Effect of physical activity showcards on accuracy in self-reports: a randomized crossover study with the Global Physical Activity Questionnaire (GPAQ)

Introduction
Misperception and inaccuracy are major challenges when measuring physical activity via questionnaire. Some questionnaires like the Global Physical Activity Questionnaire (GPAQ) provide showcards illustrating physical activity of moderate and vigorous intensity. This study aims to investigate whether these showcards benefit the accuracy of self-reported physical activity data.
Methods
Fifty-four physically active volunteers (57.4 % female, mean age: 28.3±12.2 years; mean BMI: 23.2±3.1 kg/m2) were randomized to answer the German version of the GPAQ with showcards (“GPAQ+”) or without showcards (“GPAQ-“) after seven days. After the next seven days, the two study groups were reversed according to the randomized cross-over design. Objective physical activity of all participants was recorded by accelerometry (ActiGraph GT3X+) over the whole investigation period of 14 days (2x7 days). Differences between each GPAQ version and accelerometry were evaluated by Bland-Altman-analyses and Wilcoxon-tests.
Results
Neither for moderate (p=0.328) nor for vigorous intensity (p=0.406) statistically significant differences between the two groups could be found. Mean accelerometer-derived moderate and vigorous physical activity were 109.9±29.4 min/day and 12.8±9.7 min/day, respectively. The average discrepancy between GPAQ and accelerometry was -23.1±89.3 min/day (GPAQ-) and -21.2±117.05 min/day (GPAQ+) for moderate intensity, and 35.2±45.5 min/day (GPAQ-) and 30.3±35.4 min/day (GPAQ+) for vigorous physical activity.
Conclusion
The showcards demonstrate no significant effect on the accuracy of self-reported physical activity in this study. Both groups underestimate the time of moderate intensity and overestimate the duration of vigorous physical activity which indicates misclassification of self-assessed data. However, these findings should be further investigated with other target groups and larger sample sizes. Moreover, other strategies to improve questionnaire accuracy should be elaborated.