Background: Decades of psychiatric epidemiology have shown a wide variation in prevalence rates, but a consistent relationship between psychiatric disorder and sociodemographic variables. In this repeated cross-sectional survey, the prevalence of psychiatric disorders and their distribution in the general population of the same area was assessed in 1983 and 1997.
Methods: With an interval of 14 years, 2 two-phase studies of psychiatric prevalence were carried out among the inhabitants of a Dutch Health Area (Nijmegen). In phase 1, a random sample of persons answered the General Health Questionnaire (GHQ-30). In phase 2, the respondents were interviewed using a clinical semi-structured interview. Only phase 1 data will be reported here.
Results: The mean overall GHQ-score changed significantly from 3.1 (+/-1.0) in 1983 to 4.6 (+/-1.8) in 1997. On a bivariate level, higher score rates were found consistently in the age categories > or = 50 years, among divorced persons, the lower educational levels, the unemployed/chronically ill and in the urban areas. On a multivariate level (second order effect), however, the variance explained by these sociodemographic variables doubled, revealing the importance of complex interactions.
Limitations: Our aim to ensure identical designs in 1983 and 1997 could not completely be achieved.
Conclusions: In the course of time, psychiatric prevalence increased in all sociodemographic categories, despite the improved socioeconomic conditions in the survey population as a whole. The increasing complexity of life apparently takes its toll, even of the socially best equipped.