Changes in Physical Inactivity Among US Adults Overall and by Sociodemographic Characteristics, Behavioral Risk Factor Surveillance System, 2020 Versus 2018

Miriam E. Van Dyke, PhD, MPH¹; Tiffany J. Chen, MSPH^1,2; Jasmine Y. Nakayama, PhD, RN¹; Latetia V. Moore, PhD, MSPH³; Geoffrey P. Whitfield, PhD, MEd¹ (View author affiliations)

Suggested citation for this article: Van Dyke ME, Chen TJ, Nakayama JY, Moore LV, Whitfield GP. Changes in Physical Inactivity Among US Adults Overall and by Sociodemographic Characteristics, Behavioral Risk Factor Surveillance System, 2020 Versus 2018. Prev Chronic Dis 2023;20:230012. DOI: http://dx.doi.org/10.5888/pcd20.230012.

PEER REVIEWED

Abstract

The COVID-19 pandemic may have disrupted people’s work–life patterns and access to places to be physically active. Behavioral Risk Factor Surveillance System data were analyzed to assess changes in self-reported leisure-time physical inactivity. The results showed that prevalence of inactivity among US adults decreased 0.7 percentage points (95% CI: −1.2 to −0.3), from 24.5% in 2018 to 23.8% in 2020, and the greatest decreases were observed among rural-dwelling women, rural-dwelling men, and non-Hispanic White women. These findings highlight a need to understand and address factors that lead to differential changes in leisure-time physical inactivity across subpopulations during public health emergencies.

Top

Objective

Physical activity has many health benefits, including reducing anxiety, improving sleep, and lowering blood pressure, as well as lowering the risk of type 2 diabetes, heart disease, and some cancers (1). Physical activity also helps prevent severe outcomes from COVID-19 (2), which the World Health Organization declared a pandemic in March 2020.

Early in the pandemic, uneven access to safe places for physical activity and shifting work–life demands may have exacerbated existing disparities in physical activity levels. These changes affected some people’s ability to be active more than others (3). For example, people who could access safe, walkable neighborhoods or who worked at home may have increased their physical activity. Understanding prevalence patterns of people who are physically inactive (or who participate in no leisure-time physical activity) before and during the pandemic can provide insight into who initiates any physical activity during large public health emergencies.

The Behavioral Risk Factor Surveillance System (BRFSS) is the only national public health surveillance system that had consistent measures of physical inactivity before and during the pandemic. This study examined changes in prevalence of physical inactivity between 2016, 2018, and 2020, with a focus on changes during 2020 relative to 2018, in the US overall and across sociodemographic groups.

Top

Methods

Data were from the 2016, 2018, and 2020 BRFSS, a national state-based system of health-related telephone surveys of the civilian, noninstitutionalized US population aged 18 years or older (4). BRFSS data for 50 US states, the District of Columbia, Guam, and Puerto Rico were analyzed. Alternating years of data were analyzed because of annual fluctuation in physical inactivity prevalence, possibly attributable to differences in question order (4,5). Participants were asked, “During the past month, other than your regular job, did you participate in any physical activities or exercises such as running, calisthenics, golf, gardening, or walking for exercise?” Participants who reported no were classified as physically inactive in leisure time (hereinafter, “inactive”), and participants with missing data or who reported “don’t know/not sure” or “refused” (0.2% each year) were excluded. A total of 484,244; 436,741; and 401,276 people were included in analyses for 2016, 2018, and 2020, respectively. The proportion of BRFSS respondents per month ranged from 6.4% to 10.9% (including during 2020).

Participants self-reported their sex, age, race and ethnicity, education, and income. Urban or rural designation of participant residence was based on the 2013 National Center for Health Statistics urban–rural classification scheme for counties (6).

Prevalence differences with 95% CIs were calculated, comparing inactivity between 2020 and 2018 across sociodemographic characteristics overall and stratified by sex. To determine if there were also changes between 2018 and 2016, prevalence differences between these years were calculated. Prevalence differences with 95% CIs that excluded zero were considered statistically significant. Analyses accounted for complex survey design and nonresponse and were conducted using SAS version 9.4 (SAS Institute, Inc) and SUDAAN version 11.0.1 (RTI International). Institutional review board approval was not required because no personal identifiers were included in the data file. The study was conducted according to applicable federal law and Centers for Disease Control and Prevention policy.

Top

Results

The prevalence of physical inactivity was 24.4%, 24.5%, and 23.8% in 2016, 2018, and 2020, respectively (Figure, Table 1). Inactivity decreased overall by 0.7 percentage points (PP) in 2020 compared with 2018 (95% CI, −1.2 to −0.3). Significant decreases were observed among people aged 45 to 64 years (−1.5 PP [95% CI, −2.3 to −0.7]) and 65 years or older (−1.3 PP [95% CI, −2.1 to −0.4]); women (−1.2 PP [95% CI, −1.8 to −0.6]); people who were non-Hispanic White (−1.5 PP [95% CI, −2.0 to −1.1]); and people living in rural counties (−2.6 PP [95% CI, –3.8 to −1.4]) and, to a lesser extent, urban counties (−0.6 PP [95% CI, –1.0 to −0.1]). Observed changes in 2020 for other groups, including racial and ethnic minority groups, were not statistically significant. No significant decreases observed from 2018 to 2020 were also observed from 2016 to 2018 (Table 1).

Figure.
Prevalence of leisure-time physical inactivity, by sociodemographic characteristics, among US adults aged ≥18 years, Behavioral Risk Factor Surveillance System, 2018 and 2020. Prevalence estimates were weighted to account for complex survey design and nonresponse. Bolded groups indicate that changes in prevalence during 2020 compared with 2018 were statistically significant. [A text description of this figure is available.]

Inactivity significantly declined among multiple subgroups of women in 2020 versus 2018 (Table 2). The largest decreases occurred among rural-dwelling women (−3.1 PP [95% CI, –4.7 to −1.5]), followed by White women (−2.1 PP [95% CI, –2.7 to –1.5]). Significant decreases were also observed among women aged 45 to 64 years (−1.6 PP [95% CI, –2.7 to −0.5]) and aged 65 years or older (−1.6 PP [95% CI, –2.7 to −0.4]), women who graduated from high school (−1.9 PP [95% CI, –3.2 to −0.7]) or from college or technical school (−0.9 PP [95% CI, –1.6 to −0.2]), women making $35,000 to less than $50,000 per year (−1.8 PP [95% CI, –3.6 to −0.03]), and women living in urban counties (−1.1 PP [95% CI, –1.7 to −0.4]). In 2020, inactivity significantly declined among 3 subgroups of men: rural-dwelling men (−2.1 PP [95% CI, –3.8 to −0.3]), men aged 45–64 years (−1.4 PP [95% CI, –2.5 to −0.3]), and White men (−0.9 PP [95% CI, –1.5 to −0.3). No significant sex-specific decreases observed from 2018 to 2020 were also observed from 2016 to 2018.

Top

Discussion

Leisure-time physical inactivity modestly decreased overall and among specific US subpopulations in 2020 compared with 2018. Overall, the observed 0.7 percentage point decrease in physical inactivity suggests that nearly 1.8 million fewer US adults in 2020 were physically inactive during leisure time, and avoiding inactivity is a key recommendation for adults in the second edition of the Physical Activity Guidelines for Americans (1).

Prior surveys (3,7–9) on physical activity changes during the pandemic produced discrepant findings, which may be due to varied methodologies (eg, device- vs questionnaire-based assessment) or a focus on different domains of activity (7). The current study describes changes in the prevalence of people participating in no leisure-time physical activity. While this study does not measure changes in quantified levels of activity, initiating any activity is an important first step given relatively stagnant levels of physical inactivity before the pandemic (5).

Studies have documented persistent disparities in physical activity across racial and ethnic and socioeconomic groups (3,9), with potential widening of disparities during the pandemic (8). The 2020 decrease in leisure-time physical inactivity (or the increase in initiation of any leisure-time physical activity) we found among some populations, but not others, may result from different physical activity opportunities and access to safe spaces (9,10) across subpopulations during the pandemic. Additional research of structural determinants, such as occupational requirements (eg, remote work) affecting availability for leisure-time physical activity, may also help to explain differential decreases. Less traffic, which is more commonly reported as a barrier to walking among rural compared with urban residents (11), or access to new or changed spaces (12), may have also helped some populations be more active. Additionally, some groups experiencing disproportionate health impacts early in the pandemic (eg, people from racial and ethnic minority groups) may have had concerns over COVID-19 exposure during some physical activities (3,13), which may partially explain the decrease in inactivity among people who are White but not people from racial and ethnic minority groups.

This study has limitations. First, data on physical activity were limited to nonoccupational, leisure-time activity. Second, data were self-reported and may be subject to social desirability and other recall biases. Third, patterns of physical inactivity may have differed across periods of 2020. Fourth, this study did not identify causal factors (eg, policies) related to the pandemic that influenced patterns of physical activity. Finally, analyses did not control for multiple comparisons, and sample sizes for some groups limited the ability to statistically detect changes.

This study highlights a need to understand and address factors influencing differential changes in leisure-time physical inactivity across subpopulations during public health emergencies.

Top

Acknowledgments

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. No external funding was used for this research. No copyrighted material was used in this manuscript.

Top

Author Information

Corresponding Author: Miriam E. Van Dyke, PhD, MPH, Physical Activity and Health Branch, Division of Nutrition, Physical Activity, and Obesity, Centers for Disease Control and Prevention, 4770 Buford Highway, Bldg 107, Atlanta, GA 30341 (mpy4@cdc.gov).

Author Affiliations: ¹Physical Activity and Health Branch, Division of Nutrition, Physical Activity, and Obesity, Centers for Disease Control and Prevention, Atlanta, Georgia. ²McKing Consulting Corporation, Atlanta, Georgia. ³Office of Science, Division of Nutrition, Physical Activity, and Obesity, Centers for Disease Control and Prevention, Atlanta, Georgia.

Top

References

1. U.S. Department of Health and Human Services. Physical activity guidelines for Americans, 2nd edition 2018. Accessed October 23, 2022. https://health.gov/sites/default/files/2019-09/Physical_Activity_Guidelines_2nd_edition.pdf
2. Hill AL, Whitfield G, Morford M, Okasako-Schmucker D, So CN, Wassef M, et al. Brief summary of findings on the association between physical inactivity and severe COVID-19 outcomes. Accessed May 25, 2023. https://www.cdc.gov/coronavirus/2019-ncov/downloads/clinical-care/E-Physical-Inactivity-Review.pdf
3. Watson KB, Whitfield GP, Huntzicker G, Omura JD, Ussery E, Chen TJ, et al. Cross-sectional study of changes in physical activity behavior during the COVID-19 pandemic among US adults. Int J Behav Nutr Phys Act 2021;18(1):91. PubMed doi:10.1186/s12966-021-01161-4
4. Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System. Accessed October 18, 2022. https://www.cdc.gov/brfss/
5. Centers for Disease Control and Prevention. Nutrition, physical activity, and obesity: data, trends, and maps. Accessed May 25, 2023. https://www.cdc.gov/nccdphp/dnpao/data-trends-maps/index.html
6. Centers for Disease Control and Prevention, National Center for Health Statistics. 2013 Urban–rural classification scheme for counties. Accessed May 25, 2023. https://www.cdc.gov/nchs/data/series/sr_02/sr02_166.pdf
7. Stockwell S, Trott M, Tully M, Shin J, Barnett Y, Butler L, et al. Changes in physical activity and sedentary behaviours from before to during the COVID-19 pandemic lockdown: a systematic review. BMJ Open Sport Exerc Med 2021;7(1):e000960. PubMed doi:10.1136/bmjsem-2020-000960
8. Sher C, Wu C. Who stays physically active during COVID-19? Inequality and exercise patterns in the United States. Socius 2021;7:2378023120987710. PubMed doi:10.1177/2378023120987710
9. Lee S, Lee C, Xu M, Li W, Ory M. People living in disadvantaged areas faced greater challenges in staying active and using recreational facilities during the COVID-19 pandemic. Health Place 2022;75:102805. PubMed doi:10.1016/j.healthplace.2022.102805
10. Community Preventive Services Task Force. Physical activity: built environment approaches combining transportation system interventions with land use and environmental design 2016. Accessed May 25, 2023. https://www.thecommunityguide.org/sites/default/files/assets/PA-Built-Environments.pdf
11. Whitfield GP, Carlson SA, Ussery EN, Watson KB, Brown DR, Berrigan D, et al. Racial and ethnic differences in perceived safety barriers to walking, United States National Health Interview Survey – 2015. Prev Med 2018;114:57–63. PubMed doi:10.1016/j.ypmed.2018.06.003
12. Webber BJ, Irani KL, Omura JD, Whitfield GP. Observation of and intention to use new places and changed spaces for physical activity during the COVID-19 pandemic — United States, June 2021. Prev Med 2022;160:107100. PubMed doi:10.1016/j.ypmed.2022.107100
13. Ahmad FB, Cisewski JA, Miniño A, Anderson RN. Provisional mortality data — United States, 2020. MMWR Morb Mortal Wkly Rep 2021;70(14):519–22. PubMed doi:10.15585/mmwr.mm7014e1

Top

Tables

^a Among people with data on physical activity, missingness in sociodemographic characteristics varied from 16.7%–19.9% for income to <3% for sex, age, race and ethnicity, education, and urban/rural status across 2016, 2018, and 2020. Values for n for demographic groups may not sum to the total N (484,244 for 2016; 436,741 for 2018; and 401,276 for 2020) due to missing demographic information; those with missing demographic information were excluded from the analytic sample only for the analyses of that demographic characteristic.
^b Participants were asked, “During the past month, other than your regular job, did you participate in any physical activities or exercises such as running, calisthenics, golf, gardening, or walking for exercise?” Participants who reported no were classified as physically inactive.
^c All estimates were weighted to account for complex survey design and nonresponse.
^d Indicates that the 95% CI excludes zero and the difference in prevalence is therefore statistically significant.
^e People of another race were included in analyses; however, estimates are not shown for this group due to the heterogeneity of the category.
^f Lower CI is 0.01 and rounds to zero.
^g Urbanicity is based on the 2013 National Center for Health Statistics urban–rural classification scheme for counties (6).

^a Prevalence estimates and differences were weighted to account for complex survey design and nonresponse. Prevalence differences with 95% CIs not containing zero were considered statistically significant. People of another race were included in analyses; however, estimates are not shown for this group due to the heterogeneity of the category.
^b Indicates that the 95% CI excludes zero and the difference in prevalence is therefore significant.
^c Lower CI is −0.04 and rounds to zero.
^d Upper CI is −0.03 and rounds to zero.

Top