Sports Medicine https://doi.org/10.1007/s40279-018-1023-0
SYSTEMATIC REVIEW
Active Commuting and Multiple Health Outcomes: A Systematic Review and Meta‑Analysis Monica Dinu1,2 · Giuditta Pagliai1,2 · Claudio Macchi3 · Francesco Sofi1,2,3
© Springer Nature Switzerland AG 2018
Abstract Background Active commuting is associated with greater physical activity, but there is no consensus on the actual beneficial effects of this type of physical activity on health outcomes. Objective To examine the association between active commuting and risk of all-cause mortality, incidence and mortality from cardiovascular diseases, cancer and diabetes through meta-analysis. Methods A comprehensive search of MEDLINE, Embase, Google Scholar, Web of Science, The Cochrane Library, Transport Research International Documentation database, and reference lists of included articles was conducted. Only prospective cohort studies were included. Results Twenty-three prospective studies including 531,333 participants were included. Participants who engaged in active commuting had a significantly lower risk of all-cause mortality [relative risk (RR) 0.92, 95% CI 0.85–0.98] and cardiovascular disease incidence (RR 0.91; 95% CI 0.83–0.99). There was no association between active commuting and cardiovascular disease mortality and cancer. Participants who engaged in active commuting had a 30% reduced risk of diabetes (RR 0.70; 95% CI 0.61–0.80) in three studies after removal of an outlying study that affected the heterogeneity of the results. Subgroup analyses suggested a significant risk reduction (− 24%) of all-cause mortality (RR 0.76; 95% CI 0.63–0.94) and cancer mortality (− 25%; RR 0.75; 95% CI 0.59–0.895) among cycling commuters. Conclusion People who engaged in active commuting had a significantly reduced risk of all-cause mortality, cardiovascular disease incidence and diabetes.
Key Points The present meta-analysis reports a beneficial role of active commuting on all-cause mortality, and cardiovascular disease. Significant reduction of cancer mortality was observed among cycling commuters. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s40279-018-1023-0) contains supplementary material, which is available to authorized users. * Monica Dinu
[email protected] 1
Department of Experimental and Clinical Medicine, University of Florence, Largo Brambilla 3, 50134 Florence, Italy
2
Clinical Nutrition Unit, Careggi University Hospital, Florence, Italy
3
Don Carlo Gnocchi Foundation Italy, IRCCS, Florence, Italy
Our findings are of clinical relevance, particularly for improving the level of physical activity in the general population.
1 Introduction Physical inactivity is a major risk factor for non-communicable diseases [1]. For health reasons, current recommendations encourage adults to perform moderate or intense physical activity for at least 30 min on most, if not all, days of the Vol.:(0123456789)
week [2]. Active commuting, such as cycling and walking to work, may be an effective way of integrating regular physical activity into a sedentary lifestyle [3]. For this reason, public health campaigns encourage people to drive less and to walk and cycle more [4]. Most studies suggest that active commuting has a positive impact on obesity and overweight [5]. With respect to other health outcomes (e.g. cardiovascular disease), some large cohort studies suggest protective effects, while others show no effects of active commuting [6–8]. Three recent systematic reviews have examined the relationship between active commuting, physical activity level, body weight and health outcomes in adults, finding a significant association between active transportation, increased physical activity and reduced body weight [5, 9, 10]. However, no previous meta-analyses have evaluated the possible association between active commuting and all-cause mortality, cancer and diabetes. The only meta-analysis that evaluated the effect of active commuting on cardiovascular health found an overall reduction in cardiovascular risk of 11%, with an analysis that combined different outcomes (e.g. incidence and mortality of coronary artery disease, hypertension, diabetes) [9]. The aim of this systematic review was to evaluate the relationship between active commuting and all-cause mortality, cardiovascular disease, cancer and diabetes.
2 Methods Studies that investigated the association between active commuting and health outcomes in adults were identified through a search of electronic databases according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement [11]: MEDLINE (source: PubMed, 1966 to February 2018), Embase (1980 to February 2018), Web of Science, The Cochrane Library, Transport Research International Documentation database, and Google Scholar. Relevant keywords relating to active commuting in combination as MeSH terms and text words (“exercise”, and “active” or “commuting” or “commuter” or “active travel” or “active transport”) were used in combination with words relating to health status and diseases: “mortality”, “circulatory disease”, “cardiovascular disease”, “ischemic heart disease”, “cerebrovascular disease”, “cancer”, “diabetes”, “health”, “health effects”, “health status” and their variants. The search strategy is provided in Electronic Supplementary Material Appendix S1. We limited the search strategy to human studies, but had no language restrictions. The reference lists of all included articles were reviewed manually. We used the most recent publication when multiple articles for a single cohort study were present. Two investigators (M.D., F.S.) independently assessed potentially relevant articles for eligibility. First, we screened
M. Dinu et al.
titles and abstracts. We obtained and screened the full text of articles where it was unclear from the title and abstract whether the article should be included or not. Discrepancies were resolved via consensus, and discussion with a third investigator (G.P.) if consensus could not be reached. Prospective cohort studies that: (1) assessed active commuting as the exposure variable, (2) reported all-cause mortality, cardiovascular disease, cancer or diabetes as an outcome, and (3) provided risk estimates (relative risk, hazard ratio) with confidence intervals or standard errors (or sufficient data to calculate them) were eligible for inclusion. Disagreements on eligibility were resolved by consensus. We included only prospective cohort studies to minimize recall and selection biases that are common in case-control studies. Eligible studies were included for review if they: (1) evaluated clinically healthy adult participants at baseline, (2) reported clear definitions of methods used to assess active commuting, and (3) reported data on active commuting levels in relation to health outcomes (all-cause mortality, cardiovascular disease, cancer and diabetes).
2.1 Data Extraction Two reviewers (M.D., F.S.) independently extracted all data using a standardized form. Disagreements were resolved by consensus, or by a third investigator (G.P.) if consensus could not be reached. We extracted: lead author; year of publication; country of the study population; number of participants; length of follow-up (years); age of the study cohort at baseline; sex; definition of the outcome of interest; method used to assess active commuting; commuting activity categories; effect size measurements and confidence intervals; and details of adjustment for confounding factors in the multivariable model.
2.2 Internal Validity Assessment Two reviewers (M.D., G.P.) independently assessed the internal validity of each included study using the Evaluation of Public Health Practice Projects Quality Assessment tool [12]. Disagreements were resolved via consensus, or by a third investigator (F.S.) if consensus could not be reached. We judged the internal validity across eight domains: selection bias, study design, confounders, blinding, data collection methods, withdrawals and dropouts, intervention integrity and analysis. Each domain was rated as weak, moderate or strong internal validity.
2.3 Statistical Analysis We used Review Manager (RevMan; version 5.3 for Macintosh; Copenhagen, Denmark) to pool results from the
Active Commuting and Health
individual studies. In this meta-analysis, hazard ratios (HRs) were deemed equivalent to relative risks (RRs) [13]. Pooled results were reported as RRs and presented with 95% confidence intervals (CIs) and two-sided P values using a random-effects model (DerSimonian and Laird method). A P value less than 0.05 was considered statistically significant. We used, when available, the results of the original studies from multivariable models with the most complete adjustment for potential confounders. The confounding variables included in this analysis are shown in Table 1. If the results were reported separately for men and women, they were included in the analysis as separate cohorts. Statistical heterogeneity was evaluated using the chisquare Cochran Q-test and the I2 statistic, which assesses the appropriateness of pooling individual study results [14]. The I2 statistic provides an estimate of the amount of variance across studies that is due to heterogeneity, rather than chance. Where I2 was greater than 50%, subgroup analyses were performed to explore the source of heterogeneity [15]. We assessed whether there were differences related to mode of active commuting (walking vs. cycling), geographical region (Northern Europe vs. all other countries), sex (men vs. women), and measurement of exposure (dichotomous vs. continuous). To establish the robustness of our results, we conducted a sensitivity analysis by removing each study one-by-one from the meta-analyses and recalculating the summary estimate (the ‘leave one out’ approach). A study whose removal either pushed the significance level of the overall association from
