Declining Overuse of Hormone Therapy for Localized Prostate Cancer: Predictors of Reimbursement Responsiveness and Emerging Patterns of Care Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 22, 2019
  • Ellis, Shellie Dawn
    • Affiliation: Gillings School of Global Public Health, Department of Health Policy and Management
  • This research examines the effects of reimbursement policy as a strategy to improve quality of care. We estimated the degree to which physician characteristics are associated with declining androgen deprivation therapy (ADT) overuse; identified the effect of reimbursement changes on ADT overuse; and, evaluated the impact of changing patterns of ADT overuse on quality of care in localized prostate cancer. We used SEER-linked Medicare claims and American Medical Association data to create three distinct longitudinal cohorts of individuals diagnosed with incident prostate cancer in the 2000s and their physicians. Multilevel logistic regression modeling controlled for patient and physician characteristics associated with overuse of medical care and prostate cancer treatment selection, and clustering of patients within physicians. In the first study, time in practice was not associated with ADT overuse, but three patterns of ADT overuse were observed. We could not distinguish urologists who increased ADT overuse from those who decreased ADT overuse after MMA based on physician characteristics. Our findings suggest that: 1) new types of interventions will be needed to address persistent overuse; 2) guidelines should underscore treatment strategies for vulnerable patients; and 3) economic theory may need to consider clinic explanations for the volume response. The second study suggests that, among urologists treating early-stage and lower grade prostate cancer, variation in reimbursement was not associated with overuse of ADT during a period of guideline stability. There was a small but significant negative association between ADT overuse and excess reimbursement relative to all treatments: urologists in favorable reimbursement climates had lower odds of ADT overuse. Multi-specialty group practice type was associated with lower odds of ADT overuse. Reimbursement cuts may not be effective strategy to reduce overuse in all clinical scenarios. Finally, physicians' pre-MMA ADT overuse was negatively associated with delivering guideline-concordant care post-MMA. High users of ADT pre-MMA were also more likely to overuse ADT and provide guideline-discordant care post-MMA. Reducing reimbursement for inappropriate therapy will not necessarily improve quality of care. Physicians unable to provide guideline-concordant care may need additional resources to align with guidelines or to adopt guideline-concordant technologies.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Carpenter, William R.
  • Doctor of Philosophy
Graduation year
  • 2013

This work has no parents.