Because fee-for-service payment is still so common in health care, there isn’t a lot of published data showing how—and how much—hospital procedure volumes could decline if risk-based payment methodologies become common.
We all know that fee-for-service medicine encourages utilization, driving costs upward and raising the potential for inappropriate care. Everyone seems to agree in principle that the country would be better served by ditching this payment methodology. Furthermore, everyone presumes that under risk-based provider payment approaches, patients will use fewer health care services than they do today under the standard fee-for-service regime.
But no one seems to know by how much.
A dangerous gap in knowledge
That’s a potentially hazardous gap in knowledge for health care providers, considering how fast many providers are rushing to embrace risk-based payment mechanisms—pay-for-performance, bundled payments, shared savings, partial capitation, and global capitation, just to name a few. The less health care providers know about expected utilization for a defined population, the less likely they will be able to get appropriate contract terms.
It could be devastating for a provider to agree to a capitated per member per month (PMPM) payment below historical norms, only to discover that it’s impossible to move the utilization dial downward. Conversely, if the provider knew it could reduce utilization quickly and significantly, it could accept a below-historical PMPM rate, knowing with confidence that it would be able to generate and pocket immediate savings.
So will risk-based payment contracts really be able to reduce utilization? History offers us some clues.
L.A., 1995: Capitation slashed cardiac volumes
Back in 1995, a group of Advisory Board researchers did a study of capitation contracting in the Los Angeles metropolitan area. During that era in California, primary care practices had consolidated and merged with other specialty practices to form large multi-specialty groups. A number of commercial payers then chose to reimburse these groups using global capitation—giving them a PMPM payment encompassing professional and facility (technical) fees for a broad swath of services.
With the global PMPM in hand, many of these multi-specialty groups contracted with other specialty practices for services they couldn’t provide themselves, paying the other groups a separate PMPM rate, an arrangement referred to a "sub-capitation."
In the course of our study, the research team tracked the procedure volumes of one four-person cardiology practice that was sub-capitated for medical, diagnostic, and invasive services, looking at the change over an eight-month period as they moved from fee-for-service to capitation. I’ve reproduced the results below.
With the shift to capitated payment, diagnostic catheterizations fell by 70%, and angioplasties (ballooning at the time) by 80%. And it wasn’t just volumes of certain procedures that dropped precipitously. Overall cardiology spending, calculated as a PMPM rate for professional and facility services, fell by 35%.
At the time, we found these numbers puzzling and shocking—and they remain so today. Rarely in medicine do you see such dramatic changes in practice over so short a period of time.
The forgone procedures: Unnecessary or appropriate?
What accounted for these precipitous changes? Honestly, we didn’t have a clue at the time, and we still don’t. But any way you cut it, the reduction in procedure volumes doesn’t bode well for someone, either physicians or their patients.
Perhaps these L.A. cardiologists decided to scale back utilization and closely follow professional guidelines. If so, their fee-for-service practice had been extremely aggressive in performing invasive procedures, either for financial gain, or in pursuing their own observation-based hunches about what worked and didn’t, or some combination of the two. Unfortunately, non-indicated interventions aren’t “free,” monetarily or clinically, given the risks of procedural complications and needlessly inciting the scarring (restenosis) response in ballooned coronary arteries.
Alternatively, they could have been performing appropriate procedures under fee-for-service, but once they had a financial incentive to do so, opted to employ less-expensive substitutes such as drugs and non-invasive imaging. If so, then withholding interventions would have denied patients the benefits associated with invasive cardiology.
We’ll never know what the true story was for this one practice, whose physicians have probably long since retired or dispersed. But revisiting this case study makes me think of two lessons we all should heed as the industry moves rapidly toward risk-based payment.
Two lessons for risk-based payment
This old case study reminds us that at least in certain circumstances, providers behave very differently when their payment incentives change—most especially when those incentives reward them directly for reducing patient utilization. That’s encouraging if our collective goal is to reduce unnecessary utilization and wasteful spending.
However, there’s also a caution here: since it’s so hard to tell whether less utilization is better or worse for patients, we need checks and balances on the impact of new payment schemes. Health systems and payers need to track outcome metrics, not just cost measures, to make sure that the changes in clinical practice that accompany payment innovations result in equivalent or superior outcomes.