Author + information
- Paul T. Vaitkus, MD, MBA, FACC⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Paul T. Vaitkus, Midwest Heart Associates, Renaissance Plaza, 1340 Charles Street, Suite 300, Rockford, Illinois 61104.
New technologies in medicine almost always bring higher costs to the medical system. Most new medical technologies do not offer cost offsets via prevention of subsequent episodes of illness or reductions in labor costs. Two papers in the current issue of the Journal(1,2) document an exception to the rule of purely increased costs of a new technology. However, despite demonstrating offsetting cost savings within the overall health-care delivery system, when placed in the context of the complicated American medical system, these two papers serve to underscore serious inequities of who bears the costs and who reaps the benefits.
Drug-eluting stents (DESs), as is now well known to cardiologists, are substantially more expensive than their predecessors. The two papers published in the current issue (1,2) provide quantitative economic data that supports the common-sense notion that DESs, by preventing recurrent cardiovascular events (primarily repeat revascularization), offer a downstream savings that warrant the up-front initial greater investment. The paper of Bakhai et al. (1) builds on observations previously published on sirolimus-eluting stents (3) and showed that, in comparison to a bare-metal stent (BMS), a paclitaxel-eluting stent reduced long-term costs as it reduced restenosis. The long-term cost benefit offset the initially higher costs of the DES. The manuscript of Elezi et al. (2) compares the two commercially available DESs and reports that insofar as a sirolimus-eluting stent may be superior to a paclitaxel-eluting stent in preventing restenosis (a notion still under debate) (4–7), it will have superior long-term cost advantages. Insofar as other recently reported trials demonstrated no relative advantage of one stent over the other (8), one would expect that cost analyses from such neutral studies will demonstrate economic neutrality as well. No doubt the results of Elezi et al. (2) will be the topic of much spin-doctoring by the two manufacturers on the opposite sides of the debate.
Interpretation of the findings of Bakhai et al. (1) and Elezi et al. (2) is entirely dependent on one’s perspective in the medical system (9). We will limit our discussion to three simplified stakeholders: payers, providers (i.e., hospitals), and vendors. Equally simply, we can consider these stakeholders in a sequential chain (payer to hospital to vendor) in which one party’s costs represent the next party’s revenues. Therein lie the nettlesome economic issues.
Elezi et al. (2) adopt the perspective of payer and consistently maintain it throughout their analysis. The fact that their data were gathered in Germany does not mean they are less relevant for the U.S. Although the cost structures of the two systems certainly differ, the directionality of their analysis is nevertheless generalizable. A repeat hospitalization prevented is a large cost avoided on either side of the Atlantic. A DES costs more than a BMS in the U.S. and the European Union alike.
Bakhai et al. (1) incorrectly state that they adopt “society’s” perspective in that the costs they account for are not “society’s” costs at all. The costs they assess for the initial episode of care are for the hospital, and the follow-up costs they consider largely reflect those of payers. These do not sum to “society’s” costs, as the latter would include such diverse elements as lost productivity for the patients and all direct and indirect costs of providing health care. The latter are two or more orders removed from the hospital’s cost structures. The perspective of Bakhai et al. (1) is thus ambiguous and does not accurately represent any player in the marketplace, but perhaps is best approximated by the payer’s.
The main difference between all of the published economic analyses of DESs (3,10,11), including the papers of Elezi et al. (2) and Bakhai et al. (1), and the real world is that in the protocol-driven environment, essentially only a single stent was implanted. In the real world, the average number of stents per case is closer to 2. Thus the disparity between the up-front costs of DESs and BMSs is substantially greater in clinical practice than in the published economic analyses. The notion that the reduction in late costs offsets the initial costs is very sensitive to even small changes in the number of stents implanted or the restenosis rates (3), and therefore the “cost neutral” conclusion of Bakhai et al. (1) may not be applicable outside the confines of the TAXUS-IV (Treatment of De Novo Coronary Disease Using a Single Paclitaxel-Eluting Stent) trial.
How DESs impact costs in the real world is driven to a great degree by multi-vessel disease. The up-front higher costs of DESs and the downstream reduction of costs are both magnified. The up-front costs of DESs, however, need to be considered relative to coronary artery bypass grafting (CABG), which is an even more expensive alternative. Although the real assessment of the relative costs of multi-vessel DESs versus CABG will have to wait until the completion of newly launched prospective clinical trials of multi-vessel revascularization (12,13), data modeling based on ARTS-II (Arterial Revascularization Therapies Study Part II: Sirolimus-Eluting Stents for the Treatment of Patients With Multivessel De Novo Coronary Artery Lesions) suggests that the stent strategy will likely have the overall economic advantage (14).
Winners and losers
Diagnosis-related group (DRG)-driven payers (such as Medicare) are winners in this game. They bear a fixed, controlled, increased up-front cost (which barely covers the incremental cost of a single DES), which is then largely offset by downstream savings of fewer repeat hospitalizations. This is illustrated in the last paragraph of the Results section in the paper by Bakhai et al. (1): Medicare saved slightly more than $200 per percutaneous coronary intervention (PCI) patient. Commercial payers who pay a discounted fee for service (usually a negotiated percentage of charges) are not as fortunate. Unlike Medicare, they help pay for most of the freight. Interventional cardiologists have demonstrated that their behaviors are generally cost insensitive. They may complain about the costs (of BMSs in 1995 or DESs in 2004), but, motivated by offering their patients the greatest technical and clinical advances, they will not hesitate to utilize the more expensive technologies. The vendors currently enjoy an oligopolistic pricing opportunity and have acted accordingly. The hospitals, on the other hand, bear the greater DES acquisition cost and also suffer a downstream opportunity loss as repeat revascularization is avoided. In the analysis of Bakhai et al. (1) of an artificial TAXUS-IV universe of single-vessel disease treated with a single DES, the hospital loses approximately $450 per PCI patient. For an institution that performs 500 PCIs per year, that’s ∼$250,000. In the real-world setting of multi-stent cases and multi-vessel disease, the DES PCI is an even bigger money-loser, and a CABG avoided (either in the first place or later downstream) is a large lost revenue for the hospital. Drug-eluting stents have a much worse impact on hospital finances than the arithmetic of Bakhai et al. (1) would suggest.
What could be done to rectify these inequities? The Center for Medicare and Medicaid Services regularly reviews data on actual costs associated with specific DRGs and adjusts payments accordingly. This process should partially address the problem confronting hospitals, albeit with a substantial time lag. Stent manufacturers, on the other hand, will largely respond only to market forces and not appeals to altruism. As new players enter the market and DESs are relegated to the status of commodities, more competitive pricing will inevitably come into play. Forward-looking vendors will anticipate this and attempt to solidify their relationships with their customers by offering creative pricing strategies. One particularly attractive option would be for the vendors to provide all DESs per PCI case, regardless of the number used, for the price of two DESs. There will still be plenty of immediate profit and extended long-term benefits to the vendors in developing lasting partnerships with providers. Whether recent merger and acquisition activity among the key manufacturers will serve to stifle competition warrants intense Federal Trade Commission scrutiny.
Finally, it deserves repeating that in undertaking analyses such as those of Bakhai et al. (1) and Elezi et al. (2), it is critical to clearly specify the perspective in health economics research, and to do one’s accounting at a corresponding level (14). This is not, however, sufficient. All such research should be extended to assess the impact on all stakeholders in the market and to undertake sensitivity analysis on the clinical and economic inputs that affect the measured outcomes. It no longer suffices to declare that a treatment modality is “cost effective” or “cost neutral” or “cost advantageous” without specifying: “For whom?” and “Under what circumstances?”
↵⁎ Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- Bakhai A.,
- Stone G.W.,
- Mahoney E.,
- et al.
- Elezi S.,
- Dibra A.,
- Folkerts U.,
- et al.
- van Hout B.A.,
- Serruys P.W.,
- Lemos P.A.,
- et al.
- Kaiser C.,
- Brunner-La Rocca H.P.,
- Buser P.T.,
- et al.,
- BASKET Investigators
- Cohen D.J.