Comparative effectiveness research, technological abandonment, and health care spending

Adv Health Econ Health Serv Res. 2012;23:103-21. doi: 10.1108/s0731-2199(2012)0000023007.


Purpose: Policymakers hope that comparative effectiveness research will identify examples of widely used therapies that are no better than less expensive alternatives and, consequently, reduce health care spending. Comparative effectiveness research is unlikely to reduce spending if physicians are quick to adopt effective treatments but slow to abandon ineffective ones.

Methodology/approach: We present a theoretical model that shows how physicians will adopt new treatments in response to positive evidence more readily than they abandon existing treatments in response to negative evidence if the marginal costs of production decline post-adoption. We report trends in the use of two common procedures, percutaneous coronary intervention (PCI) for patients with stable angina and routine episiotomy in vaginal childbirth, where comparative effectiveness research studies have failed to find evidence of a benefit.

Findings: Use of PCI and episiotomy have declined over time but are still excessive based on the standards implied by comparative effectiveness research. PRACTICAL IMPLICATIONS (IF APPLICABLE): The findings suggest that comparative effectiveness research has the potential to reduce costs but additional efforts are necessary to fully realize savings from abandonment.

Originality/value of chapter: There is a large literature on technological adoption in health care, but few studies address technological abandonment. Understanding abandonment is important for efforts to decrease health care costs by reducing use of ineffective but costly treatments.

MeSH terms

  • Biomedical Technology
  • Comparative Effectiveness Research*
  • Diffusion of Innovation*
  • Episiotomy / statistics & numerical data*
  • Evidence-Based Medicine
  • Female
  • Health Expenditures*
  • Humans
  • Male
  • Models, Theoretical
  • Percutaneous Coronary Intervention / statistics & numerical data*
  • Practice Patterns, Physicians'
  • United States