DEPRESSION AND ANXIETY 25:27–37 (2008) Research Article USE OF HEALTH SERVICES FOR MAJOR DEPRESSIVE AND ANXIETY DISORDERS IN FINLAND J. Hämäläinen, M.D., M.A.,1,2 E. Isometsä, M.D., Ph.D.,1,3 S. Sihvo, Ph.D.,4 S. Pirkola, M.D., Ph.D.,4 and O. Kiviruusu, B.Soc.Sc1 Factors associated with people suffering from major depressive disorder (MDD) or anxiety disorders seeking or receiving treatment are not well known. In the Health 2000 Study, a representative sample (n 5 6005) of Finland’s general adult (Z30 years) population was interviewed with the M-CIDI for mental disorders and health service use for mental problems during the last 12 months. Predictors for service use among those with DSM-IV MDD (n 5 298) or anxiety disorders (n 5 242) were assessed. Of subjects with MDD, anxiety disorders, or both, 34%, 36%, and 59% used health services, respectively. Greater severity and perceived disability, psychiatric comorbidity, and living alone predicted health care use for MDD subjects, and greater perceived disability, psychiatric comorbidity, younger age, and parent’s psychiatric problems for anxiety disorder subjects. The use of specialist-level mental health services was predicted by psychiatric comorbidity, but not characteristics of the disorders per se. Perceived disability and comorbidity are factors influencing the use of mental health services by both anxiety disorder and MDD subjects. However, still only approximately one-half of those suffering from even severe and comorbid disorders use health services for them. Depression and Anxiety 25:27–37, 2008. & 2007 Wiley-Liss, Inc. Key words: health care utilization; major depressive disorder; anxiety disorders; general population INTRODUCTION Depressive and anxiety disorders are common and have a substantial impact on functioning and quality of life [Lopez and Murray, 1998]. The efficacy and effectiveness of pharmacological and psychotherapeutic treatments for these disorders have been established and numerous evidence-based practice guidelines for treatment exist [Hyler, 2002; NHS, Depression 2004; NHS, Anxiety, 2004]. However, it is similarly well established by epidemiological studies worldwide that about half of the subjects suffering from severe forms of these disorders and more than half of those with moderate or mild disorders do not seek or receive treatment from health services [Kessler, 2005]. Previous studies [Leaf et al., 1988; Bassett et al., 1998; Lefebvre et al., 1998; Ten Have et al., 2002] have suggested that severity of the psychiatric symptoms, psychiatric comorbidity, available resources, and different health behavior patterns influence seeking r 2007 Wiley-Liss, Inc. 1 National Public Health Institute, Department of Mental Health and Alcohol Research, Helsinki, Finland 2 Helsinki City Health Department, Eastern Health Centre, Department of Psychiatry, Helsinki, Finland 3 Department of Psychiatry, University of Helsinki, Helsinki, Finland 4 STAKES, National Research and Development Centre for Welfare and Health, Helsinki, Finland Contract grant sponsor: Helsinki City Health Department. Correspondence to: Juha Hämäläinen, National Public Health Institute, Department of Mental Health and Alcohol Research, Mannerheimintie 166, 00300 Helsinki, Finland. E-mail: Juha.Hamalainen@ktl.fi Received for publication 14 February 2006; Revised 5 July 2006; Accepted 18 July 2006 DOI 10.1002/da.20256 Published online 19 January 2007 in Wiley InterScience (www. interscience.wiley.com). 28 Hämäläinen et al. treatment. Also, different sociodemographic factors may have an influence on help-seeking [Olfson and Klerman, 1992; Crow et al., 1994; Lin and Parikh, 1999]. Although several recommendations and consensus statements have been produced in order to define the role of general and specialist psychiatric health care services [Alonso et al., 2004], little is known about the factors affecting the actual flow of patients in the health care system. The availability of pharmaco- and psychotherapies for depressive and anxiety disorders have improved over the last few years. Epidemiological information on treatment therefore will rapidly become outdated. Treatment-seeking behavior is affected by changes in both treatment provision practices and public awareness of acceptability, accessibility, and usefulness of treatments. As these factors vary from one setting to another and over time, information from different countries with differing health care systems is useful for a comprehensive epidemiological view. For purposes of public health evaluation, only relatively recent information on service provision is valuable. Previous Finnish studies have reported that in 1996 the proportion of subjects in the general population with a major depressive episode who used any health services for their depression during the past 12 months was only 28% [Laukkala et al., 2001; Hämäläinen et al., 2004]. Whether this finding remains true is unknown. Furthermore, studies focusing on individual disorders may not provide the whole picture, as comorbidity is common and may influence help-seeking behavior. Knowledge about possible urban/rural residence and other sociodemographic differences in the use of health care services for mental disorders is important for developing these services. To date, however, only a few nationwide studies on health service use for comorbid depressive and anxiety disorders have been carried out [Regier et al., 1993; Bebbington et al., 2000; Alonso et al., 2004; Wang et al., 2005]. Our aim was to investigate patterns of use of health care services for major depressive disorder (MDD) and anxiety disorders, and possible differences between those two groups, in Finland using a nationally representative survey from 2000/2001. Our particular interest was in investigating the impact of disorder severity, mental disorder comorbidity, concurrent somatic illnesses, sociodemographic factors, and patterns of health behavior on health service use for mental health problems. We also investigated factors influencing the distribution of patients between primary and specialized care. MATERIALS AND METHODS This study is based on a multidisciplinary epidemiological study, the Health 2000 Study, conducted in 2000–2001 in Finland. The two-stage stratified clustersampling frame comprised 8028 adults age 30 years and over living in mainland Finland. This frame was Depression and Anxiety DOI 10.1002/da regionally stratified according to the five university hospital regions, each serving approximately one million inhabitants. The data collection phase started in August 2000 and was completed in March 2001, with a total of 7415 subjects attending one or another phase of the study. Data were collected by home interviews and examinations, telephone interviews, and health questionnaires, followed by a clinical health examination including a structured mental health interview. Details and the methodology of the project have been published elsewhere [Aromaa and Koskinen, 2003; Pirkola et al., 2005]. CIDI INTERVIEW A Finnish-translation of the German computerized version of the CIDI (Composite International Diagnostic Interview) (M-CIDI) [Wittchen et al., 1998] was performed on 6038 subjects (95% of those attending the comprehensive health examination phase), 33 of whom were subsequently excluded for interviewee unreliability (e.g., mental retardation, self-expressed intention to lie). The total number of included interviews was thus 6005, which is 75% of the original sample. The mental health interview was carried out at the end of the comprehensive health examination. Interviews were conducted to determine the 12-month prevalences of major depressive episodes and disorder, dysthymia, general anxiety disorder, panic disorder with or without agoraphobia, agoraphobia, social phobia, alcohol abuse, and alcohol dependence. For a detailed description of the CIDI procedure, see Pirkola et al. [2005]. MENTAL DISORDERS AND THEIR COMORBIDITY The available DSM-IV diagnoses were grouped into the two categories of MDD and anxiety disorders (including generalized anxiety disorder, agoraphobia, panic disorder with or without agoraphobia, and social phobia) [Pirkola et al., 2005]. Information about MDD included its severity, duration, recurrence, and resulting subjective disability. For anxiety disorders, information included severity and subjective disability. In addition to MDD and four different anxiety disorders, we collected data on dysthymia, alcohol disorders, and smoking status. We studied comorbidity of MDD and an anxiety disorder within the last 12 months. The severity of MDD was assessed on the basis of the number (5–9) of depressive symptoms described by the subject and divided into three categories: mild (5), moderate (6, 7), and severe (8, 9). As we had no information about the duration of the current major depressive episode, we used the duration of the most severe lifetime episode (in weeks) and the duration of depressive symptoms (in years) as proxy variables. The severity of anxiety disorders was classified on the basis of both the number of pertinent anxiety symptoms and the presence of avoidance into three grades (mild, Research Article: Health Service Use for Depression and Anxiety few symptoms and small avoidance; moderate, many symptoms or great avoidance; and severe, many symptoms and great avoidance). Further, information was collected on the respondent’s subjective disability related to MDD or anxiety disorders (four categories: major, moderate, slight, and none). SOCIODEMOGRAPHIC AND HEALTH BEHAVIOR FACTORS Information on basic sociodemographic variables was collected in the interview. These variables included age, sex, marital status, current employment status, education, whether respondents reside in a rural or urban location, presence and number of chronic somatic disorders, and whether the subjects parents had psychiatric problems (e.g., schizophrenia, other psychosis, or depression) in his/her childhood. Level of education and employment status were classified into three categories and professional status into four. Subjects with alcohol use disorders fulfilled the diagnostic criteria of alcohol dependence or alcohol abuse during the last 12 months. Subjects were considered smokers if they reported current daily smoking of cigarettes, cigars, or a pipe and having smoked at least 100 times during their lifetime. The variables used to assess different diagnoses, sociodemographic factors, health behaviors, and somatic health (self-informed presence and number of somatic long-term illnesses) have been described in detail previously [Pirkola et al., 2005]. USE OF MENTAL HEALTH SERVICES Questions about health service use for mental problems during the past year (yes/no) covered use of specialist-level mental health services (including municipal services of psychiatric outpatient clinics, mental health centers, psychiatric hospitals, and private psychiatrists) and primary health care services (including among others, municipal health centers and occupational health services). Persons who had used both psychiatric and primary health services were classified under special mental health services. Information was also collected on the use of primary health care services for any reason during the past year. STATISTICAL METHODS In the statistical analyses, correlates for use of health services were first analyzed by basic bivariate analyses, including the chi-square test. To control for confounding factors, binomial logistic regression analyses were performed. Values of Po.05 were considered statistically significant. A weighting adjustment was used in the analyses to take into account the sampling design and nonparticipation [Aromaa and Koskinen, 2003; Pirkola et al., 2005]. The STATA statistical package (College Station, TX, v. 8.0) was used in analyses. 29 Logistic regression analyses with manual backward elimination procedures was used to choose variables that would best associate with any health service use for MDD and anxiety disorders. The use of general versus special mental health services was analyzed in a similar manner. Independent variables in the initial models included sociodemographic factors, severity of MDD and anxiety disorders, related disability, and all possible psychiatric comorbidities. RESULTS The sample of subjects with MDD (n 5 298) comprised 93 men (31%) and 205 women (69%). Of these subjects, 91 (31%) suffered from mild, 155 (52%) from moderate, and 52 (17%) from severe MDD, and the median duration of the most severe lifetime episode was 3, 3, and 12 weeks, respectively. The sample of subjects suffering from any anxiety disorder (n 5 242) comprised 95 men (39%) and 147 women (61%). The number of sufferers of different disorders were: panic disorder 114 (40%), generalized anxiety disorder 75 (27%), social anxiety disorder 60 (21%), and agoraphobia 33 (12%). Of persons suffering from any anxiety disorder, 71 (29%) suffered from mild, 106 (44%) from moderate, and 65 (27%) from severe disorder. Sixty-six individuals (22% of persons with MDD, 27% of persons with any anxiety disorder) had both MDD and any of the anxiety disorders. Of the comorbid group, 15 (23%) suffered from mild, 31 (47%) from moderate, and 20 (30%) from severe MDD. For anxiety disorders, the respective figures were 17 (26%), 29 (44%), and 20 (30%). USE OF HEALTH SERVICES The proportion of subjects classified as having MDD who used any health services for mental problems during the past 12 months was 30% for men and 36% for women. The corresponding proportions for anxiety disorder were 38% and 35%. For persons having both MDD and an anxiety disorder the proportion was 58% (Fig. 1). The relation between use of health services and sociodemographic factors differed for MDD and anxiety disorders (Table 1). For MDD and anxiety disorders, the reported use was higher (32–41%) in the younger age groups, but somewhat lower (11–21%) in the older (Z60 years) age group. The difference was statistically significant both for anxiety disorders (chisquare, P 5.031) and for MDD (chi-square, P 5.000). The median duration of MDD among all subjects was 4 weeks, 8 weeks in those using health services, and 3 weeks in nonusers (significant difference, Mann–Whitney U test, P 5.028). The median severity of the MDD was moderate in both users and nonusers. The median subjective disability related to depression was moderate in those using health services and mild in nonusers. The median severity of anxiety disorders was Depression and Anxiety DOI 10.1002/da 30 Hämäläinen et al. Cases with MDD and/or Anxiety DO MDD only MDD and Anxiety DO Anxiety DO only No use of health services for mental problems n 231 66 175 % 100 100 100 MDD only MDD and Anxiety DO Anxiety DO only n 167 28 126 % 72 42 72 Use of health services for mental problems MDD only MDD and Anxiety DO Anxiety DO only MDD only MDD and Anxiety DO Anxiety DO only n 64 38 49 % 28 58 28 General only Both general and specialist Specialist only n % n % n % n 29 13 8 12 17 10 14 6 12 18 18 10 21 9 18 27 14 8 35 15 30 45 32 18 Total specialist % Figure 1. Use of health services for mental problems with pure major depressive disorder (MDD), persons with MDD and an anxiety disorder (Anxiety DO) and persons with a pure anxiety disorder during the past 12 months. moderate in both users and nonusers. The median subjective disability related to anxiety disorders was moderate in those using health services and mild in nonusers. For specific anxiety disorders, the use of services differed slightly; for panic disorder, the use was 33% (38/114), for social phobia 51% (30/59), for agoraphobia 39% (13/33), and for generalized anxiety disorder 43% (32/75). In the final logistic regression models explaining the use of health services for mental health reasons (Tables 2, 3), subjective disability and comorbidity remained significant for both MDD and anxiety disorders. ROLE OF SPECIFIC SYMPTOMS For MDD, specific symptoms of depression were significantly associated with the use of health care services. In the preliminary model, an association was found between use of services and increased appetite, increased weight, increased sleep, feelings of guilt, feelings of inferiority, decreased self-confidence, poor memory, difficulty in making decisions, thoughts of death, wishing to die, thoughts of suicide, and plans for suicide. In a logistic regression model explaining contact with health care services, after adjusting for severity of MDD, feelings of inferiority (odds ratio (OR) 2.01, 95% confidence interval (CI): Depression and Anxiety DOI 10.1002/da 1.24–3.46) and plans for suicide (OR 17.26, 95% CI: 2.10–141.67) remained significant. Due to the small number of cases, we could not include suicide attempts in the analysis. USE OF SPECIALIST-LEVEL SERVICES Specialist-level mental health services accounted for 64% (71% for men and 61% for women) of all health service use associated with MDD and 71% (72% for men and 71% for women) of use associated with anxiety disorders. For both groups the age distribution of use was almost identical, with specialist-level services being used least by the oldest (Z60 years) age group. In a logistic regression model for use of specialist-level mental health services, (vs. use of general health services only), the only significant factor in the final model for MDD was comorbidity with an anxiety disorder (OR 3.08, 95% CI: 1.21–7.45). For anxiety disorders comorbidity with MDD was not significant (OR 1.99, 95% CI: 0.75–5.29). While specialists tended to see cases that were more severe, of longer duration, caused more subjective disability, and generally had more psychiatric comorbidity, the results were not significant after adjusting for other factors. Research Article: Health Service Use for Depression and Anxiety 31 TABLE 1. Relation of sociodemographic, disorder-specific, and comorbid psychiatric disorder factors and use of different levels of psychiatric health services for major depressive disorder (MDD) and anxiety disorders during the past 12 months No use Characteristic % MDD Sex Male 70 Female 64 Age, years 30–39 68 40–49 62 50–59 61 60– 79 Marital status Unmarried 58 Married 71 Cohabiting 65 Divorced 49 Widowed 84 Employment Employed 68 Unemployed 55 Retired 68 Student, etc. 62 Education Low 64 Medium 64 High 70 Living environment Urban 63 Semi-urban 66 Rural 70 Chronic somatic disorders No 69 1–2 58 3– 63 Smoking Yes 59 No 68 Severity of MDD Mild 76 Moderate 67 Severe 44 Duration of MDD, weeks 0–4 76 5–16 58 17–32 55 33– 54 Subjective disability Mild 78 Moderate 61 Severe 46 Recurrence of MDD Single 64 Recurrent 68 Alcohol disorder Yes 63 No 67 Dysthymia Yes 49 No 69 General level Specialist level Total n % n % n % n P 65 130 9 14 8 29 22 22 20 45 100 100 93 204 0.369 55 64 42 34 11 12 16 12 9 12 11 5 21 27 23 9 17 28 16 4 100 100 100 100 81 104 69 43 0.318 19 110 17 27 21 12 9 19 20 12 4 14 5 11 3 30 21 15 31 4 10 32 4 17 1 100 100 100 100 100 33 156 26 55 25 0.031 124 22 40 08 14 18 7 13 25 7 4 1 19 28 25 31 34 11 15 4 100 100 100 100 183 40 59 13 0.426 74 61 60 10 16 12 12 15 10 26 20 19 30 19 16 100 100 100 116 95 86 0.551 75 78 40 11 14 14 13 16 8 26 20 16 31 24 9 100 100 100 119 118 57 0.592 135 38 22 10 15 20 20 10 7 21 27 17 41 18 6 100 100 100 196 66 35 0.276 43 151 16 11 12 25 25 21 18 46 100 100 73 222 0.323 69 103 23 6 14 19 5 22 10 19 19 37 17 29 19 100 100 100 91 154 52 0.002 107 28 16 26 8 13 14 19 11 6 4 9 16 29 31 27 22 14 9 13 100 100 100 100 140 48 29 48 0.035 107 62 26 9 14 20 12 14 11 14 26 34 19 26 19 100 100 100 138 102 56 0.001 105 90 15 10 24 13 22 22 36 29 100 100 165 132 0.465 38 157 7 14 4 33 30 20 18 46 100 100 60 236 0.087 26 169 13 12 7 30 38 18 20 45 100 100 53 244 0.006 Depression and Anxiety DOI 10.1002/da 32 Hämäläinen et al. TABLE 1. Continued No use Characteristic % Panic disorder Yes 46 No 67 Social phobia Yes 29 No 69 Agoraphobia Yes 38 No 67 GAD Yes 44 No 68 Any anxiety disorder Yes 42 No 72 Parental psychiatric problems Yes 51 No 69 Anxiety disorders Sex Male 65 Female 62 Age, years 30–39 61 40–49 58 50–59 59 60– 89 Marital status Unmarried 60 Married 70 Cohabiting 46 Divorced 52 Widowed 92 Employment Employed 67 Unemployed 46 Retired 68 Student, etc. 69 Education Low 58 Medium 57 High 73 Living environment Urban 62 Semi-urban 60 Rural 77 Chronic somatic disorders No 66 1–2 63 3– 59 Smoking Yes 60 No 66 Severity Mild 70 Moderate 65 Severe 55 General level Specialist level Total n % n % n % n P 14 184 8 13 2 35 46 20 11 54 100 100 24 273 0.013 6 187 5 13 1 34 67 19 14 50 100 100 21 271 0.000 3 191 0 13 0 36 63 21 5 60 100 100 8 297 0.027 12 172 19 12 5 29 37 21 10 52 100 100 27 253 0.158 28 167 12 13 8 29 46 15 30 35 100 100 66 231 0.000 21 171 12 13 5 32 37 18 15 45 100 100 41 248 0.024 96 58 10 11 15 10 25 28 36 26 100 100 147 94 0.839 37 49 35 33 15 10 9 8 9 8 5 3 25 32 32 3 15 27 19 1 100 100 100 100 61 84 59 37 0.018 25 85 11 22 11 12 9 21 10 0 5 11 5 4 0 29 21 33 38 8 12 25 8 16 1 100 100 100 100 100 42 121 24 42 12 0.074 80 18 45 11 12 15 8 0 14 6 5 0 22 39 24 31 26 15 16 5 100 100 100 100 120 39 66 16 0.169 42 43 69 11 13 7 8 10 7 31 29 19 22 22 18 100 100 100 72 75 94 0.182 58 64 30 10 12 8 9 13 3 28 28 15 26 30 6 100 100 100 93 107 39 0.408 86 45 23 10 9 15 13 6 6 24 28 26 32 20 10 100 100 100 131 71 39 0.793 52 102 10 10 9 16 30 23 26 36 100 100 87 154 0.528 49 69 36 11 11 8 8 12 5 19 24 37 13 25 24 100 100 100 70 106 65 0.163 Depression and Anxiety DOI 10.1002/da Research Article: Health Service Use for Depression and Anxiety 33 TABLE 1. Continued No use Characteristic General level Total n % n % n % n P 12 138 10 10 3 21 48 21 14 43 100 100 29 202 0.027 19 134 9 11 5 20 56 17 30 32 100 100 54 186 0.000 9 141 21 9 6 19 46 23 13 47 100 100 28 207 0.001 % Alcohol disorder Yes 41 No 69 Dysthymia Yes 35 No 72 Parental psychiatric problems Yes 32 No 68 Specialist level GAD, generalized anxiety disorder. Pearson chi-square test. TABLE 2. Logistic regression model of use of health services for mental problems among respondents with major depressive disorder (MDD) during the past 12 months TABLE 3. Logistic regression model of use of health services for mental problems among respondents with anxiety disorder during the past 12 months Variable Variable OR Sex Female (reference) 1.00 Male 0.62 Age, years 0.98 (continuous) Living alone No (reference) 1.00 Yes 1.82 Severity of major depressive episode Mild (reference) 1.00 Moderate 1.44 Severe 2.44 Subjective disability Mild (reference) 1.00 Moderate 1.80 Severe 2.81 Comorbid anxiety disorder No (reference) 1.00 Yes 3.29 95% CI 0.35–1.12 0.96–1.01 1.00–3.29 0.76–2.74 1.10–5.44 0.98–3.28 1.37–5.77 1.81–5.96 OR Sex Female (reference) 1.00 Male 1.17 Age, years 0.97 (continuous) Living alone No (reference) 1.00 Yes 2.02 Subjective disability 1.00 Mild or moderate (reference) Severe 2.36 Comorbid major depressive disorder No (reference) 1.00 Yes 3.03 Parental psychiatric problems No (reference) 1.00 Yes 4.18 95% CI 0.62–2.20 0.94–1.00 1.05–3.92 1.26–4.42 1.57–5.85 1.59–11.00 OR, odds ratio; CI, confidence interval. OR, odds ratio; CI, confidence interval. DISCUSSION We found that among individuals suffering from MDD or an anxiety disorder in Finland, those with subjectively more disabling and comorbid disorders used health services more frequently. Use of services did not, however, differ according to sociodemographic factors. Of individuals suffering from one disorder, a considerable proportion (about 2/3) did not receive treatment, and even for the most severe comorbid cases only 60% of individuals received treatment. These poor treatment rates are disappointing in view of the majority of persons with anxiety and depression having contacted primary health services for other reasons. The continuing low rate of treatment-seeking for mental health problems is also dismaying given that effective treatments do exist and their availability has increased. In Finland the personnel resources of health care system have increased in recent years (e.g., the number of psychotherapists) and also the variety of different pharmacotherapies. In earlier studies, wide variation has emerged in the rates of treatment-seeking. For depression, the proportion of treated individuals has ranged from 17.0–77.8% [Bristow and Patten, 2002]. In the most recent European studies included in the multinational ESEMeD project [Alonso et al., 2004], use for mood Depression and Anxiety DOI 10.1002/da 34 Hämäläinen et al. disorder subjects was 36.5% and for anxiety disorder subjects 26.1%. In the Dutch NEMESIS study, by contrast, 63.8% of persons with mood disorders received some form of help, with the corresponding proportion of anxiety disorder subjects being 40.5% [Bijl and Ravelli, 2000]. Thus, for persons with an anxiety disorder the proportion seeking treatment may be lower overall, although considerable variation may be present among different disorders. In an earlier Finnish study [Laukkala et al., 2001; Hämäläinen et al., 2004] the proportion of subjects in the general population with a major depressive episode who used any health services for their depression during the past 12 months was slightly smaller, 28%. Due to lack of information about comorbidity and some methodological differences, these estimates are not fully comparable. Earlier reports have indicated that for panic disorder and generalized anxiety disorder the use of medical services is higher than for social and specific phobias [Bijl and Ravelli, 2000; Olfson et al., 2000]. In our study, medical service usage by our social phobia subjects was relatively high. This may be related to low prevalence estimates and possibly more severe social phobia in our subjects [Pirkola et al., 2005]. IMPACT OF SEVERITY AND COMORBIDITY Relationships between use of health services and characteristics of depressive episodes have been investigated in a few previous studies, which have revealed that more severe and disabling depression is more likely to be treated [Leaf et al., 1988; Bassett et al., 1998; Lefebvre et al., 1998; Ten Have et al., 2002]. Consistent with these studies, we found that the level of functional impairment had an independent effect on the use of health services. Also for anxiety disorder subjects, the more severe and disabling the disorder, the more frequent the use of the services. Comorbid psychiatric disorders may also affect the likelihood of treatment for other disorders. The influence may increase [Goodwin and Andersen, 2002], as in comorbid affective and anxiety disorders, or decrease the likelihood of treatment [Roy-Byrne et al., 2000; Goodwin and Andersen, 2002; Koenen et al., 2003]. In our data, comorbidity of MDD and anxiety disorder significantly increased the use of services from about 20% to 50%. Internationally, comparable results have been reported. In the Netherlands [Bijl and Ravelli, 2000], use of any form of services by subjects with one disorder in the past 12 months was 23.3%, and by subjects with two or more disorders 55.5%. In Canada [Kessler et al., 1997] the corresponding figures were 17.8% and 39.4%, and in the US [Kessler et al., 1999] 18.8% and 33.9%. Interestingly, the impact of an alcoholic disorder increased the likelihood of treatment if the person also had an anxiety disorder, but not MDD. Earlier, there have been similar findings of perceived need for help [Mojtabai et al., 2002]. Only a few studies [Bucholz and Robins, 1987; Dew et al., Depression and Anxiety DOI 10.1002/da 1991; Du Fort et al., 1999] have examined whether the presence of certain specific symptoms of depression is related to help-seeking; these findings are somewhat inconsistent. Du Fort et al. [1999] found that suicidal ideation and psychomotor retardation were associated with seeking treatment. Our finding of suicidal planning being significant, even after adjusting for the severity of MDD, suggests the importance of at least severe suicidal symptoms in the use of health services. SOCIODEMOGRAPHIC FACTORS The likelihood of depression and anxiety disorders being treated has been linked to a variety of factors, including being female, older, separated, divorced or widowed, unemployed, and better educated [Olfson and Klerman, 1992; Crow et al., 1994; Lin and Parikh, 1999]. In our subjects, help-seeking for mental problems was constant up to the age of 60 years, after which a distinct decrease was observed, consistent with earlier findings [Roness et al., 2005]. Probable reasons for the lack of help-seeking later in life include lack of identification of mental problems, a tendency to somatize problems, more negative attitudes toward psychiatry, and lack of specialized mental health resources in the geriatric health care system. In our final analysis of health care usage by MDD or anxiety disorder subjects, no other sociodemographic factor other than living alone remained statistically significant. At least three previous studies [Lin et al., 1996; Lefebvre et al., 1998; Bijl and Ravelli, 2000] report higher service usage among single householders and unmarried people. Other studies have found either no such association [Sherbourne, 1988; Golding et al., 1990; Phillips and Murrell, 1994] or an increasing use of mental health services but not primary health services [Ten Have et al., 2002]. It has also been suggested that both the amount of and the satisfaction with the social support received may have a protective effect against service use [Pescosolido et al., 1998; Ten Have et al., 2002]. The low impact of sociodemographic factors (income, education, etc.) compared with some previous studies [Kessler et al., 2001] may reflect the low financial threshold and relatively broad coverage of mental health services in Finland. To our knowledge, only a few earlier findings have been made about an adult subject’s childhood family environment on subsequent use of services for psychiatric symptoms. Higher rates of perceived need of treatment have been reported for participants with a maternal history of mental illness [Mojtabai et al., 2002]. For MDD, the probability of treatment-seeking in the proband was observed to be significantly increased only if the affected relative had themselves sought treatment for their depression [Kendler, 1995]. We also found a clear connection between one or both parents having had a psychiatric disorder and the use of services by anxiety disorder. For MDD, a weaker Research Article: Health Service Use for Depression and Anxiety nonsignificant association was present in the final model. Theoretically, psychiatric disorders in persons in the childhood environment may give subjects a model of how to get treatment for mental suffering (‘‘social learning’’ model [Bandura, 1986]). In further analyses, the episodes of depression in this subgroup of MDD persons proved to be more severe and started at an earlier age. No such differences were observed in anxiety disorder persons. SPECIALIST-LEVEL MENTAL HEALTH SERVICES Few studies have compared the characteristics of people using general or special mental health services for MDD or anxiety disorders, and the results have been mixed. These studies have reported 25–40% use of special mental health services [Kessler et al., 1994; Wang et al., 2002]. The specialist-level mental health system has been speculated to treat people with more severe and complex disorders. The findings, however, are contradictory. In our study the use of special mental health services was quite high (2/3 for both MDD and anxiety disorders), although comparable with the results of the ESEMeD project [Alonso et al., 2004]. Only psychiatric comorbidity, not disorder severity or subjective disability, increased the use of special mental health services markedly (for MDD from 1/7 to 1/2), a finding consistent with earlier reports. Generally, one might conclude that the mental health care sector may be overmeeting needs, i.e., dealing with too many mild cases that could better be treated in primary care [Bijl and Ravelli, 2000]. In fact, there is no clear uniform definition of ‘‘need for services’’ in the population. Particularly in cases with nonchronic psychiatric disturbance, people can sometimes successfully cope with their symptoms and functional disabilities with support from their nonprofessional social network. STUDY STRENGTHS AND LIMITATIONS We used CIDI diagnoses in a relatively large nationwide subject pool investigating several different depressive and anxiety disorders, which allowed us to estimate the effect of comorbidity on the use of services. The study, however, did have some limitations. As in many other epidemiological studies, we used self-reports of use of mental health services. For a number of reasons (including recall bias), self-reported estimates of use may be lower than the corresponding data from administrative records [Golding et al., 1988; Rhodes et al., 2002]. Moreover, we collected only general information on the ‘‘use of health services for mental problems.’’ We had no direct information about the duration of the major depressive episode in the last year, only about the duration of the most severe lifetime episode. Nevertheless, the length of different depressive episodes over the lifetime appears consistent [Solomon et al., 1997; Spijker et al., 2002]. Due to the small number of cases, we could not perform analyses 35 on different anxiety disorders to detect possible intergroup variability. Further, we had no specific information on the ‘‘attitude’’ factors likely to influence an individual’s decision to seek professional help for psychiatric problems (willingness to disclose problems, fear of stigma, negative stereotypes of treatments, and other cultural factors) [Christiana et al., 2000; Collins et al., 2004], or on the delay between symptom onset and first consultation with a professional (often 6–14 years across anxiety and mood disorders) [Kessler et al., 1998; Christiana et al., 2000]. We also did not have information on persons under 30 years. In young adults the prevalence of disorders and the use of services may differ from the older age groups studied. Finally, we did not have information on detection of mental disorders by general practitioners, although they are the gatekeepers to mental health services [Ormel et al., 1991]. The vast majority of individuals with anxiety and depression contacted primary care at least once in the past year [Ohayon et al., 2000]. Of our subject pool, 84% visited a doctor at least once in the last year, providing an opportunity to detect and diagnose MDD or anxiety disorder. REFERENCES Alonso J, Angermeyer MC, Bernert S, Bruffaert R, Brugha TS, Bryson H, de Girolamo G, Graaf R, Demyttenaere K, Gasquet I, Haro JM, Katz SJ, Kessler RC, Kovess V, Lepine JP, Ormel J, Polidori G, Russo LJ, Vilagut G, Almansa J, Arbabzadeh-Bouchez S, Autonell J, Bernal M, Buist-Bouwman MA, Codony M, Domingo-Salvany A, Ferrer M, Joo SS, Martinez-Alonso M, Matschinger H, Mazzi F, Morgan Z, Morosini P, Palacin C, Romera B, Taub N, Volleberg WA; ESEMeD/MHEDEA 2000 Investigators, European Study of the Epidemiology of Mental Disorders (ESEMeD) Project. 2004. Psychotropic drug utilization in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatr Scand Suppl 420:55–64. Aromaa A, Koskinen S (eds.). 2003. Health and functional capacity in Finland. Baseline results of the Health 2000. Health Examination Survey. Helsinki: Publications of the National Public Health Institute. Bandura A. 1986. Social foundation of thought and action: a social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall. Bassett SS, Chase GA, Folstein MF, Regier DA. 1998. Disability and psychiatric disorders in an urban community: measurement, prevalence and outcomes. Psychol Med 28:509–517. Bebbington PE, Brugha TS, Meltzer H, Jenkins R, Ceresa C, Farrell M, Lewis G. 2000. Neurotic disorders and the receipt of psychiatric treatment. Psychol Med 30:1369–1376. Bijl RV, Ravelli A. 2000. Psychiatric morbidity, service use, and need for care in the general population: results of the Netherlands Mental Health and Incidence Study. Am J Public Health 90: 602–607. Bristow K, Patten S. 2002. Treatment-seeking rates and associated mediating factors among individuals with depression. Can J Psychiatry 47:660–665. Bucholz KK, Robins LN. 1987. Who talks to a doctor about existing depressive illness? J Affect Disord 12:241–250. Christiana JM, Gilman SE, Guardino M, Mickelson K, Morselli PL, Olfson M, Kessler RC. 2000. Duration between onset and time of Depression and Anxiety DOI 10.1002/da 36 Hämäläinen et al. obtaining initial treatment among people with anxiety and mood disorders: an international survey of members of mental health patient advocate groups. Psychol Med 30:693–703. Collins KA, Westra HA, Dozois DJ, Burns DD. 2004. Gaps in accessing treatment for anxiety and depression: challenges for the delivery of care. Clinl Psychol Rev 24:583–616. Crow MR, Smith HL, McNamee AH, Piland NF. 1994. Considerations in predicting mental health care use: implications for managed care plans. J Mental Health Admin 21:5–23. Dew MA, Bromet EJ, Schulberg HC, Parkinson DK, Curtis EC. 1991. Factors affecting service utilization for depression in a white collar population. Soc Psychiatry Psychiatr Epidemiol 26:230–237. Du Fort GG, Newman SC, Boothroyd LJ, Bland RC. 1999. Treatment seeking for depression: role of depressive symptoms and comorbid psychiatric diagnoses. J Affect Disord 52:31–40. Golding JM, Gongla P, Brownell A. 1988. Feasibility of validating survey self-reports of mental health service use. Am J Community Psychol 16:39–51. Golding JM, Karno M, Rutter CM. 1990. Symptoms of major depression among Mexican-Americans and non-Hispanic whites. Am J Psychiatry 147:861–866. Goodwin R, Andersen RM. 2002. Use of the behavioral model of health care use to identify correlates of use of treatment for panic attacks in the community. Soc Psychiatry Psychiatr Epidemiol 37: 212–219. Hyler SE. 2002. APA online CME practice guideline for the treatment opatients with major depressive disorder. J Psychiatr Pract 8: 315–319. Hämäläinen J, Isometsä E, Laukkala T, Kaprio J, Poikolainen K, Heikkinen M, Lindeman S, Aro H. 2004. Use of health services for major depressive episode in Finland. J Affect Disord 79: 105–112. Kendler KS. 1995. Is seeking treatment for depression predicted by a history of depression in relatives? Implications for family studies of affective disorders. Psychol Med 25:807–814. Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, Wittchen HU, Kendler KS. 1994. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry 51:8–19. Kessler RC, Frank RG, Edlund M, Katz SJ, Lin E, Leaf P. 1997. Differences in the use of psychiatric outpatient services between the United States and Ontario. N Engl J Med 336:551–557. Kessler RC, Olfson M, Berglund PA. 1998. Patterns and predictors of treatment contact after first onset of psychiatric disorders. Am J Psychiatry 155:62–69. Kessler RC, Zhao S, Katz SJ, Kouzis AC, Frank RG, Edlund M, Leaf P. 1999. Past-year use of outpatient services for psychiatric problems in the National Comorbidity Survey. Am J Psychiatry 156:115–123. Kessler RC, Berglund PA, Bruce ML. 2001. The prevalence and correlates of untreated serious mental illness. Health Serv Res 36: 987–1007. Kessler RC, Demyttenaere K, Bruffaerts R, Posada-Villa J, Gasquet I, Kovess V, Lepine JP, Angermeyer MC, Bernert C, de Girolamo G, Morosini P, Polidori G, Kikkawa T, Kawakami N, Ono Y, Takeshima T, Uda H, Karam EG, Fayyad JA, Karam AN, Mneimneh ZN, Medina-Mora ME, Borges G, Lara C, de Graaf R, Ormel J, Gureje O, Shen Y, Huang Y, Zhang M, Alonso J, Haro JM, Vilagut G, Bromet EJ, Gluzman S, Webb C, Kessler RC, Merikangas KR, Anthony JC, Von Korff MR, Wang PS, Brugha TS, Aguilar-Gaxiola S, Lee S, Heeringa S, Pennell BE, Zaslavsky AM, Ustun TB, Chatterji S; WHO World Mental Health Survey Consortium 2004. Prevalence, severity and unmet need for Depression and Anxiety DOI 10.1002/da treatment of mental disorders in the World Health Organization World Mental Health Surveys. JAMA 291:2581–2590. Kessler RC, Demler O, Frank RG, Olfson M, Pincus HA, Walters EE, Wang P, Wells KB, Zaslavsky AM. 2005. Prevalence and treatment of mental disorders 1990 to 2003. N Engl J Med 352: 2515–2523. Koenen KC, Goodwin R, Struening E, Hellman F, Guardino M. 2003. Posttraumatic stress disorder and treatment seeking in a national screening sample. J Trauma Stress 16:5–16. Laukkala T, Isometsä E, Hämäläinen J, Heikkinen M, Lindeman S, Aro H. 2001. Antidepressant treatment of depression in the Finnish general population. Am J Psychiatry 158:2077–2079. Leaf PJ, Bruce ML, Tischler GL, Freeman DH, Weissman MM, Myers JK. 1988. Factors affecting the utilization of specialty and general medical mental health services. Med Care 26:9–26. Lefebvre J, Lesage A, Cyr M, Toupin J, Fournier L. 1998. Factors related to utilization of services for mental health reasons in Montreal, Canada. Soc Psychiatry Psychiatr Epidemiol 33:291–298. Lin E, Parikh SV. 1999. Sociodemographic, clinical, and attitudinal characteristics of the untreated depression in Ontario. J Affect Disord 53:153–162. Lin E, Goering P, Offord DR, Campbell D, Boyle MH. 1996. The use of mental health services in Ontario: epidemiological findings. Can J Psychiatry 41:572–577. Lopez AD, Murray C. 1998. The global burden of disease, 19902020. Nat Med 4:1241–1243. Mojtabai R, Olfson M, Mechanic D. 2002. Perceived need and helpseeking in adults with mood, anxiety or substance use disorders. Arch Gen Psychiatry 59:77–84. NHS, Depression. Management of depression in primary and secondary care. Clinical Guideline 23, December 2004 (www.nice. org.uk/CG022NICEguideline). [Accessed 15 December 2004.] NHS, Anxiety. Management of anxiety (panic disorder, with or without agoraphobia, and generalized anxiety disorder) in adults in primary, secondary and community care. Clinical Guideline 22, December 2004 (www.nice.org.uk/CG022NICEguideline). [Accessed 22 December 2004.] Ohayon MM, Shapiro CM, Kennedy SH. 2000. Differentiating DSM-IV anxiety and depressive disorders in the general population: comorbidity and treatment consequences. Can J Psychiatry 45: 166–172. Olfson M, Klerman GL. 1992. Depressive symptoms and mental health service utilization in a community sample. Soc Psychiatry Psychiatr Epidemiol 27:161–167. Olfson M, Guardino M, Struening E, Schneier FR, Hellman F, Klein DF. 2000. Barriers to the treatment of social anxiety. Am J Psychiatry 157:521–527. Ormel J, Koeter MW, Van Den Brink W, Van De Willige G. 1991. Recognition, management, and course of anxiety and depression in the general practice. Arch Gen Psychiatry 48:700–706. Pescosolido BA, Wright ER, Alegria M, Vera M. 1998. Social networks and patterns of use among the poor with mental health problems in Puerto Rico. Med Care 36:1057–1072. Phillips MA, Murrell SA. 1994. Impact of psychological and physical health, stressful events and social support on subsequent mental health help seeking among older adults. J Consult Clin Psychol 62: 270–275. Pirkola SP, Isometsä E, Suvisaari J, Joukamaa M, Poikolainen K, Koskinen S, Aromaa A, Lonnqvist JK. 2005. DSM-IV moodanxiety- and alcohol use disorders and their comorbidity in the Finnish general population. Soc Psychiatry Psychiatr Epidemiol 40:1–10. Regier DA, Narrow WE, Rae DS, Manderscheid RW, Locke BZ, Goodwin FK. 1993. The de facto US Mental and Addictive Research Article: Health Service Use for Depression and Anxiety Disorders Service System: Epidemiologic Catchment Area prospective 1-year prevalence rates of disorders and services. Arch Gen Psychiatry 50:85–94. Rhodes AE, Lin E, Mustard CA. 2002. Self-reported use of mental health services versus administrative records: should we care? Int J Met Psych Res 11:125–133. Roness A, Mykletun A, Dahl AA. 2005. Help-seeking behaviour in patients with anxiety disorders and depression. Acta Psychiatr Scand 111:51–58. Roy-Byrne P, Stang P, Wittchen H, Ustun B, Walters EE, Kessler RC. 2000. Lifetime panic-depression comorbidity in the National Comorbidity Survey. Br J Psychiatry 176:229–235. Sherbourne CD. 1988. The role of social support and life stress events in use of mental health services. Soc Sci Med 27:1393–1400. Solomon DA, Keller MB, Leon AC Mueller TI, Shea MT, Warshaw M, Maser JD, Coryell W, Endicott J. 1997. Recovery from major depression. A 10-year prospective follow-up across multiple episodes. Arch Gen Psychiatry 54:989–991. Spijker J, De Graaf R, Bijl RV, Beekman AT, Ormel J, Nolen WA. 2002. Duration of major depressive episodes in the general 37 population: results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Br J Psychiatry 181: 208–213. Ten Have M, Wollebergh W, Bijl R, Ormel J. 2002. Combined effect of mental disorders and low social support on care service use for mental health problems in the Dutch general population. Psychol Med 32:311–323. Wang PS, Demier O, Kessler RC. 2002. Adequacy of treatment for serious mental illness in the United States. Am J Public Health 92:92–98. Wang PS, Lane M, Olfson M, Pincus HA, Wells KB, Kessler RC. 2005. Twelve-month use of mental health services in the United States. Arch Gen Psychiatry 62:629–640. Wittchen HU, Lachner G, Wunderlich U. 1998. Test-retest reliability of the computerized DSM-IV version of the MunichComposite International Diagnostic Interview (M-CIDI). Soc Psychiatry Psychiatr Epidemiol 33:568–578. Young AS, Klap R, Sherbourne CD, Wells KB. 2001. The quality of care for depressive and anxiety disorders in the United States. Arch Gen Psychiatry 58:55–61. Depression and Anxiety DOI 10.1002/da View publication stats