For too long there were heated debates between those who believed that mental healthcare should be largely provided from mental hospitals, and those who took the opposite ideological view, inspired by the charismatic leaders of the time, that community care should fully replace hospitals. In economically developed countries, the overall picture throughout the past two decades has been for a progressive reduction in hospital beds, along with an unbalanced, inadequate and slow investment in community services. Reference Thornicroft, Semrau, Alem, Drake, Ito and Mari1 Interestingly, more recently there has been a partial trend in the opposite direction in some European countries to increase the number of psychiatric beds, with the latest EUROSTAT data showing such increases for 6 of the 27 European Union member states (Bulgaria, Germany, Hungary, The Netherlands, Romania and Turkey) between 2008 and 2009. 2 Indeed, intriguing recent evidence is emerging that where more psychiatric beds are provided, demand increases to ensure full bed occupancy. Reference Watts, Shiner, Klauss and Weeks3 In this case, whether one is considering where to place more investment or where to decommission services (in periods of economic austerity), the key question therefore remains: what balance to strike between hospital-based and community-based services? We discuss this question in light of both the published research and clinical experience.
Elements of the balanced care model
In a previous paper we reviewed the relevant literature published until 2004. Reference Thornicroft and Tansella4 We have updated this with a literature review for 2004–2011, including cross-referencing with the results of review papers written by international experts who contributed chapters to a recent international textbook on community mental health. Reference Thornicroft, Szmukler, Mueser and Drake5 We went on to test this approach in a structured survey with experts in mental health system change from 31 countries worldwide. Reference Thornicroft and Tansella6
It would be naive to assume that a single version of the balanced care model could apply across the world, given the huge variations in resources available at national and local levels. The model is therefore structured according to the three World Bank country resource categories, and proposes mental health service components relevant for low-, middle- or high-income (or resource) countries, as shown in Fig. 1. It also needs to be appreciated that the balanced care model may also apply to settings more local than the country level, for example to sectors with different levels of resources within countries.
Low-income settings
In low-income settings, most of the available provision is by staff in primary healthcare and community facilities. Reference Eaton, McCay, Semrau, Chatterjee, Baingana and Araya7 The key service roles are summarised in Fig. 1, and the most pressing challenges are how to shift the provision of care to staff who are most widely available on the ground (‘task shifting’), and how to decrease the gap between true and treated prevalence of mental disorders (‘scaling up’). Reference Dua, Barbui, Clark, Fleischmann, van Ommeren and Poznyak8
Medium-income settings
In medium-income settings there is still a requirement for strong primary care to treat the people with the more common mental disorders and, in addition, as resources allow, the balanced care model indicates that five elements of general adult mental health services are needed.
Out-patient/ambulatory clinics
Evidence is surprisingly limited about out-patient services, although this is a basic building block for care provision in many countries. Nevertheless, there is a strong clinical consensus in many countries that they are a relatively efficient way to organise the provision of assessment and treatment. Since these clinics are simply methods of arranging clinical contact between staff and patients, the key issue is the content of the clinical interventions, namely whether they allow the delivery of treatments that are effective and which confer patient benefit. Recent examples of such intervention content are now available suitable for high-, medium- and low-resource settings, published for example by the World Health Organization. Reference Dua, Barbui, Clark, Fleischmann, van Ommeren and Poznyak8,Reference Patel and Thornicroft9
Community mental health teams
Community mental health teams (CMHTs) are core components of community mental health services. The simplest model of provision of community care is for generic (non-specialised) CMHTs to provide the full range of interventions, staffed by multidisciplinary personnel. These often prioritise adults with severe mental illness, using a case management system, for a locally defined geographical catchment area. At the same time it needs to be recognised that for patients not able or not willing to go to health facilities, the mobility of CMHTs is necessary,
Fig. 1 Mental health service components relevant to low, medium and high resource settings
but may not be sufficient for effective care. What mobile teams can facilitate is better continuity for patients who would otherwise have intermittent or discontinuous treatment. When such clinical encounters do take place, once again the critical issue is whether the content of treatment is effective.
Acute in-patient care
There is also a very weak evidence base about most aspects of in-patient care. There are few systematic reviews in this field, the processes of care are poorly understood (e.g. the most effective length of stay), and systematic data on outcomes are rare. More generally, although in practice there is a consensus that acute in-patient services are necessary both to diagnose and to treat patients, the number of beds needed is highly contingent on which other services exist locally, and on local social, economic and cultural characteristics. Since acute in-patient care commonly absorbs most of the mental health budget, reducing the numbers of patients admitted or the average length of stay may therefore be important system goals, especially if the resources released in this way can be used to pay for the development of other service components.
Long-term community-based residential care
Alongside acute psychiatric beds, it is also important to know whether patients with severe and long-term disabilities should be cared for in larger, traditional institutions or be transferred to long-term community-based residential care. Although there is no strong evidence on this question from low-income settings, the evidence from medium- and high-income settings is reasonably clear. When deinstitutionalisation is carefully carried out, when patients who have previously received long-term in-patient care for many years are discharged to community care, then the outcomes are neutral or favourable for the majority. Reference Thornicroft and Tansella6 Nevertheless, the range and capacity of community residential long-term care that will be needed in any particular area is also highly dependent on which other services are available locally, and on social and cultural factors, such as the amount of family care that is available.
Work and occupation
Rates of unemployment among people with mental disorders are generally much higher than in the general population, especially in times of economic recession. Traditional methods of occupational rehabilitation have not been shown to be effective in leading to open market employment. For areas with medium levels of resource, in the absence of relevant evidence, it is reasonable to make pragmatic decisions about the provision of work and occupation services, especially taking into account the priorities and preferences of service users and family members.
High-income settings
In high-income settings, superimposed upon the primary care system, and also in addition to the provision of general adult mental health services, the balanced care model proposes that a series of specialised services should also be provided, as resources allow, in each of these five component categories, in order to provide more intense/expert interventions, specifically targeted to identified areas of poorly met need (see Fig. 1). In fact, however, experience shows that it is often the case that specialised services are developed either in the absence of the first two layers of service, or independently, to create an unbalanced system.
Examples of specialised out-patient/ambulatory clinics include those for people with eating disorders, treatment-resistant affective disorders, people with comorbid psychotic and substance misuse/dependence disorders, or for mentally ill mothers. Specialised CMHTs, for example, can include assertive community treatment or early intervention teams. Additional/alternative specialist acute in-patient care can refer to acute day hospitals, crisis houses or home treatment/crisis resolution teams.
Three basic types of specialised long-stay community residential care have been identified as relevant to high-income settings: (a) 24-hour staffed residential care (high-staffed hostels, residential care homes or nursing homes, depending on whether the staff have professional qualifications); (b) day-staffed residential facilities (hostels or residential homes which are staffed during the day); and (c) accommodation with lower levels of staffing support. Finally, in relation to specialised forms of work and occupation, evidence is now accumulating from high-income settings in favour of supported employment approaches.
How can the balanced care model be implemented?
It is clear that formulating a mental health strategy or plan, for example one based on the balanced care model, may be necessary but not sufficient to ensure that service improvement will be put into practice. In recent years an increasingly detailed appreciation has developed about the barriers which impede the implementation of evidence-based policies and practices, and about methods which can be used to successfully overcome these barriers. Reference Tansella and Thornicroft10 In the future, therefore, it will be necessary not only to have models available which guide planning, but also models which equally clearly guide implementation.
Taking into account the combined information from both this literature review and this synthesis of expertise, what are the implications? Our conclusions are that there is no strong evidence that a comprehensive system of mental healthcare can be provided by hospital-based care, but nor is there strong evidence that it can be provided by community-based services. Rather, a balance is necessary which includes both hospital and community components. Although specific recommendations have been developed in some countries to ensure that at least a half of all mental healthcare expenditure is on community services, Reference Whiteford, Buckingham and Manderscheid11 in fact the relative mixture of service components needed depends very much on specific local circumstances, and where the blend is likely to change over time.
Acknowledgements
We wish to acknowledge the people who contributed to this paper including: Abdul Aziz Abdullah, Thomas Becker, Chee Kok Yoon, Fiona Crowley, Cecilia Cruz Villares, Nicolas Daumerie, Iris De Coster, Melvyn Freeman, Nikos Gionakis, Peykan G. Gökalp, Sergiu Grozavu, Lars Hansson, Judit Harangozo, Ulrich Junghan, Yiannis Kalakoutas, Alisher Latypov, Burul Makenbaeva, Graham Mellsop, Roberto Mezzina, Pětr Nawka, Jean Luc Roelandt, VesnaSvab, Maris Taube, Radu Teodorescu, Rita Thom, Chantal Van Audenhove, Jaap van Weeghel, Kristian Wahlbeck, Richard Warner and Stefan Weinmann who have directly contributed to the review of experience gained in making mental health service changes. We also thank the members of the World Psychiatric Association Task Force (2009-2010) on ‘Steps, obstacles and mistakes to avoid in the implementation of community mental health care’, namely, Atalay Alem, Renato Antunes Dos Santos, Elizabeth Barley, Robert E. Drake, Guilherme Gregorio, Charlotte Hanlon, Hiroto Ito, Eric Latimer, Ann Law, Jair Mari, Peter McGeorge, R. Padmavati, Denise Razzouk, Maya Semrau, Yutaro Setoya, R. Thara and Dawit Wondimagegn. We also acknowledge the important contributions to this paper of Elizabeth Barley and Maya Semrau.
eLetters
No eLetters have been published for this article.