Introduction

Coronavirus disease 2019 (COVID-19), an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2),¹ has widely spread over the world since the beginning of 2020.^1,2 In the absence of an effective vaccine or treatment, classical public health measures - including social distancing, isolation, quarantine and closure of services - were implemented by most governments to curb the pandemic.^1-3 These measures affected individuals’ lifestyles and, possibly, their physical activity (PA) levels.^1-5 In fact, Ammar A et al³ concluded that the COVID-19 home confinement has had a negative effect on all PA levels (vigorous, moderate and overall), being associated with an increase in the daily sitting time. Furthermore, Fitbit, Inc.,⁴ an American company that develops wearable devices that track individuals’ PA, has recently shared data that reveals a substantial reduction in average daily step counts during one week in the quarantine period, as compared with the same period last year, especially in European countries. Nevertheless, to the best of our knowledge, no studies have explored the effect of the COVID-19 pandemic in the compliance with the World Health Organization (WHO) aerobic PA recommendations.⁶ The effect of COVID-19 in the different dimensions of PA has also been insufficiently explored. Such assessment, however, is particularly important for devising preventive measures in case isolation measures are required to be re-implemented. In this study, we aimed to assess changes in Portuguese PA levels during COVID-19 lockdown, including changes in aerobic PA, daily sitting time, and weekly walking time.

Methods

Results

We obtained a total of 1321 responses, of whom 40 were excluded (28 for lack of consent, and 12 for being from non-Portuguese citizens). Therefore, a total of 1281 participants were included in this study. The characteristics of participants are presented in Table 1. Of the 1281 participants, 466 (36.4%) were males. Most participants were employed (n=977; 76.3%), of whom 927 (94.8%) maintained their professional activity, mostly by telework (n=592; 63.9%).

Most participants (n=776; 60.6%) underwent over 21 days of lockdown. Half of the participants (n=641; 50.0%) met the aerobic levels of PA recommended by the WHO before the lockdown. However, such frequency decreased significatively during the lockdown, where only 38.8% (n=497) met those recommendations (p<0.001). The multivariable regression models (Table 2) identified that male participants had higher odds of ceasing to comply with aerobic PA levels during lockdown (adjusted OR=1.4; 95%CI=1.0-1.9; p=0.023). An opposite trend - with increased odds of complying with aerobic PA levels - was observed for participants aged under 20 years old (adjusted OR=2.3; 95%CI=1.2-4.3; p=0.008).

We found a significant increase of the daily sitting time during the lockdown period (p<0.001). Multivariable regression models (Table 3) have identified that male participants had non-significant lower chances to either decrease (adjusted OR=0.7; 95%CI=0.5-1.1; p=0.117) or increase (adjusted OR=0.8; 95%CI=0.6-1.0; p=0.078) daily sitting time.

Lockdown associated with an average reduction of 24.8 minutes (25.5%) in walking time (average walking time before and after the lockdown were of 97.2 and 72.4 minutes, respectively; p<0.001). Except for the number of days with PA before home confinement, no variables were identified to be significantly associated with walking time changes in multivariable models (Table 4).

Discussion

This study presents data from an online survey comparing PA before and during home lockdown in the context of COVID-19. We observed that lockdown was associated with significant decreases in aerobic PA levels and total weekly walking time, as well as with increases in daily sitting time. Despite the recommendation that home lockdown should not hinder people from being physically active,⁶ our results suggest that there was a decline in PA levels among the Portuguese population during that period. This finding is in accordance with other studies suggesting a negative effect of lockdown in sports and PA engagement.^3,4,11 For example, Ammar et al³ reported a 24% decrease in the average number of days per week in which participants practiced PA, a 34% decrease in the daily number of walking minutes, and a three-hour average increase (from 5 to 8 hours) in the daily sitting time. Furthermore, the devices company Fitbit, Inc.⁴ has recently shared PA data that pointed to a substantial reduction in average daily step counts during the lockdown period, as compared to the same period last year (ranging from 7% to 38% across different countries). The decrease in PA levels may go beyond the lockdown period itself - Goethals et al¹¹ observed that even before lockdown was enforced, the COVID-19 epidemic had already negatively affected the number of seniors attending group PA programs in two different French study territories, possibly due to the fear of potentially contacting infected people.

We were not able to identify a set of characteristics associated with greater propensity to change PA levels during lockdown. This could be explained by the multiple ways by which COVID-19 pandemic may affect PA (thus affecting individuals with different characteristics/profiles), including (i) the fact that the closure of nurseries and schools forced many of the parents to stay at home to take care of their children; (ii) the loss of work activity and subsequent loss of income (loss of work was observed in one-quarter of respondents in our sample); and (iii) the greater social isolation allied to the closing of the gyms^2,11 it was in the latter that almost half respondents performed their PA).

There are several limitations of this study that warrant discussion. Firstly, this study included a convenience sample, in which younger and more-schooled individuals were overrepresented and its timeframe reflected a particular week during lockdown when more severe restrictions were imposed by the Portuguese government. Nevertheless, our data on the daily sitting time and on the setting where individuals practiced PA before lockdown are close to those published in 2017 in the Sport and Physical Activity Report⁷ (the other outcomes we assessed were not evaluated in this reports), which assessed a representative sample of the Portuguese population and, to the best of our knowledge, is the study which assessed the largest number of Portuguese individuals about the PA levels. The authors also acknowledge that the reported levels of parameters such as sitting time reflect the perception of the individuals that answered to the survey and therefore may somehow differ from the effective times. Another important limitation concerns the possibility of information bias, particularly the possibility (i) of overestimation or underestimation of PA levels before the lockdown (recall bias), and (ii) of providing incorrect answers subsequent to misunderstanding of questions (e.g., in the questions and examples provided for the respondent to distinguish between moderate vs vigorous PA). Finally, we collected data from only one country, which had already low PA levels,⁷ potentially limiting the generalization of our results. Nevertheless, this study has also several strong points, including its large sample size, the use of validated definitions and the simultaneous evaluation of multiple outcomes. Another strong point concerns its novelty, being one of the first and largest studies assessing the impact of COVID-19 pandemic on PA. In addition, the authors built multinomial logistic regression models to try to identify characteristics associated with greater chances of changing PA levels during lockdown.

Conclusion

In this study, we observed that home lockdown during COVID-19 pandemic was associated with a significant reduction in PA levels, including a decrease in total weekly walking time, and an increase in daily sitting time. These results are particularly relevant considering the negative impact of physical inactivity on cardiovascular health.² Healthcare providers must be aware of available home fitness options, to increase the chances that patients maintain their PA levels in case of new enforcement of lockdown measures in the wake of an eventual new wave of COVID-19.