Evaluating the cross-cultural validity of the Polish version of the Four-Dimensional Symptom Questionnaire (4DSQ) using differential item functioning (DIF) analysis

• Published in: Family practice Vol. 29; no. 5; pp. 609 - 615
• Main Authors: Czachowski, Sławomir, Terluin, Berend, Izdebski, Adam, Izdebski, Paweł
• Format: Journal Article
• Language: English
• Published: England 01-10-2012
• Subjects: Adult; Anxiety-Depression; Cross-Cultural Comparison; Family Practice; Female; Humans; Language; Logistic Models; Male; Mental Disorders - diagnosis; Middle Aged; Netherlands; Poland; Primary health care; Psychiatric disorders; Psychological distress; Somatization; Surveys and Questionnaires - standards; Symptoms; Poland; Netherlands
• ISSN: 0263-2136; 1460-2229
• DOI: 10.1093/fampra/cms016

The original Dutch Four-Dimensional Symptom Questionnaire (4DSQ), which measures distress, depression, anxiety and somatization, has been translated into Polish with the aim of providing primary health care with a good screening instrument for the detection of the most prevalent mental health problems (anxiety, somatization, depression and distress).
To check if the Polish version is cross-culturally valid so that the scores of Polish subjects can be compared with the scores of Dutch subjects and the Dutch cut-off points can be used in Polish subjects.
4DSQ data were collected from a mixed sample of students and primary care attendees. The Polish data were compared with the 4DSQ data of a matched sample of Dutch students and primary care attendees. Two methods of differential item functioning (DIF) analysis, ordinal logistic regression and generalized Mantel-Haenszel, were used to detect items with DIF, and linear regression analysis was used to estimate the scale-level impact of DIF.
Four items showing DIF were detected in the distress scale, one in the somatization scale and one in the anxiety scale. The DIF in distress caused Polish subjects with moderate scores to score circa 1 point less than their Dutch counterparts.
The results of the DIF analyses suggest that the Polish 4DSQ measures the same constructs as the Dutch 4DSQ and that the Dutch norms can be used for the Polish subjects, except for distress: the first cut-off point should be one point lower.