Regional Correlates of Psychiatric Inpatient Treatment

International Journal of Environmental Research and Public Health Article Regional Correlates of Psychiatric Inpatient Treatment Taina Ala-Nikkola 1,2,3,*, Sami Pirkola 4, Minna Kaila 1,3, Samuli I. Saarni 5, Grigori Joffe 1, Raija Kontio 1,6, Olli Oranta 5, Minna Sadeniemi 2,7 and Kristian Wahlbeck 2 1University of Helsinki and Helsinki University Hospital, Välskärinkatu 12, Helsinki FI-00029, Finland; [email protected] (M.K.); [email protected] (G.J.); [email protected] (R.K.) 2Unit for Mental Health, National Institute for Health and Welfare (T.H.L.); Mannerheimintie 168, Helsinki FI-00270, Finland; [email protected] (M.S.); [email protected] (K.W.) 3Public Health Medicine, University of Helsinki and Helsinki University Hospital, Helsinki FI-000014, Finland 4University of Tampere School of Health Sciences, and Tampere University Hospital, Lääkärinkatu 1, Tampere FI-33014, Finland; [email protected] 5Turku University Hospital and University of Turku, Kiinanmyllynkatu 4-8, Turku FI-20520, Finland; [email protected] (S.I.S.); [email protected] (O.O.) 6University of Turku, Turku FI-20014, Finland 7Department of Social Services and Health Care, City of Helsinki, Helsinki FI-00099, Finland; [email protected] *Correspondence: [email protected]; Tel.: +358-50-345-8130 Academic Editor: Paul B. Tchounwou Received: 21 September 2016; Accepted: 30 November 2016; Published: 5 December 2016 Abstract: Current reforms of mental health and substance abuse services (MHS) emphasize community-based care and the downsizing of psychiatric hospitals. Reductions in acute and semi-acute hospital beds are achieved through shortened stays or by avoiding hospitalization. Understanding the factors that drive the current inpatient treatment provision is essential. We investigated how the MHS service structure (diversity of services and balance of personnel resources) and indicators of service need (mental health index, education, single household, and alcohol sales) correlated with acute and semi-acute inpatient treatment provision. The European Service Mapping Schedule-Revised (ESMS-R) tool was used to classify the adult MHS structure in southern Finland (population 1.8 million, 18+ years). The diversity of MHS in terms of range of outpatient and day care services or the overall personnel resourcing in inpatient or outpatient services was not associated with the inpatient treatment provision. In the univariate analyses, sold alcohol was associated with the inpatient treatment provision, while in the multivariate modeling, only a general index for mental health needs was associated with greater hospitalization. In the dehospitalization process, direct resource re-allocation and substituting of inpatient treatment with outpatient care per se is likely insufficient, since inpatient treatment is linked to contextual factors in the population and the health care system. Mental health services reforms require both strategic planning of service system as a whole and detailed understanding of effects of societal components. Keywords: psychiatry; mental health services; hospitalization; integrative medicine 1. Introduction Mental health and substance abuse services (MHS) across the globe are undergoing changes in governance, structure, and content [ 1 – 3 ]. Since the 1990s, there has been a major shift from institutional care towards outpatient community care, with the aim of creating balanced care models [4–6] and alternatives for acute inpatient care [ 7 – 9 ]. The exclusive use of specialized health care has been Int. J. Environ. Res. Public Health 2016,13, 1204; doi:10.3390/ijerph13121204 www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2016,13, 1204 2 of 16 associated with lower quality of life and fewer social contacts and networks [ 10 ]. The reasons for hospitalization are linked with affective symptoms, suicidal intoxication, and poor medication compliance in psychosis patients [ 11 ]; social withdrawal and conflicts with family members are also often reasons for hospitalization [ 12 ]. Severe mental health illnesses like schizophrenia are associated with longer hospitalizations in general [ 9 , 13 , 14 ]. In primary care, social networking can be supported differently, so joint use of specialized and primary care has been recommended [10]. How community and hospital care are best combined depends on specific local circumstances related to mental health policy and existing service structures [ 10 , 15 , 16 ]. According to the World Health Organization’s (WHO) Mental Health Action Plan 2013–2020, policy level decision-making and strong leadership are needed to accomplish a transformation [1]. Finland is a sparsely populated country of 5.5 million inhabitants, where municipalities are responsible for social and health care services. Each municipality is free to provide the public services as a municipal activity, or to purchase the services from an external provider, e.g., another municipality, a joint municipal authority or even a private provider. For specialized public health services, including specialized mental health care, the municipalities join to form hospital districts. Finland is divided into 21 hospital districts. The municipal health centres are the main providers of primary care services, and the hospital districts, which are owned and governed by the municipalities, are the main providers of specialist services. Current reforms in Finland are aimed at a profound integration of the primary and specialized mental health as well as social services (altogether further referred to as MHS) [17]. Likewise, in Finland, the MHS strategy 2010–2015 is aimed at developing more patient-centered, community-based and integrated psychiatric and somatic services, while further limiting the use of hospital-based services [ 18 – 21 ]. Despite the policy of investing in community care and the deinstitutionalization process, most MHS resources in Southern Finland are still allocated to hospital and non-hospital residential services, while low-threshold outpatient services are scarce [21–23]. Previously we reported that MHS diversity positively correlates with catchment area population size. We also noted that catchment areas with an outpatient-based MHS structure have fewer personnel (in full-time equivalents) than areas with hospital-centered services [ 20 ]. These findings have raised interest in the interrelationships between community-oriented care and inpatient care. In essence, it is important to understand which factors in the MHS structure and its client pool explain the currently large variation in psychiatric hospital use. For the purposes of the current reforms, including an ongoing dehospitalization process with increasing outpatient care and the integration of psychiatry with general hospitals, more knowledge on the preconditions for reducing the relatively high level of psychiatric hospitalization is needed [19]. We hypothesized that more diverse structure in mental health services with a relative emphasis on outpatient and day care services would be associated with a reduced number of inpatient treatments in acute and non-acute psychiatric facilities [4]. The aim of the present study is to explore whether the MHS structure (diversity of services and balance of personnel between community and hospital care) and indicators of mental health needs are associated with the provision of acute and time-limited non-acute psychiatric hospital treatment in a regional comparison. 2. Methods 2.1. The Study Area REFINEMENT (REsearch on FINancing systems’ Effect on the quality of MENTal health care) is a collaborative project of nine EU countries (for the second phase, internationally referred to as CEPHOS-link, six EU countries: Italy, Norway, Romania, Slovenia, Austria, and Finland), currently led by the Finnish National Institute for Health and Welfare (THL) and partly funded by the EC Seventh Research Framework. The study is contributing to the evidence base needed to plan reforms in financing and the integration of the MHS and overall health care system. Details of the FIN-REFINEMENT, the Finnish part of project, have been described previously [ 20 , 24 , 25 ]. Briefly, in this study, the study area included four hospital districts (regions) in the southernmost part Int. J. Environ. Res. Public Health 2016,13, 1204 3 of 16 of Finland: the Hospital District of Helsinki and Uusimaa, Kymenlaakso (Carea), Etelä-Karjala (Eksote), and the Hospital District of South-Western Finland [ 20 ]. These hospital districts consist of municipalities (n= 67) that form 13 non-overlapping catchment areas, each equipped with psychiatric inpatient services. The total population in the study area is 2.3 million people, with 1.8 million adults, which is approximately 43% of the Finnish adult population (aged 18+ years). The adult population varied considerably by catchment area, from approximately 18,200 (Turunmaa) to 500,000 inhabitants (Helsinki). 2.2. Data Collection 2.2.1. Structure We classified MHS by means of the European Service Mapping Schedule (ESMS-R). The ESMS-R allows for a standardized description of the key features of mental health service structures and provision, including those services provided by primary care and social services [ 20 , 24 – 28 ]. The data collection and instrument has been described previously [ 20 , 23 , 24 ]. Briefly, mental health services are classified into 89 different “Main Types of Care” (MTC) in the ESMS-R classification. The MTC is the main descriptor of the care function (e.g., mobile acute team or acute hospital care). The MTC are organized by the “Basic Stable Input of Care” (BSIC); i.e., the organizational units that provide the services (e.g., acute ward or day care center). MTC are allocated to six main branches of the ESMS-R: (1) information for care; (2) accessibility to care; (3) self-help and voluntary help; (4) outpatient care, (5) day care; and (6) residential care [28,29]. In addition, we collected the numbers of admission rates and of days spent in a hospital or residential services in one year as a means to map the inpatient treatment provision in the areas. The data were collected between 2012 and 2014 by trained researchers from public corporation data sources for the years 2012 and 2013, with co-operation from local stakeholders. Data on MHS provided by third sector and private providers were collected by structured questionnaires and followed up where necessary by e-mail and telephone contact. 2.2.2. Service Structure Variables The diversity of the MHS structure was measured by counting the main types of care (MTC) in outpatient and day care branches, the hypothesis being that diversified outpatient and day services diminish the need for hospital beds. The used terms are definite on Refinement Glossary [ 30 , 31 ]. “Outpatient services: Setting in which mental health services are provided on an outpatient basis, without overnight stay, either mobile (when the facility is capable of being moved to different locations) or fixed (when the person seeking care must travel to a fixed service site). There is contact between staff and service users for some purpose related to management of their condition and its associated clinical and social difficulties. These services are not provided as a part of the delivery of day care services, and they have at least some qualified health care professionals as staff members. Day care: Care provision (i) is normally available to several consumers at a time (rather than delivering services to individuals one at a time); (ii) provide some combination of treatment for problems related to long-term care needs: e.g., providing a structured activity, or social contact and/or support; (iii) have regular opening hours during which they are normally available: and (iv) expect consumers to stay at the facilities beyond the periods during which they have face-to-face contact with staff (i.e., the service is not simply based on individuals coming for appointments with staff and then leaving immediately after their appointments). The care delivery is usually planned in advance”. The ESMS-R tool’s residential service branch includes a total of 21 different MTC (Figure 1). Acute wards are defined by the ESMS-R as high and medium intensity acute care facilities with 24-h physician cover in a registered hospital. Non-acute wards are defined as time-limited facilities where a fixed maximum period of residence is routinely specified (temporary stay). A facility should be classified as time-limited if a maximum length of stay is fixed for at least 80% of those entering the facility [28,30]. Int. J. Environ. Res. Public Health 2016,13, 1204 4 of 16 Int.J.Environ.Res.PublicHealth2016,13,12047of17  classifiedastime‐limitedifamaximumlengthofstayisfixedforatleast80%ofthoseenteringthe facility[28,30].  Figure1.TheresidentialcareclassificationoftheESMS‐Rtool[28].Wardsincludedinthestudyare giveninbold. Asanindicatorofacuteandsemi‐acutepsychiatricinpatienttreatment,weusedthetotal numberofusedbedsonacutepsychiatrichospitalwards(ESMS‐RcategoriesR1andR2)andnon‐ acute,time‐limitedpsychiatrichospitalwards(R4).Thecalculationofthepercapitanumberof inpatienttreatmentbedsbyeachcatchmentareapopulationincludingtheinpatienttreatmentbeds physicallylocatedbothinsideandoutsideofthecatchmentareas.Wealsoexploredtheproportion ofbedsusedatothertypesofhospitalwards(e.g.,acutenon‐physiciancover)andnon‐hospital residentialserviceunits(e.g.,supportedhousing). 2.2.3.SocioeconomicandHealthFactorsRelatedtoMentalHealthNeeds Thesocioeconomicfactorsincludedarethosecommonlyrelatedtomentalhealthneeds: educationindex(educationyearsafterprimaryschool),alcoholsales(litersof100%alcoholper person),unemploymentrateintheworkingagepopulation,andtheproportionofsingle‐person households(%). Thementalhealthindex(MHI)isanindicatorofpopulationmentalhealthstatuscalculatedfor eachcatchmentareausingthreeyearsofdataon(1)numberofsuicidesandsuicideattempts;(2) personseligibleforspecialreimbursementforantipsychoticmedication;and(3)personsondisability pension(18–64yearsold)duetomentaldisorders.TheMHIforthewholeofFinlandissetto100.An MHIsmallerthan100indicatesabetterthanaveragestateofmentalhealth.Thesocioeconomicdata from2011–2012werecollectedfortheperiod2012–2013fromStatisticsFinlandandfromtheIndicator BankSotkanet(www.sotkanet.fi). 2.2.4.PersonnelResourceFactors Theallocatedpersonnelfull‐timeequivalents(FTE)perusedacuteandsemi‐acutebedindicate howwellthewardswereresourced,thehypothesisbeingthatbetterresourcingwouldleadtoshorter hospitalizations.Communityorientationwasoperationalizedastheproportionofallpersonnel Figure 1. The residential care classification of the ESMS-R tool [ 28 ]. Wards included in the study are given in bold. As an indicator of acute and semi-acute psychiatric inpatient treatment, we used the total number of used beds on acute psychiatric hospital wards (ESMS-R categories R1 and R2) and non-acute, time-limited psychiatric hospital wards (R4). The calculation of the per capita number of inpatient treatment beds by each catchment area population including the inpatient treatment beds physically located both inside and outside of the catchment areas. We also explored the proportion of beds used at other types of hospital wards (e.g., acute non-physician cover) and non-hospital residential service units (e.g., supported housing). 2.2.3. Socioeconomic and Health Factors Related to Mental Health Needs The socioeconomic factors included are those commonly related to mental health needs: education index (education years after primary school), alcohol sales (liters of 100% alcohol per person), unemployment rate in the working age population, and the proportion of single-person households (%). The mental health index (MHI) is an indicator of population mental health status calculated for each catchment area using three years of data on (1) number of suicides and suicide attempts; (2) persons eligible for special reimbursement for antipsychotic medication; and (3) persons on disability pension (18–64 years old) due to mental disorders. The MHI for the whole of Finland is set to 100. An MHI smaller than 100 indicates a better than average state of mental health. The socioeconomic data from 2011–2012 were collected for the period 2012–2013 from Statistics Finland and from the Indicator Bank Sotkanet (www.sotkanet.fi). 2.2.4. Personnel Resource Factors The allocated personnel full-time equivalents (FTE) per used acute and semi-acute bed indicate how well the wards were resourced, the hypothesis being that better resourcing would lead to shorter hospitalizations. Community orientation was operationalized as the proportion of all personnel allocated to outpatient and day care services, the hypothesis being that better community Int. J. Environ. Res. Public Health 2016,13, 1204 5 of 16 resourcing would lead to lower hospital use. The personnel allocation for community services is counted as two different variables: (1) the sum of total FTE allocated to outpatient and day care; and (2) the community-based service ratio (outpatient and day care FTEs/divided residential FTEs), where 100% is an equal allocation. 2.3. Data Analysis The SPSS statistics program version 22 was used for the analyses. Scatterplots were used to explore and illustrate the associations between indicators. Spearman correlation analysis was used to investigate the association between acute and semi-acute inpatient treatment provision (independent variable) and the explanatory MHS service structure variables: diversity of outpatient and day care services, personnel FTE allocation per bed, and community orientation. The correlations between acute and semi-acute beds and other hospital beds and non-hospital beds were also analyzed. The main outcome in the analyses was the provision of acute and semi-acute inpatient treatment, counted by used beds per 1000 (18+) on acute and semi-acute wards. The dependences between the explanatory variables and the provision of inpatient treatment were analyzed using linear regression analysis. First, univariate analyses were performed for each explanatory factor separately. Next, analyses were controlled for MHI. The significance level was set to p< 0.05. The analyses were carried out with Statistical Package for Social Sciences (SPSS) version 22 (IBM, Armonk, NY, USA). 3. Results 3.1. The Inpatient Treatment Provision and the Mental Health Service Structure 3.1.1. Inpatient Treatment Provision and Community (Outpatient and Day Care) Service Diversity The provision of treatment in different ward levels is presented in Table 1. The provision of acute and semi-acute hospital beds varied four-fold, between 0.27 and 1.00 beds per 1000 adults. The provision of other hospital beds varied from 0.0 to 0.08. The median length of stay was 18.7 days (mean 22.7, SD 7.7, range 20.02). The highest provision of acute and semi-acute beds was found in the small Turunmaa area (1.00), but notably this is based on only one ward situated in the area, and the per capita total of hospital provision by bed (1.06) is close to the average level per 1000 adults (mean 0.78). In scatterplots, the smallest district of Turunmaa, which also used the acute ward for long-term care, appeared to be more or less an outlier. Therefore, we also analyze and discuss its role separately where appropriate, while excluding it from all linear regression analyses. Figure 2a–c present the scatterplot and regression line between per capita acute or semi-acute bed provision and the availability of different outpatient and day care and total community (outpatient and day care) services. There was a weak direct relationship between the diversity of day care services and the provision of hospital beds (Figure 2b). The same effect is shown for total community services and the provision of hospital beds (Figure 2c). The service structure of the smallest Turunmaa area, with only one ward for all residential care, affected the model, intensifying the significance (visible in Figure 2a,c). A sensitivity analysis was performed, excluding the Turunmaa area, in which no significant relations were found. Int. J. Environ. Res. Public Health 2016,13, 1204 6 of 16 Table 1. Provised beds on different types of psychiatric wards and non-hospital services per 1000 (18+) and community service diversity. Catchment Area LänsiUusimaa (1) Lohja (2) Hyvinkää (3) Porvoo (4) Helsinki (5) Jorvi (6) Peijas (7) Carea (8) Eksote (9) Turku (10) Salo (11) VakkaSuomi (12) Turunmaa (13) SD Weighted Mean Size of catchment area 18+ population 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 Beds on acute and semiacute wards (R2, R4) * 0.44 0.27 0.44 0.32 0.53 0.33 0.28 0.42 0.46 0.66 0.29 0.53 1.00 0.20 0.45 Beds on other hospital wards (R3, R6) 0.44 0.80 0.54 0.39 0.44 0.22 0.48 0.26 0.14 0.42 0.11 0.00 0.06 0.22 0.33 Beds on non-hospital services ** 2.30 4.15 4.26 5.23 7.07 3.48 3.70 3.62 3.65 3.67 2.08 3.93 1.65 1.38 3.75 Total beds in all categories 3.18 5.22 5.24 5.94 8.04 4.03 4.46 4.30 4.25 4.75 2.48 4.46 2.71 1.44 4.54 Community service diversity *** 10 8 12 8 14 12 12 13 14 17 15 9 1 * R1 Wards were not found; ** ESMS-R codes: R0, R3.1, R3.1.1, R3.1.2, R5, R7, R8, R8.1, R8.2, R9.1, R9.2, R10.1, R10.2, R11, R12, R13, R14; *** Different services (main types of care) on community services (outpatent and day care services). Int. J. Environ. Res. Public Health 2016,13, 1204 7 of 16 Int.J.Environ.Res.PublicHealth2016,13,12047of17  Inscatterplots,thesmallestdistrictofTurunmaa,whichalsousedtheacutewardforlong‐term care,appearedtobemoreorlessanoutlier.Therefore,wealsoanalyzeanddiscussitsroleseparately whereappropriate,whileexcludingitfromalllinearregressionanalyses.Figure2a–cpresentthe scatterplotandregressionlinebetweenpercapitaacuteorsemi‐acutebedprovisionandthe availabilityofdifferentoutpatientanddaycareandtotalcommunity(outpatientanddaycare) services.Therewasaweakdirectrelationshipbetweenthediversityofdaycareservicesandthe provisionofhospitalbeds(Figure2b).Thesameeffectisshownfortotalcommunityservicesandthe provisionofhospitalbeds(Figure2c).TheservicestructureofthesmallestTurunmaaarea,withonly onewardforallresidentialcare,affectedthemodel,intensifyingthesignificance(visibleinFigure 2a,c).Asensitivityanalysiswasperformed,excludingtheTurunmaaarea,inwhichnosignificant relationswerefound.  a b Int.J.Environ.Res.PublicHealth2016,13,12047of17   Figure2.Associationbetweenacute‐andsemi‐acute(time‐limited)inpatienttreatmentprovisionand different;(a)outpatient‐,(b)daycare‐,and(c)totalcommunityservices‐available.Thenumbersin bracketsaftercatchmentareasindicatethedatacollectionorderandwereusedinapreviousarticle [20].  Theprovisionofotherresidentialservicesandotherhospitalbedsandnon‐hospitalbedswere notsignificantlyassociatedwithacuteandsemi‐acuteinpatienttreatmentprovision(Figure3a,b).  Figure3.Associationsbetweenacute‐andsemi‐acute(time‐limited)inpatienttreatmentprovision and;(a)otherhospitalbed‐and(b)non‐hospitalbed‐provisionper1000adults(18+).Thenumbers inbracketsaftercatchmentareasindicatethedatacollectionorderandwereusedinapreviousarticle [20]. 3.1.2.TheInpatientTreatmentProvisionandPersonnelResourceFactors ThehighestpercapitauseoftotalpersonnelresourceswasfoundinLänsi‐Uusimaa(5.1)and thelowestinJorvi(2.1)(range2.1–5.1,mean3.3).Thehighestpercapitauseofcommunity‐based personnelresources,countedasthesumofoutpatientanddaycareresources,wasfoundinLänsi‐ Uusimaa(1.9)andthelowestinTurunmaa(0.8)andLohja(0.9)(range0.8–1.9,mean1.3).Thegrade ofcommunityorientationvariedfrom0.9to0.3(Table2).  c ab Figure 2. Association between acuteand semi-acute (time-limited) inpatient treatment provision and different; ( a ) outpatient-, ( b ) day care-, and ( c ) total community servicesavailable. The numbers in brackets after catchment areas indicate the data collection order and were used in a previous article [ 20 ]. Int. J. Environ. Res. Public Health 2016,13, 1204 8 of 16 The provision of other residential services and other hospital beds and non-hospital beds were not significantly associated with acute and semi-acute inpatient treatment provision (Figure 3a,b). Int.J.Environ.Res.PublicHealth2016,13,12047of17   Figure2.Associationbetweenacute‐andsemi‐acute(time‐limited)inpatienttreatmentprovisionand different;(a)outpatient‐,(b)daycare‐,and(c)totalcommunityservices‐available.Thenumbersin bracketsaftercatchmentareasindicatethedatacollectionorderandwereusedinapreviousarticle [20].  Theprovisionofotherresidentialservicesandotherhospitalbedsandnon‐hospitalbedswere notsignificantlyassociatedwithacuteandsemi‐acuteinpatienttreatmentprovision(Figure3a,b).  Figure3.Associationsbetweenacute‐andsemi‐acute(time‐limited)inpatienttreatmentprovision and;(a)otherhospitalbed‐and(b)non‐hospitalbed‐provisionper1000adults(18+).Thenumbers inbracketsaftercatchmentareasindicatethedatacollectionorderandwereusedinapreviousarticle [20]. 3.1.2.TheInpatientTreatmentProvisionandPersonnelResourceFactors ThehighestpercapitauseoftotalpersonnelresourceswasfoundinLänsi‐Uusimaa(5.1)and thelowestinJorvi(2.1)(range2.1–5.1,mean3.3).Thehighestpercapitauseofcommunity‐based personnelresources,countedasthesumofoutpatientanddaycareresources,wasfoundinLänsi‐ Uusimaa(1.9)andthelowestinTurunmaa(0.8)andLohja(0.9)(range0.8–1.9,mean1.3).Thegrade ofcommunityorientationvariedfrom0.9to0.3(Table2).  c ab Figure 3. Associations between acuteand semi-acute (time-limited) inpatient treatment provision and; ( a ) other hospital bedand ( b ) non-hospital bedprovision per 1000 adults (18+). The numbers in brackets after catchment areas indicate the data collection order and were used in a previous article [ 20 ]. 3.1.2. The Inpatient Treatment Provision and Personnel Resource Factors The highest per capita use of total personnel resources was found in Länsi-Uusimaa (5.1) and the lowest in Jorvi (2.1) (range 2.1–5.1, mean 3.3). The highest per capita use of community-based personnel resources, counted as the sum of outpatient and day care resources, was found in Länsi-Uusimaa (1.9) and the lowest in Turunmaa (0.8) and Lohja (0.9) (range 0.8–1.9, mean 1.3). The grade of community orientation varied from 0.9 to 0.3 (Table 2). 3.2. The Inpatient Treatment Provision and Mental Health Need Indicators There were some differences between catchment areas regarding socioeconomic factors (Table 3). The average MHI was 90.3 (SD 21.9), indicating that in the study area as a whole, the need for mental health services may be lower than in Finland overall. The areas located in the Helsinki and Uusimaa Hospital District, near Finland’s capital, had a lower MHI than the national average. Further away from the capital, i.e., in Kymenlaakso, Etelä-Karjala, and South-Western Finland (Areas 10–13), the MHI was higher than the national average. In Spearman’s correlation analysis, the sold alcohol in liters per adult was significantly associated with the provision of acute and semi-acute hospital beds ( ρ = 0.606, p= 0.028). The education indices or number of single person households were not significantly associated with the per capita provision of acute and semi-acute beds (Table 4). Int. J. Environ. Res. Public Health 2016,13, 1204 9 of 16 Table 2. Personnel full time equivalents (FTE) resources on mental health services per 1000 (18+). Catchment Area LänsiUusimaa (1) Lohja (2) Hyvinkää (3) Porvoo (4) Helsinki (5) Jorvi (6) Peijas (7) Carea (8) Eksote (9) Turku (10) Salo (11) VakkaSuomi (12) Turunmaa (13) Study Area Size of catchment area 18+ population 35,316 70,192 138,973 74,079 497,814 227,605 185,984 141,085 107,612 151,616 128,039.2815 81,392 18,200 1,857,907 Personnel FTE resources per 1000 (18+) Day care (D) FTE per 1000 0.6 0.2 0.4 0.3 0.3 0.1 0.1 0.2 0.6 0.4 0.4 0.3 0.0 0.3 Outpatient care (O) FTE per 1000 1.3 0.7 0.9 0.7 1.1 0.9 0.9 0.9 0.8 1.4 1.2 1.0 0.8 1.0 Residential care (R) FTE per 1000 3.2 2.6 2.9 1.8 2.0 1.1 1.5 2.8 1.6 2.9 1.7 2.4 2.2 2.0 Community-based services = FTE D + O per 1000 1.9 0.9 1.3 0.9 1.4 1.0 1.0 1.1 1.4 1.8 1.6 1.2 0.8 1.3 Community orientation (FTE ratio outpatient/residential) 0.6 0.3 0.5 0.5 0.4 0.9 0.7 0.4 0.9 0.4 0.5 0.3 0.3 0.5 Total FTE per 1000 5.1 3.5 4.2 2.7 3.4 2.1 2.6 4.0 3.1 4.7 3.3 3.6 3.0 3.3 FTE: full-time equivalent. The numbers in brackets after catchment areas indicate the data collection order and were used in a previous article [20]. Int. J. Environ. Res. Public Health 2016,13, 1204 16 of 16 35. Rezvyy, G.; Oiesvold, T.; Parniakov, A.; Ponomarev, O.; Lazurko, O.; Olstad, R. The Barents Project in psychiatry: A systematic comparative mental health services study between northern Norway and Archangelsk County. Soc. Psychiatry Psychiatr. Epidem. 2007,42, 131–139. [CrossRef] [PubMed] 36. Jarman, B.; Hirsch, S.; White, P.; Driscoll, R. Predicting psychiatric admission rates. BMJ 1992 ,304, 1146–1151. [CrossRef] [PubMed] 37. Zanjani, F.; Smith, R.; Slavova, S.; Charnigo, R.; Schoenberg, N.; Martin, C.; Clayton, R. Concurrent alcohol and medication poisoning hospital admissions among older rural and urban residents. Am. J. Drug Alcohol Abuse 2016,42, 422–430. [CrossRef] [PubMed] 38. White, A.M.; Hingson, R.W.; Pan, I.J.; Yi, H.Y. Hospitalizations for alcohol and drug overdoses in young adults ages 18–24 in the United States, 1999–2008: Results from the nationwide inpatient sample. J. Stud. Alcohol Drug 2011,72, 774–786. [CrossRef] 39. Perala, J.; Suvisaari, J.; Saarni, S.I.; Kuoppasalmi, K.; Isometsa, E.; Pirkola, S.; Partonen, T.; Tuulio-Henriksson, A.; Hintikka, J.; Kieseppa, T.; et al. Lifetime prevalence of psychotic and bipolar I disorders in a general population. Arch. Gen. Psychiatry 2007,64, 19–28. [CrossRef] [PubMed] 40. Myklebust, L.H.; Olstad, R.; Bjorbekkmo, S.; Eisemann, M.; Wynn, R.; Sorgaard, K. Impact on continuity of care of decentralized versus partly centralized mental health care in northern Norway. Int. J. Integr. Care 2011,11, e142. [CrossRef] [PubMed] 41. Marshall, M.; Crowther, R.; Almaraz-Serrano, A.; Creed, F.; Sledge, W.; Kluiter, H.; Roberts, C.; Hill, E.; Wiersma, D.; Bond, G.R.; et al. Systematic reviews of the effectiveness of day care for people with severe mental disorders: (1) Acute day hospital versus admission; (2) vocational rehabilitation; (3) day hospital versus outpatient care. Health Technol. Assess. 2001,5, 1–75. [CrossRef] [PubMed] 42. Douzenis, A.; Seretis, D.; Nika, S.; Nikolaidou, P.; Papadopoulou, A.; Rizos, E.N.; Christodoulou, C.; Tsopelas, C.; Mitchell, D.; Lykouras, L. Factors affecting hospital stay in psychiatric patients: The role of active comorbidity. BMC Health Serv. Res. 2012,12. [CrossRef] [PubMed] 43. Lloyd-Evans, B.; Slade, M.; Jagielska, D.; Johnson, S. Residential alternatives to acute psychiatric hospital admission: Systematic review. Br. J. Psychiatry 2009,195, 109–117. [CrossRef] [PubMed] 44. Johnson, S.; Gilburt, H.; Lloyd-Evans, B.; Osborn, D.P.; Boardman, J.; Leese, M.; Shepherd, G.; Thornicroft, G.; Slade, M. In-patient and residential alternatives to standard acute psychiatric wards in England. Br. J. Psychiatry 2009,194, 456–463. [CrossRef] [PubMed] 45. Stulz, N.; Bielinski, D.; Junghan, U.M.; Hepp, U. Heavy use of psychiatric hospitals and the use of outpatient services in Switzerland. Psychiatr. Prax. 2012,39, 332–338. [PubMed] 46. Zhang, J.; Harvey, C.; Andrew, C. Factors associated with length of stay and the risk of readmission in an acute psychiatric inpatient facility: A retrospective study. Aust. N. Z. J. Psychiatry 2011 ,45, 578–585. [CrossRef] [PubMed] 47. Harris, M.G.; Buckingham, W.J.; Pirkis, J.; Groves, A.; Whiteford, H. Planning estimates for the provision of core mental health services in Queensland 2007 to 2017. Aust. N. Z. J. Psychiatry 2012 ,46, 982–994. [CrossRef] [PubMed] 48. Dieterich, M.; Irving, C.B.; Park, B.; Marshall, M. Intensive case management for severe mental illness. Cochrane Database Syst. Rev. 2010,10. [CrossRef] 49. Wahlbeck, K. Public mental health: The time is ripe for translation of evidence into practice. World Psychiatry 2015,14, 36–42. [CrossRef] [PubMed] © 2016 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/).