. >> let's hear from aaron lyss. >> i think matthew brought up a critical issue that i think has been understated throughout the morning, which is the importance of benchmark setting and target price setting and getting that right and having that be -- having those target prices and benchmarks be equitable. there was one element i didn't hear that i think is a very significant concern. i was wondering how the panel would propose addressing this. that is the degree of innovation in therapies and in diagnostics that we've seen over the past several years is fundamentally different than a lot of the data on which these models were based. and i think -- i'd be interested in hearing your perspectives on how to account for that. to me, it just seems like that is -- it requires a degree of technical expertise in various specialties provided by these that is hard to get and hard to account for in the design of these models. >> i'll share a couple of thoughts but to some extent, this is just a hard technical problem we don't have answers to. but i think there are two broad approaches you can t