The practice of psychiatry in the second half and particularly in the last decade of the 20th century has fundamentally changed [1–4]. During the 19th and the first half of the twentieth centuries, more and more asylums for psychiatric patients were established in the industrialised or Western world [5–7]. Patients were admitted to facilities far away from urban areas and, in many cases, stayed there for the rest of their lives [6–8]. The number of patients that lived in institutions increased in the first half of the 20th century: psychiatric asylums in Europe reached populations of up to 3000 inmates in the middle of the 20th century [5, 9–14]. State mental hospitals in the United States of America became villages with 3000–10,000 inhabitants [6–8]. Independently of diagnosis, patients were taken from their families, from the street, the prisons and poorhouses to the psychiatric asylums [6]. One explanation for the enormous growth in psychiatric patients in asylums is partly the increasing incidence and prevalence of specific psychiatric disorders, such as neurosyphilis, epilepsy, partly the way in which society dealt with people with these disorders and—related to that—a therapeutic pessimism [6]. Better therapies, particularly the introduction of chlorpromazine in 1954, the conviction that asylums created a high level of dependence, the need for a more humane attitude towards psychiatric patients, and the immense costs of maintaining the asylums, which were largely built in the 19th century, formed the basis for changes in the second half of the 20th century [6, 15–17].

In Europe, the USA, Canada and Australia, rigorous changes started around 1950 [6, 7, 9–15]. This was also the case in some Central and Eastern European countries [18]. In the last decennia of the twentieth century, the socio-political strategy focused on non-institutional treatment of psychiatric patients: idealistic convictions, psychopharmacological drugs, GPs, and medical and psychiatric social work made possible this move away from institutions. Clinical admissions became limited and, if still deemed necessary, their duration had to be as short as possible [6, 10, 19]. Soon after this, a more systematic dehospitalisation policy was implemented in the USA, Canada, Australia and different regions of European countries. This led to community mental health care, supplied by community mental health teams, GP consultants and community-based living and day care facilities serving by and large as alternatives for in-patient treatment [7, 9–15, 20, 21]. The most spectacular decline in beds in mental hospitals in the USA shows a decrease from 559,000 beds in 1955 to 138,000 in 1980, a decline of 75% [22]. Admissions increased from 150,000 in 1955 to 400,000 in 1970 [23]. The mean length of stay decreased from 20 years in 1955 to 7 months in 1975 [24]. In England and Australia, closure of the large asylums has largely been accomplished [22, 25]. Different authors underline the complexity of this reform process, moving away from the asylums and into society [1–7, 9–14, 16–23].

Becker and Vázquez-Barquero [5] summarise the interrelated aspects of psychiatric reform:

“Psychiatric reform is not just about abolishing the old-fashioned psychiatric institutions but also concerns a number of issues such as: legislation, attitudes of society towards psychiatry, the choice of the scale of the catchment area for alternative facilities, the realisation of new facilities, the roles of other care suppliers such as the GP, the welfare sector, the general health care services, the balance and financing of the care, the fate of the patients coming from the old-fashioned institutions, the way to cope with the ever-increasing demand for psychiatric help and finally the actual quality of psychiatric help.”

Over the last hundred years, mental health care in the Netherlands has undergone an enormous process of development. In the first half of the century, the development of hospital care and psychiatric hospitals brought about a particularly striking change, while in the second half, partly due to the introduction of the first psychopharmacological drugs and new forms of psychotherapy, the change has been the introduction of community-based care, partial hospitalisation and sheltered housing facilities [8, 25]. Up to the time of World War II, German psychiatry predominated in the Netherlands, but US psychiatry became the most influential afterwards [25]. The century has seen the emphasis on mental health care undergo a gradual shift from providing an asylum and custodial care to providing assessment, treatment and possibly a cure [25]. Phases of the reform process in mental health care services during the last 25 years have been marked by the integration of ambulatory services in the early 1980s, subsequent implementation of community mental health centres (RIAGGs), differentiation of target populations, dehospitalisation of patients, differentiation within the field of sheltered housing accommodation, and the merger process of the above three entities into integrated regional mental health care organisations [26]. Current issues in the development of services include the growing demand for mental health care, special programmes for defined target populations, legislation and patient rights, rehabilitation and empowerment [26]. Table 1 gives a summary of development of Dutch mental health care.

In the Netherlands, the number of beds in psychiatric institutions increased in the first half of the 20th century, from 5000 in 1884 to 20,000 in 1928, peaking at 28,000 in 1955 [24]. After 1955, the number of beds decreased to 22,800 in 1990 and stayed the same until 1996. This is an absolute decline of 10% and 17% corrected for population growth [24]. The decline is mainly due to the decrease in the number of patients admitted for more than five years; a third of these long-stay patients was transferred to psychogeriatric institutions and institutions for people with mental disabilities [8]. We can therefore hardly speak of psychiatric reform in this period. The decline is at least modest in relation to the USA with its 66% decline in institutional beds in the same period [8]. The first concerted effort on the part of the Dutch government to reduce the number of beds in psychiatric hospitals stems back to 1974: the creation of tiers, in which access to institutions (the third level) could only be reached after first going through the first and second levels of general and ambulatory mental health care [27]. The second level, the community mental health centres was supposed to provide a solution to the above by offering facilities such as: types of ambulatory psychiatric care, psychotherapy, psychosocial assistance, crisis intervention and pre2vention [27]. The development of ambulant mental health care from 1974 to 1984 is described in Van der Grinten's thesis [28].

The formation of Regional Institutions for Mental Health Care did not develop into authoritative units in comparison to psychiatric hospitals, and the number of beds decreased little in the period from 1980–1996 [29, 30]. Radical reform was now being called for in the Netherlands and resulted in the Nieuwe Nota Volksgezondheid (‘New Report for National Public Health’): this report was a reaction both to the moratorium published in 1983 that called for rebuilding psychiatric hospitals as well as confirmation that the Regional Institutions for Mental Health Care did not provide a counterbalance to the existing monopoly of psychiatric hospitals [31]. The concept of the multifunctional unit (MFU) was therefore introduced in this report as ‘a collective policy direction with regard to admission, treatment, discharge, and aftercare [31]. During the course of almost two decades that followed this report, the concept in various recommendations and policy documents changed on several points. The starting point was that facilities for social psychiatry, outpatient treatment, part-time treatment, and short-term clinical treatment would be integrated [32, 33]. Gradually, the aims were defined as follows: to offer patient-oriented care, small-scale facilities and to help avoid admissions [34, 35]. There were different points of view on whether or not to formalise the framework for the mental health partners on such issues as the role of the psychiatric departments of general hospitals in relation to the multifunctional units, the size of the catchment area of a multifunctional unit (more or less than 100,000 inhabitants) and the capacity of the MFUs (more or less than 60 treatment spaces) [34, 35]. The first MFU was not opened until 1988. It was soon predicted that approximately 80 MFUs would be built in the Netherlands within a decade [36, 37].

Subsequently, extramuralisation and socialisation have been the policy issues that underlie the current reforms in the Netherlands [38, 39]. The number of care renewal projects increased fivefold from 1991 to 1995 [40–42]. In 1996, there were a total of 583 care renewal projects in the GGZ mental health care institutions, which amounted to an average of fourteen projects in each psychiatric hospital (APZ) at the end of 1995; the APZ was the initiator in 84% of these cases [40, 41]. The projects cover a wide range of topics and were aimed at improving the following care issues for chronic and non-chronic patients: cooperation between institutions, needs assessment, diagnostics and patient treatment/support, integration into society and alternative treatments for clients other than admission [40, 41]. In relation to the above, the first merger between an APZ and a RIAGG took place in 1993 [43].

The most recent comparative review was published by Becker and Vàzquez-Barquero [5] and is merely for the European situation. Research into the outcome of reforms in psychiatry is particularly difficult according to these and other authors [5, 19, 45]. This is also the case for the Netherlands: Wiersma et al. conclude that there is no systematic research available in the Netherlands on the effects of extramuralisation (i.e. day treatment, psychiatric home care, supported living houses) and dehospitalisation; these data would provide insight into the effect of reforms and the quality of mental health [44]. This seems to be of great importance: the lessons learned from psychiatric reform in other countries focus on the counterpart of deinstitutionalisation, whereby it is warned that the quality of patient care will be threatened if the implementation of alternative community care is neglected [45]. The same warnings are heard in the Netherlands, especially in association with increasing compulsory admission [46]. The decreasing availability of beds in (secured) wards caused by closing down the old mental hospitals, while the realisation of ambulatory mental health care alternatives soaks up the budgets for mental health care, could lead to a shortage in secured ward capacity [47].

There was a relatively low, mainly theoretical dehospitalisation rate until the mid-1990s. Major mental health care renewal activities first appeared in the early 1990s. The period from 1993 to the present appears to be particularly relevant, considering that major effects of change seem to begin in 1993 with the first two mergers of psychiatric hospitals and RIAGGs. Mergers between ambulatory and clinical mental health care institutions and the deinstitutionalisation process in the Netherlands seem to be intertwined [25, 32, 43, 48].

Two recent, related studies performed by the author of this article focused on how Dutch mental health care specifically developed toward deinstitutionalisation from 1993 to 2004 [49]. The first study involves the size, nature, aims and effects of mergers; the second, the development of regional mental health centres. The results of these studies are summarised below.

The first study showed that the 41 psychiatric hospitals included at the onset of the study in 1993 were operating at a regional level. They differed in terms of size of catchment area, the number of beds per hospital, the number of personnel and financial turnover. All hospitals were certified for general psychiatric care and some were also certified for specialist treatment such as psychogeriatrics, child and adolescent psychiatric care, drug abuse care and detention by hospital order.

In the period from 1993 to 2004, almost all general psychiatric hospitals were involved in mergers with at least one regional institution providing ambulatory mental health care (regional ambulatory mental health care institutes). Other merger partners included regional institutions providing sheltered housing, alcohol and drug abuse centres and other general psychiatric hospitals. In the mid-1990s, it was expected that all the original general psychiatric hospitals would merge with one or more of the other ambulant mental health care providers by 2000. Of the 41 general psychiatric hospitals registered on January 1, 1993, a total of 38 had merged with at least one regional ambulatory mental health care institute ten years later (December, 2004). In ten cases, two regional ambulatory mental health care institutes were involved and in once case four regional ambulatory mental health care institutes were involved. At the end of 2004, there were 36 mental health care organisations, three of which had not merged with a regional ambulatory mental health care institution.

The main lines of the change process focused on building new facilities, creating care circuits and setting up care programmes. Care circuits (networks) are organisational units where similar treatment programmes or care facilities for a particular target group are combined. The circuits of merged or non-merged institutions differentiate between short-term treatment of adult psychiatric patients, chronic psychiatric patients and geriatric patients. Some of these treatment circuits are accommodated in regional mental health care centres and/or are provided with new premises and located in subregions.

Nearly all general psychiatric hospitals were involved in developing an innovation strategy, consisting of improving the continuity of care by removing existing hindrances, particularly in the transition of ambulatory care to clinical care and vice versa. One aim was to reduce clinical care, if enough ambulatory care could be offered. Another idea was that patients would benefit socially if the care could be offered to them closer to home. Finally, the goal was greater differentiation, less overlap and fewer gaps in care supply. Beneath this layer of concrete motives was the need for new buildings in Dutch psychiatry and the unequal division of admission capacity, both between and within the various provinces and large towns. The need for new buildings could only come about through cooperative initiatives on the part of mental health care partners.

Dehospitalisation, individualisation, rehabilitation and decentralisation were the key concepts for the above processes. In practice, these concepts are translated into replacement of clinical facilities by part-time clinical treatment or completely extramuralised forms of treatment, such as home care and assisted ambulatory housing. An estimated 10,000 beds will be reallocated within 10 years. It is expected that almost all of the capacity for short-term care patients will be transferred to regional mental health care centres. Operating costs connected to 50% of the beds will be invested in day clinics, home care etc. The expected benefits of the changing processes are better distribution of care and improved accessibility as well as more treatment continuity and greater attention for long-term inpatients through the introduction of a rehabilitation philosophy. The separation of housing and treatment/care will lead to a situation where the bed as a basis for treatment will become less important.

Other consequences are of organisational importance, like the involvement of GPs, cross-sectional links between mental health care organisations and hospitals, nursing homes and institutions for people with a mental disability.

The reduction in the number of beds in merged organisations has a possible relationship to a shortage of admission possibilities, particularly closed beds and seclusion rooms. Respondents estimate that the asylum facility eventually required will be 6000 beds, about half of the original general psychiatric hospital asylum capacity.

There is a close relationship between these mergers and the building of new, smaller, integrated regional mental health centres. The capacity of the old mental hospitals ‘in the dunes and the woods’ has been proportionally reduced in favour of the formation of an estimated 80 regional mental health care centres in various agglomerated parts of the catchment area.

The second study results showed that of the 31 regional mental health care centres built between 1988 and 2000, regional mental health care centre catchment area sizes vary from 80,000 to 400,000 inhabitants, and their capacities vary from 34 to 150 treatment spaces. Table 2 shows the main characteristics of these mental health care centres.

Some regional mental health care centres have turned out to be larger than intended in policy documents. As a starting point, regional mental health care centres have ‘basic psychiatric functions’, which means generalised, non-specialised ambulatory and outpatient treatment for adult and elderly patients with mental disorders, and a small part of the regional mental health care centres offer help to children, adolescents or patients with addictions. Eighty-seven percent of the regional mental health care centres have a mobile team and 24-hour care and three-quarters of the regional mental health care centres offer home-based care. The number of outpatient visits has risen, the amount of admissions fallen and the duration of admissions decreased in recent years. Seven regional mental health care centres reported some sort of vacancy. On the other hand, five regional mental health care centres had too little capacity for admission. This was mostly the case in the larger cities. In ten organisations, the surplus of spaces has been transformed into day-care spaces, which gives less flexibility when more admission capacity is suddenly needed. Most regional mental health care centres are not situated near a general hospital.

Regional mental health care centres differed from the size and scale originally intended by the government and capacities were not based on a rational analysis of mental health needs, but were derived from existing numbers of admissions and estimated capacity needs.

This was caused by significant external influence: downsizing forces from national government legislation in combination with the need for new buildings and local mergers. The lack of legislation aimed at the desired situation had resulted in different conceptualisation of regional mental health care centres. The consequences for the new, merged and deconcentrated institutions were also financial: on the one hand, reduction of costs was made possible by reducing overhead by limiting the numbers of directors, management staff and administrators, while on the other hand, small-scale units like regional mental health care centres generated more costs due to such things as an increase in the average number of nurses per patient and higher overhead costs due to multiplied equipment and infrastructure. The lack of legislation aimed at financing the desired situation resulted in an economically unstable organisation of regional mental health care centres.

We have mentioned two main discussion points arising from these studies. In the first place, there has been an actual, rapid decrease in psychiatric beds in the Netherlands. Recently the Dutch ‘Board for the building of hospital facilities’ reported a decrease of 2200 beds, 10% of the total psychiatric beds from 1996 to 2003 [50], a dramatic contrast with the stable period from 1986 to 1996 [51]. This is still more than 1 bed per 1000 inhabitants (1, 25 promille), which maintains a high bed ratio in the Netherlands in comparison to most Western countries with ratios between 0.5 and 1.0. Another expected decrease in the Netherlands is based on the opinions of managers of psychiatric hospitals who estimate that the asylum facilities eventually required will be about half of the original general psychiatric hospital asylum capacity: a closed ward would be the best place for only 400 patients [40]. Regional mental health care centres are taking over the admission function of mental hospital admission wards, a process now partly realised and partly in progress, which will lead to a fair percentage of the old-fashioned admission wards to empty or be demolished. Taking the above into account, the emerging scenario for the Netherlands is a further, more rigorous move than ever away from the old institutions or what is left of them. So far, most of these institutions have not been dismantled as in other countries, and the question now is whether they should be: there are different signs and facts that the role of the (still existing) premises of old mental hospitals and institutions should be reconsidered, although many of the original buildings were or will be removed and the premises sold to real estate project developers wanting to build villas on them. There are signs that these old shoes should not be readily discarded.

In countries where the old mental hospitals have been completely abolished, there is a growing concern about the effectiveness of alternative community services, especially concerning the fate of patients coming from the dismantled institutes and that of new, young chronic patients, who live in the community but are in fact in need of a closed ward [45, 52]. This is the case as long as there is no integrative approach to developing alternative community services, especially assertive community treatment and as long as prevention and early detection of psychiatric disorders remains in an early phase. The law on compulsory admissions does not provide any opportunity to treat non-compliant patients, but this law is expected to change soon [53]. Some authors have recently reported a trend towards reinstitutionalisation: e.g. the increasing number of forensic beds in the UK, the attitudes towards compulsory admissions of psychiatric patients and the increase in compulsory admissions [54–56], a trend that is also taking place in the Netherlands [57]. There seems to be two reasons for this increase: an increasing need for safety in the community and good patient care [53, 58]. Studies on mental disorders and crime show an association between psychiatric disorders and criminal convictions, which demonstrates that the process of dehospitalisation of the relevant diagnostic categories, has consequences for the safety of society [59]. Hansen et al. report a higher suicide rate after deinstutionalisation in Norway [60]. In the Netherlands, an association has recently been made between the increasing incidence of psychiatric disorders in detention populations and dehospitalisation in psychiatry [61].

The second point for discussion is what should be the future role of the new regional mental health care centres, particularly in relation to capacity needs in the new, smaller catchment areas? The underlying and yet unanswered issue regards the minimum scale size of a catchment area on which to base the amount of basic psychiatric care facilities and exactly what basic facilities should be available. During the formation of regional mental health care centres ‘the building should correspond to the care’ and not the other way round, which has been the case. The reasons for this conclusion are that on a national level the formation of regional mental health care centres has focused too much on other interests, such as a reduction of capacity in psychiatric hospitals, refurbishment of old buildings and the need for new buildings. On a regional level, the regional mental health care centre acts like a ball bouncing between efforts to arrive at synergy in the subdivided mental health care supply. Such processes contrast with the issue of whether and how the regional mental health care centres will be able to meet actual and future care needs in the region by collaborating with first line and welfare organisations. The most striking example of the diversity in regional mental health care centres is the size of the catchment area, varying from 80,000 to 400,000, a discrepancy with respect to the internationally accepted size of 50,000–150,000 inhabitants for community-based mental health care [64]. In France (70,000) England (30–80,000) and Spain (55–100,000), an explicit choice has been made for smaller, independent and responsible regions [20, 62]. The choice of such scale size is motivated by the possibility of providing for care demands that are strongly localised and indeed may be dependent on age distribution, income, degree of urbanisation and level of development of a neighbourhood or region. Ultimately, demographic and epidemiological data should be the basis for this kind of choice [20]. Such preliminary epidemiological studies are carried out at regional levels, determining the care needs in a certain area in a reasonably reliable way without extensive demographic and epidemiological research [20]. In the last four years in the Netherlands, only three examples of those studies have been carried out [63–65]. In these studies, an increase in the use of psychiatric capacity is forecast, also intramural in origin. Only one province in the Netherlands, Limburg, has used that study in its vision for mental health care [65].

In Scotland, there are government guidelines for mental health providers with regard to services and the quality of services [66]. The exact interpretation of how this is carried out is left up to the local partnerships formed between the various care providers. In Belgium, the development of regional mental health care centres is still in the brainstorming phase, but the emphasis there is also to concentrate on the formation of coalitions before actually rebuilding [67]. In Germany, there is already a long tradition of substitution and dehospitalisation without building too many new facilities [9].

We have concluded that the newly built regional mental health care centres are increasingly taking over the short admissions in their transmural context, leaving behind the old wards in the old-fashioned psychiatric institutions. It is unclear what the position of these regional mental health care centres is and will become in relation to the psychiatric departments of general hospitals (PDGH) and academic hospitals. In particular, in different European regions as well as non-European countries such as Australia and the United States, the process of deinstitutionalisation implies a more prominent or unique role for general hospitals, particularly the PDGHs [9–14, 21, 22]. In England, Sweden, and Italy, almost all of the acute admissions take place in a PDGH [9–11]. Spain and France are in the process of implementing this system [12, 14]. Although there are people in the Netherlands in favour of giving the PDGHs a more prominent role, the number of PDGHs has dropped from 60 in 1997 to 51 in 2003, due to the fact that these PDGHs have become part of a regional mental health care centre; this was also anticipated for another approximately 20 PDGHs [68], but the Dutch government changed the rules [69]. Consequently, it appears that psychiatric hospitals, their regional mental health care centres and PDGHs in the Netherlands can coexist and their fate is more the result of local market forces than integral policy. This ‘market economy’ is supposed to be advantageous for patients, giving them freedom of choice in a certain area. However, this seems to be merely the positive side effect of a missed chance. Considering the (predicted) shortages of help and means within mental health care, allocation of tasks would provide a better solution for shortages when cooperation can be used to direct available services to existing needs. In the present situation, it is likely that competition between mental health services will increase now that the market is allowed to develop [70].

In conclusion, it can be said that it is unclear who has decided and decides on the direction to take in reforming Dutch mental health care in order to solve the remaining problems. Of these remaining problems, we have addressed the contrast of rehospitalisation in relation to the sudden, late but rapid reduction of the old mental hospitals and their premises, and the desired small yet relatively big scale for community-based psychiatry in relation to building mental health care centres. Compared with the issues of psychiatric reform as summarised at the beginning of this article, we conclude that the main points of psychiatric reform in the Netherlands until now have been the choice of a new size of scale for new facilities and the realisation of those new facilities in association with a merging of clinical and ambulatory institutes and downsizing of old-fashioned psychiatric hospitals. Less attention has been paid to legislation, attitudes of society towards psychiatry, the roles of other care suppliers such as GPs, the welfare sector, the general health care services, the balancing and financing of care, the fate of psychiatric patients coming from old hospitals, the way to cope with the ever-increasing demand for psychiatric help and the actual quality of psychiatric help. Right now, these issues are subject to different opinions, policies and financing strategies, resulting in a fragmented overall strategy. During the last decennium, the essential and fundamental debate on such an integrative long-term policy in mental health care has not taken place in the Netherlands [71].

This more integrative approach is desirable, coherently addressing the aspects above. The very recent Mental Health Declaration and Action plan for Europe, issued at the WHO European Ministerial Conference on Mental Health in Helsinki, Finland in January 2005, gives priority to such things as access to mental health care by GPs, access to new psychopharmacological drugs and the development of community care particularly for people with severe mental disorders [72].

However, it is not necessary to view the course of development described above in a negative light: the creation of ambulatory, regional facilities before decreasing the number of psychiatric institutions is a sequence of events that was missed in other countries, such as Australia and the United Kingdom, where the abolition of the old institutions and rebuilding of new was fragmented and resulted in negative effects. Counting our blessings on that issue, however, should not be an excuse for leaning back and refraining from formulating a new coherent policy for mental health care with clear targets and an open eye to interrelated issues.

Quality: The degree to which the total aspects of a product, process or service meets the explicit or implicit needs of the user.

Deinstitutionalisation: the transfer of mentally ill people from state hospitals to the community, their diversion from hospital admission and the development of alternative community services.

Dehospitalisation: the transfer of mentally ill people from state hospitals into the community, without evaluating whether or where they have gone.

Mental health care: mental health care offers care to people with psychological problems and psychiatric disorders. This care is offered by different organisations and professionals: regional institutions for ambulant mental health care (RIAGGs), general psychiatric hospitals (APZ), including their outpatient and day treatment facilities, the psychiatric wards of general hospitals (PAAZ), regional institutions for sheltered living (RIBW), private psychiatrists and private psychotherapists. Also institutions for children and adolescent psychiatry, drugs clinics and forensic psychiatry clinics are parts of mental health care.

Health care innovation: (synonym: care renewal, care innovation) a deliberately chosen change in the care supply by an existing organisation with the aim of increasing quality or efficacy. Innovation in relation to the ‘traditional’ situation can be aimed at: site, intervention, target group, formal or informal care supplier, and method of reimbursement.

Regional: a specific demographic or geographical area of a country.

RGC (Regionaal GGZ Centrum): a regional mental health care centre. Because there is no existing consensus regarding the definition, we took as our starting point the following minimal definition characteristics that are the highest common factor for policy and advisory memoranda up to 1999:

• A care supply of social psychiatry, outpatient treatment, part-time treatment, short-term clinical treatment.

• These services are all offered in one location.

• The care supply includes at least the target group of adults (18+).

A care supply of social psychiatry, outpatient treatment, part-time treatment, short-term clinical treatment.

These services are all offered in one location.

The care supply includes at least the target group of adults (18+).

Care programme: a coherent set of care services for a defined group of patients, e.g. a diagnostic category.

Prof. Dr. C. Spreeuwenberg, MD, chair integrated chronic care, University Maastricht, the Netherlands.

One anonymous reviewer