Repository logo
HARVEST

Browsing by Author "Fuller, Daniel"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Changes in physical activity among Canadian adults more than 6 months into the COVID-19 pandemic: a secondary analysis of the INTERACT cohort study
    (BMJ Open, 2024-09) Sturrock, Shelby; Gesink, Dionne; Winters, Meghan; Kestens, Yan; Stanley, Kevin; Moineddin, Rahim; Woodruff, Sarah; Fuller, Daniel
    Objective: To estimate the effect of (a) the COVID-19 pandemic and (b) COVID-19 restriction stringency on daily minutes of device-measured moderate-to-vigorous physical activity (MVPA). Design Physical activity data were collected from theINTerventions, Equity, Research and Action in Cities Team (INTERACT) cohorts in Montreal, Saskatoon and Vancouver before (May 2018 to February 2019, ‘phase 1’) and during the pandemic (October 2020 to February 2021, ‘phase 2’). We estimated the effect of the two exposures by comparing daily MVPA measured (a) before vs during the pandemic (phase 1 vs phase 2) and (b) at different levels of COVID-19 restriction stringency during phase 2. Separate mixed effects negative binomial regression models were used to estimate the association between each exposure and daily MVPA, with and without controlling for confounders. Analyses were conducted on person-days with at least 600min of wear time. Effect modification by gender, age, income, employment status, education, children in the home and city was assessed via stratification. Setting: Montreal (Quebec), Saskatoon (Saskatchewan) and Vancouver (British Columbia), Canada. Main outcome measure: Daily minutes of MVPA, as measured using SenseDoc, a research-grade accelerometer device. Results: Daily minutes of MVPA were 21% lower in phase 2 (October 2020 to February 2021) compared with phase 1 (May 2018 to February 2019), controlling for gender, age, employment status, household income, education, city, weather and wear time (rate ratio=0.79, 95%CI 0.69, 0.92). This did not appear to be driven by changes in the sample or timing of data collection between phases. The results suggested effect modification by employment, household income and education. Restriction stringency was not associated with daily MVPA between October 2020 and February 2021 (adjusted rate ratio=0.99, 95%CI 0.96, 1.03). Conclusions: Between October 2020 and February 2021, daily minutes of MVPA were significantly lower than 2 years prior, but were not associated with daily COVID-19 restriction stringency.
  • Thumbnail Image
    Item
    Predicting lying, sitting, walking and running using Apple Watch and Fitbit data
    (BMJ Open Sport & Exercise, 2021) Fuller, Daniel; Rahimipour Anaraki, Javad; Simango, Bongai; Rayner, Machel; Dorani, Faramarz; Bozorgi, Arastoo; Luan, Hui; Basset, Fabien
    Objectives This study’s objective was to examine whether commercial wearable devices could accurately predict lying, sitting and varying intensities of walking and running. Methods We recruited a convenience sample of 49 participants (23 men and 26 women) to wear three devices, an Apple Watch Series 2, a Fitbit Charge HR2 and iPhone 6S. Participants completed a 65min protocol consisting of 40min of total treadmill time and 25min of sitting or lying time. The study’s outcome variables were six movement types: lying, sitting, walking self-paced and walking/running at 3 metabolic equivalents of task (METs), 5 METs and 7 METs. All analyses were conducted at the minute level with heart rate, steps, distance and calories from Apple Watch and Fitbit. These included three different machine learning models: support vector machines, Random Forest and Rotation forest. Results Our dataset included 3656 and 2608min of Apple Watch and Fitbit data, respectively. Rotation Forest models had the highest classification accuracies for Apple Watch at 82.6%, and Random Forest models had the highest accuracy for Fitbit at 90.8%. Classification accuracies for Apple Watch data ranged from 72.6% for sitting to 89.0% for 7 METs. For Fitbit, accuracies varied between 86.2% for sitting to 92.6% for 7 METs. Conclusion This preliminary study demonstrated that data from commercial wearable devices could predict movement types with reasonable accuracy. More research is needed, but these methods are a proof of concept for movement type classification at the population level using commercial wearable device data.