It seems that everywhere you look “value-based purchasing” (or value-based care) is hailed as the savior of U.S. health care. With it, comes the promise of a system that pays for quality rather than volume. Following the money tells the same story – cloud-based EHR giant Athena Health, for instance, just beat Q2 revenue estimates by about $3 million, and saw revenue up nearly 27 percent from the same period last year. More broadly, hospitals appear to be investing heavily in tracking the value of the care they deliver: a Leap Frog Group survey notes that the number of hospitals using computerized physician order entry has increased from 10 percent in 2009 to 43 percent in 2013.
If nothing else, however, the prospect of value-based care is an appealing message. After all, for decades, American health care has been defined by opaque prices far removed from actual costs, a fragmented care delivery system, and for the most part, a dearth of information about what we actually get for our money.
First, it pays to understand what value-based purchasing means. While hundreds of pages can be (and have) written on this subject, we'll stick to a very simple definition from Deloitte:
“[VBP is] a payment methodology that rewards quality of care through payment incentives and transparency.
In short, value-based payments are those that offer incentives for quality and transparency, while offering disincentives for overtreatment and opacity. Value-based care, then, is just an extension of this – in theory, all medical care has “value” (though the marginal value of some care can be zero or even negative). But the idea is that value-based care responds to incentives created by value-based payments, to maximize the value of care delivered to a given patient.
In recent years (and months), we've finally seen examples of value-based payments and care – and while the results are mixed, it appears that both providers and payers are starting to recognize and invest in the coming shift.
For instance, a recent survey found that 90 percent of payers and 81 percent of providers are already using some form of value-based purchasing. Both expect fee-for-service to decline dramatically over the coming years. A 2013 scorecard from the Catalyst for Payment Reform noted that around 10.9 percent of all commercial in-network payments were “value-oriented,” putting us more than halfway to their target of 20 percent by 2020. Q1 GDP numbers were also, at least partly, depressed because hospitals actually lost revenue. Causality is tough to prove, but there's some reason to think that hospitals are making future-oriented tradeoffs. Here's former OMB director, Peter Orszag:
“One factor is that the systems are being digitized, so hospital executives have a much better sense of what's going on within the hospital and can target waste. The second thing is that the payment system is changing, and it's expected to continue to change.
If hospitals and other providers (who bear significant risk in a move to value-based payments) are starting to change current business models in anticipation of more value-based payments, it's hard to argue against such a trend. Moreover, the ACA's implementation of accountable care organizations and patient-centered medical homes puts even more pressure on providers to adapt.
It's instructive to remember, though, that the move towards a value-based health care system isn't new. And going forward, there are many “lessons learned” that can inform us about what works and what doesn't – from both the commercial and the public sector.
In the public sphere, one of the shining examples has been Medicare Advantage (MA) – a program through which Medicare beneficiaries can purchase private insurance coverage as an alternative to the traditional government-run program. Under MA, plans bid to provide coverage for Medicare beneficiaries, and are paid a fixed rate per month for each enrollee. Among MA plans, the most popular tend to also be the most “managed” – HMOs make up about 70 percent of total MA enrollment, and 19 percent of total Medicare enrollment. These plans also tend to be the ones that try to implement the most value-based payments – with some even using a fully-capitated model, paying hospital systems a fixed monthly amount to take care of beneficiaries. While data is somewhat sparse, existing studies have found improvements in health outcomes, reductions in hospital readmission rates, and spillover benefits to traditional Medicare.
MA certainly isn't perfect – as it stands, the program spends more money per beneficiary than traditional Medicare would. Whether better outcomes are contingent on spending extra money isn't clear – but under the ACA, these additional payments are set to decline over time. Nevertheless, the program has demonstrated that paying for value can exist in the public sector, too.
Private Sector – Kaiser Permanente
But value-based payments have also permeated the private sector – perhaps best exemplified by a Kaiser Permanente, a California-based hospital system that has made commitments to value and transparency.
The California-based hospital system and insurer, Kaiser Permanente, is often help up as a gold standard, and rightfully so. One study found that with improved “a comprehensive delivery system redesign and expanded and integrated existing clinical information systems, decision support, work flows, and self-management support,” Kaiser's Southern California branch averaged a 13 percent improvement on 51 health metrics, versus a median 5.5 percent nationally. And Kaiser is also credited with not-insignificant health care savings – with a disease registry program saving about $30 million annually.
Critical to Kaiser's success has been significant investments in health IT – with a powerful electronic health record system (KP HealthConnect) that helps avoid duplicative testing and treatment, while also helping to shine a light on inefficiencies in care. The lesson here is about the power of EHRs designed deliberately for improving care rather than simply paying lip service to an IT trend.
Lessons To Be Unlearned
Despite some shining examples of success, value-based payments have a nastier side as well. Most notorious perhaps is the uncertainty about whether Managed Medicaid (a form of Medicaid contracting where states pay managed care organizations to provide care for beneficiaries) has managed to deliver any cost savings or improvements in health outcomes. In a synthesis of numerous studies, Michael Sparer of Columbia University explains:
“The peer-reviewed literature finds little savings on the national level…There is some evidence that Medicaid managed care improves access to care, but the scope of the improvements is state-specific and depends on how access is measured.
Indeed, there is enormous variation in state Medicaid programs, which could help explain some of the variation in results. But overall, Sparer notes a number of factors that limit the potential for cost savings or improved outcomes from Managed Medicaid including: the fact that Medicaid already has very low rates and that increasing access to care may lead to higher costs itself. When it comes to outcomes, Sparer focuses on the difficulty of attributing health outcomes (which are functions of a web of complex factors) to any one particular cause. But in assessing coverage that requires significant upfront investment – any value-based system will incur a high fixed cost – it would seem that “stickiness” or volume are crucial. And the Medicaid market is generally more volatile than the private insurance market or Medicare. According to the 2010 Medical Expenditure Panel Survey, for instance, about 86 percent of people with private insurance had coverage throughout the year, while only 68 percent of those with Medicaid coverage retained it for all twelve months. Not only is this stickiness important for offsetting the cost, but it is also very difficult to manage the health of a population that bounces in and out of coverage.
None of this is to say that value-based care isn't possible in Medicaid – it is. But for the Medicaid population, managed care may not be the right solution.
The Healthy Indiana Plan, for instance, gave beneficiaries a health savings account with which to pay for initial medical costs, but ensured that preventive services were covered with no cost-sharing. Unfortunately, Medicaid's very low reimbursement rates, as Sparer pointed out, may be a long-lasting obstacle to better value – improving beneficiary outcomes may very well require increasing reimbursements in the future.
Outside of broad government programs and particular hospital systems, there are many other highly successful examples of unique approaches to wrangle more value out of the health care system.
In what is becoming a highly-cited example, CALPERS, the California public employee retirement system, started using “reference-pricing” (a flat rate) for hip and knee replacement surgeries. Members were provided with information on hospitals that would cover the surgery at the reference price, but were given a choice of hospitals – members were free to go to higher-cost hospitals, but the cost above the reference price would come out-of-pocket. Analysis of the experiment has found savings of about $2.8 million in one year alone, and a significant shift in volume from high-cost to low-cost settings.
The narrow network plans available on the ACA's public exchanges are also becoming an important attempt to pay for value. Many plans, even PPOs, have excluded many higher-cost providers from their networks to be able to price more competitively. You might think, of course, that using lower-cost providers would result in worse outcomes. But a McKinsey study noted that there was no difference in quality between hospitals participating in broad versus narrow networks. Cost differences were significant, though – with savings between 13 and 17 percent depending on the plan tier.
A report from the BlueCross BlueShield Association also details various efforts to reduce spending and improve quality through value-based payments. One PCMH initiative offered fee increases based on performance, and actually saw savings of nearly $100 million. Another initiative “[implemented] key capabilities like offering after-hours access to care, implementing processes for following up on test results, and using registries that flag care gaps,” resulting in fewer ER visits and lower use of high-tech radiology services.
Given that, according to PWC, there are about 600 public and private ACOs around the country covering 18 million people – we're likely to see even more experiments with value-based payments. These experiences will be important for understanding what works and what doesn't.
It's clear then, that value-based payments will become an ever-more important part of the health care system. But what effects should we expect to see? Will spending decline or at least continue to grow slowly?
Maybe. There is no doubt that the U.S. health care system is rife with waste and inefficiency, with low-end estimates at around 20 percent of total spending. Weeding out the waste will make the system more efficient, but there's only so much waste you can eliminate. Some “steady-state” level exists, beyond which, eliminating waste may cost more than it's worth.
But there are also reasons to think that value-based payments may increase spending. For one, value costs money. And while there's little correlation between cost and outcomes in health care, the opportunities for actually cutting spending while improving outcomes will likely be the relatively low-hanging fruit, and many of these savings are likely to be small in magnitude. For instance, investments in health IT like EHRs will cost money. And while EHRs may help to eliminate duplicative testing and help clinicians make more effective treatment decisions, it isn't clear that this will actually reduce spending. For people who don't receive enough treatment, EHRs can increase spending. Moreover, as RAND researchers point out, poorly designed EHRs can “act more like a “frequent flyer card” designed to enforce customer loyalty to a particular hospital, rather than an “ATM card” that enables you and your doctor to access your health information whenever and wherever needed.” Even worse, providers with less than noble intentions can use EHRs to the system and upcode patients, making them look sicker than they really are, and raking in higher payments.
There are, of course, going to be cases where higher spending may not be pernicious – instead, it may deliver significant value to patients. For instance, Gilead Sciences' hepatitis C drug, Sovaldi, costs about $96,500 per treatment – significantly higher than prior treatments – but costs only a bit more per “sustained virologic response” (a proxy for cure rates). The benefit to patients that avoid a liver transplant (or just even the terrible side-effects of interferon) can be huge, and well above the pecuniary cost of the drug. But using the drug will almost certainly increase health spending (at least in the short run), even though it may very well be “value-based care.” Similarly, a hypothetical cure for cancer would be enormously expensive – but the social value would likely be significantly greater, estimated at around $50 trillion by University of Chicago researchers. Again, this would represent an increase in health care spending, and would likely cost more than a round of chemotherapy – but would deliver significantly more value.
When it comes down to it, however, the success or failure of value-based payments should not be judged solely on a cost-savings basis. This would be a mistake. If we can vastly improve outcomes with extra spending, such an investment may still be warranted.
This piece originally appeared in Forbes