By Cindy Atoji
Feb. 19, 2008 | On the eve of the annual national P4P conference, Feb. 27-29, Tom Williams, a major pay-for-performance backer, admits that results for physician incentive programs have been mixed. Williams, CEO of the Integrated Healthcare Association (IHA), runs California’s P4P program, the largest in the country. The non-profit group is backed by all the state’s major health plans and involves over 40,000 physicians. Digital HealthCare & Productivity spoke with Williams about the challenges of implementing P4P and the use of information technology, a key performance measurement.
DHP: In your opinion, is P4P working?
Williams: The results have been mixed. On the positive side, we’ve had significant engagement from the physicians and they’ve worked hard to improve the results. We’ve had a number of successes, including the collaboration of plans around a uniform measure set and the use of aggregated data, which has been very powerful. When physicians know an entire patient population is being measured, they’re much more confident in the validity and reliability of the results, so that has been a big plus. We’ve been able to publish, through state of California, a single report card using aggregated results.
We’ve had good reimbursement in terms of incentive bonuses, including over $55 million a year in payments.
But the disappointment is that to some degree we’ve seen what some people describe as “teaching to the test.” There’s improvement in things we’re measuring but we haven’t been able to translate that to a fundamental change in outcomes. We’ve seen incremental improvements in clinical results, but nothing breakthrough.
And health plans have raised concerns that we’ve haven’t demonstrated any change in cost, so that’s become a big focus of ours.
DHP: Research evidence has not been very strong in support for P4P, but program administrators and participants continue to be enthusiastic. So the question is, who’s right?
Williams: That’s one of the big questions. Are the researchers right and we just haven’t caught up with it out in the field, or is the field right and researchers haven’t caught up with it? The sense from those on the ground is that it’s somewhere in between. Some programs are getting significant results but haven’t had enough time yet to publish these favorable outcomes.
I believe generally the results are incremental; they’re not dramatic. A big piece is how much incentive is put behind a program, and bigger incentives will see bigger effects. Measurement science is just catching up, and as we’re able to measure things in a way that’s easier for the physicians, we’ll have the ability to move the dial more. I think we’re going to see more positive research come out in the next three years, but it’s going to continue to be mixed. Often it will be moderate results versus breakthrough.
DHP: What is the work that remains to be done?
Williams: One of the big findings has been dramatic regional variations, so when you look at our results regionally, there are communities where the dial is moving little, and others where it’s moving a lot in the positive. One of the things that needs to be done is a more comprehensive focused attention on those geographies. Just putting measures and bonuses out are not going to be enough.
DHP: Last summer you reported that physician groups with the best record of implementing information technology also had clinical quality scores 18 percent better than groups that did not meet IT standards. Could you elaborate?
Williams: There has been a fair amount of bonus dollars paid toward IT adoption. We’re seeing a strong consistent correlation between groups who score well on clinical quality; they also have good results on information technology. But what we can’t say is that information technology is driving the improved clinical results, because there are other factors that we’re not measuring, such as leadership and reimbursement. But clearly, groups that are doing well in clinical quality are doing well in IT implementation. And that continues to be demonstrated.
DHP: What information technologies are physician groups adopting, and which are having the greatest impact on clinical quality?
Williams: In our case, we have incentives for 10 different types of information technologies, including population management and point-of-care technologies such as electronic prescribing or electronic checking for drug indications. On the population management side, the biggest uptake we’ve seen is in patient registries.
On the point of care technology side, we’ve been surprised that the ePrescribing uptake was less than we anticipated. What we learned is that historically, ePrescribing has been harder than people anticipated, but I think we’re going to see that change as government has really pushed on ePrescribing.
The most dramatic improvement has been electronic access to lab results. This has been really helpful in disease and population management. We’ve also seen uptake in electronic messaging between patient and doctor, electronic access to clinical notes, and electronic retrieval of patient reminders.
DHP: What is the future of P4P? Some suggest that it is only a temporary tool for improving quality.
Williams: Is P4P a transitional tool or something more fundamental? The one thing we’re seeing is that P4P is building a foundation for performance measurement. Ultimately, we need fundamental payment reform. Right now we pay for activity, and not for outcomes and we have to find ways to pay for the outcomes we want, not just more activity. And you can’t do that if not measuring things. I see this as major stepping stone to something more fundamental, something more dramatic.
DHP: How can organizations achieve a real return on investment (ROI) from P4P?
Williams: Many of the measures have been around primary care and preventions: immunizations, mammographies, and cancer screenings. It’s hard to show an ROI on these and it’s hard to adjust for all factors that might affect the outcome. But as we get more directly in the space of true outcome measures then we can more clearly show ROI.