Published in:
01-08-2015 | Original Paper
Cross-sectional time trends in psychological and somatic health complaints among adolescents: a structural equation modelling analysis of ‘Health Behaviour in School-aged Children’ data from Switzerland
Authors:
Michelle Dey, Anthony F. Jorm, Andrew J. Mackinnon
Published in:
Social Psychiatry and Psychiatric Epidemiology
|
Issue 8/2015
Login to get access
Abstract
Purpose
This study examined cross-sectional time trends in health complaints among adolescents living in Switzerland, including differences between population subgroups and sources of differential response to items.
Methods
Swiss data were analysed from the Health Behaviour in School-aged Children (HBSC; including 11–15 years old) from 1994 (n = 7008), 1998 (n = 8296), 2002 (n = 9066) and 2006 (n = 9255). Structural equation modelling was used to assess (1) the structure of the HBSC Symptom Checklist (HBSC-SCL; questionnaire, which asks about the frequency of eight health complaints) and (2) associations between the HBSC-SCL with year of data collection and demographic characteristics of the participants.
Results
Two correlated factors fitted the data better than a single factor. The psychological factor included the items ‘feeling low,’ ‘irritability and bad temper,’ ‘nervousness’ and ‘difficulties in getting to sleep,’ and the somatic factor the items ‘headache’, ‘backache’, ‘stomach ache’ and ‘dizziness’. Relative to 1994, lower levels of psychological health complaints were experienced in 1998, 2002 and 2006. However, the changes were only minor. In contrast, somatic health complaints increased monotonically over the years of the survey. Experiencing psychological and somatic health complaints was more pronounced with age among females relative to males and was associated with living in particular language regions of Switzerland.
Conclusions
Different cross-sectional time trends were identified for the psychological and somatic latent variables, indicating that both factors should be investigated when studying period effects.