Validation of activity questionnaires in patients with cystic fibrosis by accelerometry and cycle ergometry
1 Department of Pediatrics, University of Wuerzburg, Josef-Schneider-Strasse 2, Würzburg, Germany
2 Division of Respiratory Medicine, University Children's Hospital, University of Zurich, Steinwiesstrasse 75, Zürich, Switzerland
3 Swiss Tropical and Public Health Institute, University of Basel, Socinstrasse 57, Basel, Switzerland
BMC Medical Research Methodology 2012, 12:43 doi:10.1186/1471-2288-12-43Published: 3 April 2012
The objective of this study was to validate physical activity questionnaires for cystic fibrosis (CF) against accelerometry and cycle ergometry.
41 patients with CF (12-42 years) completed the Habitual Activity Estimation Scale (HAES), the 7-Day Physical Activity Recall questionnaire (7D-PAR) and the Lipid Research Clinics questionnaire (LRC) and performed an incremental exercise test according to the Godfrey protocol up to volitional fatigue. Time spent in moderate and vigorous physical activity (MVPA) assessed objectively by accelerometry was related to the time spent in the respective activity categories by correlation analyses and calculating intraclass correlation coefficients (ICC). Furthermore, the results of the exercise test were correlated with the results of the questionnaires.
Time spent in the categories 'hard','very hard' and 'hard & very hard' of the 7D-PAR (0.41 < r < 0.56) and 'active' (r = 0.33) of the HAES correlated significantly with MVPA. The activity levels of the LRC were not related to objectively determined physical activity. Significant ICCs were only observed between the 7D-PAR activitiy categories and MVPA (ICC = 0.40-0.44). Only the LRC showed moderate correlations with the exercise test (Wmax: r = 0.46, p = 0.002; VO2peak: r = 0.32, p = 0.041).
In conclusion, the activity categories 'hard' and 'very hard' of the 7D-PAR best reflected objectively measured MVPA. Since the association was at most moderate, the 7D-PAR may be selected to describe physical activity within a population. None of the evaluated questionnaires was able to generate valid physical activity data exercise performance data at the individual level. Neither did any of the questionnaires provide a valid assessment of aerobic fitness on an invidual level.