In 2008 the WPA General Assembly approved the Action Plan of the Association for the triennium of the Presidency of Professor Mario Maj. One of the items of the Plan is the production of guidelines on practical issues of interest to psychiatrists worldwide (1,2. The present document, providing guidance on lessons learned and mistakes to avoid in the implementation of community mental health care, is part of that project. In subsequent publications we shall describe in more detail the particular challenges and solutions identified in the various regions worldwide. Mental health problems are common, with over 25% of people worldwide developing one or more mental disorders at some point in their life 3. They make an important contribution to the global burden of disease, as measured by disability-adjusted life years (DALYs). In 2004, neuropsychiatric disorders accounted for 13.1% of all DALYs worldwide, with unipolar depressive disorder alone contributing 4.3% towards total DALYs. In addition, 2.1% of total deaths worldwide were directly attributed to neuropsychiatric disorders. Suicide contributed a further 1.4% towards all deaths, with 86% of all suicides being committed in low- and middle-income countries (LAMICs) each year 4. A systematic review of psychological autopsy studies reported a median prevalence of mental disorder in suicide completers of 91% 5. Life expectancy is lower in people with mental health problems than in those without (in some countries dramatically so) also due to their higher levels of physical illnesses 6. Mental health problems, therefore, place a substantial burden on individuals and their families worldwide, both in terms of diminished quality of life and reduced life expectancy. The provision of high-quality mental health care is vital in reducing some of this burden 7. In this context, the aim of this report is to present guidance on the steps, obstacles and mistakes to be avoided in the implementation of community mental health care, and to make realistic and achievable recommendations for the development and implementation of community-oriented mental health care worldwide over the next ten years. It is intended that this guidance will be of practical use to psychiatrists and other mental health and public health practitioners at all levels, including policy makers, commissioners, funders, non-governmental organizations (NGOs), service users and carers. Although a global approach has been taken, the focus is mainly upon LAMICs, as this is where challenges are most pronounced. There are wide inconsistencies between, and even within, countries in how community-oriented care is defined and interpreted. Historically speaking, in the more economically developed countries, mental health service provision has been divided into three periods 8: • The rise of the asylum (from around 1880 to 1955), which was defined by the construction of large asylums that were far removed from the populations they served. • The decline of the asylum or "deinstitutionalization" (after around 1955), characterized by a rise in community-based mental health services that were closer to the populations they served. • The reform of mental health services according to an evidence-based approach, balancing and integrating elements of both community and hospital services 8–10. Within a "balanced care model", most services are provided in community settings close to the populations served, with hospital stays being reduced as far as possible, and usually located in acute wards in general hospitals 11. Differing priorities apply to low, medium and high resource settings: • In low-resource settings, the focus is on establishing and improving the capacity of primary health care facilities to deliver mental health care, with limited specialist back-up. Most mental health assessment and treatment occurs, if at all, in primary health care settings or in relation to traditional/religious healers. For example, in Ethiopia, most care is provided within the family/close community of neighbours and relatives: only 33% of people with persistent major depressive disorder reach either primary health care or traditional healers 12,13. • In medium-resource settings, in addition to primary care mental health services, an extra layer of general adult mental health services can be developed as resources allow, in five categories: outpatient/ambulatory clinics; community mental health teams; acute inpatient services; community-based residential care; and work, occupation and rehabilitation services. • In high-resource countries, in addition to the above-mentioned services, more specialized ones dedicated to specific patient groups and goals may be affordable in the same five categories described for medium-resource settings. These may include, for instance, specialized outpatient and ambulatory clinics, assertive community treatment teams, intensive case management, early intervention teams, crisis resolution teams, crisis housing, community residential care, acute day hospitals, day hospitals, non-medical day centres, recovery/employment/rehabilitation services. It is this balanced care model approach that has been taken here in considering community-oriented care. In low-resource settings, community-oriented care will be characterized by: • A focus on population and public health needs. • Case finding and detection in the community. • Locally accessible services (i.e., accessible in less than half a day). • Community participation and decision-making in the planning and provision of mental health care systems. • Self-help and service user empowerment for individuals and families. • Mutual assistance and/or peer support of service users. • Initial treatment by primary care and/or community staff. • Stepped care options for referral to specialist staff and/or hospital beds if necessary. • Back-up supervision and support from specialist mental health services. • Interfaces with NGOs (for instance in relation to rehabilitation). • Networks at each level, including between different services, the community, and traditional and/or religious healers. Community-oriented care, therefore, draws on a wide range of practitioners, providers, care and support systems (both professional and non-professional), though particular components may play a larger or lesser role in different settings depending on the local context and the available resources, especially trained staff. Underpinning the successful implementation of community-oriented mental health care is a set of principles that relate on the one hand to the value of community and on the other to the importance of self-determination and the rights of people with mental illness as persons and citizens 14,15. Community mental health services emphasize the importance of treating and enabling people to live in the community in a way that maintains their connection with their families, friends, work and community. In this process it acknowledges and supports the person's goals and strengths to further his/her recovery in his/her own community 16. A fundamental principle supporting these values is the notion of people having equitable access to services in their own locality in the "least restrictive environment". While recognizing the fact that some people are significantly impaired by their illness, a community mental health service seeks to foster the service user's self-determination and his/her participation in processes involving decisions related to his/her treatment. Given the importance of families in providing support and key relationships, their participation (with the permission of the service user) in the processes of assessment, treatment planning and follow-up is also a key value in a community model of service delivery. Various conventions identify and aim to protect the rights of service users as persons and citizens, including the recently ratified United Nations (UN) Convention on the Rights of Persons with Disability (UNCRPD) 17 and more specific charters such as the UN Principles for the Protection of Persons with Mental Illness and for the Improvement of Mental Care adopted in 1991 18. The above-mentioned and other international, regional and national documents specify the right of the person to be treated without discrimination and on the same basis as other persons; the presumption of legal capacity unless incapacity can be clearly proven; and the need to involve persons with disabilities in policy and service development and in decision-making which directly affects them 18. This report has been written to explicitly align with the requirements of the UNCRPD and associated treaties and conventions. This guidance has been produced by taking into account the key ethical principles, the relevant evidence, and the combined experience of the authors and their collaborators. In relation to the available evidence, systematic literature searches were undertaken to identify peer-reviewed and grey literature concerning the structure, functioning and effectiveness of community mental health services or obstacles to their implementation. These literature searches were organized for most of the World Health Organization (WHO) Regions, reflecting the context of the report's main authors. There are limitations to this approach, in particular the WHO Eastern Mediterranean Region was not fully represented, and this report focuses upon adult mental health services. Accordingly, this guidance does not address the service needs of people with dementia or intellectual impairment, and of children with mental disorders. Searches varied according to local expertise and resources. Medline was searched for every region. Other databases searched were EMBASE, PsycINFO, LILACS, SciELO, Web of Knowledge (ISI), WorldCat Dissertations and Theses (OCLC) and OpenSigle. Searches, adapted for each database, were for M.E.S.H. terms and text words relating to community mental health services and severe mental illness. Other electronic, non-indexed sources, such as the WHO, Pan American Health Organization (PAHO), WPA, other mental health associations, and country-specific Ministry of Health websites, were also searched. Google was searched for PDFs published in European and African countries which contained the words "community mental health". Searches were limited to articles published in the languages spoken by the authors covering each WHO Region, and authors sought relevant advice from WHO Regional Advisors. Electronic searches were supplemented by searches of the reference lists of all selected articles. Hand searches of issues from the past five years of three key journals relevant to Africa (African Journal of Psychiatry, South African Journal of Psychiatry, and International Psychiatry) were also conducted. In addition, key texts were identified: these included relevant papers and book chapters published by authors of the current work 19–24 and a special edition of the Lancet on Global Mental Health 25–29. WHO publications which provide information regarding community mental health services worldwide were also sourced 7,31–33. For the Africa Region, original research was conducted in order to supplement published data. Twenty-one regional experts completed a semi-structured, self-report questionnaire concerning their experience in implementing community mental health care in sub-Saharan Africa 34. The experts were from 11 countries and one NGO active in several countries across sub-Saharan Africa. International and inter-cultural differences can play a significant role in shaping what mental health services are needed and possible within local settings (most particularly, the level of financial resources available 28). Nevertheless, in preparing this report, we have been surprised to find that the most fundamentally important themes (both in terms of challenges and lessons learned) apply to many countries and regions. We therefore discuss next each of these key themes in turn. One challenge common to many countries worldwide is the difficulty in putting community mental health intentions into practice. We distinguish here between: • National policy (or provincial or state policy in countries where health policy is set at that level): an overall statement of strategic intent (e.g., over a 5–10 year period) that gives direction to the whole system of mental health care. • Implementation plan: an operational document setting out the specific steps needed to implement the national policy (e.g., what tasks are to be completed, by whom, by when, with which resources, and identifying the reporting lines, and the incentives and sanctions if tasks are completed or not completed). • Mental health programmes: specific plans either for a local area (e.g., a region or a district) or for a particular sector (e.g., primary care) that specify how one component of the overall care system should be developed. According to WHO's Mental Health Atlas 31, 62.1% of countries worldwide had a mental health policy, and 69.6% had a mental health programme in place in 2005 (with 68.3% and 90.9% of the global population covered respectively). Many of the countries without such policies were LAMICs. Even where comprehensive evidence-based mental health policies are in place, problems in implementing these policies are common 33,35). Some of the reasons may include health staff not complying with policies due to difficulties in accepting and implementing changing roles 33, the lack of accessible evidence-based information or guidelines for health staff, inadequate funding mechanisms, inadequate training of health care personnel, the lack of mechanisms for training and coaching health staff, poor supervision and support, and an overall lack of human resources 35. Detailed and highly practical implementation plans (taking into account available resources) are therefore necessary in enabling effective community mental health care provision. A further challenge that needs to be addressed worldwide is the massive gap between population needs for mental health care (true prevalence of mental illness) and what is actually provided in mental health care (treated prevalence) 7, highlighting the importance of scaling up services for whole populations. The evidence concerning the substantial burden of mental disorders has not been translated into adequate investments in mental health care 29. The treatment gap is particularly pronounced in LAMICs, where commonly over 75% of people with mental disorders receive no treatment or care at all, and less than 2% of the health budget is spent on mental health 7. Whilst the high-income countries of the world have an average of 10.50 psychiatrists and 32.95 psychiatric nurses per 100,000 population (median figures), in low-income countries there are only 0.05 and 0.16 respectively 31. Furthermore, even within countries, the quality and level of services often vary greatly according to, for instance, patient group, location (with service provision usually being higher in urban areas), or socio-economic factors 3. Similarly, only 10% of global mental health research is directed to the health needs of the 90% of population living in LAMICs, and only a fraction of this research activity is concerned with implementing and evaluating interventions and services (36). Methods to estimate resource needs are necessary in scaling up services. A systematic methodology for setting priorities in child health research has been developed taking into consideration that interventions should be effective, sustainable and affordable to reduce the burden of disease 37. A similar methodology was applied by the Lancet Global Mental Health Group, which focused on four groups of disorders whilst setting priorities for global mental health research: depressive, anxiety and other common mental disorders; alcohol- and other substance-abuse disorders; child and adolescent mental disorders; and schizophrenia and other psychotic disorders 30. It was recommended that interventions should be delivered by non-mental health professionals within existing routine care settings, and specialists should play a role in capacity building and supervision 38. A comprehensive review of packages of care for six leading neuropsychiatric disorders − attention/deficit hyperactivity disorder (ADHD), alcohol abuse, dementia, depression, epilepsy and schizophrenia − have also recently been proposed as means to extend treatment in LAMICs 20–24. An extensive set of treatment guidelines, also suitable for LAMICs, will be published by the WHO in 2010 as a part of their mhGAP programme. A survey of availability and feasibility of various treatments for the most prevalent mental disorders in the various age groups has been recently carried out by the WPA with its Member Societies 39. A further common barrier in identifying and treating mental disorders worldwide is the lack of awareness about them within communities, with stigma towards, and discrimination against, people with mental health problems being widespread. This is important, because effective awareness-raising campaigns can result in increased presentation of persons with mental illness to primary health care 40. Three main strategies have been used to reduce public stigma and discrimination: protest, education, and social contact 41. Protest, by stigmatized individuals or members of the public who support them, is often applied against stigmatizing public statements, such as media reports and advertisements. Many protest interventions, for instance against stigmatizing advertisements or soap operas, have successfully suppressed negative public statements and for this purpose they are clearly very useful 42. However, it has been argued 41 that protest is not effective for improving attitudes toward people with mental illness. Education interventions aim to diminish stigma by replacing myths and negative stereotypes with facts, and have reduced stigmatizing attitudes among members of the public. However, research on educational campaigns suggests that behaviour changes are often not evaluated. The third strategy is personal social contact with persons with mental illness 43. For example, in a number of interventions in secondary schools, or with the police, education and personal social contact have been combined 44,45). Social contact appears to be the more efficacious part of the intervention. Factors that create an advantageous environment for interpersonal contact and stigma reduction may include equal status among participants, a cooperative interaction, and institutional support for the contact initiative 46. For both education and contact, the content of programmes against stigma and discrimination matters. Biogenetic models of mental illness are often highlighted because viewing mental illness as a biological, mainly inherited, problem may reduce shame and blame associated with it. Evidence supports this optimistic expectation (i.e., that a biogenetic causal model of mental illness will reduce stigma) in terms of reduced blame. However, focusing on biogenetic factors may increase the perception that people with mental illness are fundamentally different, and thus biogenetic interpretations have been associated with increased social distance 47. Therefore, a message of mental illness as being "genetic" or "neurological" may be overly simplistic and unhelpful for reducing stigma. Indeed, in many LAMICs, conveying a message emphasizing the heritable nature of mental illness fuels stigma, for instance making marriage more difficult. Anti-stigma initiatives can take place nationally as well as locally. National campaigns often adopt a social marketing approach, whereas local initiatives usually focus on target groups. An example of a large multifaceted national campaign is Time to Change in England 48. It combines mass-media advertising and local initiatives. The latter try to facilitate social contact between members of the general public and mental health service users as well as target specific groups such as medical students and teachers. The programme is evaluated by public surveys assessing knowledge, attitudes and behaviour, and by measuring the amount of experienced discrimination reported by people with mental illness. Similar initiatives in other countries, such as See Me in Scotland 49, Like Minds, Like Mine in New Zealand 50, or the WPA anti-stigma initiative 51, along with similar programmes in other countries, including Japan, Brazil, Egypt and Nigeria, have reported positive outcomes 40. In sum, there is evidence for the effectiveness of measures against stigma and against discrimination 52. On a more cautious note, individual discrimination, structural discrimination and self-stigma lead to innumerable mechanisms of stigmatization. If one mechanism of discrimination is blocked or diminished through successful initiatives, other ways to discriminate may emerge 53,54. Therefore, to substantially reduce discrimination, stigmatizing attitudes and behaviours of influential stakeholders need to change fundamentally. The collaborative engagement of a wide variety of supportive stakeholders is critical to successful implementation of community-oriented mental health care. It is important to have a systemic view of the change process. The support is needed of politicians, board members and health managers whose primary focus may not be on mental health, clinicians, key members of the community including NGO providers, service users and their families, and traditional and religious healers. To involve them in the imperative for change will require different strategies and a change management team that includes people with a variety of expertise. Overall, having clear reasons and objectives for the shift to community-oriented care is essential. Messages should be concise, backed by evidence and consistent. Developing consensus for change requires a lot of work in meeting and communicating with people. The main means of communication need to include written material and opportunities to meet with stakeholder groups. Politicians and administrators will require a compelling business case. However, others will need summaries of plans, slide presentations and the opportunity to meet and work through proposals and concerns. E-mails and website information and surveys are now valuable supplements to the process. The emphasis must be on a willingness to communicate in good faith and to do so openly and honestly doing "what it takes" to convince people of the benefits of the change process. It is important to bear in mind that in some cases prejudice and self-interest will have to be confronted. It is helpful, at the beginning of the process, to identify both those who are likely to support change, and those who are likely to oppose it. A willingness to listen to concerns and to find ways of incorporating them, if possible, into the planning and implementation process is essential because, when such an attitude is communicated, there is an opportunity for people to feel included in the process. That done, boldness and firmness will communicate to remaining detractors the seriousness of the intent to implement change; it will also encourage supporters to believe that their aspirations for better mental health care will be realized, and thus embolden them in turn. Engaging stakeholders requires both formal and informal opportunities to meet, receive advice and work through issues. The establishment of reference groups early on in the change process is a key formal mechanism to achieving this. These should include all the key stakeholders, in particular service users, families, clinicians and service providers, with the latter being essential to facilitate integrated systems of care further on in the process. While it is important to structure the overall process with formal meetings and communications, it is also important to be willing to convene informal meetings upon request to "trouble-shoot" situations of concern. The consultation process should result in an amalgam of "bottom-up" and "top-down" contributions to the change process. Reports on progress are an essential way of maintaining trust and building excitement to the process of successful implementation. It is also important to bear in mind that good mental health services have established processes for ensuring that the voices of service users, their families and community providers are heard on an ongoing basis. The aim is not simply to achieve discontinuous change, but to promote an ongoing quality improvement in which consumers of mental health services know they have a major stake. Without such effective and united consortia, policy makers may find it easy to disregard the different demands of a fragmented mental health sector, and instead respond positively to health domains (e.g., HIV/AIDS) which demonstrate the self-discipline of united approach with a small number of fully agreed priorities. Several key mistakes are commonly made in the process of attempting to implement community mental health care. First, there needs to be a carefully considered sequence of events linking hospital bed closure to community service development. It is important to avoid closing hospital-based services without having successor services already in place to support discharged patients and new referrals, and also to avoid trying to build up community services while leaving hospital care (and budgets) intact. In particular, there needs to be at each stage of a reform process a workable balance between enough (mainly acute) beds and the provision of other parts of the wider system of care that can support people in crisis. A second common mistake is to attempt system reform without including all the relevant stakeholders. Such initiatives especially need to include psychiatrists, who may otherwise feel subject to "top-down" decision making and react, either in the interests of patients or in their own interests, by attempting to delay or block any such changes. Other vital stakeholders to be directly included in the process will often include policy makers and politicians, health service planners, service users and carers, service providers including those in state and private practice, national and international NGOs, and those working in alternative, complementary, indigenous and religious healing traditions, and relevant national and professional associations. Typically, those groups not fully involved in a reform process will make their views known by seeking to undermine the process. A further common mistake is linking inappropriately the reform of mental health care with narrow ideological or party political interests. This tends to lead to instability, as a change of government may reverse the policies of their predecessors. Such fault lines of division or fragmentation may also occur, for example, between service reforms proposed by psychologists and psychiatrists, or between socially and biologically oriented psychiatrists, or between clinicians and service user/consumer groups. Whatever the particular points of schism, such conflicts weaken the chance that service reforms will be comprehensive, systemic and sustainable, and they also run the risk that policy makers will refuse to adopt proposals that are not fully endorsed by the whole mental health sector. Additional issues that may compromize the integrity of community based services include: a) an exclusive focus of community services on psychotic conditions, so that the vast majority of people with mental disorders are neglected or dealt with by professionals who do not have the appropriate expertise; and b) the neglect of patients' physical health. A fundamental component in the successful implementation of mental health service provision is that of funding 10. As indicated above, funding for mental health services in LAMICs tends to be very low. This may be due in part to a stigmatizing attitude toward mental disorders, and to an absence of the recognition of the economic benefits that can accrue from improved mental health care. Ideally, the share of its health funding that a country devotes to mental health care will be informed by careful consideration of the comparative health benefits of spending on alternative forms of care. The data needed to carry out such an analysis are, however, typically not available in LAMICs. Furthermore, whatever funding there is also tends to be concentrated on inpatient services. Correcting this is, initially, a matter of budgetary re-allocation: using resources that could have been used for other purposes to increase funding for community-oriented care. The issue then arises of how to pay public providers (hospitals, stand-alone programs, and possibly independent individual providers such as psychiatrists) for the services that they render. The simplest forms of payment are global budgets for facilities and programs, which may be carried over from year to year with minor adjustments for inflation, and salaries for individual providers. These simple payment mechanisms have the advantage of administrative simplicity. At the same time, they have at least two important drawbacks. First, they provide no incentive for increasing either the quantity or the quality of service provision. Second, population shifts are likely to cause the demand for the services of different providers to evolve and, without taking changes in local demand into account, inequities in payment across providers are likely to emerge and grow over time. This in turn will compromize access to overburdened providers, while possibly resulting in overprovision (e.g., excessive lengths of stay) by other providers. Accordingly, countries with the technical and administrative capacity to introduce more complex payment systems should consider doing so. For hospitals, a fairly simple alternative which
