Cross-cultural validation of a simple self-report instrument of physical activity in immigrants from the Middle East and native Swedes. Issue 3 (May 2014)

• Record Type: Journal Article
• Authors: Arvidsson, Daniel; Leijon, Matti; Sundquist, Jan; Sundquist, Kristina; Lindblad, Ulf; Bennet, Louise
• Is Part Of: Scandinavian journal of public health. Volume 42:Issue 3(2014:May)
• Page Start: 255
• Page End: 262
• Publication Date: 2014-05
• Subjects: Accelerometer -- accuracy -- agreement -- ethnicity -- gender -- GT1M -- migration; Public health -- Periodicals; Social medicine -- Periodicals; 614
• DOI: 10.1177/1403494813514143
• Languages: English
• ISSNs: 1403-4948

Abstract

Aim: To investigate cross-cultural validity of a simple self-report instrument of physical activity intended to be used in Swedish health care. Methods : A validation study performed in 599 Iraqis (58% men) and 553 Swedes (53% men) aged 30–75 years living in the city of Malmö, Sweden. The self-report instrument by the Swedish National Board of Health and Welfare was compared to corresponding measures assessed from accelerometry as reference. Results : The agreement between the methods in assessing the participants as sufficiently/insufficiently physically active (cut-point 150 min/week) was 65% in the Iraqis and 52% in the Swedes ( p <0.001). The proportion disagreement where the self-reported physical activity was sufficient but insufficient according to the accelerometry was 26% and 45% in Iraqis and Swedes, respectively. Physical activity time (min/week) was overestimated by self-report compared to accelerometry by 71% in the Iraqis and 115% in the Swedes ( p <0.001). The smallest and largest overestimation was seen in Iraqi (57%) and Swedish (139%) women, respectively. The deviation of the self-report instrument compared to accelerometry was related to the physical activity level, as the overestimation mainly occurred at lower physical activity. Conclusions : The self-report instrument proposed by the Swedish National Board of Health and Welfare may overestimate the proportion sufficiently physically active, but to an extent depending on cultural background and gender.