Assessing the National Health, Education, and Air Quality Benefits of the United States Environmental Protection Agency's School Bus Rebate Program: A Randomized Controlled Trial Design.

Introduction: Approximately 25 million children ride buses to school in the United States. While school buses remain the safest school transport from a traffic accident perspective, older buses can expose students to high levels of diesel exhaust. These exposures can adversely affect health, which might cause missed school days and reduced learning. To hasten the transition to cleaner, lower-emission vehicles, the US Environmental Protection Agency's (US EPA) ongoing School Bus Rebate Program randomly allocated over $27 million to replace older, higher-emission school buses with cleaner, lower-emission alternatives between 2012 and 2017. Here, we evaluated the effectiveness of this national program.

Methods: Leveraging the randomized allocation of rebate funding, we assessed the impacts of the US EPA's 2012-2017 School Bus Rebate Programs on attendance, educational achievement, emergency department (ED) visits for respiratory causes among children in Medicaid, and community air pollution levels. We analyzed all districts linked to applications with complete data using modified intention-to-treat (ITT) modeling for randomized controlled trials, comparing changes in school-district levels of each outcome, after versus before each lottery year, by funding selection status. We also examined the heterogeneity of effects by model years of the replaced buses and by quartiles of estimated ridership on applicant buses.

Results: Of the 3,019 applications that met our inclusion criteria, 406 were randomly selected for funding. The districts that were linked to these applications were similar in terms of size, demographic makeup, funding requests, and socioeconomic status to the districts linked to applications that were not selected for funding. The districts that were linked to applications selected for funding that replaced the oldest buses had improvements in attendance, educational performance, and ambient particulate matter ≤2.5 µm aerodynamic diameter (PM_2.5) concentrations in the year after the lottery, compared with districts linked to applications that were not selected for funding. Districts that replaced pre-1990 model year buses had the largest gains, with 0.45 percentage points (pp) and 95% confidence interval (CI) of 0.26 to 0.65 higher attendance (equivalent to 45 additional students attending school each day in an average-size school district of 10,000 students), 0.06 standard deviation (SD) higher reading and language arts (RLA) (95% CI: 0.05 to 0.07), 0.03 SD higher math test scores (0.01 to 0.04), and -1.0 µg/m³ (-1.5 to -0.5) lower ambient PM_2.5 concentrations compared with districts not selected for funding. The replacement of model year 2000 and newer buses showed almost no effect on these outcomes. Districts replacing the oldest buses had suggestively higher ED visit rates, but these findings were not statistically distinguishable from no association and were sensitive to differing model specifications.

Based on the attendance improvements observed alone, we estimate that the total investment of $27 million by the US EPA for the 2012-2017 lotteries may have resulted in $350 million of benefits per year, although these benefits could not be distinguished from no benefit. Further investment of funds to replace all school buses manufactured before the year 1990 could lead to an additional $400 million of economic benefits per year and replacing all school buses manufactured before the year 2000 could lead to an additional $1.3 billion of economic benefits per year.

Conclusions: We conclude that the US EPA's School Bus Rebate Program investments to remove very old buses from the fleets have positively affected communities.