Physician Ordering of Screening Ultrasound: National Rates and Association With State-Level Breast Density Reporting Laws

Abstract Purpose To describe factors associated with screening ultrasound ordering and determine whether adoption of state-level breast density reporting laws was associated with changes in ordering rates. Materials and Methods We performed a cohort study using National Ambulatory Medical Care Survey data for 2007 to 2015. We included preventive office visits for women aged 40 to 74 years without breast symptoms and signs or additional reasons requiring ultrasound ordering. Multivariate logistic regression was used to identify changes in ultrasound ordering rates pre- versus post-state-level density reporting laws, accounting for patient-, physician-, and practice-level characteristics. Analyses were weighted to account for the multistage probability sampling design of National Ambulatory Medical Care Survey. Results Our sample included 12,787 visits over the 9-year study period. Overall, 28.9% (3,370 of 12,787) of women underwent a breast examination and 22.1% (2,442 of 12,787) had a screening mammogram ordered. Only 3.3% (379 of 12,787) had screening ultrasound ordered. Screening ultrasounds were ordered more frequently for younger women (rate ratio [RR] 0.8 per 10-year increase in age, 95% confidence interval [CI]: 0.6-0.9, P  = .003), at urban practices (RR 2.3, 95% CI: 1.1-5.0, P  = .028), and less frequently in practices with computer reminders for ordering screening tests (RR 0.6, 95% CI: 0.3-0.9, P  = .024). In multivariate analyses, the rate of ultrasound ordering did not change after adoption of density notification laws (RR 0.7, 95% CI: 0.3-2.0, P  = .57). Conclusion The rate of screening ultrasound ordering remains low over time. There was no observed association between adoption of state-level density reporting laws and overall changes in ultrasound ordering.