Drug-eluting stents

[One of the authors responds:]

We appreciate the interest in our recent article.¹ We echo the concern of Liana Falcone and Navdeep Tangri that the cost-effectiveness of drug-eluting stents should be scrutinized; our desire to understand the variability in research conclusions prompted our study.

We thank Stéphane Rinfret and Erick Schampaert for their observation that publicly funded studies were of higher quality, but we find it difficult to reconcile this statement with their specious suggestion of a bias pertaining to authors' undisclosed relationships with government agencies as this completely lacks face validity. We can only speculate how they were able to identify such funding. We identified their study in our literature search but excluded it as it involved only a subgroup of patients with drug-eluting stents whereas our outcome was the recommendation (or not) of widespread use. Their assertions that the authors of publicly funded studies are unlikely to encourage widespread adoption of an intervention unless it is expected to save costs and allow responsible policy statements to be produced reflect a misunderstanding of the role of these agencies. Very few medical advances save costs; the metric for this form of health services research is not cost savings but value for investment. Moreover, such research seeks to inform policy-making, not usurp its role in decision-making.

Rinfret and Schampaert also worry that our quality rating was biased by knowledge of the studies' conclusions and source of funding. Our quality rating was based on the clear, unambiguous and objective criteria found in the appendices of our article. The 4 evaluators of the conclusions and 1 of the 2 quality evaluators were blinded to the source of funding, and there were few discrepancies among the evaluators. We invite others, including Rinfret and Schampaert, to validate our findings.

In addition, they state that as a consequence of our publication “the independence and validity” of the work of researchers with industry support is compromised, “even in cases in which the support is unrestricted and the research is performed without any direct input from the funder.” We had no way of assessing unrestricted funding and consequently made no inferences about this issue. The only ones questioning the independence of these particular health researchers are Rinfret and Schampaert.

We also appreciate David Cohen and Ameet Bakhai's clarifications that some of their articles were not original cost- effectiveness studies, but this seems slightly disingenuous as these articles were reported in the electronic databases and have been referenced by others. A reanalysis of our data with their additional information would strengthen our overall conclusions. Their comment about not requiring a statement about the source of funding reinforces our general message of caveat lector or caveat emptor.