Evaluating the validity of the French version of the Four-Dimensional Symptom Questionnaire with differential item functioning analysis

Background. GPs are daily confronted with mental disorders and psychosomatic problems. The Four-Dimensional Symptom Questionnaire (4DSQ), measuring distress, depression, anxiety and somatization, was purposively developed for primary care. It has been translated into 12 languages and is commonly used in several countries. It was translated into French in 2008, by forward and backward translation, but it has not been validated for a primary care population. Aim. This study aimed to establish whether the French 4DSQ measured the same constructs in the same way as the original Dutch 4DSQ. Method. Two samples of French general practice patients were recruited during routine care to obtain as much variability as possible. One sample included consecutive patients, from the waiting room of rural GPs, over a period of 2 weeks and the other sample included patients with suspected psychological problems or unexplained symptoms. This population was compared to a matched Dutch sample using confirmatory factor analysis (CFA) and differential item functioning (DIF) analysis. Results. A total of 231 patients, from 15 French GPs, completed the questionnaire (Dutch reference group: 231). Mean age was 42.9 years (Dutch: 42.1); females numbered 71% in both samples. The multigroup CFA assessed configural invariance of one-factor models per 4DSQ scale. Thirteen of the total of 50 items in the 4DSQ, in three scales, were detected with DIF. However, DIF did not impact on the scale scores. Conclusion. French 4DSQ scales have the same latent structures and measure the same traits as the original Dutch 4DS