Physical activity pattern before and during the COVID-19 pandemic and association with contextual variables of the pandemic in adults and older adults in southern Brazil

Abstract

Aim

To compare the physical activity pattern before and during the COVID-19 pandemic and verify the association with contextual, behavioral, and health variables related to the pandemic in adults and older adults from southern Brazil.

Subject and methods

This is a panel-type, population-based study in Rio Grande-RS and Criciúma-SC, with 4290 individuals. The physical activity pattern (dependent variable) was measured using the International Physical Activity Questionnaire-IPAQ. In addition, contextual, behavioral, and health aspects related to the pandemic (independent variables) were assessed by questionnaires. Fisher’s exact test was used for bivariate analyses and Poisson regression with robust variance to calculate crude and adjusted prevalence, with their respective 95% confidence intervals.

Results

There was a 72% reduction in commuting physical activity and a 145% increase in physical inactivity when compared before and during the pandemic. Social distancing, excessive search for information about COVID-19, fear of the pandemic, and COVID-19 infection were all factors that contributed to the decline in physical activity during the pandemic. The home office was a protective factor for physical inactivity.

Conclusion

The COVID-19 pandemic has negatively affected the pattern of physical activity in the general population, except for those who switched to working from home.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10389-022-01789-x.

Introduction

In response to the COVID-19 outbreak, governments in several countries implemented non-pharmacological public health interventions, such as containment strategies, to limit the spread of the virus, and prevent the population from being harmed by this virus (Garcia and Duarte 2020). Among these interventions, social distancing was presented as the most effective measure for preventing COVID-19 (Nussbaumer-Streit et al. 2020). For example, in Brazil, social distancing measures were promptly adopted by states and municipalities, such as encouraging remote work and implementing digital services, restricting the use of public transport, interrupting activities in schools and universities, and closing non-essential businesses and services (Garcia and Duarte 2020; Aquino et al. 2020).

Although necessary in times of a pandemic, these measures caused a significant change in daily life and affected the usual processes and routines, including physical activity (Wunsch et al. 2022).

Regular and adequate levels of physical activity are widely known to have beneficial effects on the immune system (Laddu et al. 2021) and to reduce the risks for many comorbidities such as obesity, diabetes mellitus, and coronary heart disease (Cleven et al. 2020), as well as depression (Schuch et al. 2018) and anxiety (Schuch et al. 2020). Furthermore, a recent meta-analysis concluded that physically active people had less chance of hospitalization and mortality from COVID-19 (Rahmati et al. 2022), and the regular practice of physical activity could be an important component for promoting, maintaining, and restoring mental health in the restrictive period (Matias et al. 2020; Marconcin et al. 2022).

Despite these benefits, recent systematic reviews with meta-analyses (Stockwell et al. 2021; Wunsch et al. 2022) found that the social restrictions imposed by the pandemic have reduced the population’s practice of physical activity and that psychological depreciation (Schuch et al. 2020) have also contributed to the decline of this behavior.

Although there is no doubt that the containment measures implemented to reduce the spread of COVID-19 have impacted the practice of physical activity by the population around the world, most studies evaluated the change in physical activity retrospectively (Ammar et al. 2020; Malta et al. 2020; Wilke et al. 2021; Caputo et al. 2021; Füzéki et al. 2021a, b). Prospective comparisons of this behavior “before” and “after” the pandemic are still scarce in the literature. In addition, it is crucial to analyze the effect of the pandemic on the different domains of physical activity and investigate what other contextual variables related to the pandemic (e.g., fear of infection, infodemic; home office) may be associated with the pattern of physical activity.

There are indications in the literature that more time at home during the pandemic (Knell et al. 2020) and physical symptoms of COVID-19 (Smith et al. 2020) negatively influenced physical activity behavior during the pandemic.

Method

Study design and location

To fill the knowledge gaps, the present study was carried out with the aim of comparing the pattern of physical activity before and during the COVID-19 pandemic and verifying the association with contextual, behavioral, and health variables related to the pandemic in adults and elderly adults in southern Brazil.

This is a panel-type study, which uses data from three cross-sectional studies, entitled “Health of the Rio Grande a population” and “Health of the adult and older population of Criciúma”, carried out in 2016 and 2019, respectively, and replicated in 2021, with the “Mental-COVID” survey, in order to compare the pattern of physical activity before and during the COVID-19 pandemic. “The studies “Health of the Rio Grande a population” and “Health of the adult and older population of Criciúma” aimed to evaluate the population’s health, and further methodological details about these studies can be found in other publications (Dumith et al. 2018; Saes-Silva et al. 2021). The Mental-COVID study sought to assess the impact of COVID-19 on the mental health of the adult and older population of these two municipalities in southern Brazil.

The municipality of Rio Grande is located in the southern region of the state of Rio Grande do Sul (RS) and Criciúma is located in the extreme south of the state of Santa Catarina (SC). The municipalities have similar characteristics in terms of population size and Human Development Index (HDI) (IBGE 2010).

Population and sample

The target population consisted of individuals aged 18 years or over, living in the urban areas of Rio Grande and Criciúma. Individuals who were institutionalized or who had physical and/or cognitive incapacity to answer the questionnaire were considered ineligible.

The samples of the three studies were obtained through a process of random sampling methods, which was carried out in two stages, based on data from the last Demographic Census of 2010 (IBGE 2010). The first stage included the census sectors, and the second stage, the households. In 2016, approximately 1600 eligible individuals were found, in the year 2019, approximately 1200 individuals, and in 2021, approximately 3000 individuals. All persons aged 18 years and over, residing in the selected households, were invited to participate in these studies.

The instrument, with an average completion duration of 30 minutes, was applied by previously trained interviewers. The interviews were carried out in front of the homes of those eligible, using a pre-coded and standardized questionnaire, mostly composed of closed questions. In the 2016 and 2019 studies, the interviews were applied through printed questionnaires. In 2021, the instrument was applied using tablets and was built using RedCap® software (acronym for Research Electronic Data Capture), followed by data transfer to the computer. Data collection took place between October 2020 and January 2021, with the interviewers wearing the appropriate Personal Protective Equipment-PPE.

Variables

The dependent variables of this study were: Some leisure-time physical activity (No/Yes); Some commuting physical activity (No/Yes); Leisure and commuting physical activity according to the recommendations of 150 minutes per week (World Health Organization 2010) (No/Yes); and Leisure and commuting physical inactivity (No/Yes). Those who responded that they did not perform physical activity on any days per week were considered inactive. These variables were evaluated using the leisure and commuting section of the International Physical Activity Questionnaire (IPAQ), long version, validated by Matsudo et al. (2012). The practice of physical activity in the week prior to the interview (in the previous seven days) was assessed through questions about weekly frequency and duration for walking (also for cycling in the commuting section) and the practice of moderate and vigorous physical activities.

The independent variables were: adherence to social distancing (No/Yes), search for information about the pandemic several times a day (No/Yes); started working remotely during the pandemic (No/Yes); has had symptoms of COVID-19 (No/Yes), has been infected with COVID-19 (No/Yes); and has had contact with someone infected with COVID-19 (No/Yes). These variables were obtained through self-report. In addition, fear of the pandemic was evaluated through the Fear of COVID-19 (FCV-19S) scale, adapted and validated for the Brazilian adult population by Medeiros et al. (2021).

Based on this scale, a score in quintiles was generated for this study, with the last quintile being isolated and considered the group with the greatest fear of being infected by the disease. Subsequently, the groups were dichotomized into “no” (from the first to the fourth quintile, group with less fear of COVID-19) and “yes” (last quintile, group with greater fear of COVID-19). Participants were also asked about the following symptoms of COVID-19: feverish sensation or fever ≥37.8°C, cough without phlegm, difficulty breathing, sore throat, muscle pain or more tiredness than normal, diarrhea, decreased taste, decreased sense of smell, tremors or chills, and headache. This variable was dichotomized into “no” (no symptoms) and “yes” (some symptoms).

To control for possible confounding factors, the following covariates were included in the analysis: sex (male/female), age group (collected in complete years and categorized as 18 to 39 years/40 to 59 years/60 years or more), schooling (elementary/secondary/higher), socioeconomic status (lowest – lowest socioeconomic level/intermediate/highest – highest socioeconomic level), stress (lowest stress level/intermediate/highest – highest stress level), and regular or poor perception of health (no/yes). Stress was assessed using the 14-item Perceived Stress Questionnaire (Siqueira Reis et al. 2010), with the score divided into tertiles.

In addition to the variables mentioned, the following were collected for exploration purposes: skin color, marital status, smoking, excessive alcohol consumption, BMI, health insurance, regular or poor sleep quality, depressive symptoms (Santos et al. 2013), feelings of sadness (Andrews and Withey 1976), arterial hypertension, diabetes mellitus, heart disease, and chronic back pain. These variables are detailed in Supplementary Table 1.

Statistical analysis

Univariate analysis was performed using absolute and relative frequencies to describe the characteristics of the sample. Bivariate analysis was performed to calculate the prevalence of the outcome according to the independent variable, using the Fisher’s exact test.

Multivariate analysis was performed using Poisson regression, which was used to calculate crude and adjusted prevalence ratios (PR) and their corresponding 95% confidence intervals (95%CI). The Wald ratio test for heterogeneity (dichotomous or nominal exposures) was used. The level of significance was set at 5% for two-tailed tests.

It is noteworthy that interaction was tested with the years of research, with statistical significance (p value <0.10) between all the aforementioned covariates.

All statistical procedures were performed using the Software for Statistics and Data Science (STATA), version 16.1 using the prefix svy, which considers the complexity of the sampling process and the effect of the study design. The figure was constructed using the Excel® program.

Results

A total of 2170 subjects participated in the study in the year 2021, with a response rate of 72%. Of these, the majority were female (59.7%), 31.2% were 60 years of age or older, and more than 70% of the population did not have higher education. Almost a quarter of individuals reported seeking information about the pandemic several times a day, 7.7% reported having switched to working remotely during the pandemic, more than 20% reported having had symptoms of COVID-19, and less than 10% had been infected by the disease (Table 1). The distribution of the sample in relation to demographic, socioeconomic, behavioral, and health variables before and during the pandemic was similar (see Supplementary Table 2).

Table 1.

Demographic, socioeconomic, health, and pandemic context characteristics of the sample (2021 assessment) (N = 2,170)

Variables	N	%
Sex
Male	875	40.3
Female	1295	59.7
Age group
18–39	729	33.6
40–59	763	35.2
60 or +	678	31.2
Schooling
Elementary	921	42.5
Secondary	692	31.9
Higher	555	25.6
Asset index (income tertiles)
1 (lowest)	719	34.7
2 (intermediate)	673	32.5
3 (highest)	680	32.8
Level of stress (tertiles)
1 (lowest)	760	35.3
2 (intermediate)	718	33.3
3 (highest)	676	31.4
Regular or poor perception of health
No	1624	74.9
Yes	545	25.1
Social distancing
No	1768	81.5
Yes	402	18.5
Infodemic
No	1692	78.0
Yes	478	22.0
Fear of the pandemic
No	1740	80.9
Yes	412	19.1
Switched to working remotely during the pandemic
No	2002	92.3
Yes	168	7.7
Symptoms of COVID-19
No	1667	76.9
Yes	500	23.1
Infection by COVID-19
No	2023	93.2
Yes	147	6.8
Had contact with infection
No	1617	74.5
Yes	553	25.5

N, Absolute frequency; %, Relative frequency

Figure 1 shows the comparison of the pattern of physical activity before and during the pandemic. It was found that commuting physical activity was 3.5 times higher before the pandemic and during the pandemic it decreased to 18.7%, representing a drop of 72.0%. Before the pandemic, almost half of the sample reached the recommendations of leisure and commuting physical activity for health (>=150min/week) and during the pandemic, only a quarter reported meeting the recommendations (>=150min/week). Approximately 24.4% of subjects reported being physically inactive before the pandemic, while during the pandemic almost 60% were classified as inactive. A change in the prevalence of leisure-time physical activity was also observed when analyzing before and during the pandemic; however, the percentage delta variation was smaller when compared to the other outcomes.

Fig. 1.

Comparison of the physical activity pattern before and during the COVID-19 pandemic in adults and older adults in the urban areas of Rio Grande, RS (2016/2021) and Criciúma, SC (2019/2021) (N = 4,290). LPA, Leisure Physical Activity; CPA, Commuting Physical Activity; LCPA, Leisure and Commuting Physical Activity; ∆, Delta Percentage= percentage difference in prevalence from before to during the pandemic

Regarding the changes from before to during the pandemic in associated factors, the reduction in leisure-time physical activity was greater for males, individuals aged 60 years or older, with less schooling and with a lower level of stress (Supplementary Table 3). The greatest reduction in commuting activity was observed for male individuals, with lower economic status, who perceived their health as fair or poor and had a low level of stress (Supplementary Table 4). Older, less educated, and low-income people who perceived their health as fair or poor reported greater reductions in leisure-time physical activity and commuting in line with health recommendations (>=150min/week) (Supplementary Table 5). Considering physical inactivity, the greatest increases were observed in individuals with a low level of stress, without depressive symptoms, and without feelings of sadness (Supplementary Table 6).

With respect to the association of demographic, socioeconomic, and health variables with pandemic variables, it can be observed that women adhered more to social distancing, had more infodemic behavior, and greater fear of the pandemic when compared to men. Individuals with higher levels of schooling adhered less to social distancing, had less infodemic behavior, and more COVID-19 infections, more contact with someone infected with the disease, and more remote work than individuals with lower levels of schooling. Those with the highest levels of stress had less infodemic behavior and greater fear of the pandemic, more symptoms of COVID-19, and more contact with someone infected with the disease than those with the lowest level of stress (Supplementary Table 7).

The prevalence of leisure-time physical activity during the pandemic was 17.0% (95%CI 13.6; 21.0) for those who adhered to social distancing versus 30.8% (95%CI 28.7; 33.0) for those who did not adhere. For commuting physical activity during the pandemic, the prevalence for those who sought information about the pandemic several times a day was 12.2% (95%CI 10.0; 15.5) versus 20.5% (95%CI 18 .6; 22.5) for those who did not seek information about the pandemic as often. Considering leisure and commuting physical activities according to health recommendations (>=150min/week), it was found that the prevalence for those who complied with social distancing was 14.8% (95%CI 11.6; 18.6) versus 26.9% (95% CI 24.9; 29.0) for those who did not comply. The prevalence of physical inactivity during the pandemic was 41.0% (95%CI 33.7; 48.7) for those who started working remotely during the pandemic versus 61.4% (95%CI 59.2; 63.5) for those who continued working as before (Table 2).

Table 2.

Adjusted analysis between pandemic variables and physical activity measures in adults and older adults in the urban areas of Rio Grande, RS (2021) and Criciúma, SC (2021) (N = 2170)

Variables	LPA		CPA		LCPA		Physical inactivity
	%	PR (95%CI)*	%	PR (95%CI)**	%	PR (95%CI)***	%	PR (95%CI)***
Social distancing
No	30.8	1.00	18.8	1.00	26.9	1.00	57.3	1.00
Yes	17.0	0.78 (0.61; 1.00)	18.0	1.14 (0.87; 1.48)	14.8	0.77 (0.59; 1.01)	70.7	1.06 (0.96; 1.17)
Infodemic
No	28.9	1.00	20.5	1.00	25.0	1.00	57.8	1.00
Yes	25.9	0.93 (0.76; 1.15)	12.2	0.74 (0.55; 0.98)	23.4	0.99 (0.79; 1.25)	66.9	1.11 (1.01; 1.22)
Fear of the pandemic
No	29.4	1.00	18.3	1.00	26.0	1.00	58.7	1.00
Yes	22.6	0.93 (0.77; 1.12)	18.3	0.81 (0.63; 1.03)	17.3	0.78 (0.64; 0.96)	66.1	1.10 (1.01; 1.20)
Switched to working remotely during the pandemic
No	26.4	1.00	18.0	1.00	23.1	1.00	61.4	1.00
Yes	50.6	1.14 (0.95; 1.37)	26.8	1.37 (1.02; 1.84)	44.0	1.16 (0.92; 1.47)	41.0	0.87 (0.72; 1.04)
Symptoms of COVID-19
No	28.5	1.00	16.5	1.00	24.8	1.00	60.9	1.00
Yes	27.5	0.97 (0.80; 1.18)	25.7	1.55 (1.25; 1.92)	24.4	1.06 (0.85; 1.33)	56.2	0.89 (0.80; 0.99)
Infection by COVID-19
No	28.0	1.00	18.8	1.00	24.6	1.00	59.8	1.00
Yes	32.0	0.88 (0.63; 1.24)	16.3	0.42 (0.25; 0.69)	26.0	0.75 (0.49; 1.15)	60.3	1.34 (1.14; 1.58)
Had contact with infection
No	26.8	1.00	17.6	1.00	23.6	1.00	61.7	1.00
Yes	32.6	1.00 (0.84; 1.19)	21.7	1.29 (1.01; 1.65)	27.9	1.01 (0.84; 1.20)	54.3	0.92 (0.83; 1.03)

Bold entry indicate statistically significant associations

%, Prevalence, PR, Prevalence Ratio

*Adjusted PRs were calculated using Poisson regression, with robust adjustment for variance. The adjustment variables were: sex, age, education, asset index, stress, regular or poor perception of health, and commuting physical activity

**Adjusted PRs were calculated using Poisson regression, with robust adjustment for variance. The adjustment variables were: sex, age, education, asset index, stress, regular or poor perception of health and leisure-time physical activity

***Adjusted PRs were calculated using Poisson regression, with robust adjustment for variance. Adjustment variables were: sex, age, education, asset index, stress and regular or poor perception of health

During adjusting for possible confounding factors, it was observed that the probability of practicing physical activity during leisure time during the pandemic was lower in those who adhered to social distancing (PR = 0.78; 95%CI 0.61;1.00) when compared to those who did not. Regarding physical activity while commuting, the highest probability of practicing physical activity while commuting during the pandemic period was among those who started working remotely during the pandemic (PR = 1.37; 95%CI 1.02;1.84), who had symptoms of COVID-19 (PR = 1.55; 95%CI 1.25;1.92), and who had contact with someone infected with the disease (PR = 1.29; 95%CI 1.01;1.65) when compared to their peers. On the other hand, those who reported infodemic behavior and COVID-19 infection were 26% (95%CI 2 to 45%) and 58% (95%CI 31 to 75%), respectively, less likely to be active in commuting during the pandemic when compared to those who did not report infodemic behavior and who had the COVID-19 infection. As for the practice of physical activity during leisure time and commuting, according to health recommendations (>=150min/week), it was observed that the probability was 22% (95%CI 4 to 36%) lower in those who reported greater fear of the pandemic compared to those who were not so afraid. The highest probability of physical inactivity during the pandemic was for those who sought information about the pandemic several times a day (PR = 1.11; 95%CI 1.01;1.22), were more afraid of the pandemic (PR = 1.10; 95%CI 1.01;1.20), and had the COVID-19 infection (PR = 1.34; 95%CI 1.14;1.58) when compared to their peers. On the other hand, those who reported having symptoms of COVID-19 were 11% (95%CI 1 to 20%) less likely to be inactive when compared to those who had no symptoms of the disease (Table 2).

Discussion

This study compared the pattern of physical activity before and during the COVID-19 pandemic in adults and older adults and analyzed its association with contextual, behavioral, and health variables related to the pandemic in two municipalities in southern Brazil. The results showed that the pattern of physical activity underwent an unfavorable change during the pandemic period compared to the period before the restrictions, as found in other investigations that identified the effect of social isolation induced by the pandemic on physical activity behavior worldwide (Stockwell et al. 2021; Wunsch et al. 2022). These results were already expected since the “stay at home order,” together with working from home, and the ban on organized sports groups (Wunsch et al. 2022), probably influenced the reduction in this behavior during the pandemic.

It was found in our study that approximately one in four individuals were completely physically inactive before the pandemic, and during the pandemic period more than half became inactive. In the study by Silva et al. (2021), with 39,693 Brazilian adults, a 26% increase in physical inactivity was found during the pandemic. These data point to a worrying scenario, as in 2017, the World Health Organization developed a global plan called the “Global Action Plan on Physical Activity 2018–2030” (World Health Organization 2018) to prevent and control physical inactivity and promote physical activity. One of the goals of this plan was to reduce physical inactivity by 10% by 2025 and 15% by 2030 (World Health Organization 2010). However, instead of reducing the prevalence of physical inactivity, which before the pandemic was already high in adults (27.5%) (Guthold et al. 2018), the social distancing imposed by the pandemic influenced the even greater increase in its prevalence worldwide (Stockwell et al. 2021; Wunsch et al. 2022). Added to this is the fact that patients infected by COVID-19 who were physically inactive before infection had greater chances of hospitalization (OR = 2.26; 95%CI 1.81;2.83), admission to the intensive care unit (OR = 1.73; 95%CI 1.18;2.55), and death (OR = 2.49; 95%CI 1.33;4.67) as a result of this disease (Sallis et al. 2021).

In the current study it was observed that the greatest reduction in physical activity measures occurred in the commuting domain and not, as expected, in the leisure domain, which was also reported in studies carried out with the population of Italy (Füzéki et al. 2021a) and Germany Füzéki et al. 2021b), and can be explained by the closing of schools, universities, businesses, and non-essential commerce, as well as the worldwide adherence to the home office. Of all the physical activity measures evaluated in this study, leisure-time physical activity suffered the smallest decline during the pandemic when compared to the other measures, which could be attributed to the fact that individuals in population groups that were more active in this domain before the restrictions (such as individuals with higher education and higher income) continued to be more likely to practice leisure-time physical activities even with the restrictions imposed.

As verified in our findings, the virus infection that causes COVID-19 was not the only factor responsible for negatively influencing the behavior of physical activity, but the pandemic context, such as social distancing, the excessive search for information about COVID-19, and the greater degree of fear of the pandemic were also responsible for the decline in this behavior in the analyzed period. Social restrictions, as expected, reduced the probability of practicing physical leisure activities in the investigated population. This was also found in the study by Knell et al. (2020), in a sample of 1809 adults residing in the United States, indicating that those who spent more time at home were 1.06 times (95%CI 1.02; 1.09) more likely to report decreased physical activity. This may be linked to restrictions in access to parks and places conducive to sports and outdoor physical activities during the first waves of the pandemic (Wunsch et al. 2022).

In the current study, the excessive search for information about COVID-19 and a greater degree of fear of the pandemic were risk factors for total physical inactivity during the pandemic. These findings are in line with other studies that found that infodemic behavior and pandemic fear are associated with unfavorable health outcomes, such as anxiety, depression, stress, panic, fear, and tiredness (Rahman et al. 2020; Rocha et al. 2021) and increased risk behaviors (alcohol and smoking) (Rahman et al. 2020). In view of these convergences, it is suggested that the negative influence of the excessive search for information about COVID-19 and the greater fear of the pandemic were extended to behaviors related to physical activity in our study, as they may have been considered important barriers to the practice of physical activities during the pandemic period by the investigated population.

As found in our results, the study by Smith et al. (2020), carried out with 911 adults from England, Northern Ireland, Scotland, and Wales, to investigate the levels and correlates of physical activity during social distancing in a sample of UK adults, found that those who reported having physical symptoms of COVID-19 were less likely (OR = 0.31; 95%CI 0.21; 0.46) to be physically active during the pandemic. The direction of these associations was predictable, as this disease often causes physical complications such as fatigue, myalgia, arthralgia, reduced physical capacity, and declines in physical function, usual care, and daily activities (Shanbehzadeh et al. 2021), which can easily lead to greater unwillingness to practice physical activities.

While containment strategies may have introduced barriers to physical activity for some people, the requirement to work from home may actually have provided opportunities for physical activity for others, as seen in our study. These findings are consistent with another Brazilian population-based study carried out in two municipalities in Minas Gerais-Brazil, with a sample of 1750 individuals aged 18 years and over, which found that those who worked from home had reduced chances (OR = 0.45; 95%CI 0.24;0.85) and (OR = 0.51; 95%CI 0.29;0.88) of being physically inactive during leisure time, from March to August 2020 and October to December 2020, respectively (Moura et al. 2021).

Finally, this study must be interpreted considering its limitations and strengths. First, the cross-sectional design of the study does not allow a temporal relationship to be established, and thus may be subject to reverse causality bias. For example, it is not known whether those individuals who were infected with COVID-19 did not already have some debilitating physical condition that limited them in practicing physical activity before the outbreak of the pandemic, and the COVID-19 infection further aggravated this framework. Second, self-report of positive COVID-19 testing and physical activity measures may be subject to recall bias, as well as underdiagnosis, since the individual may have been infected but not aware of this.

As strengths, it should be noted that the present study was population-based and carried out in person, in the households of the interviewees. This is also a differential when compared to other studies on the subject that used online platforms, as the data collection method employed enabled the inclusion of individuals without internet access. In addition, it should be noted that our study differs from others in that we compared the pattern of physical activity before and during the pandemic with the same population and not retrospectively. Finally, we mention the fact that we evaluated different measures of physical activity (leisure physical activity, commuting physical activity, physical activity according to health recommendations (>=150min/week), and total physical inactivity) and that we identified the repercussions of the pandemic for each of these outcomes.

Information on the influence of the COVID-19 pandemic on physical activity pattern, as well as understanding how different contextual, behavioral, and health factors related to the pandemic influenced different measures of physical activity are valuable and can contribute to planning and targeting specific actions and strategies to improve each measure of physical activity behavior evaluated herein. We emphasize the importance of suggesting longitudinal studies to monitor this population, as well as interventions that promote the practice of physical activity, in order to mitigate the impacts of the pandemic on the present and future health of these individuals.

Conclusion

We concluded that the COVID-19 pandemic negatively affected the pattern of physical activity of the population studied. Commuting physical activity was the measure that suffered the biggest decline during the pandemic and not leisure activity as expected. Total physical inactivity almost tripled in the comparison of before and during the pandemic. Infection with the virus that causes COVID-19 was not the only factor that contributed to negatively altering physical activity behavior during the pandemic, as social distancing, fear of the pandemic, and the excessive search for information about COVID-19 were also factors responsible for this decline. Contrary to what was expected, work from home might be a protective factor for physical inactivity.

Authors’ contribution

VSFV participated in the design of the manuscript, performed the analyses, participated in the writing of the manuscript. TSM, EGA, and YPV participated in the writing of the manuscript. TSM, FOM, and AAS critically reviewed the manuscript. SCD supervised the analyses, revised the final version of the article, and coordinated the study. All authors read and approved the final manuscript.

Funding

This study received financial support from the Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul - FAPERGS, process number 20/2551-0000277-2.

Data availability

The data that support the findings of this study are available from the corresponding author upon request.

Code availability

Not applicable.

Declarations

Ethics approval and consent to participate

The studies are approved by the appropriate Research Ethics Committees (CEP) under the following opinion numbers: (no. ^FURG 20/2016 CAAE: 52939016.0.0000.5324; no. ^UNESC 3.084.521; no. ^FURG 4.055.737). All ethical principles established by the National Health Council in Resolution 466/12 were respected. Those who agreed to participate in the study informed this decision during reading and signing the Free and Informed Consent Form.

Consent to publication

Not applicable.

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.