I'll tackle that as two aspects, and I'll pull them apart a bit.
One is whether we should be going after all the places where there could be risk or whether we take a more risk-based approach. I think that's the question you're asking. The other is a specific methodology to project what we think the amounts of the ineligible payments were. I'll leave that aside for future questioning.
In terms of partially disagreeing, our perspective on it is we generally—almost exclusively—take an approach within the agency to focus on a risk-based assessment. We take a look at the information that we have at our disposal, try to identify where the highest risks are and go after those. It's the highest risks and potential for recovery, and not covering 100% of the people.
There is an approach where you could go after 100% and uncover everything, but we take that risk-based approach and we amend it as we move forward. As we develop evidence about what worked and what did not work in previous attempts, we adjust our risk parameters based on business intelligence that we have. That may be leads that we have or some of the information we have at our disposal.
I think that's the core of the issue, in terms of partially disagreeing. We didn't want to commit to reviewing every claim that came in.