(Re)examining the factorial sstructure of the generalized anxiety disorder-7 in a college students sample

Introduction Generalized anxiety in young adults during college career is a serious public-health problem that untreated has a chronic course. Research has shown that the self-report questionnaire generalized anxiety disorder-7 (GAD-7) is a reliable and valid measure to assess generalized anxiety symptoms severity in heterogeneous psychiatric samples. However, GAD-7 is not available for non-clinical populations and their factor structure has not been re-examined. Objectives Our objective was to examine factor structure and measurement invariance of the GAD-7 among college students testing two alternatives models. Aims The original model fit of single-factor was compared to two-factor model that considered in comprehensive approach of generalized anxiety the assessment of cognitive-emotional nature and somatic symptoms. Methods In this cross-sectional study the GAD-7 was administrated to college students ( n  = 1031) recruited in the six schools which compose the Polytechnic Institute of Coimbra, Portugal. Confirmatory factor analysis was used testing two models. Results Among college students, 32.8% reported significant generalized anxiety symptoms. The original unidimensional structure of GAD-7 was confirmed but the two-factor model comprising cognitive–emotional and somatic factors presented better fit to the data ( χ 2 (1) = 21.01, P r  = 0.51, P α  = 0.85 and α  = 0.86 for cognitive-emotional and somatic items, respectively). The invariance factor of two-factor model across gender was also confirmed. Conclusions Results indicate that within college communities the GAD-7 integrates two stable generalized anxiety factors related but independent structure. The GAD-7 can be an adequate measure to detected generalized anxiety symptoms in this population.