Innovation is the lifeblood of successful businesses. “Moving fast and breaking things” can provide rapid access to new advances and great potential rewards. However, with innovation often comes uncertainty, and the consequences of product failure can be severe. Clever managers and regulators try to balance these inherent risks and benefits. It is a difficult task, and undesirable outcomes are both inevitable and highly publicised. For example, in the medical technology markets that we study, regulators are alternately accused of killing patients or killing innovation. But are we getting the balance right on average?

This is a challenging question to answer, and one that business and political leaders have grappled with in the context of recent regulatory reforms. In the United States, the 21st Century Cures Act pushed for greater access and more reliance on post-approval vs. pre-approval data on product safety and efficacy. In the European Union, recent rewriting of medical device regulations will require more clinical information from firms seeking to introduce new medical technologies to the marketplace. These policies are important to get right because the standard that a regulatory body imposes has the potential to fundamentally alter market outcomes by requiring testing that firms would not undertake themselves. In a recently published research paper, we develop a theoretical and empirical framework to measure these trade-offs: higher testing standards can create value through generating information that decreases uncertainty, but this benefit comes with the potential cost of fewer available products due to delayed entry and higher entry costs from more testing.

We use our framework to conduct an empirical case study of optimal regulatory policy for cardiac stents, a blockbuster device used to treat blockages of the coronary arteries. Important to the study is that, until recently, the US and the EU have taken very different approaches to regulating high risk medical devices. FDA approval in the US requires a large randomised clinical trial while EU approval can often be obtained with bench testing or a small trial with no control group. In large part due to these differences in regulatory approval, the EU sees entry of more innovative new devices than the US. US physicians have 11 stents available to implant while their EU counterparts have 39 from which to choose. 81 per cent of products (accounting for 23 per cent of stents used) in the EU never enter the US. Conditional on the product entering the US, EU physicians have access to the product almost a year earlier.

However, EU patients who may benefit from stent implantation also face greater performance uncertainty as stents enter with less evidence on product efficacy. A clinical trial has been published for only 20 per cent of EU-only available devices. In contrast, 85 per cent of FDA approved devices have undergone a published clinical trial. Also, conditional on publishing a clinical trial, average sample sizes for the FDA-approved devices are 1,313 patients versus 280 for the EU-only devices. This extra evidence comes at a cost as the additional subjects are associated with an extra 9 months in trials (due to subject recruitment time). This time is costly in terms of delayed access for patients as well as raising fixed costs of entry

In order to assess and quantify the trade-off between device performance information provision and entry costs, we study detailed hospital-stent level stent usage and pricing in the EU and the US, linked to clinical performance data. We find significant variation in stent quality across stents, implying large returns to early product testing. Our estimated model predicts that without any EU testing, the market for stents would shrink significantly – without proper testing to valid product quality, innovations are not adopted. Further, the required US testing in excess of EU requirements substantially decreases the uncertainty of using an inferior product and thus significantly increases consumer wellbeing. It also implies that the EU enjoys positive spillovers from US testing – if US testing were equal to the EU, welfare in the EU would decline by 6.4 percent.

The heart of the paper is a calibration of the optimal regulatory policy that balances risk versus access to new devices. Our estimates imply that EU surplus would increase by as much as 3-7% by requiring more pre-market testing for stents. Indeed, total societal welfare is maximised when the premarket trials are at least 16-17 months longer than the old EU requirements. Thus, at least when it comes to the historical experience with cardiac stents, our estimated model suggests that the increased clinical requirements recently imposed in the EU are a step in the right direction.

Extrapolating from these results with stents to broader policy conclusions should be done with care. In this direction, we conduct optimal policy calibrations under a variety of alternative model parameters. We hope that our results and framework provide a starting point that, with modifications as appropriate, can be used to think more carefully about getting regulatory policy for innovative new products right.

♣♣♣

Notes:


Matthew Grennan is an assistant professor of health care management at the University of Pennsylvania’s Wharton School and a faculty research fellow at the National Bureau of Economic Research. His research and teaching touch on firm strategy and public policy issues in innovation, product adoption, and pricing in medical technology and pharmaceutical markets.

 

Robert Town is the James L. and Nancy Powell centennial professor of economics at the University of Texas and a research associate at the National Bureau of Economic Research. His research focus on the industrial organisation of health care markets examining competitive behaviour across providers, insurers and medical technology sectors.