Identifying local and centralized mental health services—The development of a new categorizing variable

International Journal of Environmental Research and Public Health Article Identifying Local and Centralized Mental Health Services—The Development of a New Categorizing Variable Taina Ala-Nikkola 1,2,3,*ID , Sami Pirkola 4, Minna Kaila 1,3, Grigori Joffe 1, Raija Kontio 1,5,6, Olli Oranta 7ID , Minna Sadeniemi 1,2,8, Kristian Wahlbeck 2ID and Samuli I. Saarni 7 1Clinic of Psychiatry and Clinic of Public Health Välskärinkatu 12 and Stenbäckinkatu 9, University of Helsinki and Helsinki University Hospital, FI-00029 Helsinki, Finland; [email protected] (M.K.); [email protected] (G.J.); [email protected] (R.K.); [email protected] (M.S.) 2Unit for Mental Health, National Institute for Health and Welfare (T.H.L.), Mannerheimintie 168, FI-00270 Helsinki, Finland; [email protected] 3Public Health Medicine, University of Helsinki and Helsinki University Hospital, FI-000014 Helsinki, Finland 4University of Tampere School of Health Sciences and Tampere University Hospital, Lääkärinkatu 1, FI-33014 Tampere, Finland; [email protected] 5University of Turku, FI-20014 Turku, Finland 6Lohja Hospital Area, Sairaalakatu 8, 08200 Lohja, Finland 7Turku University Hospital and University of Turku, Kiinanmyllynkatu 4-8, FI-20520 Turku, Finland; [email protected] (O.O.); [email protected] (S.I.S.) 8Department of Social Services and Health Care, City of Helsinki, FI-00099 Helsinki, Finland *Correspondence: [email protected]; Tel.: +35-85-0345-8130 Received: 16 April 2018; Accepted: 28 May 2018; Published: 31 May 2018   Abstract: The challenges of mental health and substance abuse services (MHS) require shifting of the balance of resources from institutional care to community care. In order to track progress, an instrument that can describe these attributes of MHS is needed. We created a coding variable in the European Service Mapping Schedule-Revised (ESMS-R) mapping tool using a modified Delphi panel that classified MHS into centralized, local services with gatekeeping and local services without gatekeeping. For feasibility and validity, we tested the variable on a dataset comprising MHS in Southern Finland, covering a population of 2.3 million people. There were differences in the characteristics of services between our study regions. In our data, 41% were classified as centralized, 37% as local without gatekeeping and 22% as local services with gatekeeping. The proportion of resources allocated to local services varied from 20% to 43%. Reclassifying ESMS-R is an easy way to compare the important local vs. centralized balance of MHS systems globally, where such data exists. Further international studies comparing systems and validating this approach are needed. Keywords: mental health care; health service research; integrated care; European Service Mapping Schedule-Revised 1. Introduction The global ongoing reforms in mental health and substance abuse services (MHS) are defined in terms of balancing and integrating institutional and community care. The balance of services is being moved from hospitals, so that most services are provided in community settings close to the populations served, and hospital stays are reduced as far as possible. The integration of health services means that mental health services should be functionally integrated with other services; for example, mental health with primary care, and acute wards within general hospitals [1]. Int. J. Environ. Res. Public Health 2018,15, 1131; doi:10.3390/ijerph15061131 www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2018,15, 1131 2 of 16 The evidence suggests that compared to hospital-centered systems, community-based health systems reach more patients [ 2 – 5 ], human rights are better respected [ 6 – 8 ], and de-stigmatization is achieved [ 9 , 10 ]. Diversified community-based MHS structures are associated with lower suicide rates than traditional hospital-based systems [ 11 ]. Community-based services are generally, although not always, more easily accessible to patients, and without gatekeeping, and the peer-support that is often offered by third-sector providers is more respecting of patient’s autonomy and self-determination, thus furthering de-stigmatization [ 7 , 12 , 13 ]. In addition, a more preventive approach is often seen as more cost-effective in the long run than traditional institutional and inpatient-based systems [ 5 ]. Community-based care has been found to be cost effective if the quality of institutional care is simultaneously developed [ 1 , 14 , 15 ]. In addition to mental health services, patients also need physical health services and social support (for education, work, accommodation). Meeting this need requires an integrated system in which community level primary services, secondary level specialized services, and tertiary level services function as a whole [ 5 , 16 , 17 ]. The fragmentation of physical and mental health services needs to be reversed for improved equality and outcomes of care for persons with mental disorders. Barriers such as administrative, financial and clinical hurdles need to be identified before successful integration of MHS. Kilbourne et al. (2008) suggested that strategies to overcome barriers to integrated care may require cooperation across different organizational levels, including administrators, providers and health care financers in order for integrated care to be established and sustained over time [18]. An adequate division of responsibilities between secondary and primary care (vertical collaboration and integration) and between social and health care organizations (horizontal integration) are also needed in Finland, where major reforms of social and health services are planned. (http: //alueuudistus.fi/en/frontpage) [ 19 , 20 ]. On a horizontal scale, organizations operate within their own substance area (health service, social service or mental health service) at the same level of specialization, and value expansion is achieved by cooperation. On the vertical scale, the level of specialization increases when moving to the next stage of organization, e.g., from primary care to the secondary or tertiary care level. The reform aims to fully integrate all social and health services into user-oriented services, in effect (at least in theory) making the traditional organizational divisions between primary care, secondary care and social care obsolete. In general, such development could provide a more comprehensive and more easily manageable service structure, regardless of the funding and steering option. The MHS cost should be seen as a whole, because intensive local services could be more expensive than even long-term hospital care, but may still be seen as more cost-effective because of better outcomes. The local administrators and budget controllers need to engage in joint planning in order to develop effective and cost-effective care [ 21 ]. Investigating mental health system structures and their relationship to health outcomes is important. As this is often also very complex, investigating the effects of organizational sub-components (such as integration of care) on health, or the effects of strategy-level decisions on organizational sub-components or other intermediate outcomes (such as resource shifts), is important. Some service settings can even be viewed as treatments (e.g., “partial hospitalization”), whereas some treatments are always embedded in a service matrix (e.g., assertive community treatment) or organizationally combined (e.g., “integrated treatment” for co-occurring mental disorder and substance abuse). Ideally, for example, studies would focus on horizontal and vertical integration, primary care vs. secondary care, and local vs. centralized mental health authorities—Each of which could be conceptualized as a health care technology, with empirical studies to assess its effectiveness. [ 22 ]. All these studies need comparable, reliable instruments for classifying health care systems. More generally, for all evidence-informed reforms, a fact-based view of the current state of affairs must be obtained, along with a view of the future and measurement instruments that can indicate progress. Thus, an instrument is needed that can be used to track changes in mental health services, including the balance between community (local) and centralized services. The European Service Mapping Schedule-Revised (ESMS-R) was designed to map mental health services, to describe their Int. J. Environ. Res. Public Health 2018,15, 1131 3 of 16 major characteristics, the provision of services, as well as resource allocation. The ESMS-R instrument allows for a good-quality common description of the socioeconomic profile of the population of a specified area, alongside key features of mental health service provision, including those provided by primary care and social services [ 17 , 23 – 25 ]. The first version of the ESMS has been used previously in Finland [11,26] and other European countries [27–29] as well as in Chile [30]. However, the ESMS-R as a mapping tool does not differentiate between services that should be provided locally and those that can be centralized, or between services without gatekeeping and those where gatekeeping is used. This kind of information would be valuable when reforming the organizations that provide health and social care, as is currently planned in Finland. We thus set out to develop a definition to differentiate local services and to apply that categorization to the ESMS-R instrument. In the study area, centralized services are mainly organized by hospital districts or specialized private or third sector organizations. Local services can be reached with or without gatekeeping, referral or other prior specialist consulting. All services can be organized by public, private or third sector (e.g., foundations or associations) organizations. The specific aims of the study were: 1. To create a new coding variable for the ESMS-R mapping tree to categorize MHS into local and centralized categories, for the use of developmental activities in different settings globally. 2. To test the feasibility of this new variable as a potential quality indicator for MHS, using a Finnish dataset representing a publicly managed Western service. 3. We also set out to test whether a quality indicator for MHS could be developed for ESMS-R, based on the hypothesis that when more MHS without gatekeeping are locally available, less centralized services are required. 2. Methods The study methods consist of two parts: creation of the new coding variable for ESMS-R and then testing the new “Local service” variable. We used a modified Delphi procedure with alternating theoretical meetings and practical classification phases [ 31 – 33 ] to create the new variable, and a Finnish dataset covering a population of 2.3 million within 13 different catchment areas for testing. The creation process is reported in Methods and feasibility testing in Results. 2.1. The European Service Mapping Schedule-Revised Instrument and Dataset The ESMS-R is derived from the previous European Service Mapping Schedule and the Description and Evaluation of Services and Directories for Long-Term Care in Europe coding system [ 17 , 23 , 34 ]. In the used version of ESMS-R, mental health services are classified into 89 different “main types of care” (MTC). The MTC is the main descriptor of the care function (for example, mobile acute team or acute hospital care). The MTCs are organized according to the “basic stable input of care” (BSIC), such as the organizational units that provide the services (for example, an acute ward or a day care center). The MTCs are categorized into six main branches: information for care (I), accessibility to care (A), self-help and voluntary help (S), outpatient care (O), day care (D), and residential care (R) [23,35]. The whole ESMS-R system is described in Supplementary Figure S1. We used the ESMS-R data collected by trained researchers from Southern Finland between 2012 and 2014. The data collection has been described previously [ 24 , 25 , 36 ]. Briefly, the study area included four hospital districts: Helsinki and Uusimaa, Carea, Etelä-Karjala, and Varsinais-Suomi. The districts are further divided into thirteen non-overlapping health care areas. The size of the adult population varied within the areas, from 18,200 to 500,000 inhabitants, the median being 128,000. In total, the study area covers 2.3 million people, with 1.8 million adults (18+), in 67 municipalities and representing 43% of the Finnish population. The study area is much more densely populated than the average for the whole country (174 vs. 16 inhabitants per square kilometer). Each study area has its own psychiatric in-hospital care, with some coordination at the hospital district level. Psychiatric hospital Int. J. Environ. Res. Public Health 2018,15, 1131 4 of 16 care is integrated into general hospitals in some study areas, but most areas still have free-standing psychiatric hospitals. The ESMS-R service mapping covers all municipalities in the study area and is aimed to include all adult (18+) mental health and substance use services (in primary, secondary and tertiary health care) and social services located in the catchment areas. The personnel resources allocated to each service unit were measured based on full-time equivalents (FTE). The services were classified by their vertical organizational level: primary care, secondary care or integrated care. Horizontally, the services were classified by their legal status: public, third sector, or private companies [24,35]. 2.2. Creating the New “Local Service” Variable on Local Versus Centralized Services (Part One) A modified Delphi technique [ 31 , 33 ] was used to develop the criteria for local services and the new “local service” variable for the ESMS-R service mapping system. The Delhi technique is used in various fields and is suitable e.g., for policy determination, such as decisions concerning which services are better arranged locally, with or without gatekeeping. The Delhi process in this study concentrated on seeking consensus for ESMS-R service classification in live meetings and independent practice classifications. We invited an expert panel consisting of eleven mental health professionals (researchers and administrators) familiar with the ESMS-R, the ongoing research project and the current Finnish MHS system. The panel consisted of two senior administrative psychiatrists, two senior administrative nurses, a research professor, and researchers with work experience in the areas under study (Supplementary Table S1). In the first part of the study the panel worked in four phases: Phase 1. Two theoretical meetings Phase 2. Individual classification (round one) Phase 3. Consensus meeting and practical classification Phase 4. Individual classification (round two) and final decisions 2.2.1. Phase One: Theoretical Meetings Two meetings were held to discuss how “local” and “centralized” services should be conceptualized and defined in MHS. The main idea was that services that are needed often should be arranged in local settings, in contrast with services that are needed seldom or need more resources or special equipment, which are better to arrange in a centralized setting. The background material included a draft developed by the Ministry of Social Affairs and Health for a new legal conceptualization of local and centralized health and social services [19]. The baseline proposal for the panel was made by T.A.N., K.W. and S.S., suggesting that all services on the main branches of “information for care”, and “accessibility to care” should be classified as local services, and all services on the main branch “residential care” should be classified as centralized services. Thus, only services under the branches “self-help and voluntary care”, “outpatient care” and “day care” would need to be re-classified. The first theoretical meeting concluded that the initial definition of local versus centralized MHS was insufficient, as it did not recognize the potential difference between organizations or patients’ viewpoints regarding local or centralized services. The second theoretical meeting elaborated on this and concluded that in addition to physical locality, other factors influencing access to services should be considered. Thus, two new viewpoints should be addressed: (1) the needs of patients to access services with no or low barriers to access (distance, gatekeeping, costs etc.) and (2) the needs of organizations to recognize some services as so complex, rare or expensive that they require a specialist assessment or some other kind of gatekeeping. Thus, two categories were added: “local services with gatekeeping” and “centralized services without gatekeeping”, changing the initial dichotomy into Int. J. Environ. Res. Public Health 2018,15, 1131 5 of 16 four categories (Figure 1). This classification was accepted for empirical testing. It was agreed that “residential care” would be classified as centralized services with gatekeeping. Int. J. Environ. Res. Public Health 2018, 15, x FOR PEER REVIEW 5 of 17 Figure 1. Quadrangle for ESMS-R* main types of care (MTC) new variable classification (Round One). *European Service Mapping Schedule–Revised. 2.2.2. Phase Two: Individual Classification (Round One) The quadrangle shown in Figure 1 was used as the basis for re-classifying the 46 (n = 89) types of services in ESMS-R branches O and D and validating the baseline proposal on ESMS-R branches I, A and S-H. The baseline proposal on ESMS-R branch R (centralized with gatekeeping) was agreed without re-classification. Six panelists participated in the exercise. The panelists were in 100% agreement on the category “information for care”, in 90% agreement on “self-help services” and in 80% agreement on “accessibility to care”. Their responses were more variable regarding the “outpatient” (n = 24) and “day care” main branches (n = 22) (Supplementary Table S2). 2.2.3. Phase Three: Consensus Meeting and Preliminary Testing of the Classification The aim of the Delphi meeting was to reach consensus and make final decisions on how the 89 MTCs should be divided into the four categories. The classifications in which at least half of the experts agreed were accepted and only the remaining 14 of 89 MTC were further discussed (Supplementary Table S3). Discussions were continued until consensus was reached. The resulting classification divided the different MTCs relatively equally between centralized with gatekeeping (37%), local services with gatekeeping (31%), and local services without gatekeeping (27%). However, only four MTC were classified in the category centralized services without gatekeeping (5%) (Figure 2a). Investigation of the empirical dataset consisting of 987 organizational units (i.e., BSICs) indicated that centralized services with gatekeeping were most frequent (40.1%), followed by local services without gatekeeping (37%), local services with gatekeeping (19%) and centralized services without gatekeeping (3%) (Supplementary Table S3). Local Centralized eCare, ICT-service, HUS mental health house (I1.1) Crisis center (O4.2) Without gatekeeping (no referral needed) Peer support group (I1.3) Psychiatric emergency unit (O3.1) Voluntary activities (S1.4) Health centre (O10.1) Psychiatric outpatient clinic (O9.1) Acute admission ward (R2) With gatekeeping (referral or other specialist assessment needed) Psychiatric and substance abuse care (O7.1) Rehabiliation ward (R4) Therapy centre (D8.1) Supported housing service (R11) Nursing home (R12) Figure 1. Quadrangle for ESMS-R* main types of care (MTC) new variable classification (Round One). *European Service Mapping Schedule–Revised. 2.2.2. Phase Two: Individual Classification (Round One) The quadrangle shown in Figure 1was used as the basis for re-classifying the 46 (n= 89) types of services in ESMS-R branches O and D and validating the baseline proposal on ESMS-R branches I, A and S-H. The baseline proposal on ESMS-R branch R (centralized with gatekeeping) was agreed without re-classification. Six panelists participated in the exercise. The panelists were in 100% agreement on the category “information for care”, in 90% agreement on “self-help services” and in 80% agreement on “accessibility to care”. Their responses were more variable regarding the “outpatient” (n= 24) and “day care” main branches (n= 22) (Supplementary Table S2). 2.2.3. Phase Three: Consensus Meeting and Preliminary Testing of the Classification The aim of the Delphi meeting was to reach consensus and make final decisions on how the 89 MTCs should be divided into the four categories. The classifications in which at least half of the experts agreed were accepted and only the remaining 14 of 89 MTC were further discussed (Supplementary Table S3). Discussions were continued until consensus was reached. The resulting classification divided the different MTCs relatively equally between centralized with gatekeeping (37%), local services with gatekeeping (31%), and local services without gatekeeping (27%). However, only four MTC were classified in the category centralized services without gatekeeping (5%) (Figure 2a). Investigation of the empirical dataset consisting of 987 organizational units (i.e., BSICs) indicated that centralized services with gatekeeping were most frequent (40.1%), followed by local Int. J. Environ. Res. Public Health 2018,15, 1131 6 of 16 services without gatekeeping (37%), local services with gatekeeping (19%) and centralized services without gatekeeping (3%) (Supplementary Table S3). Int. J. Environ. Res. Public Health 2018, 15, x FOR PEER REVIEW 6 of 17 (a) (b) Figure 2. Proportions of different main types of care (MTCs) classified as local or centralized; initial and final classifications. Initial classifications (a) and final classifications (b) represented 2.2.4 Phase Four: Individual Classification (Round Two) and Final Decisions The results of the classification were considered problematic, as there were only four types of care (ESMS-R codes 0.3.1, 0.3.2, 0.4.1 and 0.4.2) that were classified as centralized services without gatekeeping. Furthermore, it was considered that centralization of services can often in itself effectively result in gatekeeping, caused for example by the distance to services. The panel decided that these services should be merged into the other three categories. A second individual classification round was organized to reclassify these four types of care (Shown with red colour in Supplementary Table S2). The final local or centralized services variable thus included three categories: (1) local services without gatekeeping, (2) local services with gatekeeping and (3) centralized services (Figure 2b). The final classification of every ESMS-R MTC is shown in Supplementary Figure 1. In sum, MTCs belonging to the “information for care”, “accessibility to care” and “self-help and voluntary care” categories were mostly classified to local services without gatekeeping (7/9, 3/5 and 7/10, respectively), “outpatient care” to local services with gatekeeping (19/24) and “day care” and “residential care” to centralized services (12/22 and 19/19, respectively). 24 28 4 33 Initial classification in four categories 1 = Local without gatekeeping 2 =Local with gatekeeping 3 = Centralized without gatekeeping 4 = Centralized with gatekeeping 20 18 23 Final classification with three categories 1= Local without gatekeeping 2= Local with gatekeeping 3= Centralized Figure 2. Proportions of different main types of care (MTCs) classified as local or centralized; initial and final classifications. Initial classifications (a) and final classifications (b) represented 2.2.4. Phase Four: Individual Classification (Round Two) and Final Decisions The results of the classification were considered problematic, as there were only four types of care (ESMS-R codes 0.3.1, 0.3.2, 0.4.1 and 0.4.2) that were classified as centralized services without gatekeeping. Furthermore, it was considered that centralization of services can often in itself effectively result in gatekeeping, caused for example by the distance to services. The panel decided that these services should be merged into the other three categories. A second individual classification round was organized to reclassify these four types of care (Shown with red colour in Supplementary Table S2). The final local or centralized services variable thus included three categories: (1) local services without gatekeeping, (2) local services with gatekeeping and (3) centralized services (Figure 2b). The final classification of every ESMS-R MTC is shown in Supplementary Figure 1. In sum, MTCs belonging to the “information for care”, “accessibility to care” and “self-help and voluntary care” Int. J. Environ. Res. Public Health 2018,15, 1131 7 of 16 categories were mostly classified to local services without gatekeeping (7/9, 3/5 and 7/10, respectively), “outpatient care” to local services with gatekeeping (19/24) and “day care” and “residential care” to centralized services (12/22 and 19/19, respectively). 2.3. Testing the New European Service Mapping Schedule-Revised-Local Service Variable (Part Two) The practical usability of the new ESMS-R-Local service variable was tested by addressing the following questions in the dataset: 1. The balance between local and centralized services was explored by comparing: (a) The proportion of service units classified as local or centralized. (b) The proportion of resources measured as full-time equivalents allocated to local services. 2. The differences in proportion regarding services provided as local without gatekeeping (in BSIC and FTE) between the areas were explored and considered as a quality indicator. 3. The types of services provided by public (primary or secondary health care), private, or third sector providers were explored in order to estimate how different types of local or centralized services integrate horizontally and vertically with other health services. Descriptive statistics were used to explore associations, while Spearman rank correlations and linear regression modeling were used for analyzing associations. 3. Results: Testing the New “Local Service” Variable 3.1. The Balance between Local and Centralized Services Our dataset included services classified into 61 different MTCs (out of the 89 different possibilities in the used version of ESMS-R) delivered at 986 service units (BSIC) with a total full-time equivalent (FTE) personnel of 6785. The distribution of these BSICs and FTE to local or centralized services is shown in distribution of service units (BSICs) and mental health personnel (FTE) to local or/and centralized services Table 1. Of the service units identified, 41% were classified as centralized, 37% as local services without gatekeeping and 22% as local services with gatekeeping. Of the personnel, 67% worked in centralized services, 11% in local services without gatekeeping and 22% in local services with gatekeeping. Table 1. Distribution of service units (basic stable input of care) and mental health personnel (full time equivalent) to local or/and centralized services. ESMS-R Main Branch Local Services Without Gatekeeping Local Services With Gatekeeping Centralized Services BSIC Found n(%) Different MTC Found/Possible * Information for care 7 2 0 22 (2) 6/9 Accessibility for care 3 2 0 6 (1) 1/5 Self-help and voluntary care 7 1 2 191 (19) 6/10 Outpatient care 3 19 2 279 (28) 17/24 Day care 4 6 12 157 (16) 17/22 Residential care 0 0 19 331 (34)) 14/19 Different BSIC found n(%) 367 (37.2) 213 (21.6) 406 (41.2) 986 Different MTC found / possible * 20/24 18/30 23/35 61/89 Percentage of personnel ** (%) 11 22 67 * ESMS-R includes 89 categories of the main types of care (MTCs), of which our data includes 61. ** Personnel converted to full-time equivalents (FTE). Int. J. Environ. Res. Public Health 2018,15, 1131 8 of 16 3.2. The Difference between Study Areas in the Proportion of Services Provided as Local Services without Gatekeeping as a Potential Quality Indicator The full time equivalent (FTE) personnel per 1000 adults (18+) by provider status (a.), organizational level (b.), local versus centralized services (c.) is shown in Table 2and the difference in service units (BSIC) is shown in Supplementary Table S4. The personnel allocated to any local (with or without gatekeeping) services varied widely, from 0.31 per 1000 adults in the Carea study area to 1.08–1.13 persons per 1000 in Turku and Salo (Table 2). The proportion of total resources allocated to local services varied less, but the range was still high, from 20% to 43% (mean 31%). The number of personnel allocated to local services without gatekeeping again varied more, from 0.06 per 1000 (Kymenlaakso) to 1.01 (Salo) (mean 0.5). The proportion of total resources allocated to local services without gatekeeping varied from 1.4% (Kymenlaakso) to 30.5% (Salo) (mean 13.1%). The most populous area (Helsinki) had the largest total number of BSICs (196), whereas Turunmaa had the smallest number (13). Conversely, the Salo and Carea areas had the highest numbers of service units relative to the population (0.8 units per 1000 adults), whereas the smallest number was found in the Jorvi area (0.3). There was variation between the areas in the number of local services without gatekeeping relative to the population (range 0.10 to 0.49, mean 0.25 per 1000 adults), local services without gatekeeping (0 to 0.26, mean 0.10) and centralized services (0.12 to 0.35, mean 0.23). There were no significant associations between the number of personnel (per 1000 adults) and the proportion of personnel allocated to local services, or between the number of personnel and the number of local service units. Table 3shows the correlations between allocated full time equivalents per 1000 18+ in local vs centralized services. There is a strong correlation between total personnel and personnel allocated to centralized services, but not with personnel allocated to other types of services. To investigate the issue further, linear regression models were created with total personnel resources as a dependent variable and the proportion of personnel allocated to centralized services as an independent variable, while controlling for population size, service needs indicator (mental health index) or both. The relative proportions of resources allocated to centralized services were not correlated with the total resources in the regression models, with or without controls. Only the mental health index correlated significantly with total resources. Int. J. Environ. Res. Public Health 2018,15, 1131 9 of 16 Table 2. The full time equivalent (FTE) personnel per 1000 adults (18+) by provider status (a.), organizational level (b.), local versus centralized services (c.). Catchment Areas * LänsiUusiMaa Lohja HyvinKää Porvoo Helsinki Jorvi Peijas KymenLaakso Eksote Turku Salo VakkaSuomi TurunMaa Sum Weighted Average SD Mental health index (Finland = 100) 92.3 94 92.9 89 90 77.2 89.6 106.2 102.7 109.7 101 102.9 101.3 96 8.9 Size of catchment areas adult (18+) population (2012) 35,296 70,379 139,734 74,611 501,929 230,005 187,332 143,265 109,379 151,616 128,039 81,392 18,200 1,871,178 128,000 (median) 122,759 Total personell (FTE) Total personnel FTE per 1000 3.64 4.10 2.8827 3.10 4.01 2.80 3.46 4.17 2.98 4.82 3.31 3.67 2.93 3.63 0.61 a) Providers status Public personnel FTE per 1000 2.58 2.43 1.9642 1.52 2.69 2.06 2.51 2.38 2.43 2.58 2.03 1.87 1.94 2.35 0.35 Third sector personnel FTE per 1000 0.19 0.43 0.3766 0.97 1.26 0.38 0.39 1.11 0.14 0.95 0.59 0.18 0.00 0.73 0.40 Private company personnel FTE per 1000 0.87 1.24 0.5419 0.61 0.05 0.36 0.56 0.68 0.41 1.29 0.69 1.62 0.99 0.55 0.43 b) Organizational level 6784.72 3.62 Primary health care personnel FTE per 1000 0.59 1.31 1.3203 1.14 1.28 2.05 2.04 0.72 0.23 3.22 1.90 1.85 0.28 Primary healthcare personnel FTE per 1000 2.14 2.24 1.2661 1.92 1.87 1.30 1.78 2.37 1.01 2.51 1.98 2.37 1.22 1.83 0.49 Secondary healthcare personnel FTE per 1000 1.50 1.87 1.6165 1.17 2.14 1.50 1.68 1.80 0.18 2.31 1.33 1.30 1.71 1.69 0.52 Integrate primary and secondary healthcare personnel FTE per 1000 1.79 195.61 c) Local vs centralized service level Local without gatekeeping FTE per 1000 0.51 0.33 0.4835 0.20 0.22 0.17 0.37 0.06 0.25 0.92 1.01 0.71 0.77 0.46 0.30 Local with gatekeeping FTE per 1000 0.46 0.90 0.6343 0.44 1.07 0.80 0.97 1.05 0.78 0.78 0.41 0.29 0.00 0.66 0.32 Total local resources (without and with gatekeeping) per 1000 0.97 1.23 1.1178 0.64 1.29 0.97 1.33 1.11 1.03 1.70 1.41 1.00 0.77 1.12 Int. J. Environ. Res. Public Health 2018,15, 1131 16 of 16 32. Tuomisto, L.; Erhola, M.; Kaila, M.; Brander, P.E.; Puolijoki, H.; Kauppinen, R.; Koskela, K. Asthma Programme in Finland: High consensus between general practitioners and pulmonologists on the contents of an asthma referral letter. Prim. Care Respir. J. 2004,13, 205–210. [CrossRef] [PubMed] 33. Campbell, S.M.; Braspenning, J.; Hutchinson, A.; Marshall, M. Research methods used in developing and applying quality indicators in primary care. Qual. Saf. Health Care 2002,11, 358–364. [CrossRef] [PubMed] 34. Salvador-Carulla, L.; Alvarez-Galvez, J.; Romero, C.; Gutierrez-Colosia, M.R.; Weber, G.; McDaid, D.; Dimitrov, H.; Sprah, L.; Kalseth, B.; Tibaldi, G.; et al. Evaluation of an integrated system for classification, assessment and comparison of services for long-term care in Europe: The eDESDE-LTC study. BMC Health Serv. Res. 2013,13, 218. [CrossRef] [PubMed] 35. Refinement-Project, 2013. Available online: http://www.refinementproject.eu/ (accessed on 15 May 2018). 36. Ala-Nikkola, T.; Pirkola, S.; Kontio, R.; Joffe, G.; Pankakoski, M.; Malin, M.; Sadeniemi, M.; Kaila, M.; Wahlbeck, K. Size matters—Determinants of modern, community-oriented mental health services. Int. J. Environ. Res. Public Health 2014,11, 8456–8474. [CrossRef] [PubMed] 37. Thornicroft, G.; Tansella, M. Components of a modern mental health service: A pragmatic balance of community and hospital care: Overview of systematic evidence. Br. J. Psychiatry 2004 ,185, 283–290. [CrossRef] [PubMed] 38. Tello, J.E.; Jones, J.; Bonizzato, P.; Mazzi, M.; Amaddeo, F.; Tansella, M. A census-based socio-economic status (SES) index as a tool to examine the relationship between mental health services use and deprivation. Soc. Sci. Med. 2005,61, 2096–2105. [CrossRef] [PubMed] 39. Tucker, C.A.; Escorpizo, R.; Cieza, A.; Lai, J.S.; Stucki, G.; Ustun, T.B.; Kostanjsek, N.; Cella, D.; Forrest, C.B. Mapping the content of the Patient-Reported Outcomes Measurement Information System (PROMIS(R)) using the International Classification of Functioning, Health and Disability. Qual. Life Res. 2014 ,23, 2431–2438. [CrossRef] [PubMed] 40. Madden, R.H.; Dune, T.; Lukersmith, S.; Hartley, S.; Kuipers, P.; Gargett, A.; Llewellyn, G. The relevance of the International Classification of Functioning, Disability and Health (ICF) in monitoring and evaluating Community-based Rehabilitation (CBR). Disabil. Rehabil. 2014,36, 826–837. [CrossRef] [PubMed] 41. Prodinger, B.; Cieza, A.; Oberhauser, C.; Bickenbach, J.; Ustun, T.B.; Chatterji, S.; Stucki, G. Toward the International Classification of Functioning, Disability and Health (ICF) Rehabilitation Set: A Minimal Generic Set of Domains for Rehabilitation as a Health Strategy. Arch. Phys. Med. Rehabil. 2016 ,97, 875–884. [CrossRef] [PubMed] 42. WHO Aims 2.0. Available online: http://www.who.int/mental_health/evidence/AIMS_WHO_2_2.pdf (accessed on 15 May 2018). 43. Saxena, S.; Lora, A.; Morris, J.; Berrino, A.; Esparza, P.; Barrett, T.; van Ommeren, M.; Saraceno, B. Mental health services in 42 lowand middle-income countries: A WHO-AIMS cross-national analysis. Psychiatr. Serv. 2011,62, 123–125. [CrossRef] [PubMed] 44. McPherson, P.; Krotofil, J.; Killaspy, H. What Works? Toward a New Classification System for Mental Health Supported Accommodation Services: The Simple Taxonomy for Supported Accommodation (STAX-SA). Int. J. Environ. Res. Public Health 2018,15, 190. [CrossRef] [PubMed] 45. De Silva, M.J.; Lee, L.; Fuhr, D.C.; Rathod, S.; Chisholm, D.; Schellenberg, J.; Patel, V. Estimating the coverage of mental health programmes: A systematic review. Int. J. Epidemiol. 2014 ,43, 341–353. [CrossRef] [PubMed] 46. Brieger, P.; Wetzig, F.; Bocker, F.M. Institutions and services of psychiatric care in Saxony-Anhalt: Assessment with the European Services Mapping Schedule. Eur. Psychiatry 2003,18, 145–147. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).