Comments

Health Affairs encourages readers to engage in discussion via comments on our Web site.

• To RESPOND to a particular article: Click on the link "Submit a response to this article" in the box at the top right-hand corner of the article.
• To READ responses to a particular article: Click on the link "View responses" in the box at the top right-hand corner of the article.

Comments to:

Shelby D. Reed, Kevin J. Anstrom, Damon M. Seils, Robert M. Califf, and Kevin A. Schulman, Use Of Larger Versus Smaller Drug-Safety Databases Before Regulatory Approval: The Trade-Offs, Health Affairs Web Exclusive, August 5, 2008 [Abstract] [PDF] [Appendix and Disclosures] [Reprints & Permissions]

Comments:Submit a response to this article

Comments published:

Hidden Costs To Increased Sample Size

Stephen A. Williams   ( 19 August 2008 )

Hidden Costs To Increased Sample Size 19 August 2008
Stephen A. Williams, Partner Decisionability LLC Send comment to journal: Re: Hidden Costs To Increased Sample Size stephen.a.williams{at}decisionability.com Stephen A. Williams	I applaud the approach taken by Reed et al. in their investigation of the cost-effectiveness of larger vs. smaller safety databases. This type of evaluation is, in my opinion, far preferable to an emotionally based opinion by committee. However, there are several issues that lead me to question the results of the specific cost-effectivness ratios. First, the cost of increased sample size in the paper appears to represent only the per patient costs of clinical trials. There are other substantial costs that should ideally be included. Larger trials and larger NDA submissions are almost inevitably slower than small ones, and therefore the benefits to patients for novel drugs are delayed; this factor should be included in the costs, and as it affects every program (not just the ones with safety issues), the disease costs to patients could be substantial. Second, raising the cost of a drug program has an effect on the economics of drug development, leading to fewer medicines, most obviously for smaller medical populations where some drugs may never be put into development if the trial size requirements are too high. This is difficult to model, maybe, but nonetheless another substantial cost that is not included or mentioned. Finally, there is a limitation in deducing rates of societal tolerance of safety issues from prior decisions that have not included cost-effectiveness evaluations. Both Vioxx and rotavirus vaccines were mentioned as helping to determine societal acceptance of safety issues -- but certainly for rotavirus, and possibly also for Vioxx, had a cost-effectivness appoach been used, it would have been highly positive in favor of the therapy. It seems a bit of a paradox to use numbers from opinion-based decisions to determine the limits for cost-effectiveness based ones. If drug regulation moves towards a more quantitative approach like the one advocated in this paper, then perhaps people will tolerate different rates of adverse events because they can have the confidence that a competent calculation has been done and that -- on average -- the benefits outweigh the risks.