Accountable care organizations and physician antibiotic prescribing behavior

Document Type


Publication Date




Publication Title

Social Science & Medicine


Background: Physician accountable care organization (ACO) affiliation has been found to reduce cost and improve quality across metrics that are directly measured by the Medicare ACO programs. However, little is known about potential spillover effects from this program onto non-measured physician behavior such as antibiotic prescribing. Methods: Using a two-part structural selection model that accounts for selection into treatment (ACO group), and non-treatment (control group), we compare physician antibiotic prescribing across these groups with adjustment for volume, patient, physician and institutional characteristics. We also estimate heterogeneous treatment responses across specialties, focusing on physicians with a primary specialty of internal medicine, family or general practice, nurse practitioners, as well as general and orthopedic surgeons. Results: We find that ACO affiliation helps reduce antibiotic prescribing by 20.4 (95%CI = −26.65 to −14.16, p-value<0.001) prescriptions (about 19.5%) per year. We show that each additional hospital and practice affiliation increases prescriptions by 1.6 (95%CI = 1.27 to 1.95, p-value<0.001) and 1.7 (95%CI = 1.00 to 2.47, p-value<0.001), respectively. However, the use of electronic health records and high-quality medical training is associated with a decrease in antibiotic use of 7.9 (95%CI = −8.79 to −7.07, p-value<0.001) and 3.6 (95%CI = −4.47 to −2.73, p-value<0.001) claims, respectively. The treatment effects are found to vary with specialty, where internal medicine physicians experience an average decrease of 23.6 (95%CI = −29.98 to −17.20, p-value<0.001), family and general practice physicians a decrease of 22.1 (95%CI = −28.37 to −15.77, p-value<0.001), nurse practitioners a decrease of 7.1 (95%CI = −13.99 to −0.77, p-value = 0.028), general surgeons a decrease of 9.6 (95%CI = −16.02 to −3.25, p-value = 0.003), and orthopedic surgeons a reduction of 8.1 (95%CI = −14.84 to −1.42, p-value = 0.018) in their antibiotic prescribing per year. Conclusions: In assessing the impact of Medicare ACO programs it is important to account for spillover effects. Our study finds that ACO affiliation has had a measurable impact on physician antibiotic prescribing.

Link to Published Version