Associations between Bike Score and frequency of bicycle ridership among women of reproductive age in the Puget Sound Region of Washington state

Background
Physical activity is linked to countless health benefits, including reduced risk for chronic disease. Women, minority groups, and persons with lower socioeconomic status are at increased risk for physical inactivity. Chronic disease during pregnancy is associated with increased risk for adverse pregnancy and neonatal outcomes. Bicycling is uniquely positioned to mitigate risk of developing chronic disease. Nonetheless, few Americans bicycle, and ridership is lower among women than men. Prior research indicates that women of reproductive age may increase ridership with the development of bicycle infrastructure. Scant research has examined this theory on a wide scale in practice, however. Our study aimed to address this gap in research.
Methods
The centroid of the residential census tract of eligible women of reproductive age who participated in the 2017 or 2019 Puget Sound Regional Council Household Travel Survey was matched with Redfin’s Bike Score, a proxy for infrastructure. We fit a log-linear regression model to investigate the association between Bike Score and bicycle ridership for 15 minutes or more at least once in the past 30 days (versus less than that) among women of reproductive age in the Puget Sound Region of Washington state, adjusting for age and seasonality. We assessed this association for effect modification by race/ethnicity, a proxy for structural and interpersonal racism.
Results
The study population comprised 3,204 women of reproductive age. This included 1,582 participants who lived in regions with Bike Scores less than 70 (the median Bike Score in this study population) and 1,622 participants who lived in regions with Bike Scores greater than or equal to 70. The risk of having ridden a bicycle recently was 33% higher for each 20 points of Bike Score, adjusting for age and seasonality (RR=1.33; 95% CI: 1.21, 1.45). None of the interaction terms were statistically significant at alpha-level=0.05. Excluding participants who identified as non-Hispanic White alone, the risk of having ridden a bicycle recently was 35% higher for each 20 points of Bike Score, adjusting for age and seasonality (RR=1.35; 95% CI: 1.11, 1.64).
Conclusion
Bike Score was positively associated with bicycle ridership among women of reproductive age, including women of reproductive age of color. The development of bicycle infrastructure may foster an increase in ridership among women of reproductive age that may result in reduced incidence of chronic disease and adverse pregnancy and neonatal outcomes. The lower baseline probability of riding among women of reproductive age of color compared to non-Hispanic White alone women of reproductive age was not explained by infrastructure. This finding highlights the need for additional research on non-infrastructural barriers and facilitators to bicycle ridership among women of reproductive age of color, with a focus on antiracism work and police reform, to reduce the racial disparity in ridership.