Main topics in transcultural psychiatric research in the Netherlands during the past decade

Introduction

In 2011, one out of seven inhabitants of the Netherlands was of non-Western descent, that is, had at least one parent born in a country outside the European Union—a number that increased from 1.4 million (10.8% of the total population of the Netherlands) in 2000, to 1.9 million (14.4%) in 2011. This number does not include asylum seekers and undocumented residents. These figures include first- as well as second-generation immigrants, in accordance with the definition used in demographic statistics in the Netherlands. The three largest ethnic minority groups are from Turkey, Morocco, and Surinam, each comprising more than 300,000 people, followed by immigrants from the former Netherlands Antilles, with about 140,000 people. Refugees and asylum seekers together currently account for about 1.5% (approximately 250,000) of the national population. The largest refugee groups, from Iran, Iraq, and Somalia, each comprise between 30,000 and 50,000 people. The yearly inflow of asylum seekers has decreased from 43,000 in 2000, to 11,600 in 2011, partly because of more restrictive government policy. As these statistics demonstrate, over the past 15 years, the population of the Netherlands has become increasingly diverse in terms of country of origin, ethnicity, culture, and religion.

The international image of the Netherlands has long been one of a peaceful, efficient, and well-off country: socially tolerant, easy-going, and supportive toward immigration and multiculturalism. However, during the past 10 to 15 years, anti-immigrant and anti-Muslim attitudes have become more apparent, and have drawn broader public support among the majority ethnic Dutch population of the Netherlands. The terrorist attacks in New York and Washington on September 11, 2001, and the murder in Amsterdam of Theo van Gogh, a Dutch film producer, by a Muslim fundamentalist in November 2004, marked the beginning of a new political era in the Netherlands, characterized by an increasing anti-immigrant and anti-Muslim atmosphere in the public domain. A populist political movement espousing a nationalistic and anti-immigrant agenda won 15.5% of the votes in the 2010 national elections. And because the government depended on its support, this movement has achieved a strong influence on national policy decisions. During the same period, the overall interest of mental health professionals and researchers in issues linked to transcultural psychiatry has grown steadily, and the effects of social defeat—including (perceived) discrimination and social exclusion—on the mental health of the different ethnic minority groups living in the Netherlands has become a research topic.

The Netherlands has a long tradition in transcultural psychiatry, mainly focussing on ethnic disparities in epidemiology and treatment, which dates back to the 1980s and is rooted in the work of transcultural psychiatrists (Rob Giel, Frank Kortmann, Joop de Jong, Annechien Limburg-Okken), cross-cultural psychologists (David Ingleby, Fons van der Vijver, Ype Poortinga), and medical anthropologists (Annemiek Richters, Els van Dongen). Their contribution to the development of culturally competent mental health services in the Netherlands can hardly be overestimated. Since the mid-1990s, transcultural expertise found its way to textbooks in transcultural psychiatry and cross-cultural psychology that are widely used in teaching and practice in the Netherlands (Borra, van Dijk, & Rohlof, 2002; de Jong & Colijn, 2010; de Jong & van den Berg, 1996; Kortmann, 2006).

In this paper, we briefly review the main topics of transcultural psychiatry research in the Netherlands during the past decade and discuss the relationship between the research findings and relevant social and political developments in the Netherlands. We include scientific research by psychiatrists, psychologists, and anthropologists on aspects of epidemiology and treatment of mental health problems among migrants, refugees, and asylum seekers in the Netherlands in Dutch or English, published between 2000 and 2011. We focus on general psychiatry, excluding forensic and addiction psychiatry and studies done by Dutch researchers on international mental health topics. Moreover, clinical case studies and papers reflecting on culturally competent mental health care were excluded. As most of the transcultural psychiatric research in the Netherlands is done as part of doctoral research projects, we selected all relevant PhD theses (N = 27) defended in this period and the related publications in scientific journals. Screening of the major Dutch scientific journals in the field of transcultural psychiatry and medical anthropology resulted in additional publications, mostly in Dutch. Additional searches on PubMed (with keywords including: Netherlands, AND mental health, psychiatry, immigrants, refugee, asylum seeker, ethnic, cultural) yielded 86 additional publications fitting the criteria.

By far the greatest number of studies were epidemiological. Scientific research into the quality and effects of treatment was scarce, as previously noted by Bekker and van Mens-Verhulst (2008). Three major topics were identified through the review: (a) the prevalence of psychiatric disorders and their relation to migration issues, especially anxiety and depression, and schizophrenia and other psychotic disorders; (b) the social position of refugees and asylum seekers, and its effects on mental health; and (c) patterns of health-seeking behaviour and use of mental health services. Among the less common topics that will not be discussed in this paper are suicide and suicidal behaviour in migrants, ethnic disparities in involuntary admission to clinical facilities, the use of interpreters in mental health care, and postdisaster mental health problems in migrants. In addition, we address anthropological perspectives in transcultural psychiatric research. We conclude with some remarks about future directions of transcultural psychiatry research in the Netherlands.

Common mental disorders among immigrants: Ethnicity and socioeconomic status

A major problem in getting a clear understanding of mental disorders found in ethnic minority groups is the fact that these populations have not been routinely included in national population-based epidemiological mental health studies in the Netherlands. The two major national epidemiological studies (Netherlands Mental Health Survey and Incidence Study [NEMESIS I, 1996; Bijl, van Zessen, & Ravelli, 1998] and [II, 2009; De Graaf, ten Have, van Gool, & van Dorsselaer, 2012]) explicitly excluded non-Dutch speaking respondents: immigrants were only included on the condition that they were sufficiently fluent in Dutch. Nevertheless, in the past 10 years transcultural psychiatry research has increased knowledge about the prevalence of common mental disorders, in particular anxiety and affective disorders, in ethnic minority groups. Most studies have been carried out in the main cities of Amsterdam, The Hague, and Rotterdam.

In 2002, Bengi-Arslan, Verhulst, and Crijnen assessed the prevalence for minor psychiatric disorders in randomly selected samples of 262 male and 523 female Turkish immigrants. Using the General Health Questionnaire (GHQ-28) they found higher prevalence rates in females (36.1%) compared to males (27.9%). These rates are higher than the prevalence normally found in general population samples (25%). Educational level (used as a proxy for socioeconomic status [SES]) did not contribute significantly to any of the GHQ scales and total scales.

A study by van der Wurff et al. (2004) comprising 933 subjects, Turkish (n = 330), Moroccan (n = 299), and ethnic Dutch (n = 304) elderly (55–75 years of age) found a significantly higher prevalence of (current) depressive symptoms among elderly of Moroccan and Turkish descent, compared to the ethnic Dutch respondents; Moroccans 33.6%, Turkish 61.5%, and ethnic Dutch 14.5%. In all three samples, depressive symptoms (measured with the Center for Epidemiologic Depression Scale [CES-D]) were associated with gender, chronic physical illnesses, and limitations in activities of daily living (ADL). The association with female sex was significantly stronger in the Dutch sample compared to the Moroccan and Turkish samples (odds ratios = 3.14, 1.84, and 1.77, respectively). Turkish and Moroccan immigrants suffered significantly more often from limitations in ADL and the number of chronic illnesses was higher in the Turkish elderly, explaining part of the increased prevalence rates. In multivariate analysis, ethnic origin was uniquely associated with the presence of clinically significant depressive symptoms. Low socioeconomic status (SES) did not reach a statistically significant association in the analyses. The authors developed an instrument to measure acculturation and attitudes towards return migration and found that only a few items had a significant association with depressive symptoms. They suggest that less acculturated Turkish and Moroccan immigrants with high rates of exposure to the risk factors associated with both poverty and depression may have been overrepresented in the study.

In a study of Surinamese (n = 311), Turkish (n = 648), and Moroccan (n = 102) immigrants in Amsterdam, Kamperman, Komproe, and de Jong (2005) demonstrated the importance of using more than one instrument to measure the extent and the burden of mental illness. Psychiatric morbidity was measured with the Composite International Diagnostic Interview (CIDI) and subjective complaints and disability with the Medical Outcome Study, short form, 36 items (MOS-SF-36). They found differences in levels of psychiatric disorders and levels of subjective complaints and disability across the groups: for example, the Surinamese group had the least subjective symptoms and disabilities and the highest number of psychiatric disorders. In the Turkish group the lifetime prevalence of an anxiety disorder was 6.9% for men and 9.1% for women and of a depressive disorder 6.6% and 9.5%.

In this study, the Surinamese and Moroccan samples were not representative of the Surinamese and Moroccan Amsterdam population with regard to gender (Surinamese), mean age, and martial status (Moroccan), so another study (De Wit et al., 2008) was done with the same ethnic minority groups (Turkish [n = 231], Moroccan [n = 191], Surinam/Antilleans [n = 87]). An ethnic Dutch sample (n = 321) was included, so that more accurate interethnic comparisons could be made and compared with the majority ethnic Dutch population. The CIDI was used to assess lifetime, 12-month, and current prevalence rates. The lifetime prevalence of both disorders was significantly higher in the Turkish population (35.3%) than in the ethnic Dutch population (28.8%). Turkish women had the highest risk for a current depressive or anxiety disorder (odds ratio = 5.3; 95% CI [2.1, 13.1]). The risk of a current disorder also was not significantly increased for Turkish men and Moroccan men compared to ethnic Dutch men and there was no increased risk for men and women of Surinamese or Dutch-Antillean descent. Although socioeconomic factors were related to the prevalence of depressive and anxiety disorders, they could not explain the ethnic differences. An interesting finding was ethnic differences in the ratios between lifetime, 1-year, and current prevalence. Among the ethnic Dutch, about 40% of those with a lifetime disorder also had a disorder in the last year, and 40% with a disorder in the last year also had a disorder in the last month. Among the Turkish and Moroccan groups, these proportions were much higher (between 64 and 72%). These findings may be a result of differences in recall influenced by cultural factors or the attribution of earlier episodes to external stress. They could also indicate that mental health problems among immigrants have a more chronic course.

An epidemiological study in general practice in Dutch urban areas showed that the 1-year prevalence rate of common mental disorders was significantly higher among Turkish immigrants (5.22%) than among ethnic Dutch (4.43%), Moroccans (4.25%), and Surinamese (3.24%; Fassaert et al., 2010). There were significant ethnic differences in the prevalence of depression, but not in the prevalence of anxiety. Another study by the same research group in the Amsterdam catchment area (Fassaert et al., 2008), using CIDI and Kessler Psychological Distress Scale (K10), showed that the perceived need for mental health care among Turkish immigrants was higher than among Moroccan immigrant and ethnic Dutch respondents.

Kamperman, Komproe, and de Jong (2007) and Fassaert et al. (2011) tried to assess the effect of different patterns of acculturation on mental health with the help of the Lowlands Acculturation Scale (Mooren, Knipscheer, Kamperman, Kleber, & Komproe, 2001). The effect of cultural traditionalism, defined as higher maintenance of traditional culture and identity, proved to be complex in both studies. According to Kamperman and colleagues, more culturally traditionalist respondents expressed less psychiatric morbidity, but also reported lower levels of mental well-being. The authors offer two explanations: first, culturally traditionalist immigrants may “translate” psychosocial and mental problems as poor mental well-being, but not as psychiatric symptoms. Another explanation is that cultural traditionalism buffers against stress, thereby preventing severe psychiatric problems from becoming apparent. Moreover, Kamperman and colleagues found that immigrants who experienced more feelings of cultural loss reported more psychiatric morbidity. On the other hand, respondents with less practical skills for adapting to life in Dutch society (e.g., limited fluency in Dutch language and limited familiarity with Dutch society and [mental health care] institutions) reported lower levels of mental well-being.

Fassaert et al. (2011) concluded that lack of skills for adapting to life in Dutch society, including limited language fluency, was associated with psychological distress among Turkish and Moroccan immigrants. On the other hand, among Moroccans and female Turkish immigrants, strong cultural traditionalism was related to less distress. In the case of male Turks, more traditional values and norms were associated with more psychological distress. The conclusion of both the Fassaert et al. (2011) and the Kamperman et al. (2007) studies is that successful contact and participation in Dutch society (including mastery of the Dutch language) and maintenance of traditional culture and identity are important for the mental health status and sense of well-being of immigrants. Since the present Dutch government favours assimilation—or, alternatively, suggests that immigrants “go back to your own country”—these conclusions are very relevant.

While in the studies by Bengi-Arslan et al. (2002), De Wit et al. (2008), and van der Wurff et al. (2004) socioeconomic differences could not explain the difference in prevalence rates, Kamperman et al. (2005) stated that the differences in mental health cannot be explained by culture or ethnicity alone. Kamperman stresses the importance of the classical determinants of mental health: sociodemographic characteristics, coping strategy, and somatic health. Fassaert et al. (2011) found that including variables like socioeconomic status (SES), including highest education level achieved, did not change the results.

Several studies have been undertaken with immigrant youth, mainly focusing on prevalence rates of psychopathology and behaviour problems. The study with the largest number of subjects was part of the so-called Tracking Adolescents’ Individual Lives Survey (TRAILS). A sample of 2,230 11-year-old children in the Netherlands participated in TRAILS. Approximately 10% of these children (n = 230) belonged to immigrant families of non-Western background (Vollebergh et al., 2005). Mental health problems were measured by the Youth Self-Report (YSR) and the Child Behaviour Check List (CBCL) completed by the parents. In addition to yielding a total score of mental health problems, the instruments measure internalizing problems (withdrawn behaviour, somatic complaints, and anxious/depressed syndromes) and externalizing problems (delinquent behaviour and aggressive behaviour syndromes). “Social problems”, “thinking problems” (predominantly containing items on hallucinations, delusions, and obsessive compulsive problems), and “attention problems” are not part of either the internalizing or externalizing categories (Achenbach, 1991a, 1991b). A new instrument was developed to allow teachers to report on problem behaviour in a less laborious fashion, the Teacher’s Checklist of Psychopathology (TCP). Vollebergh et al. (2005) found that children (11 years old) from immigrant families did not appear to experience more mental health problems than their ethnic Dutch peers. Zwirs et al. (2007), in studies among children from 6 to 10, reached the same conclusion. However, parents from immigrant families did report more problems—especially more internalizing problems, social problems, and attention problems in their daughters—than did ethnic Dutch parents. By contrast, teachers reported lower levels of internalizing, social, and thinking problems in particular in boys, and higher levels of externalizing problems in both immigrant boys and girls. The authors stress the possibility that teachers overreported externalizing problems, and argue that perhaps they were influenced by media attention to criminal behaviour in older immigrant youth. Further research is needed to clarify the significance of these findings. Overall, children from lower SES families appeared to be at higher risk for all types of problems. The effect of SES was particularly strong for externalizing problems in boys.

Van Oort et al. (2007) reported the result of a longitudinal study, comparing the course of mental health problems among Turkish and Dutch adolescents over a period of 5 years. The results showed that both adolescent (aged 11 to 18) and adult migrants (aged 21 to 28) reported more internalizing and externalizing problems than Dutch, most pronounced for internalizing problems. Disparities decreased from adolescence into adulthood for both problems, independently of gender, age, and also of parental socioeconomic position. The changes in disparities were due to a more favourable development among Turkish migrants than among Dutch.

The studies reviewed above show that there is still discussion about the weight of ethnicity and culture on the one hand, and socioeconomic status on the other hand, in explaining disparities in psychopathology within and between ethnic groups.

Explaining the elevated prevalence of psychotic disorders among immigrants

An important series of studies started with the publication by Selten and Sijben (1994) on first psychiatric admission rates for schizophrenia among immigrants in the Netherlands. They found evidence of an increased incidence in Surinamese and Dutch-Antillean immigrants and in male Moroccan immigrants, but not in Turkish immigrants. Additionally, the incidence of schizophrenia in second-generation immigrants was shown to be higher than in first-generation immigrants. Several other studies followed, corroborating these clinical findings, in both inpatient and outpatient settings (e.g., Schrier, van de Wetering, Mulder, & Selten, 2001). However, a follow-up study by Veen et al. (2004) showed that after a 30-month follow-up period, the relative risk of schizophrenic disorder in Turkish-born males was also significantly increased. According to Veen et al., in the case of Turkish respondents the diagnosis of schizophrenia often was missed in the first diagnostic assessment; often a “psychotic disorder not otherwise specified” diagnosis was made at baseline. Veen offers as explanation that, as a result of increasing confidence in their clinicians, patients might have offered additional information important in the diagnostic process. Language limitations might have been one of the problems.

Adriaanse, van Domburgh, Veling, and Doreleijers (2011) screened 1,563 nontreated children aged 9 to 16 using, among other instruments, the Strength & Difficulties Questionnaire (SDQ) and the Kiddie Schedule for Affective Disorder and Schizophrenia-Present and Lifetime (K-SADS-P). They compared children of ethnic Dutch (N = 693) and Moroccan descent (N = 408). There was in general no difference in prevalence of psychiatric disorders in both groups, in line with the findings of Zwirs et al. (2007). But in this study Moroccan children reported psychotic-like experiences twice as often as did ethnic Dutch children, which could mean that they are at greater risk to develop a psychotic disorder at a later age.

A meta-analysis of population-based incidence studies concerning immigrants by Cantor-Graae and Selten (2005) led to the conclusion that a personal or family history of migration is a significant risk factor for schizophrenia. The relative risk for schizophrenia of those with first- or second-generation migrant status was 2.9 (95% CI [2.5, 3.4]). In the discussion, the authors note that this risk is lower than the risk of having a family member with the disease, but higher than the risk of urbanization. The authors challenged their own findings, discussing, among other limitations, issues of diagnostic bias and selective migration. They argued for rejecting both of these possible explanations, and suggested that discrimination and social defeat—defined as the chronic stressful experience of “a subordinate position” or “outsider status” (Selten & Cantor-Graae, 2007)—might be important explanatory factors. In a recent study, Selten, Laan, Kupka, Smeets, and van Os (2011) reached similar conclusions. The risk for receiving treatment for a nonaffective psychotic disorder proved to be very high for the Turkish-Dutch (RR = 8.7, 95% CI [5.5, 13.9]), Moroccan-Dutch (RR = 7.2, 95% CI [4.5, 11.8]), and Surinamese-Dutch (RR = 6.5, 95% CI [4.3, 9.9]), of the second generation. No heightened risks were found for migrants from the countries of the European Union. According to these researchers, the stressful position of non-Western migrants in Dutch society has negative consequences for their mental health.

Veling et al. (2007) studied the incidence of psychosis among immigrants in The Hague over a 7-year period and rated the degree of perceived discrimination. They found that if immigrants felt that they were exposed to a high level of discrimination, the incidence rates of psychosis were higher: the age- and gender-adjusted incidence rate ratios of schizophrenic disorders for ethnic minority groups exposed to high, medium, low, and very low discrimination were 4.00 (95% CI [3.00, 5.35]), 1.99 (95% CI [1.58, 2.51]), 1.58 (95% CI [1.10, 2.27]), and 1.20 (95% CI [0.81, 1.90]), respectively. Furthermore, when not only schizophrenia but all psychotic disorders were included in the analysis, the results were similar (Veling, Hoek, & Mackenback, 2008). In another analysis using the same data, Veling, Susser, et al. (2008) showed that the incidence of psychotic disorders was elevated most significantly among immigrants living in neighbourhoods where their own ethnic group comprised only a small proportion of the population. The authors considered ethnic density to be a buffering factor that can diminish the risk of immigrants developing severe psychiatric symptoms and disorders. The study emphasized the importance of including analysis of the broader social context, in addition to the factors that act on the level of the individual.

The issue of diagnostic bias as an explanation for the higher prevalence rates of schizophrenia in migrants was examined by another Dutch research group. Zandi et al. (2010) developed a culturally sensitive version of the Comprehensive Assessment and Symptoms History (CASH), the instrument that was used in the previously mentioned studies by Selten and Sijben (1994) and Veen et al. (2004), and compared the two versions. They found that first-contact incidence of schizophrenia among Moroccan immigrants was not significantly higher than among ethnic Dutch if a culturally sensitive diagnostic instrument was used during the diagnostic interview. The prevalence of schizophrenia appeared to be lower when using the CASH-CS than when using the nonculturally adapted CASH. These findings generated lively discussion in the journal Schizophrenia Research (Selten et al., 2010). The debate will probably continue, as Zandi and colleagues (2011) have now published their findings from a 30-month follow-up study, which showed that the diagnostic stability using the standard interview instrument was high for ethnic Dutch (92%), but low for Moroccan subjects (27%), whereas diagnostic stability using the culturally sensitive version was high for both groups (85% and 81%, respectively). The authors point out that these results lead to serious doubts about the validity of previous studies showing an increased incidence of schizophrenia in immigrants using standard diagnostic procedures. The issue of the diagnostic bias was also addressed by Mulder, Koopmans, and Selten (2006) in a study of 720 patients (aged 18–65) referred to emergency psychiatric services in Rotterdam. The authors found that non-Western migrants, Turkish migrants excepted, were more frequently admitted compulsorily than were migrants from Western countries and native Dutch (RR = 1.4–3.6). The authors conclude that non-Western immigrants may have “a different clinical presentation causing more frequent compulsory admissions” (2006, p. 391).

In summary, over the last decade the prevalence of psychotic disorders in migrants became a topic of major research interest in the international literature of cultural psychiatry, psychiatric epidemiology, clinical psychiatry, and public health policy. Dutch researchers have been active participants in the research and debate around this issue. Research has gradually revealed the complex relation between psychotic disorders and ethnicity/migration, and has also demonstrated the explanatory potential of contextual variables including social defeat, social cohesion, and ethnic density.

Asylum seekers and refugees: Social exclusion as a risk factor for mental illness

Asylum seekers and refugees throughout the world are considered a high-risk group for mental health problems. Adverse life events, forced migration, and the prolonged asylum procedures, in addition to the complexity of the acculturation process, can all contribute to higher levels of psychopathology. While the number of asylum seekers applying for residence permits has decreased steadily over the last 10 years, there is still a considerable number of asylum seekers in the Netherlands. Furthermore, the number of asylum seekers who have been denied residence permits but are still living in the Netherlands was estimated to be 100,000 in 2009 (van der Heijden, Cruyff, & van Gils, 2011). A series of studies of asylum seekers and refugees in the Netherlands started in the early 1990s, with a mixed patient/population study, using a nonrandomized sample (Hondius, van Willigen, Kleijn, & van der Ploeg, 2000). The study showed a high level of anxiety and depressive complaints (measured with the Hopkins Symptom checklist [HSCL-25]), but surprisingly low rates of posttraumatic stress disorder (PTSD; clinical diagnosis, based on DSM-III-R criteria).

Gerritsen et al. (2006a) compared randomly selected samples of asylum seekers (n = 232) and refugees (n = 178) of Iranian, Somali, and Afghan origin and assessed prevalence rates of PTSD and depression/anxiety problems. Using the HSCL-25 and the Harvard Trauma Questionnaire (HTQ) they found that, compared with refugees, asylum seekers more often had symptoms of PTSD (10.6% and 28.1%, respectively) and depression/anxiety (39.4% and 68.1%, respectively). Respondents from Afghanistan, and particularly those from Iran, had a higher risk for PTSD and depression/anxiety compared with respondents from Somalia. Overall, asylum seekers reported much higher levels of postmigration stress, as well as higher levels of physical complaints, compared to refugees who had been given residence permits.

Laban, Gernaat, Komproe, Schreuders, and de Jong (2004) conducted a population-based study among a randomly selected sample of Iraqi asylum seekers (n = 294). The sample was prestratified into two groups, according to how long they had been living in the Netherlands—less than 6 months and more than 2 years—because of the special focus of the study on the relationship between psychopathology (using the CIDI) and asylum seekers’ length of stay in the Netherlands. Overall prevalence of psychiatric disorders was 42% in the first group and 66.2% in the second. The prevalence rates of anxiety, depressive, and somatoform disorders were significantly higher in the group that had been living in the Netherlands for more than 2 years (14.0% vs. 30.5%; 25.2% vs. 43.7%; 4.9% vs. 13.2%, respectively). PTSD was high in both groups, but did not differ significantly by length of time living in the Netherlands (31.5% vs. 41.7%). On logistic regression of all relevant risk factors, a long asylum procedure showed an odds ratio of 2.16, indicating that the duration of the asylum procedure is an important risk factor for experiencing psychiatric problems. Further analysis (Laban, Gernaat, Komproe, van Tweel, & de Jong, 2005) on related risk factors in the postmigration period revealed that worries about lack of work, family issues, and asylum procedure stress were the most important risk factors in their relationship to psychopathology. The study also showed higher disability, lower quality of life, and higher levels of physical complaints in the group that had been living in the Netherlands for more than 2 years.

Bean, Derluyn, Eurelings-Bontekoe, Broekaert, and Spinhoven (2007) studied the mental health of unaccompanied refugee minors (URMs). The aim of the study was to assess the prevalence and nature of their psychosocial distress (measured with the HSCL-37A and the CBCL), as well as the mental health care needs of this population. Participants were 3,273 adolescents from the Netherlands and Belgium, including 920 URMs. The URM group consistently reported significantly higher scores for internalizing problems, traumatic stress reactions, and stressful life events than all other groups. Reijneveld, de Boer, Bean, and Korfker (2005) compared the mental health of adolescents in a restricted campus reception setting (n = 69) and in a setting offering more autonomy (n = 53). Respondents in a restricted reception setting reported significantly more emotional problems (M = 59.3) than their counterparts in the more autonomous group (M = 53.4). Main effects concerned increased anxiety, and girls showed larger differences than boys. In a 1-year follow-up study of the main study group (Bean, Eurelings-Bontekoe, & Spinhoven, 2007), the self-reported psychological distress of refugee minors was found to be severe (50%) and of a chronic nature (stable over 1 year), which was confirmed by reports of the subjects’ guardians and teachers. The core finding of these studies is that the asylum procedure—especially when it is prolonged—can be inherently damaging to mental health.

A different type of contribution to understanding the psychological problems of asylum seekers is found in a medical anthropological study. Tankink (2009) interviewed refugee women of Bosnian, Afghan, and Sudanese background about their response to sexual violence. The study found that all the women considered remaining silent about their traumatic experiences as the best coping strategy, because they thought that if their secret became public they would risk being rejected by their family and community. These findings are important in relation to screening interviews of asylum seekers by staff of the Immigration Department. Women who have been victims of sexual violence might be denied a residence permit, because they do not acknowledge their experience of sexual violence. Feldmann, Bensing, de Ruijter, and Boeije (2006, 2007) undertook two small-scale qualitative studies on the contacts between Afghan and Somali refugees with medically unexplained physical symptoms and their general practitioners. They found that general practitioners generally wrongly concluded that their refugee patients did not associate their somatic complaints with their stressful life circumstances. These studies, using a more qualitative methodology and an anthropological perspective, offer important insights into the help-seeking strategies of refugees and the dynamics of refugee–therapist interaction. This approach complements in a significant way the more quantitative surveys reviewed above.

Health seeking behaviour and health service use: A differentiated picture

There is a general concern that immigrants underutilize available mental health services. Unfamiliarity with community resources, including health and counselling services, in addition to cultural barriers, language problems, limited financial means, and inadequate referral and intake systems, all contribute to immigrants’ underutilization of mental health services. During the past decade several studies were undertaken in the Netherlands to determine factors related to utilization of mental health services by several different immigrant groups. Stronks, Ravelli, and Reijneveld (2001) showed that the utilization of more specialized health services (e.g., outpatient mental health centres) is lower for immigrant groups in the Netherlands, particularly for adult Turkish and Moroccan immigrants. Although underutilization of more specialized services is also seen among the lower socioeconomic groups in general in the Netherlands, the analysis indicates that this only partly explains the lower utilization of these services among immigrant groups.

Other studies present a more detailed picture. Dieperink, van Dijk, and De Vries (2007) showed considerable inter- and intraethnic differences in mental health service utilization by adult migrants. The use of mental health services in Rotterdam between 1990 and 2004 varied by patients’ country of origin, age, and gender. The overall use of these services increased, but the relative differences between and within ethnic groups remained. Unexpectedly, the utilization of mental health services by Moroccan men was comparable to that of ethnic Dutch native men. However, among Moroccan women, utilization of mental health services was considerably less than among ethnic Dutch women. In 2004, 57 per 1,000 male ethnic Dutch residents between the ages of 20 and 44 used the regional mental health care services. For Moroccan, Turkish, Surinamese, Antillean, Cape Verdean, and other migrant men in the same age category, the numbers were, respectively, 67, 51, 46, 39, 19, and 32 per 1,000; for migrant women in the same age category the numbers were 61, 94, 53, 38, 42, and 39, compared to 89 for ethnic Dutch women.

A study conducted in Amsterdam a few years later found that health care service utilization by immigrants was less problematic than expected (Fassaert et al., 2009). In a population-based study, first-generation Turkish and Moroccan immigrants, as well as ethnic Dutch subjects, were interviewed concerning the presence of common mental disorders (i.e., depression and anxiety disorders) and their use of health services during the 6 months preceding the interview. Of the respondents with a psychiatric disorder, 50.9% reported receiving care for those problems during that period; 35.0% contacted specialized psychiatric services. However, immigrants were equally likely to access specialized mental health services if a psychiatric disorder was diagnosed by their general practitioner. When the findings were corrected for age and sex differences, and for variables like SES and education, the results were unchanged. Fassaert and colleagues point to differences between subjective and objective need for care. The objective need is operationalized as the presence of a psychiatric disorder as diagnosed by a medical professional. The subjective need of Moroccan and Turkish immigrants is higher compared to ethnic Dutch subjects, but this does not necessarily lead to a higher level of utilization of the services of general practitioners. Once a psychiatric disorder is diagnosed, there are no differences in use of mental health services between Moroccan, Turkish, and ethnic Dutch patients. Seen in terms of subjective need, there is an underutilization of baseline mental health care offered by the general practitioner. These findings call for further qualitative (anthropological) research to discover underlying patterns of health-seeking behaviour. The findings of Fassaert et al.’s study are consistent with the results of a study by Knipscheer and Kleber (2005), which found that, although Moroccan immigrants of lower educational level and more recently arrived immigrants reported a more negative attitude toward consulting mental health clinicians, there were more interethnic similarities than differences among immigrant groups in their use of mental health services. Both studies were conducted in Amsterdam and more research is needed to determine if these findings can be replicated in other regions of the Netherlands and among other ethnic groups.

Little is known about the use of mental health services by migrant children and youth. However, the first results of studies based on psychiatric case register are alarming. Boon, De Haan, and De Boer (2010) found that migrant children between the ages of 0 and 19 were underrepresented in the mental health services in The Hague. According to these researchers, this underutilization was primarily associated with ethnicity, not with socioeconomic status. Ethnic density of neighbourhoods proved to be a mediating factor. Wierdsma and Kamperman (2011) came to the same conclusions for the catchment area of Rotterdam. The average yearly number of new patients in care (aged 0 to 18) in 2008–2011 was 35.8 for ethnic Dutch children, compared to 4.3 (Moroccan) and 10.6 (Antillean) for migrant children.

Health service use by asylum seekers was studied in two research projects. In the Netherlands, most asylum seekers live in reception centres and do have access to health care facilities on an equal basis with the Dutch population. The medical staff present at these centres have the responsibility to refer to more specialized care facilities, as needed. Laban, Gernaat, Komproe, and de Jong (2007) found that the use of mental health services by Iraqi asylum seekers was low compared to the prevalence of psychiatric disorders. Only 8.8% of subjects with a psychiatric disorder were seen by a mental health clinician in the previous 2 months. There was a high level of medication use among these subjects, especially pain medication. Low quality of perceived general health and functional disability were the most important predictors of services use. Another finding was that those with mental health problems were more likely to be referred to a nonpsychiatrist medical specialist than to a mental health clinician. Accordingly, there was a mismatch in the referral procedure.

In the previously mentioned study by Gerritsen et al. (2006b), it was found that there were no differences between refugees and asylum seekers in the use of health care services, although asylum seekers reported higher levels of mental health problems. Somali respondents reported fewer contacts with a general practitioner, less use of mental health services, and less medication use than respondents from Afghanistan and Iran. The authors suggest that the lower level of mental health problems among Somali respondents explains this finding. Both female gender and older age were related to more contacts with a general practitioner and a medical specialist, as well as greater medication use. In both studies the use of health services was lower compared to the general Dutch population, while the levels of health problems were much higher.

Undocumented immigrants are a special case. In the Netherlands they have no right to health insurance coverage, but are entitled to receive medically necessary care, defined as responsible and appropriate medical care as indicated by the treating doctor. About 8,000 undocumented immigrants are detained each year because of being in the Netherlands illegally. Most of them are asylum seekers whose applications for residence have been rejected and who are waiting to be sent back to their countries of origin. Dorn et al. (2011) interviewed a sample (n = 122) of these detained immigrants and found that only half of them knew how to get access to medical care in the Netherlands before they were detained, and about 25% of the care seekers reported being denied care by a health care provider. In a recent study, Schoevers, Loeffen, van den Muijsenbergh, and Lagro-Janssen (2010) interviewed undocumented women (n = 100) in order to identify obstacles they experience in accessing health care facilities. A majority (69%) reported encountering obstacles, and overall their health care utilization was low. Forty-seven women (47%) reported institutional obstacles, and 40 women (40%) reported personal obstacles such as shame, fear, and/or lack of information.

We conclude that present research results on specialized mental health service use indicate a tendency toward equal access in the case of similar psychiatric problems for certain groups of adult immigrants (Turks and Moroccans), inadequate data for other groups—especially for immigrants from Surinam and the former Netherlands Antilles—and underutilization of medical and mental health services among migrant children, asylum seekers, and most groups of refugees.

Anthropological perspectives

Most research in the field of transcultural psychiatry in the Netherlands has been done by psychiatrists and psychologists. A promising development, however, is that an increasing number of research teams now include anthropologists. In general, however, quantitative epidemiological studies still prevail and there has been little anthropological research with qualitative methodologies and an “experience near” approach. The main topics of anthropological (inspired) research related to psychiatry are perceptions of illness and disease, use of nonbiomedical health care, health-seeking behaviour, cultural identity and mental health, and patient–therapist interaction, especially the effects of ethnic matching of therapists and patients. In addition, the core transcultural psychiatric concepts of idioms of distress and explanatory models have been topics of anthropological research. We will briefly discuss these issues, and give an impression of recent anthropological inspired research in the Netherlands.

Concerning health-seeking behaviour, we have discussed the ground-breaking study by Tankink (2009; Tankink & Richters, 2007) on coping strategies of sexually violated refugee women. Beijers (2004) explored the construction of meanings of psychological distress and the pathways to accessing health care by members of the small Cape Verdean community in Rotterdam. These immigrants made considerably less use of mental health services than did ethnic Dutch, and also less than Moroccan and Turkish immigrants. The community appeared to be highly networked and not limited to city borders, but rather transnational and including Cape Verdeans in the Netherlands and abroad. Beijers noted that Cape Verdean patients had a syncretic blend of beliefs about the causes and culturally appropriate methods of treating their symptoms and illnesses, and accordingly used both standard medical procedures as well as complementary medicine, including spiritual, healing methods in the Netherlands and abroad.

Borra (2011) analysed the presentation of symptoms in a sample of 20 Turkish women with somatic and depressive complaints. Mental health workers disputed the presence of depressive disorder. Using the Diagnostic Interview for Turkish Women (Borra, 2010), 18 of the women were diagnosed with a depressive disorder. The women presented a wide range of somatic complaints. Anxiety and agitation were present as frequently as depressive complaints. Furthermore, these women used specific Turkish words to express their suffering. Borra concluded that the idiom of distress used by these women could result in the underdiagnosis of depressive and anxiety disorders.

Concerning explanatory models, Ghane, Kolk, and Emmelkamp (2010) studied the therapist–patient discrepancy in illness explanations and early outcome in intercultural psychotherapy. Greater discrepancy in psychological explanations was significantly associated with decreased therapeutic effectiveness and predicted lower psychotherapy adherence. The presentation of illness beliefs was influenced by ethnic (mis)match. In ethnically matched therapeutic relationships, patients considered interpersonal, religious, and mystical causes of their illness to be of primary importance. When this was not the case, patients adhered to medically acceptable explanations of their symptoms. Ghane et al. concluded that psychotherapy did not reduce therapist–patient discrepancy in psychological attributions, and that discrepancies in medical and magical attributions were not associated with poorer outcome. Oliemeulen and Thung (2007) also investigated the role of explanatory models in a study of psychotic patients of Dutch, Asian-Surinamese, and Turkish descent, and their care providers. They concluded that the explanatory models of patients were rarely cohesive and highly variable, and that illness beliefs had little predictive value.

During the past decade, transcultural psychiatrists and medical anthropologists have developed a promising new instrument, the Cultural Interview (Groen, 2009; Rohlof, Knipscheer, & Kleber, 2009). The interview was introduced in 2002 to assess cultural factors influencing clinical encounters and to facilitate a cultural formulation of diagnosis (CFD, DSM-IV). Initial research using this approach (Groen & Laban, 2011) showed that of the five themes considered to be core components of cultural case formulation included in the cultural interview, the theme of “cultural identity” might be more important for clinical treatment than “cultural explanations of illness.” Research is in progress to determine the effectiveness of the cultural interview and cultural case formulation in baseline mental health services.

Discussion about the role of culture related to mental health care has a long history in the Netherlands. Since the 1990s, anthropologists have pointed out that (mental) health care workers mainly interpreted culture as a static, invariable set of values, norms, and traditions of social groups, determining their health beliefs and behaviour. Anthropologists argued that culture as a set of definitions of reality is dynamic and interpersonal, and that health care is, in fact, producing culture (Bartels, 2002). Culture is not the only determinant of peoples’ beliefs and behaviour, and people make selective use of the cultural repertoires at their disposal (Hoffer, 2009; van Dijk, 1998). Van Dongen (2005) states that the way the concept of culture is used turns immigrants into classificatory anomalies or “liminal personae.” She concludes that culture is used in mental health care “to engender differences without questioning or investigating the answers people have to their health problems” (p. 194).

In a case study, van Dijk (2009) showed the negative effects of prevailing cultural stereotypes in mental health care, and how they may contribute to the neglect of the personal history and the local social world of the patient in both the diagnostic process and in treatment. In interpreting the patients’ beliefs and behaviour, some clinicians even ignored the biographical facts about the patients’ personal history, which they recorded themselves in the patients’ clinical file. The patient disappears as an individual and reappears as an anonymous representative of what is taken to be his/her culture. Van Dijk concludes that by using cultural stereotypes, therapists try to contain a deranging therapeutic process. During the past several years, there has been a growing tendency among both clinicians and health care planners to stress the individuality and uniqueness of the patient as a person, neglecting the cultural dimension. Culture seems now to become only one of the numerous dimensions of diversity. Rethinking the role of culture in mental health care by taking account of the “individualized experience of culture” (Lakes, Lopez, & Garro, 2006) as the starting point will be a challenge in Dutch transcultural psychiatry during the coming decade.

Concluding remarks

During the last decade the number of studies, publications, and dissertations in the Netherlands dealing with issues in the field of transcultural psychiatry has rapidly increased. Dutch researchers have participated more prominently in the international debate on issues of migration and mental health. In his review of international developments in transcultural psychiatry over the past 25 years, Kortmann (2008) stated that “the focus of attention has broadened to include inter-ethnic differences and social and community factors that induce and perpetuate mental disorders in various ethnic groups” (p. 149). This also was the case in the Netherlands. Assumptions about the mental health and care of migrants governing professional opinion and health policy in the previous decades were scientifically tested and often proved to be not accurate. Based on empirical and epidemiological studies, Knipscheer (2000) was one of the first who came to the now widely accepted conclusion that “the” immigrant does not exist. In the period that followed his publication, the exploration of intra- and interethnic differences in mental health and utilization of mental health services became a recurring theme in Dutch transcultural psychiatric research. To investigate risk and protective factors in relation to health disparities and differences in health service use, many studies have analysed the impact of socioeconomic status on the one hand, and culture and ethnicity on the other.

Although indicated by medical professionals in the 1980s and 1990s, research now has shown that specific sociocultural factors have a negative impact on mental health. For example, social defeat, including perceived discrimination and social exclusion, is associated with substantial higher incidence rates of psychosis, and a long asylum procedure and related postmigration problems are important risk factors for psychopathology. At the same time, successful social interaction and participation in Dutch society, along with maintenance of traditional culture and identity, and ethnic cohesion—with ethnic density as indicator—are associated with less psychological distress among immigrants. From this perspective, the present tendency in politics and public opinion to put pressure on immigrants in the Netherlands to either assimilate or leave, should be considered as a risk to their mental health.

The overall conclusion of this review is that considerable intra- and interethnic differences in psychopathology and mental health service utilization have become evident in the Netherlands during the past decade. This variability calls for further exploration to explain its causes and may lead to redefining the relationship between migration and mental health, and reconsidering underlying mechanisms. Explanations have been sought in identifying mediating sociocultural factors, such as social defeat and ethnic density, as well as bias in diagnostic instruments and procedures. At the same time, discussion continues about the significance of ethnicity and socioeconomic status in determining the prevalence of mental disorders. This means that sociocultural factors should be included in the definition of individuals or subgroups at risk, rather than defining ethnic minority groups as a whole as a group at risk.

Looking back over this first decade of the new millennium, research on culture and mental health has flourished. Many universities and institutes have established research programs focused on migrants and migration-related issues. However, there is still a lack of coordination of research projects developed by psychiatrists, psychologists, and anthropologists. Interdisciplinary coordination is still in its early stages, and the often recommended mixed-methodology approach to transcultural psychiatry research has seldom been utilized. In a “state-of-the-art” review of current research on prevalence, accessibility, and quality of care, Bekker and van Mens-Verhulst (2008) strongly recommended investing in research on the quality and effectiveness of standard therapeutic interventions in mental health. We agree with that. An international conference on transcultural psychiatry—held in June of 2010 in Amsterdam and organized together with the World Psychiatric Association (WPA) Transcultural Psychiatry Section—brought together many Dutch psychiatry, psychology, and anthropology researchers and generated increased support for the kind of interdisciplinary and mixed-methodology research also recommended by Bekker and van Mens-Verhulst (2008).

Another observation is that the research findings have not yet had much impact on political decisions. In a complex world of a multitude of scientific findings, chaotic and incident-focused behaviour of politicians and media, and increasing “fact-free politics,” researchers have a hard time getting their message through. Influencing policy takes a lot of time and endurance and it is therefore understandable that most researchers prefer to concentrate on their next study, rather than trying to change national policies toward immigrants, asylum seekers, and refugees. Unfortunately, the consequence is that these groups, who are the subjects of their studies, do not benefit from their research findings. Perhaps greater cooperative efforts with relevant nongovernmental organizations could increase the chances of the translation of scientific findings into political decisions.

The following is a list of research questions relevant to policy and practice for future consideration:

What factors can keep immigrants, refugees, and asylum seekers healthy and make them resilient, and what interventions could improve their mental health and well-being?