Development and Verification of Validity and Reliability of the International Physical Activity Questionnaire Taiwan Version

Background: Physical inactivity is a global concern, but diverse physical activity measures in use prevent international comparisons. The International Physical Activity Questionnaire (IPAQ) was developed as an instrument for cross-national monitoring of physical activity and inactivity. An International Consensus Group developed two long and two short forms of the IPAQ instruments (administered by telephone interview or self-administration, with reference periods, the“last 7 days” of recalled physical activity). Methods: Under translation, back translation and 12 sections of focus group discussion; three versions of IPAQ-Taiwan and GPAQ-Taiwan version were developed. From August to November in 2003, investigators from a community were collected reliability and validity data (141 samples) by systemic and household sampling. Test-retest reliability was assessed within 3 to 21 days. Concurrent validity was assessed at the IPAQ, GPAQ, 7-D PAR, diary and TASPA, and criterion IPAQ validity was assessed against the RT3 accelerometer. Spearman’s correlation coefficients were reported, based on the total reported physical activity. Results: The CVI of IPAQ-Taiwan versions were very high (higher than 0.9). The agreement coefficients between original versions and Taiwan versions were good (intraclass correlation coefficients = 0.72~0.93). Overall, the repeatable data which IPAQ questionnaires produced (Spearman’s ρ clustered around 0.67~0.96) was better than most of other samples (0.25~0.88), comparable data from short and long forms. Concurrent validity which was 0.63~0.88 comparisons with either two IPAQ, was similar with the other national data. Criterion validity had a median of about 0.31 to 0.41, which was comparable to most of other IPAQ validation studies (0.02~0.52). The reliability of telephone administration was better than the self-administered mode. Conclusions: The IPAQ-Taiwan versions have acceptable measurement properties, at least they are as good as other established the other versions of IPAQ and measurement of physical activity of self-reported. The study proved that the criteria validity coefficients coming from the subjects of lower education level did not decrease; therefore, IPAQ can be used to measure their level of physical activity. Considering the community base samples in this study, IPAQ-Taiwan version has reasonable measurement properties for monitoring population levels of physical activity among 18 to 65 years old adults, the short IPAQ (Taiwan version) form is recommended for national monitoring and the long (Taiwan version) form is also recommended for research.