Looking at Pay For Performance Across Industries

HCI: In all cases, you make recommendations, among them, that leaders of these PBASs should create the right units of accountability, make the rewards or penalties big enough to matter without exceeding the value of improved performance, and integrate the PBASs with existing performance databases and accounting and personnel systems. What might that mean, for example, in the healthcare context?

Stecher: The recommendation that comes first to my mind is that these systems shouldn’t be static; that the accountability system itself needs to be live, and that you need a mechanism in place for revising and updating it. Because if it’s successful in changing behavior, people will top out on the [performance] measures under the best circumstances, and won’t have any incentives to further improve. So you need an annual or more-frequently-than-annual mechanism for modifying the system, and then need to modify it as a result. So you can’t really think of this as a turnkey kind of thing. It’s hard, though, to get policy consensus; and nobody wants to go back and revisit it. One of the things we noticed in healthcare is that in a lot of these pay for performance systems, the incentives are relatively modest compared to overall reimbursement; and as a result, the changes are relatively modest. So it would seem that there’s not enough there to really grab attention.

HCI: Can you think of any lessons from other industries that healthcare leaders might want to consider?

Stecher: I can’t point to a specific case where I’d say, go look at how the transportation industry handles things differently. But I would say that more money is put at risk in the transportation and construction industry relative to overall payment. I could point to a negative comparison with education, where a lot of main incentives in No Child Left Behind are sanctions and interventions, and those have been shown to be less than effective, because they’re negative. They’re beginning to be effective in eliminating poor performers, but it’s not clear that we can replace them with better performers.

HCI: Any overall thoughts on the data and IT challenges involved?

Stecher: We noted that there are times when the accountability system itself might do a better job if it focuses on inputs rather than outcomes. And IT is a good example. If you think medical records are important, then to get people to invest in them, you might have to build incentives for them to do that, and of course, that has now taken place in healthcare [via the HITECH Act]. And you see some of that in public emergency preparedness, because you don’t have very many cases, and it’s hard to predict what will occur; but you can measure stockpiles of supplies, and you can do the tabletop or live simulations of exercises. So they’re you’re essentially, rather than measuring outcomes, measuring inputs or processes.

HCI: Any other thoughts?

Stecher: I think a continuing conversation across sectors might provide potential value. Right now, we haven’t figured out a forum for doing that. But those of us who participated thought it was remarkably insightful. We did a conference, and maybe we should do another.