Statistics from Altmetric.com
The health benefits of regular physical activity in preventing chronic disease-related morbidity and mortality have been well established.1 2 Evidence suggests that regular physical activity reduces all-cause and cardiovascular disease (CVD) mortality by 20%–30% in a dose–response manner, among healthy individuals across sex and a wide age range.1 Further, there is increasing concern over physical inactivity globally, which is a leading cause of premature death, accounting for >5 million deaths/year worldwide.3 Current Joint European Society of Cardiology and American Heart Association guidelines recommend that all healthy adults engage in ≥150 min/week of moderate intensity or ≥75 min/week of vigorous intensity physical activity; however, 23% of adults do not meet recommended activity levels globally.1 2 4 The guidelines also state that ‘any activity is better than none and more activity is better than some.’1 It has been suggested that the increased use of ‘passive’ modes of transportation contributes to insufficient physical activity.4 Research on the effect of more active modes of travel for work or other purposes on the risk of morbidity and mortality is ongoing.
In their Heart paper, Panter5 and colleagues investigate the possible health benefits (including potential lower CVD, cancer morbidity and mortality risk) associated with active modes of travel for work and other purposes, combined and separately, versus exclusive vehicle use for transportation. The prospective observational study included adults aged 37–73 years old who were participants in the UK Biobank, a large national population-based study, designed to measure and track the health of more than half a million adult residents of primarily urban areas in the UK, recruited from 2006 to 2010. At baseline, participants completed an extensive electronic questionnaire on sociodemographic, health status, and lifestyle characteristics, and modes of commute and non-commute travel. Active travel was defined as bicycling or walking alone or in combination with a vehicle for transit. Participants were followed for five main outcomes: incident and fatal CVD, incident and fatal cancer and all-cause mortality. Outcomes were ascertained from hospital records, the national cancer registry and death certificates obtained through the National Health Service. Outcomes that occurred during the first 2 years after the baseline assessment were excluded from the analysis to minimise the effects of reverse causation. All analyses were stratified on whether the participants were regular or not regular commuters (ie, unemployed, employed but travelling to work <3 times/week and working from home). Fully adjusted models included demographic, geographic and socioeconomic characteristics, other behavioural factors (eg, dietary, cigarette smoking, leisure time physical activity, sedentary behaviours) and health conditions/characteristics (eg, long-standing limiting illness/disability, body mass index, blood pressure, diabetes) measured at baseline as potential confounders.
Among the 358 799 participants, 52% (n=187 281) were characterised as regular commuters and 48% (n=171 498) were characterised as not regular commuters. Over a mean of 7 years of follow-up (SD, 1 year), for commute travel among individuals characterised as making regular commutes there were 7438 events, including 1697 all-cause deaths, 1595 incident CVD events and 4146 incident cancer events. For non-commute travel among individuals characterised as not making regular commutes, there were 12 293 events, including 3456 deaths, 2621 incident CVD events and 6216 incident cancer events. Among regular commuters, compared with those who exclusively relied on a vehicle for travel, those with more active travel had 11% lower risk of incident of CVD (HR 0.89, 95% CI 0.79 to 1.00) and 30% lower risk of CVD mortality (HR 0.70, 95% CI 0.51 to 0.95) after adjusting for all potential confounding variables. When considering all-commute and non-commute-related travel combined, those with more active travel (vs vehicle use exclusively) had 43% lower risk of CVD mortality (HR 0.57, 95% CI 0.39 to 0.85). Among participants who were not regular commuters and compared with those who exclusively relied on a vehicle for travel, participants who travelled using more active modes had 8% lower risk of all-cause mortality in fully adjusted models (HR 0.92, 95% CI 0.86 to 0.99). The authors did not find evidence that the reported associations were modified by the travel distance to work or access to a vehicle.
The current study has many important strengths: this is one of the first studies of non-commute travel and travel among individuals who do not regularly commute. The exposure (active vs passive travel) and a wide range of potential confounders were well defined based on a prospectively established protocol in the UK Biobank. Ascertainment of outcomes from national registries on deaths and hospitalisations reduced the potential for differential ascertainment of the outcomes, and measures were taken to minimise the potential for reverse causation by excluding prevalent and incident outcomes occurring during the first 2 years of follow-up. In addition, the main findings within strata of regular and not regular commuters were consistent across numerous additional analyses.
Despite the rigour of this study, there are some important limitations to consider when interpreting the findings. The outcomes associated with active travel (vs exclusive reliance on a vehicle) were inconsistent for regular commuters versus those who did not regularly commute. The risk was lower for CVD outcomes among regular commuters who engaged in active travel (both for commute and non-commute purposes) whereas the association with all-cause mortality was not statistically significant. On the other hand, among those who were not regular commuters, active travel (for non-commute purposes) was associated with all-cause mortality whereas the association with CVD outcomes was not statistically significant. The authors did not discuss potential reasons for the inconsistencies. It should be noted that among regular commuters, there was a statistically significant association for active travel for non-commuting purposes and all-cause mortality in models that only adjusted for sociodemographic and residential geographic factors, HR 0.86 (95% CI 0.78 to 0.95). Also, regular commuters (vs those who were not regular commuters) were generally younger (mean age 52 vs 61 years) and healthier (without long-standing illness/disability among 76% vs 61%) at baseline and had fewer deaths for analyses of non-commute travel (1699 vs 3456) despite very similar follow-up time (mean 7.0 vs 6.9 years). The extent to which these differences suggest that residual confounding or other factors (eg, insufficient statistical power) may have accounted for the different associations across strata of regular and not regular commuters needs further investigation. The authors commented that a substantial number of consented eligible participants (29%) were excluded due to missing data using the complete-subject analysis approach and there were key differences in excluded versus included participants. For instance, excluded participants had lower levels of physical activity, occupational classification and educational attainment compared with those who were included. It is unclear whether other approaches such as imputation or inverse-probability weighted methods to address the missing data would have resulted in similar findings as were observed using the complete-subject analysis.
The findings of previous observational studies of commute travel using UK Biobank data support those reported by Panter et al. 6 7 Among UK Biobank participants who were regular commuters, active commuting was associated with lower body mass index, per cent body fat, incident and fatal CVD and cancer, and all-cause mortality. Celis-Morales et al reported that while commuting by walking (vs commuting via car only) was associated with a lower risk of CVD incidence and death, cycling was associated with the lowest risk of these outcomes in addition to a lower risk of all-cause mortality and cancer.6 Also, study evidence supported that there were greater benefits for active commuting exclusive of passive modes (vs of a mix of both) and for longer commuting distances, above the median (6 miles/week for walkers and 30 miles/week for cyclists). Additional research is needed to further characterise physical activity for non-commute travel, including walking versus cycling, energy expended and travel distance in relation to health outcomes.
Can the findings of this study be extended beyond urban UK? Caution may be warranted, especially where sociocultural and physical environment differences exist. Access to safe bike or walking paths, weather/seasonal patterns and air quality, among others, may result in different findings. These differences in environment may partly account for the mix in findings in previous studies of active commuting and health outcomes, although the evidence seems strongest for a benefit related to cardiometabolic outcomes.8 9
In conclusion, the study by Panter et al 5 provides the best evidence to date that active (vs passive) non-commute travel is associated with better health outcomes, including among individuals who do not regularly commute. This is an important finding, as many individuals may work remotely, do not work, or only commute occasionally and need additional strategies (eg, active non-commute travel) to incorporate regular physical activity into their routines. The data further support the observation that the benefits of physical activity operate in a dose–response fashion for CVD outcomes, where authors reported lowest risk of fatal CVD mortality among commuters who used active modes of travel for both commuting and non-commuting purposes. This study provides further evidence to support efforts, including policies and guideline recommendations, that encourage active travel even if that is only possible for part of the journey.
Sincere thanks to Dr Adi Gundlapalli and Carrie Edlund for their helpful reviews of early drafts of this editorial. Resources and administrative support were provided by VA Salt Lake City Health Care System (IDEAS 2.0 Center).
Contributors AFM is the sole author of this Commissioned Editorial. She made substantial contributions to the conception of the work and interpretation of data. She drafted the work and revised it critically for important intellectual content. AFM agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Funding Dr Mohanty is supported by the US Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research and Development Project (CDA 1 IK2 RX002324-01A1).
Disclaimer The views expressed in this article are those of the author and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the US government.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.