ORIGINAL PAPER Prevalence of mental disorders and trends from 1996 to 2009. Results from the Netherlands Mental Health Survey and Incidence Study-2 Ron de Graaf ��� Margreet ten Have ��� Coen van Gool ��� Saskia van Dorsselaer Received: 16 September 2010 / Accepted: 8 December 2010 �� Springer-Verlag 2010 Abstract Objective To present prevalences of lifetime and 12- month DSM-IV mood, anxiety, substance use and impulse- control disorders from the second Netherlands Mental Health Survey and Incidence Study (NEMESIS-2), and to compare the 12-month prevalence of mood, anxiety and substance use disorders with estimates from the first study (NEMESIS-1). Method Between November 2007 and July 2009, a nationally representative face-to-face survey was conducted using the Composite International Diagnostic Interview 3.0 among 6,646 subjects aged 18���64. Trends in 12-month prevalence of mental disorders were examined with these data and NEMESIS-1 data from 1996 (n = 7,076). Results Lifetime prevalence estimates in NEMESIS-2 were 20.2% for mood, 19.6% for anxiety, 19.1% for sub- stance use disorder and 9.2% for impulse-control disorder. For 12-month disorders, these figures were 6.1, 10.1, 5.6 and 2.1%, respectively. Between 1996 and 2007���2009, the 12-month prevalence of anxiety and substance use disorder did not change. The prevalence of mood disorder decreased slightly but lost significance after controlling for differ- ences in sociodemographic variables between the two studies. Conclusion This study shows that in the Netherlands mental disorders are prevalent. In about a decade, no clear change in mental health status was found. Keywords Mental disorders Prevalence Trends Introduction The Netherlands Mental Health Survey and Incidence Study (NEMESIS-1) was the first Dutch nationally representative survey that estimated the prevalence of mental disorders [1, 2]. It showed that in 1996 the 12-month prevalence of at least one Axis-1 DSM-III-R disorder among the adult pop- ulation was 23.2%, and the lifetime prevalence was 41.2% [2]. Of those with a 12-month mental disorder, 33.9% used professional care for mental problems [3]. These results are now more than a decade old and it is the question whether these figures still hold. Therefore, the Netherlands Mental Health Survey and Incidence Study-2 (NEMESIS-2) was executed, with an expansion toward externalising (impulse- control) disorders. In the Netherlands [4���6], like elsewhere [7], there has been debate on the question whether mental disorders increased in the last decade(s). Increased substance use and earlier onset of substance use was found among adolescents 10���20 years ago in the Netherlands [8���10]. These subjects have now become adults and their (former) substance use might have contributed to an increase in adult substance use disorders, and of other mental disorders which occur secondarily [11]. On the other hand, suicide rates, which also are associated with mental disorders, have not increased since 1996 in the Netherlands (according to Statistics Netherlands http://www.cbs.nl). Furthermore, a decrease of mental disorders could be expected because of R. de Graaf (&) M. ten Have S. van Dorsselaer Netherlands Institute of Mental Health and Addiction, Da Costakade 45, 3521 VS Utrecht, The Netherlands e-mail: rgraaf@trimbos.nl C. van Gool Centre for Public Health Forecasting, National Institute for Public Health and Environment, Bilthoven, The Netherlands 123 Soc Psychiat Epidemiol DOI 10.1007/s00127-010-0334-8

an advance in treatment with medication or otherwise [12]. Worldwide, trends in prevalence of mental disorders have not often been studied. For the US, such trends are known from two comparable general population studies. Com- pared to the National Comorbidity Survey (NCS 1990��� 1992), the prevalence of any 12-month mental disorder in the National Comorbidity Survey-Replication (NCS-R 2001���2003) did not change (29.4 vs. 30.5%) [7, 13]. In Great Britain (1993���2000) and Australia (1997���2007) also no significant change in the prevalence of psychiatric dis- orders was found [12, 14]. In contrast to these studies, other research found that the prevalence of major depression in the US increased from 3.3% in 1991���1992 to 7.1% in 2001���2002 [15]. Here we describe the data of NEMESIS-2 regarding lifetime and 12-month prevalence of DSM-IV mood, anx- iety, substance use and impulse-control disorders. Trends in the 12-month prevalence of mood, anxiety and substance use disorders are studied by indirect comparison of these figures with imputed DSM-IV figures of NEMESIS-1. Also, sociodemographic correlates of disorders in NEME- SIS-2 and trends in correlates are studied. Materials and methods Samples In both NEMESIS-2 and NEMESIS-1 a multistage, strati- fied random sampling procedure was applied. First, a ran- dom sample of municipalities was drawn. Second, a random sample of addresses of private households from postal registers in these municipalities was drawn, each address with the same probability of selection. Third, based on the most recent birthday at first contact within the household, a random individual aged 18���64 years and sufficiently fluent in the Dutch language was selected to be interviewed. Addresses of institutions were excluded thus institutional- ized individuals (i.e. those living in hospices, prisons) were excluded. Those temporarily living in institutions, however, could be interviewed later during the fieldwork if they returned home. Fieldwork Both studies were approved by a medical ethics committee. In NEMESIS-2, after having been informed about the study aims, respondents provided written informed consent. In NEMESIS-1, respondents provided verbal informed con- sent, according to the prevailing Dutch law of 1996. In both studies, selected households were sent a letter by the Minister of Health, Welfare and Sport, in which (s)he explained and recommended the study. In NEMESIS-2, a brochure was accompanied explaining its goals in more detail it also referred to a website for respondents. In both studies, shortly after sending out this letter, households were contacted by telephone or visited in person if no phone number was available. At least ten visits or phone calls at different times of the day and different days of the week were done. Willingness to participate in scientific studies decreased dramatically in the Netherlands [16] and elsewhere [17] in the last decade, probably because people are requested frequently to participate in marketing surveys. To achieve a relatively high response rate, in NEMESIS-2 the recruit- ment methods were much more intensive compared to NEMESIS-1: less first contacts by phone due to less available phone numbers and more undisclosed cell phone numbers more repeated contact efforts by different inter- viewers in case of no hard refusal and higher incentive. By means of these methods it was possible to reach a relatively high response rate for the Netherlands (N = 6,646): 65.1% [18]. In NEMESIS-1 (N = 7,076) this was 64.2% [1]. In both NEMESIS-studies, younger people (especially 18���24 years) were underrepresented in the sample. To be able to generalize the results to the general population, in both datasets a weighting factor was constructed to correct for different response rates in different population groups. The following population characteristics obtained from Statistics Netherlands were used to construct these weight- ing factors: sex, age, partner status, educational level (only in NEMESIS-2) and urbanicity. Table 1 presents demographic characteristics of the weighted samples. Sample differences reflect changes that have taken place in the population at large. For a more detailed description of the design and fieldwork, see De Graaf et al. [18] for NEMESIS-2 and Bijl et al. [1] for NEMESIS-1. Diagnostic instruments In NEMESIS-2, DSM-IV disorders were assessed with the Composite International Diagnostic Interview (CIDI) 3.0 and in NEMESIS-1, DSM-III-R disorders with the CIDI 1.1. CIDI 3.0 was developed and adapted for use in the WHO- World Mental Health (WMH) Survey Initiative [19]. In the Netherlands, the CIDI 3.0 was used in the European Study on the Epidemiology of Mental Disorders (ESEMeD), which is a part of this initiative. The CIDI 3.0 was first produced in English and underwent a rigorous process of adaptation to obtain a conceptually and cross-culturally comparable Dutch version [20, 21]. The CIDI 3.0 version used in NEMESIS-2 was an improvement of the one used in the Dutch ESEMeD study. To reduce interview duration and to minimise the possibility that respondents learn how to shorten the inter- view by answering negatively to key questions if these are Soc Psychiat Epidemiol 123