularly and proactively contact them to ‘check in’. Service users felt this would be invaluable when they were experiencing poor mental health as it would reduce their isolation and reassure them that someone cared and was attentive to their needs (mechanism – response). As P17 described ‘your life can be chaotic and without much structure … it’s really important to manage things carefully, because just pull the rug out from someone, they can feel like all the other times in their life that they’ve had the rug pulled out from under them’. Participants felt this could prevent crisis and relapse and would be more cost effective than allowing them to fall into the gaps between services and relying on blue light and primary care referrals (outcome).

… like a quarterly review … just a quarterly check in where you know you get an appointment a couple of times a year … particularly when you’re really isolated, if that reassurance that someone cares enough to check in with you, and if you’re not OK, you have the opportunity to let them know and then like whatever it is that he is can be arranged on your behalf and by someone with the skills to do that, and its’s not placing all of your onus on the unwell person to do that for themselves.

P17, Case Study C

It’s picking up the phone and knowing that I can get hold of her care co-ordinator, I can actually get hold your care coordinator and speak to the psychologist. You know, if we need to do something now, for instance.

P21, (Carer) Case Study A

The realist synthesis and evaluation identified the importance of collaborative care pathways; however, the evaluation indicated that in practice often these pathways were not actively implemented, leading to lack of collaboration between organisations. This resulted in disjointed services for service users and carers. Co-ordination of COSMHAD care and communication with service users and carers was highlighted in the realist synthesis as important, but service user and carer interviews indicated consistently that co-ordination was lacking in services they had received. Service users made clear that it was important for services to reach out to them, instead of them trying to co-ordinate their own care, which was difficult when managing mental health and substance use issues.

Summary of evidence

Evidence from the synthesis reinforces the policy guidance that people with SMI who also have co-occurring substance use issues should have their needs met primarily by statutory mental health services. The rationale for this is the high prevalence of co-occurring disorders (context) within mental health services, and the need for integrated care planning for service users (mechanism – resource). Graham et al.,80 in their study of integrating COSMHAD services through the COMPASS liaison model in the UK, argue that this requires ‘integration of treatment both at the level of the clinician and service’ (p. 184) and will result in ‘a conceptual shift within the organisation and those working in it’81 (p. 586) with a single mainstream clinician being able to address the holistic needs of people with COSMHAD (mechanism – resource)80,81 (Figure 24).

FIGURE 24.

Refined PT 9: mental health as holding primary responsibility.

Staff participants recognised that there was a high prevalence of co-occurring disorders within mental health services (context) and that it was a core aspect of mental health care. As one participant stated, CMHTs were ‘recognising we’re dealing with it anyway. it’s a core part of … our patients are dual diagnosis’ (SP83, Case Study E). Some participants discussed the ‘siloed nature of healthcare and substance use specific services and … the difficulty with aligning those’ (SP20, Case Study C) and that this could be addressed in mental health services if ‘everybody should have a basic set of skills’ that could address both mental health and substance use (SP20, Case Study C) because ‘this is knowledge that everyone should have’ (SP28, Case Study D) to enable them to work with people with COSMHAD. Participants described how ‘more support in like the overall workforce’ (mechanism – resource) would lead to ‘having these competencies and skills and confidence to work with this group of people’ (mechanism – response) (Figure 25).

FIGURE 25.

Programme theory 9 context to mechanism.

However, staff also identified additional contextual factors within mental health services which impacted upon their ability to be responsible for care planning (mechanism – resource). The term ‘mainstreaming’ which originated in early COSMHAD policy17 has not been operationalised and is therefore open to interpretation. It has been thought erroneously that mainstreaming placed all responsibility for someone with COSMHAD on the care co-ordinators (under care programme approach) in mental health. However, high caseloads and pressure in mental health services (context) made it challenging for them to ‘hold in [their] mind’ (SP29, Case Study D) the various needs of people with COSMHAD (mechanism – response). A care co-ordinator in a psychosis service described how she felt that she didn’t have sufficient time to provide effective support for substance use (mechanism – resource) on top of the immediate, complex needs of the service user, particularly when they were responsible for seeking out training and advancing their knowledge on COSMHAD themselves (context). Consequently, staff felt that they were unable to provide meaningful interventions to address COSMHAD leading to service users having unresolved needs or, in some cases, being discharged from services where they do not meet the assessment criteria to alleviate the pressure on case managers (outcome).

Well, if you’ve got 50 to 100 people on your caseload as a care co-ordinator and there’s a possibility that you can say if you don’t stop using this substance, I can’t do very much with your mental wellbeing, then you’re going to say that probably, you know, just from a position of managing.

SP37, Case Study A

So from my role as a care coordinator, I find that when sort of there is so much to remember and to learn and to do and we are coordinators of care rather than being able to do everything I suppose for an individual who often has very complex needs, one of those being substance use … I can do so much in terms of what I learned through sort of training or looking up things myself … There’s too much to remember … especially when you specialize in psychosis … and as part of that as we know, it is very much often trauma related, stressful housing, financial, you name it and obviously we always talk about substances and work on the risk of relapse if they continue to use … it’s so complex. So yeah, I don’t know really. I’d love to. But in reality, it’s hard just to do the basics at the moment to be honest with you, when you’ve got 25 plus clients, very complex, somebody is being in crisis of becoming homeless … to be able to engage sort of therapeutically meaningfully and without those basics, it’s very difficult to go ‘Now let’s talk about this’.

SP1, Case Study C

Staff emphasised that in their view the focus of mainstreaming should be on building relationships with service users (mechanism – resource) which meant that the service users and carers trusted that the care co-ordinator would be able to secure support from other agencies and teams as required (mechanism – response). Some staff described that training on COSMHAD in mental health services was insufficient to sustain practice in the long term, and that access to practice-based learning (including access to expert support from an expert clinical lead) was required. Therefore, the concept of mainstreaming requires workforce development opportunities as well as expert support and collaborative integration of local services. As SP15 describes, ‘knowing where the experts live and how to use them’ (context) would ensure more effective care planning in mental health for co-occurring disorders (mechanism – resource).

It is unfair to expect one worker to be, you know, specialist in every area and to be the one to, you know, provide all that knowledge and all those interventions to their client. I think you know it’s a lot of knowledge for somebody to take on and unfortunately in [city] there’s just not enough resources for a care co-ordinator to be able to do that … And that mainstreaming in that situation for me would be that they build up that trust with a service or a worker, to the point that they're trusting them to engage with the other services which would be specialist, a lot of our clients are really suspicious and not trusting. So I can agree to a certain extent that it’s helpful to have allocated workers. But if, you know, I really do feel for care co-ordinators in [name of city] who don’t have the time or resource is too.

SP6, Case Study C

Staff also felt that mainstreaming often lacked the organisational and wider structural support needed to make it happen (context); one participant described the concept as ‘a worthy aspiration that’s dressed up as a policy direction’ without sufficient organisational direction to train a generalist workforce in COSMHAD. This was felt to be hindered by the competitive commissioning structures which existed in many substance use and non-NHS services (context), and which acted as a barrier to collaborative working (mechanism – response). As a result, staff felt people with COSMHAD continued to fall into the gaps between services (mechanism – response) because there was no organisational directive in mental health for teams to take responsibility for their care.

And in terms of the mainstreaming, I think it’s a worthy aspiration that’s dressed up as a policy direction … So those committed individuals like us to actually try and work at the margins because the people that we’re talking about are at the margins. And they always have been. And it’s, there is not any kind of organizational direction to train the workforce to be kind of generalists in a sense.

SP12, Case Study C

[P]eople just don’t want the responsibility unless the client fits in a pigeonhole … I think it’s great to see all care coordination stuff. It makes absolute sense. But we’re not talking around common sense right now. We’re talking about … commissioning bids … there’s different organizations and sometimes we don’t all work together as one organization ’cause it’s like [name of substance use service] do this and [name of substance use service] do this … because the way the bidding’s done and everyone’s in competition with each other. But actually, I think the tension is the bid isn’t fit for purpose … But I think sometimes we don’t, as [name of overarching substance use service], we don’t see each other as one organization.

SP24, Case Study A

[T]hey’re hot potatoes as everybody in mental health thinks they should be with addictions and everybody in addictions … will say … They need to get back to mental health and vice versa. And then they end up with the crisis team and the crisis team will work very well with the mental health crisis. But then it’s off in the other directions … the community mental health teams are stretched and without new ways of working model staff leave and staff change in adopting new staff, et cetera, et cetera, all the other stuff that has to be done. And I think they would just see it as they don’t retain.

SP83, Case Study E

While wider contextual factors were highlighted as a challenge for mental health clinicians taking lead responsibility for people with COSMHAD care (mechanism – resource), they still agreed that this would lead to confidence in using empirically supported treatment for co-occurring disorders (mechanism – response) and a more consistent and tailored approach to symptom reduction with non-conflicting goals (outcome) (Figure 26). As one participant described ‘you got one person who’s care co-ordinating it. It makes absolute sense ’cause there’s a lot of care to co-ordinate and the client has a single point of contact. And someone has responsibility’ (SP24, Case Study A). Participants felt that having responsibility for co-ordination meant that they could be more responsive to their needs and be more consistent.

FIGURE 26.

Programme theory 9 mechanism to outcome.

This sounds like it would be good for the client, know just in terms of consistency. You have one person who’s the point of contact, and especially in that mental health role and they can then go on and say if you might need support in these other areas and you can almost back them up. But I would imagine that be helpful. I don’t think that I don’t think we definitely don’t have that system, [city name].

SP27, Case Study A

Some noted the importance of taking a trauma-informed approach

[W]e have a duty to learn about trauma and that knowledge really has to be cascaded down and shared … How dare we try and do this job without being trauma-informed? We’re going to re-traumatise people, cause them more damage.

SP16, Case Study C

[S]ome of our service users, they don’t want to be seen by mental health and vice versa that ’cause they feel more comfortable either within substance misuse services or mental health. And it’s kind of working with what they present with. And then kind of getting a good working relationship and exploring things you know. How can we do things differently in order for you to, to get better outcomes and working together with the other services if necessary. But it’s where the service user presents and what they want to do I think is the most important thing. It’s a starting point.

SP41, Case Study D

Service users and carers valued consistent support from a skilled and empathetic member of staff who took responsibility for leading their care (mechanism – resource). They also acknowledged that staff in substance use services did not always have the expertise required to deal with mental health needs, and that having access to someone in mental health services meant they could access care for their mental health. However, waiting times were a key context which impeded this support. Service users in general reported significantly less difficulty accessing drug and alcohol services compared to accessing mental health support where they reported longer waiting times to gain access (context).

… He’s been waiting. He’s always had a CPN and he’s always been in a team … under early intervention then moved out to community about 3½ years ago … He was lucky to get the CPN, very lucky, even though we desperately just need one. But I pushed for a CPN and … he was given one. But she recognized he needed additional help, so he was put on a waiting list for the outreach team … and it’s taken two years … to get to the top of this list, and he’s had his first visit from the care team yesterday.

P36, (Carer) Case Study C

Service users and carers recognised the different philosophical stances in substance use and mental health services (context). They also acknowledged that quite often they had to have stopped using substances in order to access some mental health services which they felt was counterintuitive, especially when their substance use was exacerbated by poor mental health. This rejection from mental health services was interpreted by some as mental health services deliberately managing demand by turning away complex clients (mechanism – response).

Probably always had a mental health issue, but it’s always been about, you know, there is no mental health issue. It’s all drugs. But now it seems like … we’ve wasted all that very valuable time when he was younger and could have been helped … I think we’ve gotta stop, you know, blaming drugs for everything … I think even it’s fairly well accepted now even amongst … more forward-thinking mental health professionals. That addiction is a mental health issue.

P28, (Carer) Case Study A

I think it should be more integrated, but unfortunately the … thing I’m coming up against is that therapists don’t want to treat people who are using drugs, you know … they’ve got justification, which is, you know, if you are under the influence of something, you’re not accessing your emotions properly. But it’s ironic that it’s the therapy that is required to help with the drug use. So it’s kind of convoluted … But I guess that is the way the situation is at the moment … but I think that therapists exercise their preference in not having clients who are drug users rather than they can’t.

P18, Case Study C

Service users and carers also frequently referred to the multiple additional needs experienced by people with COSMHAD including additional physical and neurodevelopmental conditions, learning disability, trauma and housing needs. All of these issues could be better co-ordinated if mental health services were more holistic. Carers described how they felt that once their loved ones were placed in housing, they were often left without purpose or holistic support from mental health (mechanism – resource) which could lead to exacerbation of their mental health symptoms and relapse of their drug or alcohol problems (mechanism – response).

[I]t just needs to be more open-minded and … within the mental health services to offer support. You know, people are going through addiction … like we’re not, I know it’s [a] hard diagnosis, but there might be some sort of review to see what’s supposed to be put in place there and then. And even if it is only short term … If I’d have had that extra support, it might be a little bit better for me.

P19, Case Study C

In summary, practitioners concurred that in cases where the main concern was a SMI, it makes sense for those with mental health expertise to take the lead role. However, this is impacted by several complex contextual factors (e.g. overstretched workloads, resources). Service users also felt that mental health services did not want to engage with complex clients, when dealing with such pressures, and were frustrated that substance use was not being recognised as a part of mental health issues.

Summary of evidence

Meaningful quality metrics must be selected to collect data on clinical processes and health outcomes for populations with co-occurring SMI and substance use disorders. 56,82 Chichester et al. 83 found that getting departmental leadership buy-in was vital to ensure that quality improvement measures were implemented across all organisations and departments (context) and that once departmental leadership was aligned with the idea of COSMHAD being everyone’s business, ‘change flowed as if a dam had been opened’ and ‘co-occurring disorder requirements were inserted into every contract the department has with providers’. People with COSMHAD have complex and variable needs which require integration across multiple services83 and services can also experience practical challenges in tracking patient data across multiple software systems56,82,83 As Currie79 found in their work in Ohio, it is therefore important that quality improvement structures, processes and outcome measures (mechanism – resource) are built on realistic expectations from leaders across organisations, of the nature of the existing system, patient preferences and realistic expectations of incremental progress (Figure 27).

FIGURE 27.

Refined PT 10: accountability and quality improvement.

There was limited evidence within the case study data regarding evaluation of services and quality improvement projects. Staff mentioned a few examples of smaller quality improvement projects which had been undertaken in their trusts in relation to areas such as assessment, assessing training needs or evaluating training programmes. The staff focus groups confirmed the link between leaders being accountable (context) for meaningful quality improvement measures (mechanism – resource) and staff feeling that work they do is valued (mechanism – response) (Figure 28). If leaders introduced (context) mandated quality standard awards for COSMHAD (mechanism – resource), then these could help staff recognise the importance and value of the work (mechanism – response). This was seen as particularly important for staff working with people with COSMHAD, where improvements may be slower and more incremental when compared, for example, with working with someone with a single diagnosis with less complexity in a community mental health setting (outcome).

FIGURE 28.

Programme theory 10 context and mechanism.

It was clear that when staff felt quality improvement and evaluation measures were perceived as more of a ‘tick box exercise’ (SP67, Case Study F) based on mandated organisational priorities rather related to service user needs, they were less likely to feel that the work they did was valued. It was felt that quality improvement measures only resulted in better quality services when staff could see the value of the mandated audit measures on their patients: ‘you don’t like it to just be a stick … you want people to do it because it is good clinical practice to do, not because … you have to because that ultimately ends up impacting on the quality’ (SP67, Case Study F). In addition, when audit and measurement work was done, staff wanted to feel that the results of these audits were used to implement recommendations which once again demonstrated the value of their work.

We’ve done audits across all services, so community services, what they’re offering, what the teams introduce, recommendations are made, and they’re never implemented ’cause they go back to the team leads to implement and nothing happens … we don’t get funded for it. So there doesn’t seem to be any real incentive to do it.

SP40, Case Study D

What do you do? What’s your plan once you’ve got this information from the patient, what do you do with it? And there’s gaps there in terms of what do we do, what staff trained and what does the trust want us to do? What’s their expectations? What’s the corporate plan for this?

SP83, Case Study E

The accountability of leadership (context) in selecting and implementing quality improvement measures was also seen to directly impact upon whether appropriate and trauma-informed outcomes were achieved, which improved the quality of care for people with COSMHAD (Figure 29). As one participant highlighted, these measures needed to reflect the incremental and individual outcomes that represented progress for each person with COSMHAD:

FIGURE 29.

Programme theory 10 mechanism and outcome.

[T]he computers don’t run us. We should be running them. There should be a way of being able to capture the fact that, you know, it’s not the same type of treatment for everybody that there are certain cohorts that we have to have a different way of measuring their input and contacts and things. I mean, for some patients, it could be if they don’t come to the appointment it’s a good thing because you made an appointment with them and said that’s there, if you need it … it’s not just the fact that every contact is a success. It could be the fact that a patient felt well enough. They don’t need to come to see you that week and they’re gonna leave it another week … but we can’t account for that.

P83, Case Study F

In contrast, staff felt that if meaningful quality improvement measures were not put in place, then they could not help in the way that they needed: ‘I just don’t get that when you’ve got a kid in front of you struggling … Why would I wanna know what the PHQ-9 is? What I want to know is how can I help you?’ (SP38, Case Study A). If leaders were not accountable (context) for developing meaningful quality improvement measures (mechanism – resource), then it would hinder services’ ability to take a trauma-informed approach to care (outcome) (see Figure 29). The importance of compassionate leadership was noted organisationally to underpin trauma-informed care, with compassion suggested to promote psychological safety.

I mean the stuff that’s come into my consciousness over this last couple of months has been stuff like Karen Treisman and the Compassionate Mind Foundation, people who are starting to talk about leadership from a compassionate … really sort of out there looking at kind of what it’s like to work from an organizational point of view with people that are traumatized do you know what I mean? And the bottom line of it is … it seems to be a massive part of the job where any of any sort of substance misuse is kind of involved, whether it’s a little t or a big T it doesn’t matter, but kind of soaking all of that up and having some sort of acknowledgement at an organizational level again seems to be completely relevant, but completely missed and obscured by an obsession with data and figures.

SP38, Case Study A

There were no data on this PT from the service user and carer interviews.