Integrating care for people with depression: developments in the Netherlands

Introduction In this article we describe the history and present state of integrated care for people with depression in the Netherlands. The central question is: what are the developments in integrated care for people with depression in the Netherlands? Methods We describe these developments from the role of an observer, and make use of several sources: important Dutch policy documents and research documents, our own national survey carried out in 2007, a number of reports and project descriptions and searches in PubMed and Google. Also key people were contacted to supply additional information. Results In the Netherlands two separate phases can be distinguished within integrated care for people with depression. From the beginning of the 1990s, specialized secondary Mental Health Care (MHC) began to develop care programmes, including programmes for people with depression. The implementation of these care programmes has taken years. Mass usage of care programmes only went ahead once the large-scale mergers between ambulatory and clinical MHC organizations around 2000 had taken effect. An analysis of these programmes shows, that they did not lead to integration with primary care. This changed in the second phase from around 2000. Then attention was directed more towards strengthening the GP within the treatment of depression, collaboration between primary and specialized care and the development of collective integrated care packages. Discussion We relate these developments to projects in other countries and discuss the scientific basis by using evidence of international literature reviews and metastudies. Some general recommendations are given about functional costing, the physical presence of MHC specialists in the primary care sector and the use of a common national standard for both primary care and specialized MHC.


Introduction
The impact of depressive disorders on society and individual people is enormous. According to the lowest estimates, in the Netherlands some 1.3 million people suffered from depression during the last 12 months [1]. Impressive attempts have been made to improve the care for people with depression. In the Netherlands there is growing insight that there should be better cooperation between primary care and secondary Mental Health Care (MHC), in order to avoid underdiagnosis and undertreatment, and to provide the necessary stepped care [2]. In this article we describe the history and present state of integrated care for people with depression in the Netherlands. The Dutch state of affairs is interesting, because there is a highly organized network of primary care and specialized MHC. Nevertheless it took a long time before cooperation between these two parties and integrated care for people with depression got off the ground, and still there is a lot to be desired.
In 2006, 5.7% of men and 9% of women in the Netherlands were affected by depression. Life prevalence is 13.7% for men and 24.7% for women. On average, more than 19% of people suffer from depression at some point in their lives [1]. However, some estimates show an annual prevalence of a maximum of 20% and life prevalence of 25% [3].
In terms of burden of disease, expressed in Disability Adjusted Life Years (DALYs), depression and dysthymia are in fourth place both nationally and internationally (after cardiovascular diseases, anxiety disorders and CVA). Furthermore, it is estimated that depressive disorders will have risen to second place in 2020 [4]. Depression is therefore a serious disorder, with a significant impact on quality of life.
The central question of this article is: what are the developments in integrated care for people with depression in the Netherlands? The purpose of this paper is to describe the specific conditions of the Dutch situation, how they affected the organization of care for depression and how they influenced the rise of integrated care. An analysis of these conditions and comparison with international literature should lead to better understanding of determining factors in integrative care and brings us to giving some general recommendations.

Method
We provide this summary of the Dutch experience from the role of an observer, and make use of several sources: important Dutch policy documents and research documents, our own national survey carried out in 2007 [5], a number of reports and project descriptions and searches in PubMed and Google. After studying the sources, we also contacted key people involved in these initiatives to request additional information.
The content of the developments will basically be presented in a chronological order. As many of these developments run along analogous lines, we distinguish between two phases. In the description of the first phase we focus on specialized MHC and the development of specific care programmes for people with mood disorders and depression, starting in the beginning of the 1990s. In the description of the second phase, from about 2000, we focus on the cooperation between primary care and specialized MHC, when integrative care between these leading parties starts to take shape. Table 1 shows a time table with the dates of the key developments and we refer to this time table in the text.

Organization of care in the Netherlands
There are two characteristics of MHC in the Netherlands that are fundamental to understanding the development of the Dutch situation. Firstly, Dutch mental health care has a two-tiered system-primary and secondary care. The primary sector includes general practitioners (GPs), social workers, and for the last 20 years also primary care psychologists. Recently social psychiatric nurses and consultant psychiatrists have also become available for the primary sector. Secondary specialized MHC services are provided by several organizations. The specialized MHC sector is organized separately from the general health care sector, except for the relatively small psychiatric departments of general hospitals (PAAZ units) and the University Clinics. The sector now has more than 70,000 employees (55,000 FTEs) and consists mainly of regional institutes, which serve areas from 125,000 up to 800,000 inhabitants. They operate in several locations spread over the region with teams of psychiatrists, psychologists, psychiatric nurses and other types of professionals. Almost 90% of the care provided by these regional MHC Institutes consists of ambulatory therapy [1].

Treatment of depression in care programmes within the specialized mental health services
The advent of care programmes for people with mood disorders Up until the late 1980s ( Table 1), treatment of depression in the Netherlands was fragmentary and uncoordinated. It was still usual for GPs to work solo, and the existence of the primary care psychologist was still barely developed. As for secondary care, people suffering from depression were directed to a multitude of different organizations and disciplines, depending on the referral culture and familiarity with the facility. Selfemployed psychotherapists and psychiatrists, RIAGGs (Regional Institute for Ambulatory Mental Health Care), psychiatric outpatient clinics and hospitals were all engaged in the treatment and guidance of people with depression, without a clear division of responsibilities or any form of coordination. This situation nourished the need in specialized MHC for integrated care in the form of care programmes. A care programme is a coherent supply of care to a defined target group, based on the best (scientific, professional and experience-based) knowledge about the problems experienced by the target group and the wishes and expectations of this group [6]. In the late 1980s, working with care programmes was strongly propagated by MHC pioneers in the ambulatory and clinical domains; they were not in favour of the generally separate and discontinuous outpatient and clinical MHC treatments and formed the Coordination Support Centre Programmatic Collaboration (COPS). 1 In 1992 they published the strategy document Mental Health Care in programmes (Table 1) [7]. In collaboration with four Multifunctional Units (small-scale collaborations between RIAGGs and Psychiatric hospitals), the COPS developed four concise care programmes, including the first Mood Disorders Programme [8]. This was published nationally in 1997, but apparently not implemented anywhere, as far as we know.
The first more elaborate care programme for mood disorders appeared in 2001 in the Utrecht region, and was developed in a joint venture between all the MHC institutions (Table 1) operating in the region [9]. After the implementation of this care programme, many others followed, including the National Depression Basic Programme in 2005 [10]. In 2007 (Table 1), after a great deal of discussion, the Multidisciplinary Depression Guideline was decided on by the National Steering Committee for Multidisciplinary MHC Guideline Development [11]. The difference between a guideline and a programme is that a guideline mainly concentrates on evidence-based treatment methods, while a programme goes into more detail about concrete aspects and the organization of care. The National Depression Guideline was approved by all the collaborating professional associations and also patient and family organizations. However, there was one significant exception: the Pandora Foundation, an authoritative organization which protects the interests of people with psychological and psychiatric problems, withheld its approval with regard to the content (omission of patients' perspective) and for political reasons (misuse of the guideline for financing purposes). 2 Since the turn of the century, the development of care programmes by MHC organizations in the Netherlands, sometimes in a regional context, has expanded enormously. This was accompanied by large-scale mergers between RIAGGs and psychiatric hospitals. In a national inventory carried out in 2006/2007 by the Julius Center and the Trimbos Institute [5], a total of 66 care programmes for mood disorders or depressions was counted. In addition, the above-mentioned research institutes listed another 21 combined programmes, most of which (17) were developed for anxiety and mood disorders. Nowadays, the Dutch MHC organizations aim at overall use of care programmes to treat people with depression.

Dutch depression care programmes analysed
To investigate the quality levels of these care programmes, we asked the participating organizations during the national inventory in 2006/7 [5] to send us their care programmes. Of the 87 care programmes for mood disorders surveyed, we selected 20 programmes

Integration between primary health care and mental health services
So far, we have described the development of integration of care for depressed people within the MHC sector, with the aid of the time table in Table 1. We did not yet focus on the relation between Primary Health Care (PHC) and Mental Health Care (MHC). This we will do in the following paragraph; once again we will follow the time table in Table 1.
Since 2000, initiatives have been developed to bridge the gap between primary and specialized care, and to acquire greater expertise in the primary care setting in the areas of recognition and treatment of people with depression. This often takes place through forms of collaboration and integration, sometimes under the names Disease Management or Collaborative Care. The initiatives in this area usually originate from specialized MHC, but there are also initiatives from the regional GP associations, often supported by the Dutch Society of GPs (NHG). 3 The first study of collaboration between primary care and specialized MHC was carried out in 2006 under the auspices of the Dutch organization for health research and development (ZonMW). A total of 121 completed Disease Management programmes in the Netherlands were investigated, including eight for depression [14].
The conclusion here was that there is still very little experience in the Netherlands with Disease Management in general, apart from diabetes. None of the eight depression programmes mentioned contain a full treatment course. They include: the prevention course 'Dealing with depression' which has been implemented in 80% of the MHC regions, an assertiveness training course for young people and an investigation into recognition of depression in older people by GPs. The largest project concerns collaboration between GP, psychiatrist and social psychiatric nurse in the treatment of depression in the primary care setting. This project was initiated by the University Medical Cen- 3 Nederlands Huisartsen Genootschap(NHG): Dutch Society of GPs. that were completely implemented, spread throughout the country, and with a broad variety of both ambulatory and institutional MHC organizations. The analysis of these 20 care programmes is based primarily on several criteria: nationally developed criteria for care programmes, including quality [12], specific criteria for mood disorders [11] and a 'structural framework' used in the Utrecht region, indicating which basic elements a care programme should contain [13]. These criteria were all developed in consensus with multidisciplinary groups of professionals. Table 2 mentions a number of these criteria, together with the number of programmes that met these criteria.
Most of these figures will speak for themselves. Around three-quarters of the 20 care programmes analysed met the basic criteria for describing the target group and patient's problem, for a quarter that was not the case; these programmes describe only the care on offer. The omission of a section about suicidal tendencies (criterion 29) in half the programmes is remarkable, particularly where these programmes are also aimed at severe depressions.
tre Groningen, where various projects to promote GP expertise have been set up since the 1990s, based on specialized MHC expertise. In the projects analysed by ZonMW, the first therapeutic effects and cost-effectiveness could be demonstrated after six months, but it still remained unclear to what extent relapse prevention would occur [15]. A follow-up investigation provided indications that supplementing primary care with psychiatric consultation or short-term Cognitive Behaviour therapy would also bring about long-term effects [16].
Another important initiative to bridge the gap between primary and specialized care, are the national Breakthrough Depression projects 1 and 2, carried out under the auspices of the Trimbos Institute. The first of these was held from 2004-2006; the second from 2006-2008. In Project 1, nine multidisciplinary 'breakthrough teams' participated; the intention was to use a specific 'breakthrough method' to achieve rapid implementation in various regions of the Multidisciplinary Depression Guideline and the Depression standard of the Dutch Society of GPs (NHG) [17]. The aims were to reduce overtreatment of people with a non-serious depression, and undertreatment of people with a serious depression. To achieve this, a stepped care procedure was set up, rising from minimal interventions in the first case to depression-specific treatment in the second. Minimal interventions include restraint with the prescription of antidepressants and the use of a psychoeducational tool. From the evaluation study, it emerged that overtreatment decreased from 61% to 11%, that the quality of diagnosis and treatment increased and that collaboration between primary and specialized care improved considerably [2]. In the second Breakthrough Depression project from 2006 to 2008, 24 multidisciplinary teams participated [18]. The aims were the same, supplemented with items, such as the detection of depression in older people [19]. An analysis of part of the project results appeared in the International Journal of Integrated Care [20]. Here the authors are more reserved about the positive effects of the project collaboration between primary and specialized care, especially as far as severely depressed patients are concerned. This conclusion is largely due to technical limitations in the project. Nevertheless, the authors state that a stepped care approach with collaboration between PHC and MHC is achievable. On the other hand, especially the treatment of people with severe depressions needs to be further investigated.
We would like to mention the following examples of lasting structural collaboration between primary and specialized care in the area of depression, some of which are in the beginning stages.
As a practical follow-up to Breakthrough project 1, a total of 20 multidisciplinary general practice teams (consist-ing of GP, psychologist, social worker, physiotherapist, MHC organization or self-employed psychiatrist) in the southern regions of the Netherlands have continued a permanent collaboration in the area of depression. After Breakthrough project 2, approximately another 30 teams in these areas will start a structural collaboration, with support from the health insurance companies. 4 A powerful impetus is given here by the local Regional Support Structure for primary care (ROS).
Also worthy of mention is a project initiated by the MHC organization Rivierduinen. This comprehensive project, in which seven large MHC organizations are now participating, began as a Routine Outcome Monitoring (ROM) project for Mood, Anxiety and Somatoform (SAS) disorders. This was done by means of a web-based set of measuring instruments separately developed for the purpose, called QuestManager [21,22]. By now, this project has been extended with a collaboration project between specialized MHC and GPs, monitored by the University Medical Centre Leiden [23]. The aim is to support the GPs with the help of a web-based instrument called PrimQuest, when making a diagnosis of people with a SAS disorder and/or alcohol dependence. The ambition is to improve synchronization and also to allow ROM to follow and manage the course of treatment [24]. 5 Another interesting example is Vicino, a collaboration established in autumn 2008 between the Amsterdam MHC organizations, GPs and health centres, whose aims include the 'combining of expertise, joint development of primary care MHC programmes and preventive activities' [25]. This elaborates on the existing presence of social psychiatric nurses in the Amsterdam GP practice, and prepares the way for the Primary Care Nurses 6 in the area of mental health care. 7 The last example is the organization Indigo, which also fits within the perspective of introducing MHC expertise into primary care. Indigo was set up by specialized MHC and is now a collaboration between six MHC organizations. Indigo provides a generalized supply of MHC products within GP practices and community centres, and presents itself as support for the GP. It supplies a broad package of prevention, treatment, guidance and consultation to GPs and is also preparing to supply the above-mentioned MHC Primary Care Nurses. By now, the treatment of depression, collaboration between primary and specialized care and the development of collective integrated care packages. On the one hand, impetus came from universities and GP training programmes; they were pursuing a qualitative improvement in the expertise of GPs [15,16]. On the other hand, specialized MHC had identified that too heavy a load was being placed on their resources due to inappropriate referrals of people with a light depression, and that undertreatment of people with a serious depression was also taking place [2]. The changes in the Dutch health care system also affected this development, as they gave a powerful stimulus towards a commercial attitude. In various places in the Netherlands, outpatient treatment centres for various conditions including depression were developed by specialized MHC. So from diverse perspectives, collaboration and tuning of primary and specialized care received a powerful impulse. Alongside an impressive number of projects in the field of depression, the first precursors of lasting structural collaboration between primary care and specialized MHC also became known.
How do the Dutch developments in the field of integrated care of depression relate to projects in other countries? For secondary specialized MHC this is difficult to say, as the Dutch state of affairs in this respect is quite unique: by now the country is covered with a network of regional MHC centres, more or less working along programmatic lines. This means there has been considerable nationwide specialization along evidence-based lines. As far as we know, this has not been done on such a vast scale in other countries. However, this development has not resulted in integration with primary care in the Netherlands. For the projects in the second phase it is easier to make a international comparison, as they aim directly at integration and collaboration between primary depression care and specialized MHC. As such they are comparable with projects as described by Badamgarav et al. [3], Neumeyer-Gromen et al. [27], Gilbody et al. [28], Gensichen et al. [29] and Smith et al. [30]. Common characteristics of these integrated programmes include: delivery system design, regular communication between primary care and specialized practitioners, accessible electronic systems with a collectively agreed set of data, electronic support for the programme and possibilities for self-management and self-direction for the client. Although it is methodologically difficult to establish which of these factors affect the quality of care and to what extent, it does become clear that these types of integrated programmes make diagnosis, treatment and medication prescription more effective. In the aforementioned studies this has been proved rather thoroughly by mostly randomized controlled trials (RCTs). In the Netherlands, scientific around 25,000 clients are making direct or indirect use of what Indigo has to offer. 8 At the moment of submitting this paper to IJIC, many professionals, ambulatory and institutional MHC organizations are working with care programmes for people suffering from depression. Most of these programmes are based either in primary health care organizations, or in mental health care organizations. Full integration of programmes based equally in both sectors is not really available.

Discussion
Within integrated care for people with depression, two separate phases can be distinguished in the Netherlands. From the beginning of the 1990s, specialized secondary MHC began to develop care programmes, including programmes for people with depression. These care programmes were judged necessary to improve the quality of care and make it more transparent, but in particular to bridge the gap between the unique constellation of ambulatory and clinical MHC in existence at that time. The implementation of these care programmes has taken years. Only once the large-scale mergers between ambulatory and clinical MHC organizations, which took place around 2000, had taken effect and almost the entire country was covered by around 40 large regional MHC organizations, did the mass usage of care programmes go ahead. During our investigation in 2007 [5], we established that all the large MHC organizations at that time worked with care programmes in some form, which led to a general acceptance of working in an evidence-based, multidisciplinary way, guided by protocols. This was confirmed by our analysis of care programmes for people with depression, described above. In both analyses, we nevertheless ascertained that this should be developed further. At the time of the investigation, no systematic process evaluation or monitoring of results took place in one third of cases. In addition, we established that a good relationship is missing with, for example, primary care. So, the first phase of integrated care within MHC in the Netherlands is characterized principally by internal examination: in many places, the main purpose of care programmes was to put things right in their own merged MHC organization. However, this development should not be underestimated: by these mergers and care programmes, the decade-old dichotomy between hospital and ambulatory mental health care in the Netherlands was overcome [26].
In the second phase, from around 2000, attention was directed more towards strengthening the GP within of the Dutch projects is that they fail in the end, if financial incentives are discontinued after the project ends and if no way of structural financing can be found. These problems are aggravated if funding has to take place on the basis of diverse financing systems. We therefore recommend a way of functional costing to facilitate a structural continuation of these projects.
In the second place, a common characteristic in most of the Dutch collaboration projects is the physical presence of MHC specialists in the primary care sector. Usually this is done by MHC-nurses, sometimes also by psychologists. In addition, regular consultation meetings of GP's with a multidisciplinary team or psychiatrist by telephone or video are often part of these projects. Knowing each other and regular communication are vital for the success of the collaboration project. Therefore, physical presence of representatives of the specialized MHC in primary care is strongly recommended.
Our last recommendation regards the use of a common national standard for both primary care and specialized MHC. Up till now in the Netherlands, GPs and specialized MHC work with their own Standards and/ or Guidelines. In the last decade there has been a lot of discussion between representatives of primary care and specialized MHC. Thanks to these deliberations and growing scientific evidence, the initial differences concerning the content of these guidelines, have faded away. Now both parties have welcomed a measure of the Dutch government to develop a National Care Depression Standard. This will stimulate integrative care, because clear statements are expected not only about the content, but also about the process of care and the necessity of a multidisciplinary care group.  [20,23]. This also applies to some research into the efficiency and cost effectiveness of Collaborative Care for depression [31]. 9 However, in many cases described, such as the Breakthrough projects, there are no RCTs, and research is primarily directed towards the implementation. Here, the check is mainly to see if the process is working out as planned; additionally, some investigation does take place into a limited number of clinical effects by carrying out a benchmark and follow-up measurements, but without making use of control groups [2,18]. Recently a study has started into the possibility of implementation of an American collaborative care model into the Dutch situation [32]. This study draws interesting conclusions about the possibility of collaborative care, but also here the clinical effects have not been investigated so far.

Reviewers
As a result, pronouncements about the effects of integrated care on depression in the Netherlands can up to now only rely on a limited scientific basis. However, the tentative conclusions of the projects carried out in the Netherlands are in agreement with the aforementioned results of research carried out mostly abroad, in terms of more effective diagnosis, treatment and prescription behaviour. From the majority of these studies, it emerges that integrated care leads to a reduction in the seriousness of the depressive symptoms, that quality of life improves and treatment compliance increases.
For these reasons there is now international acknowledgment that collaboration between primary and specialized care should be stimulated. What can we learn in this respect from the Dutch situation? At first that the development of care programmes in the specialized Dutch MHC did not by itself lead to more integration with the primary care sector. Only once competition made its entrance in Dutch health care and several competing projects 'intruded' in the MHC-regions, Dutch MHC realized that cooperation with primary care was necessary not only for the patient's benefit, but also for the continuation of specialized services of the organization. So beside scientific evidence, the market regulation has done a lot to bring together primary care and secondary specialized MHC.
However, market regulation alone will not bring about a continuation of the desired innovation, if a solid financial basis is lacking. The experience with some