Effects of an urban light rail line on health care utilization and cost: A pre-post assessment

In September 2015, a new light rail transit (LRT) line opened in metro Portland, Oregon, USA. We used this natural experiment to conduct an interrupted time series analysis of the effects of LRT introduction on health care costs. We hypothesized that such costs would decline over time based on demonstrated health benefits of increased transit-related physical activity. We wanted to evaluate whether introducing new LRT would lower the trajectory of monthly health care costs generated by a large sample of members of an integrated health care plan living near a new LRT station (cases) relative to similar members who did not (controls). We hypothesized that new LRT would stimulate higher transit use and consequently, higher transit-related physical activity among plan members living relatively closer to the LRT. However, the new LRT did not lower health care costs as hypothesized, at least relative to prior costs and within the first 4.5 years after the LRT opening. Results were consistent across outpatient visit costs and medication costs, as well as when only the 10 percent most expensive health plan members were considered. Although we did not find evidence of transit-related reductions in health care costs, research into the potential influence of public transit on physical activity and downstream health care costs remains warranted. Mechanisms through which such influence manifests are complex. Our negative results based on extensive and objectively measured health care cost data will contribute to the accuracy and validity of future models of the societal benefits of public transit.

Automated summary

This study conducted an interrupted time series analysis to examine the effects of introducing a new light rail transit (LRT) line on health care costs. The results showed that the new LRT did not lower health care costs within the first 4.5 years after its opening. Despite the lack of evidence for transit-related reductions in health care costs, further research into the potential influence of public transit on physical activity and downstream health care costs is still warranted.