Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

The Effect of False-positive Mammograms on Antidepressant and Anxiolytic Initiation

Segel, Joel E. PhD*,†; Balkrishnan, Rajesh PhD; Hirth, Richard A. PhD§

doi: 10.1097/MLR.0000000000000760
Original Articles

Background: Despite reported increases in anxiety following a false-positive mammogram, there is little evidence the effect rises to the clinical level of initiating medication.

Objective: To analyze the effect of a false-positive mammogram on antidepressant or anxiolytic initiation and identify subpopulations most at risk.

Subjects: MarketScan commercial and Medicaid claims databases used to identify women ages 40–64 undergoing screening mammography with no prior antidepressant or anxiolytic claims.

Research Design: Using a retrospective cohort design, we estimated the effects of a false-positive relative to a negative mammogram on the likelihood of initiating antidepressants or anxiolytics using multivariate logistic models estimated separately by insurance type.

Results: At 3 months after a false-positive mammogram, the relative risk (RR) for antidepressant or anxiolytic initiation was 1.19 [95% confidence interval (CI), 1.06–1.31] for the commercially insured and 1.13 (95% CI, 0.96–1.29) in the Medicaid population. In addition, 4 subgroups were at particularly elevated risk: commercially insured women ages 40–49 (RR=1.33; 95% CI, 1.13–1.54) or whose false-positive required multiple tests to resolve (RR=1.37; 95% CI, 1.17–1.57), included a biopsy (RR=1.68; 95% CI, 1.18–2.17), or whose resolution took >1 week (RR=1.21; 95% CI, 1.07–1.34).

Conclusions: False-positive mammograms were associated with significant increases in antidepressant or anxiolytic imitation among the commercially insured. Follow-up resources may be particularly beneficial for cases taking longer to resolve and involving biopsies or multiple tests. The results highlight the need to resolve false-positives quickly and effectively and to monitor depressive symptoms following a positive result.

*Department of Health Policy and Administration, Pennsylvania State University, University Park, PA

Penn State Cancer Institute, Hershey, PA

Department of Public Health Sciences and Emily Couric Cancer Center, School of Medicine, University of Virginia, Charlottesville, VA

§Department of Health Management and Policy, University of Michigan, Ann Arbor, MI

J.E.S.: received a University of Michigan Rackham Graduate Student Travel grant to present an earlier version of this work at the 2015 AcademyHealth Annual Research Meeting in Minneapolis, MN. J.E.S. was supported by the MarketScan Dissertation Support Program that provided free access to the 2003–2004 Truven Health MarketScan Commercial Claims and Encounters Database and acknowledges helpful comments on an earlier version from Edward Norton and Thomas Buchmueller. Much of the work was completed while J.E.S. was a doctoral student at the University of Michigan in the Department of Health Management and Policy. The remaining authors declare no conflict of interest.

Reprints: Joel E. Segel, PhD, Department of Health Policy and Administration, Pennsylvania State University, 504S Ford Building, University Park, PA 16802. E-mail:

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.