Response, Willingness, and Data Donation in a Study on Accelerometer Possession in the General Population

Abstract

In this study, we investigate prevalence of smartwatches; activity trackers (e.g., Fitbits); and apps to track personal activity on smartphones in the Dutch general population. In addition, we ask for respondents’ willingness to participate in a follow-up accelerometer study and wear a professional loaned activity tracker for a week. About half of the sample owns a personal device to track physical activity; 58.0% of those respondents are willing to copy personal data from the device into a questionnaire; 40.6% are willing to upload a spreadsheet from their personal device to a research portal. About half of the respondents of the entire sample express willingness to participate in a follow-up study and wear a professional accelerometer for a week. However, once invited, only 60.0% actually consent to participation. Respondents who own a personal device to monitor physical activity are more inclined to participate in the follow-up accelerometer study than respondents who do not own a personal device. This study shows that respondents with personal activity trackers show higher levels of physical activity compared to respondents without a personal device. In addition, participants in the follow-up study show higher levels of physical activity. Hence, estimating physical activity from personal activity trackers or professional activity trackers will likely result in biased results. We do not find significant differences between respondents with a personal device versus respondents without—nor people who participate in the accelerometer follow-up study—in sitting, standing, and lying time, however. Estimating sedentary behavior from personal or professional activity trackers is likely to result in unbiased results.