All time references are in CEST
Data collection with wearable devices 1
| Dr Alexander Wenz (University of Mannheim)
Professor Christopher Antoun (University of Maryland)
|Wednesday 19 July, 14:00 - 15:00
Wearable devices, such as smart watches and activity trackers, are increasingly used for data collection in the social, behavioral, and health sciences. Equipped with a wide range of sensors, these devices allow researchers to measure physical activity, sleep behavior, and cardiovascular health, among many other things. While most previous sensor-based studies were implemented on small-scale samples of volunteers, recent studies have started to extend this approach to larger samples of the general population.
Despite the proliferation of wearable devices as tools for data collection, the potential sources of error in such data are not yet fully understood. In this session, we welcome contributions that examine and improve the quality of data collection with wearable devices, for example:
• Recruitment; nonparticipation; non-adherence
• Measurement error
• Weighting; imputation
• Fieldwork; implementation issues
• Errors when processing and interpreting sensor data
• Consent; privacy
Keywords: Sensors, Activity trackers, Accelerometers, Data Quality
Mr Tim Weihrauch (Robert Koch Institute, Department of Epidemiology and Health Monitoring, Berlin, Germany / Charité Universitätsmedizin Berlin, Germany) - Presenting Author
Dr Hannelore Neuhauser (Robert Koch Institute, Department of Epidemiology and Health Monitoring, Berlin, Germany / DZHK (German Centre for Cardiovascular Research), Berlin, Germany)
Mr Johannes Lemcke (Robert Koch Institute, Department of Epidemiology and Health Monitoring, Berlin, Germany)
Mrs Julia Büschges (Robert Koch Institute, Department of Epidemiology and Health Monitoring, Berlin, Germany / DZHK (German Centre for Cardiovascular Research), Berlin, Germany)
Mrs Jennifer Allen (Robert Koch Institute, Department of Epidemiology and Health Monitoring, Berlin, Germany)
Mr Patrick Schmich (Robert Koch Institute, Department of Epidemiology and Health Monitoring, Berlin, Germany)
Dr Birga Maier (Robert Koch Institute, Department of Epidemiology and Health Monitoring, Berlin, Germany)
Home blood pressure may be of superior prognostic value for cardiovascular events than office blood pressure and BP self-measurement is thus of interest for population surveys. However, standardization requirements, i.e. sitting and arm position, selecting a cuff adapted to the arm circumference and the required resting time are challenging. Our study investigates the feasibility of standardized BP self-measurements at home.
202 persons participated in the study. They were mailed a BP monitor with instructions on how to measure BP. Instructions were also provided in a video. Participants were asked to document 2 measurements in the morning/evening, with 2 minutes in between, on 3 days on a paper sheet. Measurements were also recorded electronically. Participants had to answer an online-survey in the beginning (baseline characteristics) and at the end of the study (usability and acceptability interview).
Out of 202 agreeing to participate, 111 (56%) completed the study. Statistically significant differences in baseline characteristics in those completing the study and those not completing it were not found. The 111 participants completing the study documented 12 measurements on paper, at the same time they recorded more measurements electronically (range: 12-65), not following the recommended time schedule in 37% of cases. In spite of these differences, mean BP measurements recorded on paper and electronically did not differ for the 111 participants: mean systolic BP: 129.7 mmHg for electronic / 129.0 mmHg for paper records; mean diastolic BP: 81.6 mmHg for electronic / 81.6 mmHg for paper records.
Our results suggest that when a high number of resting BP measurements over several days are available adherence to a strict morning/evening measurement schedule may not play a major role for the recorded blood pressure level. This finding can add flexibility in study design and increase response.
Dr Maaike Kompier (CBS) - Presenting Author
Miss Inge de Wolf (Amsterdam UMC)
Dr Anne Elevelt (CBS)
Dr Annemieke Luiten (CBS)
Mr Joris Mulder (Centerdata)
Dr Vera Toepoel (CBS)
The World Health Organization (2020) guidelines advocate a minimum of 150 minutes of moderate-to-vigorous intensity physical activity per week for adults. Adherence to these guidelines in the Netherlands is assessed using a population-based surveillance system based on a self-report questionnaire (SQUASH). Yet, self-report questionnaires often suffer from response biases, resulting in overestimation of physical activity. These issues may be overcome using wearable sensors that measure physical activity (i.e., accelerometers).
The aim of this study was to examine whether accelerometers can be used as a replacement of, or in addition to, questionnaires to monitor physical activity. Differences in physical activity measures between self-reported data from the SQUASH and accelerometer data gathered by wearing a research-grade accelerometer (ActivPal) will be examined. Furthermore, we will explore how data tracked by participants’ own devices (e.g., FitBit or apple watch) compare to the SQUASH and the ActivPal data.
In this study, 556 members of a probability-based online panel (LISS panel) wore the ActivPal for one week between February and June 2022. This group existed of 97 Fitbit users and 325 users of other commercial brands. After the study period, these participants were asked to donate their data by downloading and uploading for FitBit users or by copying the data from the tracker into a questionnaire for the users of other brands. A control group of 134 participants without own tracker were included to allow testing of differences in activity levels between the groups with and without own tracker. Prior to (October 2021) and after (July 2022) wearing the ActivPal, participants completed the SQUASH questionnaire.
Data analyses are currently ongoing. The results will be presented at the conference.
Dr Jessica Faul (University of Michigan) - Presenting Author
Increasingly it has become feasible to use wearable devices to assess physical activity, sedentary behavior, and sleep patterns among large numbers of participants in epidemiologic and cohort studies, allowing for more precise assessments of these behaviors and examination of their associations with health outcomes. In 2019, the Health and Retirement Study (HRS) used the GENEactiv wrist device to measure physical activity and sleep over 10 days among 380 pilot study participants. Devices were sent to and returned by consenting participants via mail along with a short questionnaire and sleep diary. Pilot-eligible participants and a sample of controls also answered an expanded set of activity and sleep questions in their 2018 and 2020 core HRS surveys so we could better assess selection into the pilot study and behavior effects post pilot. In this paper we describe the methods used to collect and process the accelerometer data. We also describe the quality of the accelerometer data collected, comparison of physical activity and sleep data collected from the devices with self-reported data, and discuss implications for administration in a longitudinal sample.