Introduction

Severe mental illness (SMI) is characterized by psychiatric disorders which continue over time and contribute to serious difficulties in personal and social functioning, thereby reducing the quality of life of the affected person. The prevalence of SMI and the burden related with it is high. An epidemiological study in Europe on the prevalence of SMI reports a total annual prevalence rate of 2.33 per 1,000 for all disorders [1]. People with SMI differ significantly from people with psychiatric disorders other than SMI; they are more likely unmarried, unemployed, show higher levels of psychopathology and disability and utilize services more often [2]. Patients with SMI have poorer physical health and a reduced life expectancy compared to the general population [3–7]. There are data suggesting that patients with SMI die on average between 12 and 32 years earlier than the general population [7, 8].

The burden of mental illness is high. It is estimated that 7.4 % of DALYs [sum of years of life lost (YLLs) and years lived with disability (YLDs)] worldwide are due to mental disorders. The main causes of this are unipolar depression and anxiety disorder, alcohol disorder and schizophrenia. The disease burden related to mental disorders increased from 5.4 % of all DALYs in 1990 to 7.4 % of all DALYs in 2010 [9]. A recent study using years of potential life lost (YPLL) as a measure of premature mortality showed that the mean YPLL in patients with SMI was 14.5 compared with 10.3 for the general population [10].

The consequences for the individual and for society are severe. Mental illnesses top the list of health problems leading to occupational disability and retirement [11] and cause more than €28 billion in direct illness costs and €26 billion in indirect illness costs. In Germany, mental illnesses generate about 11 % of the annual direct costs and about 18 % of all lost years of work [12]. Mental health care is only one type of service required in treating the chronically mentally ill. Support in housing, social and rehabilitation services, medical care and basic minimum maintenance needs also should be taken into account [13].

For example, severely mentally ill people suffer from chronic medical illness at a higher rate than the general population. Among persons with serious mental illness, the frequency of physical health problems is reported to be 50–90 % higher than among general psychiatric outpatients [14]. A report of a literature review on studies examining length of stay, costs of care and resource utilization in individuals hospitalized in general medical-surgical settings showed increased measures for those patients with psychiatric comorbidity [15].

Although these facts are generally recognized and addressed, individuals with severe and persistent mental illness (e.g., those individuals most in need) were identified as increasingly underserved within the German healthcare system [16]. In many instances, the focus is on acute, episode-specific single interventions such as medication which are known to have a short-term effect on symptom reduction. Increasingly, psychotherapeutic and psychosocial interventions for people with SMI are demanded and considered indispensable in mental healthcare systems [17].

However, the full spectrum of psychosocial interventions for individuals with SMI is rarely implemented [18], although psychosocial interventions, in general, have much to offer to improve treatment outcome. Healthcare professionals may consider such interventions not evidence based or difficult to implement as studies are complex or implementation requires system-level change. It is, however, the task of mental health professionals to inform healthcare planners with regard to the effectiveness of complex interventions and to give recommendations in the absence of conclusive evidence. Treatment processes are insufficiently understood in care for people with SMI, and this also applies to the question of how mental health care for this patient group is best organized and delivered. There has recently been an emphasis on implementation research [19, 20], but the effect of variations in the mode of care delivery on treatment process and patient outcome is not adequately understood. MacInnes et al. [21] and Priebe et al. [22] have studied interventions that impact on the way care is delivered by changing clinical discourse between clinicians and patients, and they have linked this approach to earlier studies of care based on patient needs and computer-aided communication. We need to combine this emphasis on treatment process with individual mental health interventions, i.e., building blocks of psychosocial care such as art therapy and also with interventions optimizing team-based care.

In order to develop an evidence-based guideline on psychosocial interventions for people with SMI, the German Association for Psychiatry, Psychotherapy and Psychosomatics (DGPPN), under the supervision of a guideline methodologist representing the Association of Scientific Medical Societies in Germany (AWMF), appraised the study evidence and developed recommendations on mental health care. Therefore, the guideline is based on the best available scientific evidence and the experience of healthcare professionals and representatives of people with SMI and their relatives.

Interventions addressed in the guideline presented in this article provide building blocks of psychiatric care, and these building blocks are firmly rooted in research evidence presented in this article.

Methods

The guideline targets all professionals treating adult patients aged 18 and over with severe and persistent mental illness. Serious mental illness is defined as follows: (1) the presence of a mental disorder (major depression, schizophrenia and other psychotic disorders, bipolar disorder, obsessive compulsive disorder, borderline personality disorder), (2) duration of mental illness and/or of service contact of 2 years or more, (3) severe social dysfunction as measured by a scale for assessing social functioning (e.g., Global Assessment of Functioning, GAF Scale) [1]. Therefore, recommendations are relevant for patients, their relatives, professionals in primary and secondary healthcare settings, mental health teams and for rehabilitation, residential, community, social and employment service providers. It could also be useful for people who are responsible for planning and providing health and social services.

Interventions

The guideline gives recommendations on psychosocial interventions other than pharmaceutical or psychotherapeutic ones. The term refers to a spectrum of psychosocial and social intervention programmes, which have been grouped into three categories (Table 1). These interventions aim to improve individual and social functioning, enhance quality of life and support the community integration of people with SMI. This paper focuses on system-level and single psychosocial interventions.

Table 1 Topics for evidence search
Full size table

Guideline development process

The guideline was developed between 2009 and 2011 and edited by the German Association for Psychiatry, Psychotherapy and Psychosomatics (DGPPN). DGPPN was founded in 1842 and currently has more than 7,000 members. It is the largest and oldest scientific association of doctors and scientists working in the field of psychiatry, psychotherapy and psychosomatics in Germany. One task of the DGPPN is the development of evidence-based guidelines and scientific statements.

The development of the guideline drew upon methods outlined by the Association of Scientific Medical Societies in Germany (AWMF). There are several basic steps in the process of developing a guideline according to AWMF classification (http://www.awmf.org/fileadmin/user_upload/Leitlinien/AWMF-Regelwerk/AWMF-Guidance_2013.pdf):

  • Form an expert and a consensus group including important stake holders and patients.

  • Define clinical questions which are considered as important for practitioners and service users.

  • Devise a thorough search strategy.

  • Conduct a systematic literature search, selection and appraisal of studies and exact documentation of results.

  • Synthesize data retrieved, guided by the clinical questions.

  • Produce evidence summaries.

  • Draft preliminary recommendations including allocation of strengths of recommendation.

  • Enact formal/structured voting procedure within consensus group meetings (nominal group technique, Delphi, structured consensus conferences).

Creation of the guideline development group (GDG)

The guideline development group (GDG) consisted of a steering group, an expert group (17 experts) and a consensus group (representatives from 40 stakeholder groups) and was composed of professionals in psychiatry, community psychiatry, geriatric psychiatry, child and adolescent psychiatry, psychotherapy, clinical psychology, nursing, general practice, occupational therapy, arts therapies, social work, sociotherapy, health insurance, vocational rehabilitation and exercise therapy; academic experts in psychiatry and psychology; and representatives of patients and relatives. The guideline development process was supported by a steering group that conducted the clinical literature searches, reviewed and presented the evidence at the consensus meetings, managed the process and contributed to drafting the guideline. The professional direction and supervision were carried out by a guideline methodologist of the AWMF. The guideline was also reviewed by an external reviewer.

Literature search strategy

The aim of the literature review was to systematically identify and synthesize relevant studies from the literature to answer the specific clinical questions developed by the GDG. Thus, clinical practice recommendations are evidence-based where possible and, if study evidence was not available, systematic consensus methods were used (good practice point, GPP). A stepwise, hierarchical approach was taken to identify the evidence. First, a search was undertaken for systematic reviews and guidelines published on each topic. We included well-conducted randomized controlled trials (RCTs), and for some interventions also non-RCTs. The search was exhaustive, using several databases and other sources. Standard health-related bibliographic databases (Cochrane Library, EMBASE, MEDLINE, PsycINFO) were used for the initial search for all studies potentially relevant to the guideline. In addition, we included specific databases (e.g., OTSEEKER.com, OTDBASE.org, CINAHL.com). We also checked the reference lists of all eligible systematic reviews and included studies, as well as the list of evidence documents submitted by stakeholders. In all databases, we included only articles written in English and German. There was no limit with regard to the publication period. The search was carried out between June 2009 and February 2011.

Development of recommendations

The selected papers were assessed with regard to their methodological quality. The methodological assessment was based on a number of questions that focused on those aspects of the study design which have been shown to influence the validity of the results reported and conclusions drawn. The result of this assessment had an influence on the level of evidence, which in turn influenced the strength of recommendation. Evidence tables summarize all the included studies from the systematic literature review relating to each key question. Evidence was classified according to an accepted hierarchy (Table 2). The GRADE approach was the basis for grading the quality of evidence and strength of recommendations [23]. Recommendations were graded A, B or 0 based on the level of the related evidence or were referred to as “good practice point” in those cases where no study evidence was possible or available (Table 3).

Table 2 Categories of evidence
Full size table
Table 3 Grading of recommendations
Full size table

All recommendations in the guideline have been approved in their present form by the consensus group using a nominal group technique applied during the consensus group meetings. The report on the methodology is publicly accessible via the Internet (see: http://www.dgppn.de/dgppn/struktur/referate/versorgung0/s3-leitlinie-psychosoziale-therapien-bei-schweren-psychischen-erkrankungen.html).

Results

System-level interventionsFootnote 1

Multidisciplinary team-based psychiatric community care

Community mental health teams (CMHT)

are the basic building block for community mental health services in the UK. The simplest model of provision of community care is for generic (non-specialized) teams to provide the full range of interventions (including the contributions of psychiatrists, community psychiatric nurses, social workers, psychologists and occupational therapists). CMHTs prioritize adults with SMI and operate within a locally defined geographical catchment area [24]. The CMHT management approach has been found to result in greater treatment satisfaction and reduced hospital admission rates compared with standard care [25] (see Table 4).

Table 4 Summary of evidence from systematic reviews and meta-analyses of psychiatric care systems
Full size table

Services that offer emergency assessment and intensive home treatment as an alternative to acute hospital admission have been developed in different places around the world using a number of names: crisis resolution teams, home treatment, crisis intervention teams or mobile crisis services, for example. “A crises resolution team provides intensive support for people in mental health crises in their own home, and stays involved until the crisis is resolved. It is designed to provide prompt and effective home treatment, including medication, in order to prevent hospital admission and provide support to informal carers” [26, p. 27]. The multidisciplinary team is available to respond 24 h a day, 7 days a week. Crisis/home care reduced the number of people leaving the study early, reduces family burden and is a more satisfactory form of care for both patients and families [27]. Moreover, crisis/home management results in reduced hospital admission rates and duration of acute inpatient care compared with hospital-based treatment [28] (see Table 4).

Assertive community treatment teams provide a form of specialized mobile outreach treatment for people with more disabling mental disorders. ACT “is an intensive mental health program model in which a multidisciplinary team of professionals serves patients who do not readily use clinic-based services, but who are often at high risk for psychiatric hospitalization. Most ACT contacts occur in community settings. ACT teams have a holistic approach to services, helping with medications, housing, finances and everyday problems in living” [29, p. 141]. Other names are assertive outreach, mobile treatment teams or continuous treatment teams. In addition to multidisciplinary staffing and team approach, further key principles of ACT are integration of services, low patient–staff rations, rapid access, assertive outreach, individualized and time-unlimited services [29]. The most important evidence of psychiatric care systems available to the date of guideline development is summarized in Table 4. ACT can substantially reduce the costs of hospital care while improving outcome and patient satisfaction [30].

There are numerous studies that have assessed the effectiveness of ACT (see Table 4). Further reviews have focused on specific target groups, for example, homeless populations with SMI [31, 32] and people with comorbid severe mental and substance use disorders [33, 34]. In randomized trials, people receiving ACT showed a greater reduction in homelessness and a greater improvement in psychiatric symptom severity compared with those in standard case management treatment [31]. It could be that an integrated ACT reduces substance use and decreases number of hospitalizations and dropping outs of treatment [34]. However, newer studies showed less consistent findings. Burns et al. [35] looked into this question and concluded that intensive case management (ICM) works best with participants who tend to use a lot of hospital care at baseline. Similarly, it works less well with infrequent hospital users. When hospital use is high, ICM can reduce the use of hospital care. In addition, the team organization has an important influence: The effectiveness of ICM teams is increased as their organization reflects the assertive community treatment model.

Despite the problems caused by the high degree of fragmentation of healthcare and health support systems in Germany, there are already examples that illustrate that a specific multidisciplinary outreach work is also possible in Germany (e.g., [36, 37]). Recently, models of financing were created in order to overcome the fragmentation in German health care. Cross-sector integrated healthcare and regional psychiatry budgeting are two models of cross-sector health care for inpatient and outpatient care in Germany. The regional psychiatry budget is a specific solution for psychiatric services, whereas integrated healthcare models can be developed for all areas of health care [38]. Recommendations concerning multidisciplinary team-based psychiatric community care are formulated in Box 1.

Box 1 Recommendations concerning multidisciplinary team-based psychiatric community care
Full size table

Case management

Case management (CM) was initially introduced in the USA as a mechanism for coordinating fragmented systems of community care [39]. CM has been defined as “the co-ordination, integration, and allocation of individualized care within limited resources” [40, p. 125]. The case manager takes primary responsibility for the severely mentally ill person. “As a minimum this responsibility includes: keeping contact with the person, assessing their needs, and ensuring that these needs are met” [41, p. 106]. A variety of different models of CM have been developed over the past decades: the broker model, the clinical case management model, the ICM model, the strengths model and the rehabilitation model (see [42]). The ICM model was developed to meet the needs of high service users. The patient to staff ratio is low [41]. One important difference between ICM models and ACT and other multidisciplinary team-based approaches is that ICM programs do not subscribe to the team approach with shared caseloads and daily team meetings [29]. Today, CM for severely mentally ill people is implemented in most modern healthcare systems, often of an intensive nature [41]. Also in Germany, CM constitutes a grown form of social psychiatric practice and an essential element of community care. In Germany, the concept of practice-oriented case management has not yet been examined in clinical trials. An older Cochrane review determining the effects of several CM approaches compared against standard care showed that CM increased the numbers remaining in contact with services and approximately doubled the numbers admitted to psychiatric hospital [43]. ICM compared with standard care resulted in higher treatment satisfaction, reduced dropouts and shortened the duration of inpatient care [44] (see Table 4). Recommendations concerning CM are formulated in Box 2.

Box 2 Recommendations concerning case management
Full size table

Vocational rehabilitation and participation in work life

People who suffer from SMI experience substantial levels of exclusion from work; they can experience educational dropout, high rates of unemployment and early retirement. Vocational rehabilitation services aim at helping mentally ill people to improve their occupational status. Traditionally, these services offer a period of preparation (prevocational training, PVT) before trying to place clients in competitive employment. A newer approach, known as supported employment (SE), tries to place clients in competitive jobs without any extended preparation. “SE has been defined as paid work that takes place in normal work settings with provision for ongoing support services” (see [45, p. 3]). A manualized variation of SE is the individual placement and support model (IPS). “The core principles of this model are (1) a focus on competitive employment, (2) eligibility based on consumer choice, (3) rapid job search, (4) integration of mental health and employment services, (5) attention to consumer preference in the job search, (6) individualized job supports and (7) personalized benefits counseling” (see [46, p. 281]). In Germany, traditional approaches of prevocational training are mainly used. Yet, there is a trend toward including elements of SE. The 2009 German Social Insurance Code (§38a SGB IX “Unterstützte Beschäftigung”) further encouraged the implementation of this approach.

There is a broad evidence base with very robust results concerning work-related outcomes which demonstrate that SE is superior to PVT [28, 46–49]. Compared to PVT, SE has been shown to offer:

  • A strong increase in employment rates on competitive job market (Ia),

  • An increase in work hours per month (Ia),

  • An increase in monthly income (Ib) and

  • An increase in the number of weeks/year which are spent in competitive jobs (Ia–Ib).

There is no significant superiority of SE concerning non-work-related outcomes [50, 51]. SE’s superiority for work-related outcomes may be less marked outside the USA [52]. A closer look at the EQOLISE study offers a similar assertion. The EQOLISE study was a randomized controlled trial comparing IPS to usual high-quality vocational rehabilitation and was conducted in six European centers, including in Ulm-Günzburg, Germany. In all six centers, IPS was more effective than vocational services for every vocational outcome (proportion of people entering competitive employment, the number of hours worked, the number of days employed and the job tenure of employed patients), but there was no significant difference in Ulm and in Groningen, Netherlands [51].

Nevertheless, not all SE participants obtain competitive employment and there are also clients who do not have vocational goals [46, 53]. SE programs may not be appropriate for all clients.

Compared to standard care (not work related), prevocational training is superior on several outcomes, especially in German-speaking countries. A German study showed that traditional vocational rehabilitation is effective with regard to employment status. Furthermore, it had positive effects on the level of functioning and the psychological well-being [54]. There is some evidence that financial incentives may result in greater effectiveness of traditional vocational rehabilitation [55]. In addition, traditional vocational rehabilitation programs, in combination with psychological intervention, can improve outcomes [56] (Box 3).

Box 3 Recommendations concerning vocational rehabilitation
Full size table

Residential care

Although the importance of stable and adequate housing for individuals with SMI is indisputable, there is little scientific research in this area. The number of studies is relatively small, and results are almost not comparable. A general finding of existing studies is that supported housing can reduce lengths of inpatient stays, especially if services are not time-limited [57]. There are also a few positive individual effects, for example, the reduction of negative symptoms and improvements of social contacts [57, 58]. Institutionalization is associated with negative effects [59–61]. Results do not indicate which patients benefit from which type of accommodation and living arrangement. Recommendations concerning residential care interventions are formulated in Box 4.

Box 4 Recommendations concerning residential care interventions
Full size table

Single psychosocial interventionsFootnote 2

Psychoeducation

The term psychoeducation was first used by Anderson et al. [62] to describe a behavioral therapeutic concept consisting of four elements (informing the patient, problem-solving training, communication training and self-assertiveness training). Inclusion of relatives is another key component. A survey of complex family psychoeducation programs is found in McFarlane et al. [63]. Parallel to the development of psychoeducation in North American and the UK, psychoeducation in German-speaking countries has emerged as an independent therapeutic program and is defined as systematic, didactic-psychotherapeutic interventions that inform patients and their relatives about the illness and its treatment, facilitating both an understanding and personal responsibility for handling the illness and supporting those afflicted in coping with the disorder [64]. Group sessions last approximately 1 h, take place once to twice a week and consist of between four and 16 meetings. Group leaders are predominantly doctors or psychologists; co-leaders can be recruited from all relevant occupational groups [65]. This dual focus program was evaluated as part of the Munich Psychosis Information Project (PIP study). This randomized multicenter study showed a significant reduction in re-hospitalization rates and a reduction of intermittent days spent in hospital within this 2-year period [65]. Even 7 years after psychoeducational group therapy, significant effects on the long-term course of the illness can be found [66]. Our evidence search focused on psychoeducational interventions and family interventions with a psychoeducational approach. We have mainly found evidence for psychoeducation in schizophrenia and related disorders. These are set out in Table 5. We also identified a few single studies for psychoeducation for bipolar disorders. The results are almost uniform and show that psychoeducational intervention in schizophrenia reduces relapse and readmission rates and improves the participants’ knowledge about the illness. It is obvious that there is a benefit from interventions including family members [67, 68]. Long-term interventions (longer than 3 months) appeared to be more successful than short-term interventions [67]. Psychoeducation for patients with bipolar disorders can lead to a higher time to relapse and a significantly lower mean number of relapses (total, manic and depressive relapses); re-hospitalization rates were also reduced (e.g., [69–72]).

Table 5 Summary of evidence from systematic reviews and meta-analyses of psychoeducation
Full size table

Recommendations concerning psychoeducation are given in Box 5. Regarding the treatment recommendations for depressive disorders, reference is made to the S3-Leitlinie/Nationale Versorgungsleitlinie Unipolare Depression [73]. The authors recommend psychoeducational interventions in depression as an additional intervention in an individualized treatment plan (Grade B).

Box 5 Recommendations concerning psychoeducation
Full size table

Social skills training

SMI is often affected by impairments in daily skills and social functions, and thus, participation in society is markedly lowered. A social skills training is an important approach in psychiatric rehabilitation. Several training models have been designed and evaluated [74]. Social skills training utilizes behavior therapy principles and techniques for teaching patients to enable them “to acquire interpersonal disease management and independent living skills for improved functioning in their communities” [75]. We have identified five meta-analyses and many RCTs demonstrating the efficacy and effectiveness of social skills training. Results suggest that patients may benefit from trainings that improve social skills and social functioning [76–78]. In addition, there is empirical evidence that when cognitive rehabilitation approaches are added, social skills training can improve neuro-cognitive and social-cognitive functioning. The training of specific cognitive functions is known as integrated psychological therapy (IPT) [78, 79]. Other specific strategies have also been proven effective [80–83]. There are ambiguous results regarding other outcomes. Beyond this, techniques are required which actively support the generalization of positive achievements to patients’ natural environments [84, 85] (Box 6).

Box 6 Recommendations concerning social skills training
Full size table

Arts therapies

Arts therapies are widely used treatment strategies for people with SMI. There are various approaches and a variety of different techniques, but they all focus on non-verbal communication and on the creation of a working therapeutic relationship in which strong emotions can be expressed and processed. A systematic search was carried out for art therapy, music therapy, drama therapy and dance movement therapy. Generally, only a few randomized trials are available; however, the studies show that arts therapies in addition to standard care reduce the amount of negative symptoms among people with schizophrenia [86–88] (see Table 6). The most significant evidence is available for music therapy. In the treatment of severe depression, it has been shown that the addition of music therapy improves depression [89]. Recommendations concerning arts therapies are formulated in Box 7.

Table 6 Summary of evidence from systematic reviews and meta-analyses of arts therapies
Full size table
Box 7 Recommendations concerning arts therapies
Full size table

Occupational therapy

Occupational therapy is one of the oldest forms of treatment of mental disorders. There is a Greek term (“ergon”) that means recovery through action and work. Occupational therapy aims to support the patient in improving his ability to act independently, thereby increasing self-dependence in everyday life (self-care, leisure and productivity). Occupational therapy covers a wide range of treatment methods. There are three fundamental methodological orientations suggesting the selection of specific therapeutic approaches and techniques: (1) competence-centered method, (2) expression-centered method and (3) interactional method [90]. Generally, the effectiveness of occupational therapy as a mental health intervention has not been systematically evaluated. Only a few randomized trials are available, sample sizes are small, and outcomes are varied. Isolated positive results were obtained from single studies (e.g., [91–93]) (Box 8).

Box 8 Recommendations concerning occupational therapy
Full size table

Exercise therapy

The use of body- and movement-related measures for the prevention and healing of illnesses has a long tradition. Apart from the various physical benefits, psychological changes have been postulated [94]. Today, movement therapy and sports are part of the routine care in Germany for individuals with SMI. In psychiatry, it is possible to distinguish between three basic approaches: (1) sport therapy approach or exercise (e.g., endurance training), (2) body-oriented psychotherapy (e.g., integrated movement therapy) and (3) educational-psychosocial approach (e.g., cooperative games). We have identified one systematic meta-analysis and some single studies assessing the effectiveness of movement therapy and sports in schizophrenia and depression. The majority of identified studies examined the effectiveness of aerobic endurance training. This form of exercise can improve mental and physical health for patients with schizophrenia [95, 96]. When comparing aerobic endurance with yoga, yoga was found to have better outcomes for mental state and quality of life [95]. Single studies on exercise in schizophrenia also show positive effects on social functioning, emotionality and motor and psychomotor skills [97–100]. Body-oriented psychotherapy for patients with schizophrenia leads to increased mental state, motility and general functioning [101, 102]. For patients suffering from severe depression, results showed that aerobic endurance training positively affect the severity of depression and anxiety, quality of life, self-esteem, dysfunctional attitudes and physical health [103–108]. Motivating the patient to participate in sports in the long term has also been shown to be effective [104]. Recommendations concerning movement therapy and sports are given in Box 9.

Box 9 Recommendations concerning movement therapy and sports
Full size table

Discussion

The German guideline on psychosocial therapies in patients with SMI is based on a comprehensive appraisal of evidence on the effectiveness of psychosocial interventions in the target group. It offers an opportunity for further improving health care for patients with SMI in Germany [109]. The guideline can be used to compare the current psychiatric healthcare situation with the guideline recommendations in order to identify and draw conclusions based on the strengths and the weaknesses of the care models in different settings. Similarly, future changes in evidence-based psychiatric health care in Germany may be guided or supported by the guideline. Moreover, the guideline also highlights existing gaps and areas for further mental health service research.

It remains, however, a challenge to develop recommendations on evidence-based practices for SMI. Evidence-based practice should be grounded in consistent research evidence that meets criteria of internal and external validity [110] as well as having been subjected to quality assessment [111].

Research evidence on psychosocial interventions is often context specific. As a consequence, it is not always consistent, and new evidence evolves which generates a need for ongoing scientific review and reconsideration of recommendations. Study results may depend on the society and its values, the country, the healthcare system and other variables. Thus, in the development of the presented guideline, study evidence was only one factor influencing recommendations. It was central, but subject to the appraisal of the consensus group with regard to relevance, transferability and cost-effectiveness in the German mental health system. This appraisal used consensus methods. Its legitimacy is strengthened by the fact that representatives of all relevant stakeholders were members of the consensus group and that the evidence search and assessment of methodological quality were performed by an independent GDG.

Individual systematic reviews demonstrated that there is good evidence for the effectiveness of most of the selected psychosocial interventions in SMI. The best available evidence exists for multidisciplinary team-based psychiatric community care (Ia–Ib). The majority of published studies on approaches such as home treatment, assertive community treatment and CMHT have been conducted in the USA or the UK. However, community-based and team-based psychiatric care is also practiced in Germany. There is also a broad evidence base on work-related outcomes which demonstrate that SE is superior to prevocational training (Ia–Ib). This research was supplemented by a randomized controlled multicenter trial in Europe comparing IPS to usual high-quality vocational rehabilitation. In all six centers, IPS was more effective than were vocational services for every vocational outcome. However, within-center comparisons resulted in no significant difference in Groningen, the Netherlands and Ulm, Germany. Unfortunately, once a guideline is published, it is already outdated. In the meantime, a recent study from Switzerland has been published which favors SE over traditional vocational rehabilitation programmes even in a Western European country with a very high threshold to the open labor market for people with SMI [112]. Thus, it is probable that the SE will be recommended with a higher level of recommendation in the next edition of the guideline. Unfortunately, there is only scarce evidence for residential care interventions. There is a need for well-designed and reported RCTs of the effects of residential care interventions for people with SMI.

With regard to single interventions, there is considerable evidence for the efficacy of psychoeducation and social skills training. Convincing effects can be found mainly for family interventions with a psychoeducational approach. The social skills training, which has been examined primarily in people with schizophrenia, clearly shows positive effects on social skills and social functioning. The transfer of the gains in social skills to the daily life, i.e., social adjustment and role fulfillment of patients in their living environments, is of central importance. For other interventions, the available evidence is either scarce or characterized by methodological limitations that do not allow a strong recommendation. For example, for arts therapies, only a few randomized trials are available and sample sizes are small. In addition, short observation periods and heterogeneous interventions restrict statements about efficacy. Therefore, a reliable statement about the effectiveness of arts therapies in SMI is premature, and conflicting results have been reported. A recent study confirmed the positive effects of music therapy as an addition to standard care for people with schizophrenia [113]. However, group art therapy for people with established schizophrenia (MATISSE study) did not significantly improve global functioning, mental health or other health-related outcomes [114]. Here again, a revision of recommendations will take this finding into account. The effectiveness of occupational therapy as a mental health intervention has not been systematically evaluated; thus, only isolated positive results could be obtained from single studies of occupational therapy. Therefore, occupational therapy was labeled with a recommendation strength B in the guideline. There is no clear demarcation between occupational therapy, vocational rehabilitation and living skills training. Often vocational rehabilitation and living skills training are components of occupational therapy. The evidence for vocational rehabilitation and living skills training is very strong. We assume that increasingly establishing occupational therapy in academic institutions will offer opportunity for research in this area. On the whole, it can be stated that the low number of high-quality studies in some areas of psychosocial health services research results from the fact that implementation of such studies (unlike pharmaceutical research) is difficult although public support is available. Despite careful research, publication bias cannot be excluded.

The new edition of the guideline will not just include recent study results, but will also consider further interventions, e.g., cognitive remediation, compliance therapy and behavioral interventions addressing healthy lifestyles of patients including smoking, diet and exercise. It will also be discussed whether relaxation techniques (that are not genuine psychosocial interventions) should find their way into the next version of the guideline. Corresponding published scientific literature is available (e.g., [115–118]).

Today, the requirements for high-quality clinical practice guidelines are defined internationally in a uniform way. In German-speaking countries, they are summarized in the Guideline Appraisal Instrument DELBI which is the result of a collaboration between AWMF, German Agency for Quality in Medicine (ÄZQ) and partners in care practice, science and health (available under: http://www.delbi.de). The three key aspects of DELBI are (1) the representativeness of the GDG, (2) the evidence base and (3) the structured consensus process. However, acceptance and effects of a guideline do not depend solely on its methodological quality but also on adequate implementation strategies [119]. Implementation is a process requiring multifaceted strategies to promote changes [119].

The guideline is available as long version in book format (ISBN: 978-3-642-30269-5). The long version and a report of methodology are also freely accessible to all via the Internet (see: http://www.dgppn.de/dgppn/struktur/referate/versorgung0/s3-leitlinie-psychosoziale-therapien-bei-schweren-psychischen-erkrankungen.html). In addition, numerous publications in relevant German language specialist journals are generated. Part of our implementation strategy is a patient version of the guideline (ISBN: 978-3-642-55267-0) and an ultra-short version for waiting rooms in outpatient settings in different languages to cross-language barriers. The patient versions will be published in cooperation with the Bundesverband der Angehörigen psychisch Kranker e.V./Familien-Selbsthilfe Psychiatrie and with the Bundesverband Psychiatrie-Erfahrener e.V. A web-based dissemination support system (see: http://www.dgppn.de/dgppn/struktur/referate/versorgung0/s3-leitlinie-psychosoziale-therapien-bei-schweren-psychischen-erkrankungen.html) is available.