Background: The clinical outcomes and costs of implantable cardioverter-defibrillators (ICDs) used for primary prevention of sudden cardiac death in nonexperimental settings are uncertain.
Objective: The purpose of this study was to measure the health outcomes and costs among a nationally representative cohort of elderly, primary-prevention ICD recipients.
Methods: We collected health-care cost and utilization data from all Medicare beneficiaries hospitalized for congestive heart failure (CHF) who had received primary-prevention ICDs between October 2003 and September 2005 as well as propensity-score-matched control Medicare beneficiaries hospitalized for CHF during the same period. A multivariable Cox proportional hazards model was fitted to the cohort, which comprised 7125 ICD recipients and 7125 controls and which was followed through December 2005. Medicare claims in the first year inclusive of the index hospitalization were used to assess differences in health-care costs.
Results: ICD receipt was associated with a significant reduction in mortality (adjusted hazard ratio = 0.62, 95% confidence interval 0.58-0.67). ICD patients had higher median hospital costs in the first 30 days after initial hospitalization (median difference = $41,542, P <.001) and at 1 year (median difference = $41,503, P <.001) as well as higher outpatient and physician costs at 6 months (median difference = $1828, P <.001).
Conclusions: ICD implantation was associated with reduced mortality in a nonexperimental, elderly, primary-prevention patient population hospitalized for CHF. The additional health-care costs of ICD implantation were substantial but comparable to published cost-effectiveness models that have projected ICDs to be cost-effective.