Teleworking as the main risk factor for sedentary behavior: the COVISTRESS international study

Background: If teleworking is already associated with prolonged sedentary behavior, the literature lacks an integrating model with all putative influencing variables, as well as a sufficient number of participants to explore extreme sitting behavior. Therefore, we aimed to investigate the influence of teleworking on sitting time, considering main sociodemographic, occupational and lifestyle factors. Methods: The COVISTRESS study is an international observational study, using self-reported questionnaires (REDCap® software). We used linear and logistic multivariate model to assess the impact of telework on sitting time concomitantly with other sociodemographic, occupational, and lifestyle variables. Results: 6,653 respondents were included. The mean sitting time was 5.6 ± 3.4h/day, with 15.2% of respondents sitting for >8-14h/day and 1.6% sitting >14 hours/day. Workers who teleworked 0%, 1–25%, 26–50%, 51–75%, 76–99%, and 100% respectively sat on average 4.9 ± 3.2h/day, 5.8 ± 3.2h/day, 6.4 ± 3.1h/day, 6.5 ± 3.8h/day, 7.6 ± 3.5h/day, and 9.1 ± 4.3h/day, and the prevalence of workers sitting >14h/day increased from 1.5%, 1.9%, 0.4%, 3.2%, 5.4% to 10.7%. Compared with workers who did not telework, the risk of sitting 8–14 h/day was multiplied by 1.9 (1.2 to 3.0) for people working 1–25% in telework, by 4.7 (2.2 to 9.7) for people working 26–50% in telework, by 4.3 (1.5 to 12.2) for people working 51–75% in telework, and by 9.2 (1.6 to 51.5) for people working >75% in telework; the risk of sitting ≥ 14h/day was multiplied by 5.8 (1.07 to 31.5) for people working 51–75% in telework and dramatically increased by 34.3 (3.71 to 317) for people working >75% in telework in multivariate model. Conclusion: Teleworking is the strongest risk factor for high sitting time, with a marked dose-response relation. While several other sociodemographic, occupational, and lifestyle factors showed association with sitting time in univariate model, many of these relationships weakened or disappeared in the multivariate models, except telework that remained the most consistent determinants of sedentary behavior when concomitantly put together with all other variables. Trial registration: ClinicalTrials.gov NCT04538586