Dr Ashish K. Jha Discusses Improvement of Bundled Payment Programs

Bundled payments are an interesting experiment, but they should have more clinically relevant time periods and include more quality measures, said Ashish K. Jha, MD, MPH, the K.T. Li Professor of Health Policy at the Harvard T.H. Chan School of Public Health and the director of the Harvard Global Health Institute.

Transcript (slightly modified)

Recently, you wrote about CMS’ bundled payment programs. What about the program has CMS gotten right and where do you see room for improvement?

I actually like the experiment with bundled payments, I think it’s a really interesting idea. In some ways you could argue we’ve been doing bundled payments for thirty years, because we started bundling all the hospital services. This is just an extension of that. Now we’re going to take a few conditions, and we’re going to extend it out to 90 days, and basically bundle that all together and hold the hospital accountable for the care that people get in that 90-day period.

So what do I like about it? I think it’s an interesting experiment. I think it has a real opportunity for eliminating some unnecessary services that happen in the post-acute setting. And maybe it has the opportunity for providing better integration of care during that time period.

The things that are funny about it, I mean, they chose 90 days for each of the three conditions. Why 90 days? We don’t really know. Is that clinically the relevant time period? It’s unclear. So one of the things that I’ve often argued is that, let’s make this clinically relevant and useful. Let’s not do 90 days for everybody or 30 days for everybody. Look at the underlying data, look at when that episode really clinically tends to end, and bundle that time period. That is going to be a much more effective and clinically reasonable and salient approach.

The other big issue that I think is really important is around quality measures. If you are getting a bundled payment, the easiest thing you can do if you want to save money, is just to deny people care. Don’t send them to a rehab, don’t send them for things that are going to be important for them. Obviously, most organizations wouldn’t behave that way, but you could worry that some organizations might, or on the margins some organizations might. And so the way to ward off against that is to have quality measures.

Of course CMS knows that, they have quality measures in their program. My big criticism is that I just think we should be doing better quality measures than we have. Right now we have for acute MI, we have mortality, I think that’s really important. We have this, excess hospital days which I think is an interesting utilization measure, really doesn’t capture quality.

But what we really need is patient-reported outcomes. Good organizations are going to help people have better functional recovery, have better pain levels at the end of 30 days, 90 days. There are a whole set of measures that we should be compiling. And yes, they are more complicated to measure than something like whether somebody’s been readmitted or not, but just because it’s harder doesn’t mean we shouldn’t do it. We really need robust quality measures if these bundled payment programs are going to lead to both better care and lower spending.