Examining urban and rural bicycling in the United States: Early findings from the 2017 National Household Travel Survey

Abstract

IntroductionBicycling has personal and population health benefits. While bicycle research has focused primarily on the urban context, the rural context is equally important. There are documented disparities in health behaviors and health outcomes in rural areas compared to urban areas. It is unknown whether bicycling is one of these health behavior disparities. MethodsThis study addresses two questions: 1) what is the prevalence of bicycling behaviors by urbanicity (urban/rural and population density category), and 2) what are the characteristics that best categorize individuals as cyclists versus non-cyclists, overall and for rural populations. We used the 2017 National Household Travel Survey, a nationally representative sample of the US non-institutionalized population (age ≥5 years). Bicycling was defined as any bicycling, bicycling for exercise, bicycle commuting, and bike share program use. Analyses used complex survey procedures to estimate unadjusted and adjusted prevalence; and, random forest to rank characteristics that best categorize respondents as bicyclists. ResultsThe unadjusted prevalence of any reported bicycling was higher in high-density urban areas (≥10,000 persons per square mile (ppsm)), 14.7% (95% CI: 13.4%–16.0%), than very low-density rural areas (<500 ppsm), 11.8% (10.4%–13.2%). However, when adjusting for covariates, the prevalence was comparable: high-density urban: 14.4% (12.9%–16.0%); very low-density rural: 12.2% (10.8%–13.7%). Unadjusted prevalence of bicycling for exercise were also similar (high-density urban: 8.2% (7.0%–9.3%); very low-density rural: 7.0% (6.1%–7.9%)). We also found that characteristics that best categorize bicyclists from the overall sample were different from the rural only sample. ConclusionsThis research suggests that bicycling prevalence overall and for exercise are similar between urban and rural areas across population densities. It also suggests that characteristics that were important for rural bicyclists were different from urban bicyclists. Urban-focused bicycle research may need modification to fit the rural context to promote bicycling and physical activity.