Active commuting is a manifestation of human transportation that potentially impacts aspects of modern living that goes beyond individuals’ healthy lifestyles. In the context of the global climate emergency characterized by uncontrolled pollution and high reliance on fossil fuels for transportation, investigating how individuals habitually move is of paramount importance to develop locally oriented, evidence-based public policies to promote cleaner, healthier, and smarter choices of commuting. In fact, active commuting is aligned with several of the United Nations Sustainable Development Goals, such as health and well-being, sustainable communities, clean energy, responsible consumption, and climate action.1 Moreover, considering that physical inactivity has achieved pandemic proportions2 and caused enormous economic and health burdens,3 active commuting may be an important strategy to improve population health.4 In fact, replacing sedentary behaviors with physical activity (PA) in any domain (including transportation) is consistent5 with health benefits, especially when PA volume is low, which is common in contemporary societies.6
Brazil is a vast country, occupying nearly half of South America’s territory. Its great diversity is expressed in many facets, such as geographical distribution, cultural dynamics, socioeconomic disparities, and varied urbanization levels. With a heterogeneous population of more than 200 million inhabitants and a territory of ∼8.5 million km,2,7 estimating the nationwide prevalence of any behavior is challenging, especially considering the great variability in terms of assessment instruments, exposure definitions, and sample’s demographics. This may further increase the underrepresentation and increase the inequality in research between the global North and South.8 In this scenario, a qualitative and quantitative synthesis of data across different regions of the largest country in Latin America can provide a more balanced view of the topic. This approach can potentially inform not only local interventions as mentioned, but also serve as a model for future regional and countrywide studies and contribute to worldwide discussions and policies, given that active commuting may have broader implications as it affects the global environment.
Therefore, we conducted a systematic review and meta-analysis to estimate the proportion of active commuting in Brazil.
Methods
Search Strategy and Selection Criteria
This review was prospectively registered (CRD42023431054). Observational studies conducted in Brazil and providing a proportion of active commuting (considering at least a combination of the most common modes of active commuting: walking and biking) for individuals older than 18 years were eligible. Eligible articles were full peer-reviewed articles, published from inception to April 2023 in English, Spanish, or Portuguese. Studies reporting proportions of only one mode of active commuting, conference abstracts, and preprints were not included. The procedures described herein followed the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses9 and the Reporting Guidelines for Meta-analyses of Observational Studies in Epidemiology.10 The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
Our search strategy comprised terms related to PA, commuting, and Brazilian geographical regions. Electronic searches were performed in 5 databases (PubMed, EMBASE, LILACS, CINAHL, and SPORTDiscus) and were conducted by one of the investigators. The primary search string was generated in PubMed and then translated into the other databases using an automated tool (The Polyglot Search Translator),11 with corrections made whenever needed. Authors were contacted whenever data availability limited the inclusion of the article. The full search strategy can be found in the Supplementary Material S1 (available online). We performed an additional backward citation search by hand in selected landmark studies and reviews identified during screening.
Duplicated records were removed using an automated tool (The Deduplicator).12 Literature screening procedures were conducted in a web-based solution (rayyan.ai),13 and eligible studies were selected in duplicate in 2 steps (titles/abstracts and full texts).
To avoid duplication of analytic units, only one representative study was elected for each duplicated data set. For example, if multiple studies analyzed data from the same population sample, we kept only one of them, based upon criteria such as sample size, completeness of information, possibilities of subgroup analyses, and so on.
Data Extraction
After an initial procedural standardization with all team members, data were extracted in duplicate, with coders blinded to each other’s data, in a prespecified electronic form. Disagreements in any of the steps above were solved by consensus, or, if needed, by a third reviewer. Our team extracted information regarding studies’ year of conduction, scope (local, regional, and national), data source, active commuting definitions, tools used to estimate commuting behavior, and proportions of individuals considered active in commuting (totaled by sex, by age, and by region).
Data Analysis
Weighted prevalence estimates, if presented in the included studies, were disregarded in the analyses since the statistical package considers the number of exposed subjects and the total sample through which it calculates crude estimates for each study. In addition to the main analysis, we also conducted subanalyses related to sex, age, and geographical region.
Overall, subgroup estimates with corresponding 95% confidence intervals (CIs) were calculated based on the reported proportions of active versus inactive commuting in each included study. We applied a random-effects inverse variance model with arcsine transformation due to the heterogeneous nature of epidemiological data.14 CIs for individual proportion estimates were calculated using the Clopper–Pearson method.15 Between-study variability was estimated with the I2 statistic and is reported along with the forest plots. All analyses were conducted using the “meta” package (version 6.0) for R (version 4.2.2). Data are expressed as percent proportion estimate and 95% CI, except otherwise stated.
Risk of Bias
Risk of bias was assessed in duplicate using a tool for observational studies developed by Hoy et al,16 comprising 10 items (4 related to internal validity and 6 to external validity) classified as low or high risk of bias, and a summary item for each study classified as low, moderate, or high risk of bias. The items on internal validity are related to sample representativeness, sampling methods, and nonresponse bias, whereas the items on external validity were related to methods, instruments, and definitions in data collection procedures.
The evaluation process was piloted and standardized with all reviewers before it began. Disagreements were resolved by consensus, and, if necessary, by a third reviewer. Risk of bias plots were built with the “robvis” package for R.17
Deviations From the Protocol
As the included studies showed heterogeneous definitions of being active in the commuting domain, we created 2 categories based on the strictness of the adopted definition: “high-volume threshold” (HIGH), where only individuals with more than 150 minutes per week of active commuting were considered active; and “low-volume threshold” (LOW), where individuals with at least 10 minutes per week of active commuting were considered active (irrespective of the definition adopted in the individual study), or where active commuting was self-reported in a binary fashion (yes or no). Data from these categories were meta-analyzed both combined, as specified in the original protocol, and separately. Additionally, we decided not to meta-analyse pooled data of HIGH and LOW in regional analyses, owing to the low number of studies available.
Results
A Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram detailing the eligibility process is described in Figure 1. After deduplication of record, we screened 2522 unique studies. Of these, 255 full texts were retrieved for eligibility check and 58 articles were considered eligible for the review. Additionally, we identified 4 records through citation searching. Since there were multiple studies derived from the same epidemiological data, we excluded 19 studies from the national analysis due to duplicated data sources. However, whenever pertinent, data sets excluded from nationwide analysis were used in subgroup analysis (eg, region-specific analysis).
—PRISMA 2020 flow diagram for new systematic reviews which included searches of databases, registers, and other sources. PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
Citation: Journal of Physical Activity and Health 22, 2; 10.1123/jpah.2024-0413
A total of 39 studies were kept in this review, 2 of which were used only for subgroup analysis. The 37 remaining were included in the main meta-analytic model, which totaled 52 individual estimates and a pooled sample of 1,266,862 subjects. In the main analysis, 22 estimates (42.3%) were derived from mixed geographical regions, whereas 14 (26.9%) were obtained in Southeast, 12 (23.1%) in South, 3 (5.8%) in Northeast and 1 (1.9%) in the Midwest. We found no estimate of active commuting proportion in the North region. Table 1 shows the main characteristics of the included studies.
Characteristics of Included Studies
| Study | How active commuting was measured | Adopted criteria for active commuting | Region | Epidemiological data source | Sample type | Analyzed sample | % men | Active commuting, % |
|---|---|---|---|---|---|---|---|---|
| Benedetti et al18 | Validated questionnaire | 150 min/wk in domain | South | — | Population based | 875 | 49.9 | 19.5 |
| Bertuol et al19 | Nonvalidated questionnaire | Yes/no | Mixed | VIGITEL 2017 | Population based | 53,034 | 36.8 | 13.2 |
| Bicalho et al20 | Validated questionnaire | 150 min/wk in domain | Southeast | — | Service/community | 567 | 48.5 | 32.0 |
| Cerveira Fronza et al21 | Nonvalidated questionnaire | Yes/no | South | — | Service/community | 623 | 45.7 | 12.0 |
| Christofoletti et al22 | Validated questionnaire | Minimum 10 min/wk | Southeast | — | Population based | 2281 | 40.9 | 64.2–70.7 |
| Coledam et al23 | Validated questionnaire | 600 MET/min/wk | South | — | Service/community | 500 | NR | 8 |
| Corseuil et al24 | Validated questionnaire | Minimum 10 min/wk | South | EpiFloripa | Population based | 1656 | 36.1 | 63.2 |
| Cruz et al25 | Validated questionnaire | Yes/no | Mixed | PNS 2019 | Census | 88,500 | 46.8 | 49.6 |
| Cunha et al26 | Nonvalidated questionnaire | Yes/no | Midwest | SIMTEL | Population based | 2002 | 38.1 | 8.5 |
| da Silva et al27 | Single question | Commuting to work on most days | Mixed | Lifestyle and leisure habits of Brazilian industry workers | Service/community | 46,981 | 70.6 | 27.2 |
| de Matos et al28 | Validated questionnaire | Minimum 10 min/wk | Mixed | ELSA Brasil | Service/community | 14,876 | 46.3 | 73.9 |
| de Rosso Krug et al29 | Validated questionnaire | 150 min/wk in domain | South | — | Population based | 343 | 8.2 | 19.5 |
| Del Duca et al30 | Validated questionnaire | Yes/no | South | EpiFloripa | Population based | 1720 | 44.7 | 49.6 |
| dos Santos et al31 | Validated questionnaire | Yes/no | South | — | Population based | 2120 | 40.6 | 66.1 |
| Ferrari et al32 | Nonvalidated questionnaire | 150 min/wk in domain | Mixed | VIGITEL 2013–2019 | Census | 359,038 | 46.0 | 12–14.4 |
| Ferrari Junior et al33 | Self-report | Yes/no | Mixed | — | Population based | 5259 | 46.9 | 23.1 |
| Figueiredo et al34 | Validated questionnaire | Minimum 10 min/wk | Southeast | ISA Capital | Population based | 4851 | 44.4 | 32.3–59.1 |
| Florindo et al35 | Validated questionnaire | 150 min/wk in domain | Southeast | — | Population based | 890 | 41.3 | 37.8 |
| Hallal et al36 | Validated questionnaire | 150 min/wk in domain | Northeast | — | Service/community | 2046 | 37.1 | 26 |
| Lima et al37 | Single question | Yes/no | Southeast | ABC Socioeconomic Research | Population based | 602 | 62.3 | 17.8–24.1 |
| Madeira et al38 | Validated questionnaire | 150 min/wk in domain | Mixed | Projeto AQUARES | Population based | 12,116 | 46.0 | 34.23 |
| Mazo et al39 | Validated questionnaire | 150 min/wk in domain | South | — | Population based | 1062 | 41.1 | 17.7 |
| Mendes et al40 | Validated questionnaire | Yes/no | South | — | Population based | 2874 | 41.2 | 51.7 |
| Mielke et al41 | Nonvalidated questionnaire | 150 min/wk in domain | Mixed | VIGITEL 2006–2011 | Census | 352,240 | 46.1 | 8.2–11.0 |
| Monteiro et al42 | Validated questionnaire | Minimum 10 min/wk | South | — | Population based | 820 | 36.2 | 63.8 |
| Mourão et al43 | Validated questionnaire | 150 min/wk in domain | Northeast | — | Population based | 319 | 30.4 | 12.5 |
| Norde et al44 | Validated questionnaire | 150 min/wk in domain | Southeast | ISA-Capital | Population based | 269 | 49.8 | 31.2 |
| Pitanga et al45 | Validated questionnaire | 150 min/wk in domain | Northeast | — | Convenience | 2305 | 39.1 | 23.7 |
| Ramos Veloso Silva et al46 | Single question | Commuting to work on most days | Southeast | — | Service/community | 15,641 | 18.1 | 26.1 |
| Rombaldi et al47 | Validated questionnaire | Minimum 10 min/wk | South | — | Population based | 972 | 57.0 | 48.1 |
| Sá et al48 | Validated questionnaire | Minimum 10 min/wk | Southeast | — | Population based | 890 | 41.3 | 85.4 |
| Sá et al49 | Nonvalidated questionnaire | 150 min/wk in domain | Southeast | Household Travel Survey | Population based | 56,261 | 47.0 | 27.6 |
| Sebastião et al50 | Validated questionnaire | Minimum 10 min/wk | Southeast | — | Population based | 1572 | 41.8 | 70.5 |
| Silva et al51 | Validated questionnaire | 150 min/wk in domain | South | — | Service/community | 225 | 100.0 | 27.7 |
| Sousa et al52 | Validated questionnaire | 150 min/wk in domain | Southeast | ISA-CAMP | Population based | 986 | 42.4 | 10.9 |
| Werneck et al53 | Nonvalidated questionnaire | 150 min/wk in domain | Mixed | PNS 2013 | Census | 60,202 | 43.1 | 30.6 |
| Werneck et al54 | Nonvalidated questionnaire | Yes/no | Mixed | PNAD 2008 | Census | 169,348 | NR | 35 |
| Included only in subgroup analyses | ||||||||
| da Silva et al55 | Nonvalidated questionnaire | 150 min/wk in domain | Midwest | VIGITEL 2006–2009 | Population based | 8045 | 38.1 | 10–16 |
| Silva et al56 | Single question | Yes/no | South | Lifestyles and Leisure-Time Habits among Industry Workers in RS | Service/community | 2265 | 56.2 | 26.5 |
Abbreviations: MET, metabolic equivalent; NR, not reported.
The pooled estimate for active commuting in Brazil was 28.4% (24.4%–32.6%). Pooled estimates using different active commuting definitions were 17.4% (15.1%–19.9%) for HIGH, and 44.2% (36.9%–51.5%) for LOW (Figure 2).
—Forest plot of national prevalences of active commuting. CI indicates confidence interval; HIGH, prevalence estimates considering at least 150 minutes per week to be considered active; LOW, prevalence estimates considering at least 10 minutes per week or self-report to be considered active.
Citation: Journal of Physical Activity and Health 22, 2; 10.1123/jpah.2024-0413
Regarding sex-specific analyses, 20 primary studies with 21 individual estimates for HIGH and 24 for LOW were included, totaling a pooled sample of ∼266,000 subjects. Proportion of active commuting using HIGH definition was based on ∼221,000 observations and was estimated to be 20.5% (16.7%–24.9%) for men and 13.8% (10.2%–17.9%) for women. When using LOW definition for active commuting, our analysis was based on ∼45,000 observations and estimates were 43.4% (27.2%–60.4%) for men and 47.6% (32.1%–63.4%) for women (Figures 3 and 4).
—Forest plot of sex subgroup prevalences of active commuting for men. CI indicates confidence interval; HIGH, prevalence estimates considering at least 150 minutes per week to be considered active; LOW, prevalence estimates considering at least 10 minutes per week or self-report to be considered active.
Citation: Journal of Physical Activity and Health 22, 2; 10.1123/jpah.2024-0413
—Forest plot of sex subgroup prevalences of active commuting for women. CI indicates confidence interval; HIGH, prevalence estimates considering at least 150 minutes per week to be considered active; LOW, prevalence estimates considering at least 10 minutes per week or self-report to be considered active.
Citation: Journal of Physical Activity and Health 22, 2; 10.1123/jpah.2024-0413
Subgroup active commuting analysis by age groups revealed a proportion of active commuting of 30.4% (26.4%–34.6%) when using HIGH and 38.8% (30.3%–47.6%) when using LOW for adults, with ∼15,000 observations, based on 5 and 8 individual estimates for HIGH and LOW respectively. Active commuting in the older individuals was estimated to be 17.6% (12.5%–23.4%) for HIGH and 34.2% (14.7%–57.1%) for LOW (Supplementary Materials S2 and S3 [available online]).
When evaluating Brazilian geographic regions separately using HIGH, we found a proportion of active commuting of 20.9% (15.9%–26.4%) in the Northeast region, 26.1% (14.3%–40.0%) for the South, 27.1% (19.3%–35.8%) in the Southeast, and 12.4% (9.9%–15.0%) in the Midwest. When using LOW, we found estimates of 49.3% (36.1%–62.4%) in the Southeast and 48.8% (35.3%–62.3%) in the South. No estimates were found using LOW for Northeast and Midwest. No data were found for the northern region. Both forest plots (HIGH and LOW) for the regional subanalysis and a heatmap showing the number of individual regional estimates are presented in the Supplementary Materials S4–S6 (available online).
Risk of bias analysis revealed that most of the 39 studies were classified as presenting a high risk of bias (n = 27, 69.2%), while 23.1% (n = 9) were considered with moderate risk of bias and 7.7% (n = 3) with low risk of bias. Plots for risk of bias are presented in the Supplementary Materials S7 and S8 (available online).
Discussion
The way people commute is highly relevant to any contemporary society. Positive impacts are expected when individuals choose to move actively, including improvements in traffic, pollution, and population’s overall health and well-being.4,57 Active commuting aligns with 17 of the United Nations Sustainable Development Goals, such as good health, sustainable communities, clean energy, responsible consumption, and climate action.1 Surprisingly, however, studies assessing commuting behavior are limited in low-income and middle-income countries, hampering potential public health policies. In this regard, this review provides a comprehensive panorama of active commuting in Brazil in all its regional diversity, producing new data that can pave the way for evidence-based interventions with potential impacts locally, nationally, and globally, considering that active transportation can be a meaningful planetary health strategy. In our quantitative synthesis, we showed that the national prevalence of active commuting in Brazil varies from 17.4% to 44.2%. Also, we observed a consistent heterogeneity in active commuting definitions adopted across the studies, which is directly linked to the wide variation in our pooled estimates. The lack of a uniform definition hinders a more accurate estimation of prevalence, thereby compromising targeted interventions.
Subgroup analyses also showed striking findings. Regarding sex, prevalence estimates did not differ between men and women using neither HIGH nor LOW. However, in HIGH, prevalences tended to be lower for women (with a minimal overlap in CIs). Women may avoid walking or biking alone in some places due to fear of crime,58 which could lead to more motorized travels59 and lower volumes of active commuting.
As for the age categories, prevalence estimates were not different between adults and older individuals when using LOW; however, there was a considerable drop in prevalence of around 50% in the latter when using HIGH. The volume of transport-related PA can diminish with age, mostly affecting retired individuals, since older adults reportedly commute closer to home.60 Additionally, in Brazil, free public transport is common for older adults, which might affect the time spent in active commuting. The natural decline in physical function and the increase in disability that often accompany aging,61 coupled with safety concerns regarding risk of falls walking in poor-quality sidewalks,62 may also contribute to the lower estimates among older adults. However, the observations herein presented are based on a lower number of studies, which limits more definitive conclusions.
In relation to the geographical analyses, differences in prevalence between regions were only detectable in HIGH, whereas there was a paucity of data for 3 out of the 5 Brazilian geographical regions when using LOW. Using HIGH, Midwest and Northeast had lower prevalences of active commuting when compared to South and Southeast. The number of analytical units was limited though. Importantly, in our review, we did not find any data for active commuting specifically in the North, while scarce data were available in the Northeast and Midwest. This fact underscores the disparities in populational assessments across Brazilian regions, with studies concentrated in the richer and more densely populated regions (Southeast and South).
Efforts to uncover the multifaceted benefits of active commuting are growing. For instance, meta-analytic data from Japan showed lower diabetes prevalence in active versus inactive commuters.63 Active commuting has been also related to better quality of life64 and satisfaction with the built environment.59 From an environmental standpoint, reducing commuting in motor vehicles decreases CO2 emissions and improves traffic.57 The overall importance of active commuting in individual, populational, and planetary health justifies the search for evidence of quality to inform policies. Our review suggests that the body of evidence pertaining to Brazil could be substantially improved, with special attention to uniform case definitions, more accurate behavior estimates (eg, device-based) and comprehensive exploration of regional differences.
Estimating complex behaviors on a populational level is challenging.4,65 Indeed, there is no consensus on how to classify an active commuter. This partially explains the high heterogeneity across the included studies, making data comparability and generalizability difficult. In a study using the UK Biobank (∼155,000 middle-aged subjects), 7 different commuting categories were assessed, including mixed active/passive commuting (eg, public transportation plus walking/biking).66 Although Brazil also has some exemplary longitudinal population-based registries on health and behaviors, with some of them being included in this review,19,25,32,41,53,54 most show superficial descriptions of active commuting, missing information on modes, combinations of active and passive transport, time spent in this behavior, and so on. As Brazil is the largest country in South America, the suboptimal quality of data on activity commuting identified herein may compromise public health policies for the continent, which is generally underrepresented in science.
Aligned with this, the underrepresentation of studies in poorer regions in Brazil, particularly in the North, is highly concerning. The lack of accurate data on commuting in regions comprising or nearby Amazon rainforest precludes accurate predictions on the potential impacts of the motor vehicle pollution on the forest ecosystems and, ultimately, on climate change. While this fact might reflect a common pitfall in Brazilian scientific production, we reason that some data sources already in use, such as information from population-wide databases mentioned above19,25,32,41,53,54 can be used to provide region-specific data. Such explorations are warranted to further develop the field.
Another gap in the literature was the paucity of studies using objective tools to measure commuting activity (eg, step counts, accelerometry or GPS-based devices).67,68 These instruments measure PA more accurately because they are not prone to self-reporting bias such as with questionnaires. To better assist fine-tuning policies and programs focused on increasing PA levels, further studies should provide objective measures of commuting activity.
This review does not come without limitations. First, the main challenge in synthesizing the results herein reported was the great heterogeneity in active commuting definitions. To overcome this issue and provide a useful quantitative synthesis of prevalence, we classified active commuters considering HIGH and LOW. The former used a very strict definition in which active commuters were defined as those who meet 150 minutes per week or more of activity in transport, which certainly underestimate the actual proportion of individuals relying on active transport. Conversely, the latter definition considered as active commuters those who reached at least 10 minutes per week of active commuting, or those who self-reported as an active commuter. In this case, the odds of obtaining overestimated rates of active commuting increase, and self-reported definition does add uncertainties to the classification. Therefore, our attempt to harmonize data for the meta-analysis did not fully resolve the issue of data heterogeneity, which was previously discussed as a limitation of the existing literature. Other factors may also have contributed to the high heterogeneity detected in our analysis (as quantitatively evidenced by the I2 statistics). Different demographic characteristics of the samples, disproportionate number of studies across regions, and limited studies for subgroup analyses likely contributed to the variability in the models. Also, most studies exhibited high risk of bias, owing to a suboptimal definition of active commuters, assessments with regional or urban/rural restrictions, and lack of probabilistic samples. This elevated risk of bias comes with no surprise considering the multiple challenges in assessing a complex behavior as active commuting in a diverse country like Brazil, as previously discussed.
However, our review has strengths, including the rigorous, preplanned methodological procedures and the report of systemic review and meta-analysis in a transparent fashion. Of relevance, our main analysis encompassed a pooled sample of more than 1.2 million subjects, which is, to our knowledge, the most robust estimation of active commuting in Brazil. Finally, the new data obtained herein can potentially inform public health stakeholders, funding agencies, and the research community by underscoring the limitations in this field and providing a broad panorama of active commuting in the country.
To conclude, this systematic review and meta-analysis compiled data at national and regional levels on active commuting in Brazil. Moreover, major gaps in literature were identified. Further studies should make efforts to standardize active commuting definition, increasing the participation of individuals from underrepresented regions (beyond South and Southwest), and use objective measures of commuting activity. Given the local, regional, and global impacts of active transport in individual and planetary health and considering the importance of Brazil as a “continental” country that shelters distinct, ecologically relevant biomes (ie, Amazon, Cerrado, Caatinga, Pantanal, Atlantic Forest, and Pampa), it is suggested that new, high-quality studies assessing commuting and its multiple potential impacts on individual and planetary health can be fostered, coping with the United Nations Sustainable Development Goals. Based on very heterogeneous literature, active commuting in Brazil varies from 17% to 44%. There were strong regional differences in data availability and a lack of objectively-measured data, which warrants further investigations.
Acknowledgments
Santos was funded by FAPESP (São Paulo Research Foundation), process number 2023/08433-5 and Azevedo was funded by CNPq (The National Council for Scientific and Technological Development; process No. 172743/2023-0). Author Contributions: Structured the research question, the protocol, conducted the search strategies, worked on the eligibility and data extraction, and performed the statistical analysis, wrote, and revised the manuscript: L.P. Santos. Conceptualized the work, worked on the eligibility and data extraction, and revised the manuscript: Azevedo, Ribeiro, Santos, and Iraha. Contributed to the protocol structure, data analysis and interpretation, and wrote and revised the manuscript: Roschel and Gualano. All authors had full access to the data and agreed upon the submission to the present journal. Data Sharing: All study data, including electronic forms and analysis code, will be shared after publication upon an email request to the first author.
References
- 2.↑
Kohl HW 3rd, Craig CL, Lambert EV, et al. The pandemic of physical inactivity: global action for public health. Lancet. 2012;380(9838):294–305. PubMed ID: 22818941 doi:
- 3.↑
Santos AC, Willumsen J, Meheus F, Ilbawi A, Bull FC. The cost of inaction on physical inactivity to public health-care systems: a population-attributable fraction analysis. Lancet Glob Health. 2023;11(1):e32–e39. PubMed ID: 36480931 doi:
- 4.↑
Dinu M, Pagliai G, Macchi C, Sofi F. Active commuting and multiple health outcomes: a systematic review and meta-analysis. Sports Med. 2019;49(3):437–452. PubMed ID: 30446905 doi:
- 5.↑
Ekelund U, Steene-Johannessen J, Brown WJ, et al. Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality? A harmonised meta-analysis of data from more than 1 million men and women. Lancet. 2016;388(10051):1302–1310. PubMed ID: 27475271 doi:
- 6.↑
Hallal PC, Andersen LB, Bull FC, Guthold R, Haskell W, Ekelund U. Global physical activity levels: surveillance progress, pitfalls, and prospects. Lancet. 2012;380(9838):247–257. PubMed ID: 22818937 doi:
- 8.↑
Ramírez Varela A, Cruz GIN, Hallal P, et al. Global, regional, and national trends and patterns in physical activity research since 1950: a systematic review. Int J Behav Nutr Phys Act. 2021;18(1):5.
- 9.↑
Page MJ, McKenzie JE, Bossuyt PM, et al. The Prisma 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71.
- 10.↑
Brooke BS, Schwartz TA, Pawlik TM. MOOSE reporting guidelines for meta-analyses of observational studies. JAMA Surg. 2021;156(8):787–788. PubMed ID: 33825847 doi:
- 11.↑
Clark JM, Sanders S, Carter M, et al. Improving the translation of search strategies using the polyglot search translator: a randomized controlled trial. J Med Libr Assoc. 2020;108(2):195–207. PubMed ID: 32256231 doi:
- 12.↑
Clark J, Glasziou P, Del Mar C, Bannach-Brown A, Stehlik P, Scott AM. A full systematic review was completed in 2 weeks using automation tools: a case study. J Clin Epidemiol. 2020;121:81–90. PubMed ID: 32004673 doi:
- 13.↑
Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan—a web and mobile app for systematic reviews. Syst Rev. 2016;5(1):210. doi:
- 14.↑
Barker TH, Migliavaca CB, Stein C, et al. Conducting proportional meta-analysis in different types of systematic reviews: a guide for synthesisers of evidence. BMC Med Res Methodol. 2021;21(1):189. doi:
- 15.↑
Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med. 1998;17(8):857–872. PubMed ID: 9595616 doi:
- 16.↑
Hoy D, Brooks P, Woolf A, et al. Assessing risk of bias in prevalence studies: modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol. 2012;65(9):934–939. PubMed ID: 22742910 doi:
- 17.↑
McGuinness L. Robvis: a package to quickly visualise risk-of-bias assessment results. Github. nd. https://githubcom/mcguinlu/robvis
- 18.↑
Benedetti TRB, Borges LJ, Petroski EL, Gonçalves LHT. Physical activity and mental health status among elderly people. Rev Saúde Pública. 2008;42(2):302–307. PubMed ID: 18327498 doi:
- 19.↑
Bertuol C, Tozetto AVB, de Oliveira SN, Del Duca GF. Sex differences in the association between educational level and specific domains of physical activity: a Brazilian cross-national survey. Can J Public Health. 2022;113(3):474–483. PubMed ID: 34988924 doi:
- 20.↑
Bicalho PG, Hallal PC, Gazzinelli A, Knuth AG, Velásquez-Meléndez G. Adult physical activity levels and associated factors in rural communities of Minas Gerais State, Brazil. Rev Saude Publica. 2010;44(5):884–893. PubMed ID: 20676590 doi:
- 21.↑
Cerveira Fronza F, Berria J, Minatto G. Exposure to simultaneous sedentary behavior domains and sociodemographic factors associated in public servants. Rev Bras Cineantropom Desempenho Hum. 2017;19(4):469.
- 22.↑
Christofoletti AEM, Goulardins GS, Orcioli-Silva D, et al. Fatores associados à mortalidade de adultos e idosos residentes no município de Rio Claro-SP: um estudo de coorte. Rev Bras Cineantropom Desempenho Hum. 2018;20(3):258–268. doi:
- 23.↑
Coledam DHC, de Arruda GA, Ribeiro EAG, Cantieri FP. Association between domains of physical activity and health among teachers: a cross-sectional study. Sport Sci Health. 2022;18(2):445–453. doi:
- 24.↑
Corseuil MW, Schneider IJC, Silva DAS, et al. Perception of environmental obstacles to commuting physical activity in Brazilian elderly. Prev Med. 2011;53(4–5):289–292. PubMed ID: 21820007 doi:
- 25.↑
Cruz DKA, Silva KS da, Lopes MVV, Parreira FR, Pasquim HM. Socioeconomics inequities associated with different domains of physical activity: results of the national health survey 2019, Brazil. Epidemiol Serv Saude. 2022;31:e2021398.
- 26.↑
Cunha IC, Peixoto M do RG, Jardim PCBV, Alexandre VP. Factors associated with physical activity in Goiania’s adult population: surveillance through telephone interviews. Rev Bras Epidemiol. 2008;11(3):495–504. doi:
- 27.↑
da Silva JA, da Silva KS, Del Duca GF, et al. Moderating effect of gross family income on the association between demographic indicators and active commuting to work in Brazilian adults. Prev Med. 2016;87:51–56. PubMed ID: 26876633 doi:
- 28.↑
de Matos SMA, Pitanga FJG, Almeida M da CC, et al. What factors explain bicycling and walking for commuting by Elsa-Brasil participants? Am J Health Promot. 2018;32(3):646–656. PubMed ID: 29121794 doi:
- 29.↑
de Rosso Krug R, Lopes MA, Balbé GP, Marchesan M, Mazo GZ. Prevalence of commuting physical activity and associated factors in long-lived older adults. Rev Bras Cineantropom Desempenho Hum. 2016;18(5):520–529.
- 30.↑
Del Duca GF, Nahas MV, Garcia LMT, Mota J, Hallal PC, Peres MA. Prevalence and sociodemographic correlates of all domains of physical activity in Brazilian adults. Prev Med. 2013;56(2):99–102. PubMed ID: 23200875 doi:
- 31.↑
Dos Santos Ferreira Viero V, Matias TS, Alexandrino EG, et al. 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. Z Gesundh Wiss. 2022;13:1–9.
- 32.↑
Ferrari G, Dulgheroff PT, Claro RM, Rezende LFM, Azeredo CM. Socioeconomic inequalities in physical activity in Brazil: a pooled cross-sectional analysis from 2013 to 2019. Int J Equity Health. 2021;20(1):188. doi:
- 33.↑
Ferrari Junior GJ, Teixeira CS, Felden ÉPG. Socioenvironmental factors and behaviors associated with negative self-rated health in Brazil. Cien Saude Colet. 2021;26(9):4309–4320. PubMed ID: 34586281 doi:
- 34.↑
Figueiredo TKF, Aguiar RG de, Florindo AA, et al. Changes in total physical activity, leisure and commuting in the largest city in Latin America, 2003-2015. Rev Bras Epidemiol. 2021;24:e210030.
- 35.↑
Florindo AA, Salvador EP, Reis RS. Physical activity and its relationship with perceived environment among adults living in a region of low socioeconomic level. J Phys Act Health. 2013;10(4):563–571. PubMed ID: 22976232 doi:
- 36.↑
Hallal PC, Reis RS, Parra DC, Hoehner C, Brownson RC, Simões EJ. Association between perceived environmental attributes and physical activity among adults in Recife, Brazil. J Phys Act Health. 2010;7(suppl 2):S213–S222. doi:
- 37.↑
Lima JS, Ferrari GLM, Ferrari TK, Araujo TL, Matsudo VKR. Changes in commuting to work and physical activity in the population of three municipalities in the São Paulo region in 2000 and 2010. Rev Bras Epidemiol. 2017;20(2):274–285.
- 38.↑
Madeira MC, Siqueira FCV, Facchini LA, et al. Physical activity during commuting by adults and elderly in Brazil: prevalence and associated factors. Cad Saude Publica. 2013;29(1):165–174. PubMed ID: 23370036
- 39.↑
Mazo GZ, Benedetti TB, Sacomori C. Association between participation in community groups and being more physically active among older adults from Florianópolis, Brazil. Clinics. 2011;66(11):1861–1866. PubMed ID: 22086514
- 40.↑
Mendes MA, da Silva ICM, Hallal PC, Tomasi E. Physical activity and perceived insecurity from crime in adults: a population-based study. PLoS One. 2014;9(9):e108136.
- 41.↑
Mielke GI, Hallal PC, Malta DC, Lee IM. Time trends of physical activity and television viewing time in Brazil: 2006-2012. Int J Behav Nutr Phys Act. 2014;11(1):101. doi:
- 42.↑
Monteiro LZ, de Farias JM, de Lima TR, Schäfer AA, Meller FO, Silva DAS. Physical activity and perceived environment among adults from a city in Southern Brazilian. Cien Saude Colet. 2022;27(6):2197–2210. PubMed ID: 35649009 doi:
- 43.↑
Mourão ARC, Novais FV, Andreoni S, Ramos LR. Physical activity in the older adults related to commuting and leisure, Maceió, Brazil. Rev Saude Publica. 2013;47(6):1112–1122. PubMed ID: 24626549
- 44.↑
Norde MM, Fisberg RM, Marchioni DML, Rogero MM. Systemic low-grade inflammation-associated lifestyle, diet, and genetic factors: a population-based cross-sectional study. Nutrition. 2020;70:110596. doi:
- 45.↑
Pitanga FJG, Lessa I, Barbosa PJB, Barbosa SJO, Costa MC, Lopes AS. Sociodemographic factors associated with different domains of physical activity in adults of black ethnicity. Rev Bras Epidemiol. 2012;15(2):363–375. PubMed ID: 22782102 doi:
- 46.↑
Silva RRV, Bastos VF, Mota GHL, et al. Active commuting to work among teachers of public basic education of the state of Minas Gerais. Rev Bras Cineantropom Desempenho Hum. 2021;23:e83277.
- 47.↑
Rombaldi AJ, Menezes AMB, Azevedo MR, Hallal PC. Leisure-time physical activity: association with activity levels in other domains. J Phys Act Health. 2010;7(4):460–464. PubMed ID: 20683087 doi:
- 48.↑
Sa TH, Salvador EP, Florindo AA. Factors associated with physical inactivity in transportation in Brazilian adults living in a low socioeconomic area. J Phys Act Health. 2013;10(6):856–862. PubMed ID: 23074086 doi:
- 49.↑
de Sá TH, Parra DC, Monteiro CA. Impact of travel mode shift and trip distance on active and non-active transportation in the São Paulo metropolitan area in Brazil. Prev Med Rep. 2015;2:183–188. PubMed ID: 26844071 doi:
- 50.↑
Sebastião E, Gobbi S, Chodzko-Zajko W, et al. The International Physical Activity Questionnaire-long form overestimates self-reported physical activity of Brazilian adults. Public Health. 2012;126(11):967–975. PubMed ID: 22944387 doi:
- 51.↑
da Silva MC, Spohr CF, Engers PB, de Moura Neto AB. Atividade física no lazer e deslocamento e fatores associados em motoristas e cobradores do transporte coletivo urbano de pelotas-rs. R Bras Ci e Mov. 2017;25(2):137–144. doi:
- 52.↑
Sousa NFS, Lima MG, Cesar CLG, Barros MBA. Envelhecimento ativo: prevalência e diferenças de gênero e idade em estudo de base populacional. Cad Saúde Pública. 2018;34(11):e00173317.
- 53.↑
Werneck AO, Stubbs B, Szwarcwald CL, Silva DR. Independent relationships between different domains of physical activity and depressive symptoms among 60,202 Brazilian adults. Gen Hosp Psychiatry. 2020;64:26–32. PubMed ID: 32086172 doi:
- 54.↑
Werneck AO, Barboza LL, Araújo RHO, et al. Time trends and sociodemographic inequalities in physical activity and sedentary behaviors among Brazilian adults: national surveys from 2003 to 2019. J Phys Act Health. 2021;18(11):1332–1341. PubMed ID: 34548416 doi:
- 55.↑
da Silva SM, Luiz RR, Pereira RA. Risk and protection factors for cardiovascular diseases among adults of Cuiabá, Mato Grosso, Brazil. Rev Bras Epidemiol. 2015;18(2):425–438. PubMed ID: 26083513 doi:
- 56.↑
da Silva SG, Del Duca GF, da Silva KS, de Oliveira ESA, Nahas MV. Deslocamento para o trabalho e fatores associados em industriários do sul do Brasil. Rev Saúde Pública. 2012;46(1):180–184. PubMed ID: 22183516
- 57.↑
Maizlish N, Woodcock J, Co S, Ostro B, Fanai A, Fairley D. Health cobenefits and transportation-related reductions in greenhouse gas emissions in the San Francisco Bay area. Am J Public Health. 2013;103(4):703–709. PubMed ID: 23409903 doi:
- 58.↑
Almanza Avendaño AM, Romero-Mendoza M, Gómez San Luis AH. From harassment to disappearance: young women’s feelings of insecurity in public spaces. PLoS One. 2022;17(9):e0272933. doi:
- 59.↑
Panter J, Griffin S, Ogilvie D. Active commuting and perceptions of the route environment: a longitudinal analysis. Prev Med. 2014;67:134–140. PubMed ID: 25062909 doi:
- 60.↑
Moran M, Van Cauwenberg J, Hercky-Linnewiel R, Cerin E, Deforche B, Plaut P. Understanding the relationships between the physical environment and physical activity in older adults: a systematic review of qualitative studies. Int J Behav Nutr Phys Act. 2014;11(1):79. doi:
- 61.↑
Justice JN, Cesari M, Seals DR, Shively CA, Carter CS. Comparative approaches to understanding the relation between aging and physical function. J Gerontol A Biol Sci Med Sci. 2016;71(10):1243–1253. PubMed ID: 25910845 doi:
- 62.↑
de Lima MCC, Fernandes da Silva A, Barbosa dos Santos R, d‘Orsi E, Bestetti MLT, Rodrigues Perracini M. How do older adults living in the community in Brazil perceive walkability in the context of sidewalks? J Aging Environ. 2024;38(1):37–55.
- 63.↑
Honda T, Hirakawa Y, Hata J, et al. Active commuting, commuting modes and the risk of diabetes: 14-year follow-up data from the Hisayama study. J Diabetes Investig. 2022;13(10):1677–1684. PubMed ID: 35607820 doi:
- 64.↑
Neumeier LM, Loidl M, Reich B, et al. Effects of active commuting on health-related quality of life and sickness-related absence. Scand J Med Sci Sports. 2020;30(suppl 1):31–40. doi:
- 65.↑
Alessio HM, Bassett DR, Bopp MJ, et al. Climate change, air pollution, and physical inactivity: is active transportation part of the solution? Med Sci Sports Exerc. 2021;53(6):1170–1178. PubMed ID: 33986228 doi:
- 66.↑
Flint E, Cummins S. Active commuting and obesity in mid-life: cross-sectional, observational evidence from UK Biobank. Lancet Diabetes Endocrinol. 2016;4(5):420–435. PubMed ID: 26995106 doi:
- 67.↑
Cerin E, Cain KL, Oyeyemi AL, et al. Correlates of agreement between accelerometry and self-reported physical activity. Med Sci Sports Exerc. 2016;48(6):1075–1084. PubMed ID: 26784274 doi:
- 68.↑
Helmerhorst HJF, Brage S, Warren J, Besson H, Ekelund U. A systematic review of reliability and objective criterion-related validity of physical activity questionnaires. Int J Behav Nutr Phys Act. 2012;9(1):103. doi:
