Public health guidelines for physical activity recommend that adults engage in at least 150 minutes of moderate-intensity physical activity per week and that this amount can be accumulated in episodes as short as 10 minutes (1). Rather than relying on recreational activity, which many working people may find difficult to engage in, active commuting, the practice of walking or bicycling to work, offers a promising means to integrate the recommended amount of physical activity into daily routines (2-5).

Much research has linked physical activity with a reduced risk of chronic diseases, including diabetes, cardiovascular disease, and several forms of cancer (1,6). Several studies have documented an association between health and active commuting. Gordon-Larsen et al found a positive association between active commuting and several cardiovascular disease risk factors (7), and similarly, a review of 8 studies demonstrated a robust protective effect of active commuting on cardiovascular outcomes, especially among women (8). A randomized controlled trial of 68 inactive middle-aged men and women revealed that active commuting to work for 1 hour daily for 10 weeks created significant improvements in VO_2max, maximal treadmill time, and high-density lipoprotein cholesterol (9). Other studies have reported associations between active commuting and obesity (10) and all-cause mortality (11). Aside from benefits to physical health, active commuting has the potential to facilitate positive social (eg, increased contact with other residents), environmental (eg, less pollution), and economic (eg, lower insurance costs) outcomes (12).

Despite these benefits, rates of active commuting in the United States are low; more than 90% of respondents to the 2001 National Household Transportation Survey indicated that an automobile was their usual mode of transportation to work (13). An additional 5% reported taking public transit, almost 3% usually walked, and a little more than 1% used some other form of transportation (including only a portion who regularly bicycled) (13). In contrast, in some European countries, the percentage of trips taken in urban areas by walking or bicycling ranges from 30% to 45% (14). Even Canada, which has arguably poorer weather conditions for nonmotorized travel, exhibits rates of active transportation that are almost double those observed in the United States (14).

Many studies have examined rates and influences on children's active commuting to school (15-17). However, surprisingly little research has examined factors associated with active commuting among adults. One study that used an 18-item self-report measure found that environmental factors (eg, variety of destinations, aesthetics, traffic safety) were positively associated with walking to work, even after controlling for several other variables (18). Another study reported that walking or bicycling to work at least 30 minutes per day was related to participants' perceptions of the utility of active commuting for avoiding parking hassles, reducing expenses, increasing one's health, and reducing air pollution (19). Other research has identified distance as a determinant of active commuting (20,21). However, in general, very little is known about definitive correlates or intervention points for promoting walking and bicycling as an alternative to using cars (22).

Although limited research exists regarding personal and community-level influences on active commuting, few, if any, studies have examined how institutional factors affect an individual's propensity to walk or bicycle to work. Social ecological models posit that intervening at multiple levels is most effective (22), and much research has suggested that institutional or workplace interventions can change health behaviors (23,24). Therefore, we examined the association of cultural and physical workplace supports for active commuting with employee active commuting behavior. Understanding how workplace actions to support active commuting are related to walking and bicycling to work can suggest promising strategies for employer- or employee-initiated worksite interventions to increase utilitarian physical activity and help prevent chronic disease.

In total, 375 people completed the survey (Table 1). The mean age of respondents was 40 years, and most participants were white, female, and highly educated, which was largely representative of the study city (25). Approximately one-quarter of the sample reported actively commuting to or from work at least once per week, and men and younger adults were more likely to do so (Table 2). Approximately three-quarters of the sample reported that their workplace possessed at least 1 cultural support for active commuting. Men were more likely than women to report the presence of at least 1 cultural support. Approximately one-third of the full sample reported their workplace had no physical supports for active commuting. Again, men were more likely than women to report the presence of at least 2 physical supports, and there were few substantial differences in perceptions of physical supports by age group (Table 2).

In the full sample, participants who reported 2 or more physical supports for active commuting were more likely to actively commute at least once per week than those who reported none, but participants reporting only 1 physical support were not (Table 3). When the sample was stratified by sex, disparate results were observed. For men, neither having 2 or more nor a single physical workplace support was related to increased odds of active commuting. In contrast, compared with women who reported no physical supports in the workplace, women who reported 2 or more physical supports were more than 10 times as likely to actively commute, and women who reported a single physical support were more than 3 times as likely to actively commute. When the sample was stratified by age, compared with participants who reported no physical supports, both younger (18-39 y) and older (≥40 y) participants were more than 3 times as likely to actively commute if 2 or more physical supports were present in the workplace. Reporting only 1 physical support was not associated with increased odds of active commuting for either age group (Table 3).

In the full sample, participants who reported 1 or more cultural supports were more than twice as likely to actively commute at least once per week as participants who reported no cultural supports (Table 3). Women who reported having 1 or more cultural supports in the workplace were more than twice as likely to walk or bicycle to work at least once per week as women who reported having no cultural supports. However, this was not true of men. No significant differences were found between participants who reported 1 or more cultural supports in the workplace and participants who reported no cultural supports when the sample was stratified by age (Table 3).

We also examined the combined and disaggregated effects of different types of supports. Among the full sample, participants who reported at least 1 cultural support and at least 1 physical support for active commuting in the workplace were more than 6 times as likely to actively commute at least once per week as participants who reported both types of supports absent (Table 3). This cumulative effect was consistent for women but not for men or either age group of adults.

For all 5 samples examined, perceiving that the employer encourages active commuting was not related to increased odds of walking or bicycling to work at least once per week, and having showers or lockers at the workplace was related to increased odds of active commuting only for women (Table 4). However, the other 3 individual supports — perceptions of the number of coworkers who actively commute, availability of bicycle parking, and having bicycle storage policies — were generally related to an increased likelihood of active commuting, although bicycle parking was not a significant factor for men or older adults.

We found that people with a workplace environment that had more cultural and physical supports were more likely to walk or bicycle to work. However, this relationship between workplace supports and active commuting behavior appeared to be moderated by sex, as such associations were significant only for women in our sample. For women, the presence of at least 1 cultural support, in particular their perceptions that other coworkers walk or bicycle to work, was associated with almost a threefold increase in the likelihood of active commuting at least once per week. Likewise, women who reported 1 physical support were more than 3 times as likely to actively commute, while women with 2 or 3 physical supports were 10 times as likely to do so. Furthermore, the cumulative effect of having at least 1 cultural and at least 1 physical workplace support was strongest for women in this study. To our knowledge, few, if any, studies have explored how influences on active commuting differ by sex. One study (26) reported that women had stronger preferences for community infrastructure for bicycling (eg, off-road routes with separation from motorized traffic), and the same may be true for workplace supports for active commuting. Additionally, women may be more concerned with cleanliness and tidiness of their appearance compared with men, explaining the importance of the availability of showers and lockers to support active commuting. Women generally report lower levels of overall physical activity and lower self-efficacy for physical activity (27), but additional qualitative and experimental research is needed to further explore the reasons why workplace supports appear to differentially benefit active commuting among women.

When the sample was stratified by age, the presence of 2 or more physical workplace supports was associated with a greater likelihood of active commuting for both younger and older adults. However, with respect to cultural supports, the significant relationship observed for the full sample failed to materialize for either the younger or older subgroups. When considered in isolation, though, greater perceptions of the number of coworkers who actively commute was related to increased active commuting behavior for both age groups. Having a larger percentage of fellow employees who walk or bicycle to work may create an environment that tolerates, fosters, or reinforces similar behavior among colleagues. Future research can shed light on the psychological (eg, modeling, affiliation) or behavioral (eg, "walkpooling" or "bicyclepooling") mechanisms behind the relevance of having coworkers who actively commute.

The primary strength of our study was its novelty in considering supports in workplace settings that may influence active commuting. However, our study does have limitations. First, our data were collected from a convenience sample that was small. Because the exact number of people who were eligible and had the opportunity to complete the survey is unknown, we were unable to calculate a response rate. We also could not entirely rule out the possibility that respondents were clustered in workplaces, although some data collected in the survey (closest intersection to workplace) suggested that the participants originated from various locations. Future studies may examine our research questions with a larger, random sample of employees, potentially in a multilevel framework. Second, we used self-report measures of both the workplace environment and active commuting behavior (although perceived measures of workplace supports may be more important than objective indicators). Third, our online survey format may have excluded potential respondents who did not have access to the Internet. Finally, our sample was well-educated and almost exclusively white, which limits the generalizability of our findings.

Our results substantiate the impact of a social-ecological approach for promoting active commuting. Many studies have examined the utility of workplace interventions for promoting physical activity, with mixed results. For example, Proper et al (28) reviewed 26 studies of worksite interventions to promote physical activity or fitness and found evidence of a positive effect of such programs on physical activity and musculoskeletal disorders and found limited to inconclusive evidence for a positive effect on fatigue, physical fitness, general health, blood serum lipids, and blood pressure. Dishman et al (29) also reviewed 26 studies that delivered physical activity interventions through worksites and concluded that the mean effect of the interventions was heterogeneous and small. However, most studies included in those reviews employed largely individual-level approaches (eg, goal setting) to promote leisure-time (rather than utilitarian) physical activity. Our findings indicate that providing a supportive physical and cultural environment that promotes active commuting is associated with higher rates of walking and bicycling to work, which in turn can have significant health benefits (7-11).

In terms of health promotion, our results provide a foundation for intervention strategies, including potential physical changes to workplaces, such as the addition of showers, bicycle racks, and covered and secure bicycle parking. Likewise, a workplace culture that supports active commuting can influence mode of travel to work. The social support that results when a large number of workers actively commute, either collectively or individually, can be reinforcing (30). Such practices may be facilitated by team challenges or other worksite events focused around active commuting. Employer and government policies may also foster a climate supportive of active commuting. For example, offering tax breaks, parking refunds, health insurance premium reductions, or other financial incentives may encourage active commuting among employees. In combination with traditional individually targeted approaches, these institutional strategies can make active commuting more attractive. Moreover, our data suggest that such strategies may be especially influential for encouraging active commuting among women, a subgroup that generally exhibits lower rates of active transportation and overall physical activity (31). As workplaces increasingly adopt such practices, documenting the costs of such investments in comparison with the savings enjoyed by both employees (eg, reduced health care costs) and employers (eg, reduced sick time) is important.

Our study adds to the small body of existing literature concerning factors associated with active commuting among adults and contributes to our understanding of social-ecological influences on this behavior beyond the individual. More research is needed to evaluate interventions that aim to promote active commuting (20), but making changes to workplace infrastructure and policies may be effective avenues for increasing rates of active commuting and improving employee health.