To the Editor A recent study by Murakami et al1 assessed the validity of 12 wearable devices in measuring total energy expenditure. However, some important methodological concerns were not addressed in the article. First, for the doubly labeled water experiment, was nonwear time of the participants recorded? If so, what was the average nonwear time per day? This is important to assess the compliance rate and the validity of the study results. Second, were the participants requested to wear all wrist-worn devices on their nondominant wrist (a common practice of most research with wrist-worn accelerometers2)? Wearing the devices on the dominant wrist may misclassify sedentary behaviors and light activities (eg, window washing) as moderate-intensity physical activity.3 Third, a major limitation of this study lies in concurrently wearing 7 devices at once on the wrist, and therefore some of them will be on their dominant wrist while some on the nondominant wrist. Although the positions of these devices were randomly assigned, unavoidably this would introduce interparticipant variability. All these concerns should be addressed as they potentially affect the conclusions of the study.
Lee PH. Remaining Questions Concerning Wearable Devices. JAMA Intern Med. 2016;176(9):1408. doi:10.1001/jamainternmed.2016.4750