Why Value-Based Care Is Not A Win-Win For Everyone
"Then the game is already rigged against those safety net institutions, and preexisting differences end up influencing who wins and who loses." -Suhas Gondi, M.D., M.B.A.
Summary:
Value-based care is a payment model where patients get billed for the value clinicians provide based on the patient’s healthcare outcomes rather than the sheer volume of treatments. Value-based care is meant to be a "win-win" for patients, providers, payers, and the government because, ideally, clinicians are spending less money on treatment and focusing more on outcomes.
Dr. Suhas Gondi thinks there are probably some areas where value-based care works well, but to think of it as a panacea for our real problems in the American healthcare system would be unwise. For instance, we should just get serious about tamping down on the price inflation that occurs, particularly in the hospital sector.
Value-based payment programs tend to more likely penalize those providers that take care of disadvantaged folks. Dr. Suhas Gondi also thinks that a lot of problems facing disadvantaged patients are far more fundamental and will require not just continuous health reform but actual new economic and social policies that strengthen the safety net, narrow income and wealth inequality, and lead to a reversal of generational disinvestment in certain communities in our country.
Terminology:
Value-based care: Payment model where patients get billed for the value clinicians provide based on patient healthcare outcomes rather than the sheer volume of treatments. Specially, where physicians rank on the value and cost of care compared to their peers.
Approaches as defined by Maitreyee Joshi:
There are several broad approaches to value-based care, each with different levels of provider risk. The broad approaches are:
Pay for Performance: Providers are compensated more if certain performance metrics are met.
Shared Savings/Shared Risk: If providers hit their assigned financial and quality metrics, they get a percentage of the savings; else, they have to pay a percentage of the losses.
Bundled Payments: Providers are paid a fixed amount to treat a patient for a specific illness, condition, or medical event (e.g., a hip replacement).
Capitation Model: Providers are paid a fixed amount per patient per unit of time to treat a patient for all their conditions.
Full Risk Insurance Model: Providers and payers integrate.
For a more detailed look into value-based healthcare, look here.
Accountable care organization (ACO): An alternative payment model which cares for a population of patients and shares in any savings it produces by providing care efficiently. In the risk-bearing variety, the ACO might accept a lump sum payment based on the number and complexity of patients in the population. If the cost of caring for the ACO’s patients exceeds the payment for that year, the providers have to swallow the losses – this is the risk they take on in order to access shared savings and other bonuses.
For greater detail on ACOs look here.
Risk adjustment models: Is a methodology that equates the health status of a person using diagnosis codes to a number, called a "risk score," to predict healthcare costs.
Center for Medicare and Medicaid Innovation (CMMI): The Center for Medicare and Medicaid Innovation is an organization of the United States government under the Centers for Medicare and Medicaid Services. It was created by the Patient Protection and Affordable Care Act, the 2010 U.S. health care reform legislation.
ACO Realizing Equity, Access, and Community Health (ACO REACH) model: Developed by the CMMI to improve the quality of care for people with Medicare through better care coordination and reaching and connecting health care providers and beneficiaries, with a particular focus on improving health equity by including those beneficiaries who are underserved.
“Gaming” the healthcare payment model system: Picking the right metrics within a value-based care system which give you better scores and give you better payments, without necessarily improving care.
Safety net institutions: A safety net hospital is a type of medical center in the United States that, by legal obligation or mission, provides healthcare for individuals regardless of their insurance status or ability to pay. This legal mandate forces safety net hospitals to serve all populations.
The Problem
How do we bill patients in a way that focuses on providing patient care that actually results in better health for the patient while keeping costs low?
Traditionally Medicare pays providers solely on a fee-for-service model. The more patients get sick, and the more they go to the hospital or emergency department, the more money providers make. The unfortunate side effect of this payment model is that it can incentivize providers to give way more treatment than is necessary. Value-based care is an emerging alternative where providers get paid based on patient healthcare outcomes rather than the sheer volume of treatments.
More recent work, however, has called to attention that value-based care for patients enrolled in Medicare has failed to meaningfully reduce health care expenditures or improve the quality of care. Having a particularly negative impact on low-income and marginalized populations. However, a new model called ACO Realizing Equity, Access, and Community Health (ACO REACH) was developed by the Center for Medicare and Medicaid Innovation, which aims to fill the gap that traditional value-based care has on low-income and marginalized populations.
To understand value-based healthcare and its impact on disadvantaged patients I had the chance to talk with Suhas Gondi, M.D., M.B.A., the lead author of a recent publication in the New English Journal of Medicine that points out the gaps value-based care has left for failing to explicitly consider equity and the impact ACO REACH can have. Dr. Gondi is a resident physician in internal medicine at Brigham and Women's Hospital committed to using policy and innovation to redesign the American health care system.
This conversation has been edited for length and clarity.
What is misunderstood about value-based care?
Brociner: What do you feel is the most widely misunderstood concept of value-based care?
Dr. Gondi: I think I would have to pick this notion that, we are moving from a fee-for-service system towards a value-based care system. And that with every step that we take towards alternative payment models and towards tying payment to quality or to cost, we are moving away from a fee-for-service system, and I think that's not necessarily accurate. I think most people understand at this point that there is a general transition effort that many people are attempting to make happen. But what is often misunderstood is that the vast majority of our value-based programs, in Medicare and Medicaid, as well as in the commercial markets, are still built on a fee for service. Meaning that for instance there may be an incarnation of bonus and penalizes based on performance on quality metrics.
While that does begin to incentivize performance on those metrics, it still preserves the base payment, which is determined by a fee for service system. So, you know, we're talking about one to 2% increases or decreases based on performance, but whether that's really changing the incentives in the system, I think, is much less clear. And I think in many cases, we still preserve the same incentives for over utilization. I also think that, and maybe this is kind of a separate point, but I think the issue with that is that we think we're making a lot of progress in payment reform, but in fact, we're actually just preserving the same incentives and increasing the complexity of these programs, which require hospitals and practices to hire more people, hire consultants, get more technical expertise to help them, you know, essentially gain these metrics and these systems.
And so I think we're masking the true progress we're making while actually creating new costs in the system, which may be a potential source of noise. And then I think the other thing I would point out is that, for those of us who are super engrossed in the system, we're often tracking some of the same metrics, like how much of Medicare payments are now tied to alternative payment models. I think that that has become the goal. The goal has become to move as much payment into risk-based models as possible without enough attention to whether doing that actually drives the outcomes that we care about. The evidence that we have in the space is actually very poor to demonstrate that the underlying rationale behind value-based payment, which is that more upfront investment in primary care and preventive care, actually reduces ED utilization and hospitalizations in the future. It's actually really quite weak. And I think we've, instead of indexing on the things that matter most to us, like better health outcomes at a more achievable price, we're now indexing on something upstream of that, which is more and more value-based payment. But the link between these two things is still not clear. And we've kind of assumed there's a causal chain there, whereas in many cases, I'm not sure there is.
Is there a causal link between value and outcomes?
Brociner: Has there been or is there work going on to connect a causal ink between delivering value and health outcomes?
Dr. Gondi: I think it would be great to, and there's been some amount of it, but the issue is that this is a hard thing to study unless you really randomize. There are just so many sources of bias. Most of the data we have tends to be observational or has other limitations that tend to preclude causality. And so I think it's just that it's a really tough area. And at this point, so many people are so deeply invested in the system that I think there'd be some resistance to honestly testing a lot of these, a lot of these models.
And, for sure the CMMI has done a good job of having some regular rigourous valuations, but, outside of their models, there's been relatively little. And I would say that the evidence from their models, on the whole, when you look at their programs as a portfolio, is actually not super promising. ACOs I would point to that as probably one bright spot where we're seeing savings that have grown over time. But for most other programs, it's kind of unclear whether they actually save more money than they cost.
Is Value-Based Care the best direction?
Brociner: You seem pessimistic, I would say towards the future value based care, is that kind of the case?
Dr. Gondi: I think there are probably some areas where it works well, but to think of it as a panacea for our real problems in the American healthcare system, particularly issues around health equity, which have really come to a head, I think it would be unwise. If our goal, for instance, is to reduce the amount of money we spend on healthcare, there are much more straightforward ways to go about that than devising complex payment systems. For instance, we should just get serious about tamping down on the price inflation that occurs, particularly in the hospital sector. This is an area where there's been a lot of work, a lot of work done to demonstrate that one of the biggest drivers of healthcare spending in this country is the increase in hospital prices.
If our goal is greater health equity, for example, a value-based payment model with a complex kind of ACO structure with benchmark adjustments that try to take into account health equity could be one way to advance the cause or at the very least protect marginalized populations from unintended consequences. But there are also much more straightforward ways to have the same impact. For instance, we should reduce payment disparities, right? Medicaid pays much, much less than Medicare. And then in commercial markets, if you look at how on a unit basis, when you think about safety net institutions that take care of a lot of Medicaid patients, those institutions are getting a lot less money than institutions that take care of less disadvantaged populations because we're paying different amounts of money for the same services. And so those sorts of things, narrowing those disparities, is a very straightforward way to help solve some of those issues that doesn't involve the complex methodology involved in most value-based payment contracts.
Legislation work arounds?
Brociner: Do you believe that many of these complex payment systems are simply a way to avoid much needed health-related legislation, such as healthcare price reform?
Dr. Gondi: Yeah, certainly the things I mentioned are politically more challenging. I think so. I mean, value-based payment is sort of a seductive ideology, right? Because it makes it seem like everybody can win, where, you know, patients win, providers win, payers win, the government wins, because we're spending less money on all these things. I just think the evidence for that is not very strong and it's probably only applicable to kind of narrow applications, as opposed to, you know, the way that we're going to totally reform our healthcare system and solve disparities, improve health outcomes, and put us on a track of greater financial sustainability.
Those things are more politically challenging to do. The things I mentioned before are more politically challenging than, you know, this sort of seductive narrative around, "oh, let's just invest more upfront and then it'll save more money later". These things that kind of intuitively make sense to people are very, very powerful. And also, they don't demand that anyone lose, right? They don't demand that hospitals necessarily lose or that doctors necessarily lose. Right. So the AHA[American Heart Association] and the AMA[American Medical Association] are on board with it and are very, very supportive. In ways that they wouldn't be of potentially narrowing payment disparities or, you know, some movement towards price controls. Which, which I think, by the way, actually does a lot more harm to patients, to communities, and to our society at large than, you know, the waste that's in the system that we're trying to get rid of.
Why can value-based care hurt unrepresentative patients?
Brociner: What was the motivation behind writing the New England Journal of Medicine publication and what are the key findings from it?
Dr. Gondi: Yeah, so we've had a whole range of value-based payment programs over the last 10 years. Most of it started with the ACA, and we're now at a point, almost 10 years later, where we have some data to help see and observe what happened as a result of those programs. And when we reviewed the literature that looked at the impact of these programs on certain populations and on certain clinicians, it became clear that there was a consistent trend, which was that these programs tend to more likely penalize those providers that take care of disadvantaged folks than those who take care of less disadvantaged folks, in essence, because a lot of these programs are budget neutral. What they end up doing is transferring resources away from the safety net institutions and the safety net practices towards the well-resourced clinics that take care of more privately insured people. It's essentially taking money away from BMC [Boston Medical Center] and giving it to the Brigham [and Women's Hospital] and Mass General [Hospital].
Why does this happen? That's the really important question. And I think kind of the primary driver or the root cause is that equity wasn't really prioritized during the design of these programs and that the budget neutral component of them is one reason. Why is that the case? If a budget neutral program, by definition, is going to create winners and losers, right? It's going to reward some people at the expense of others. And when the people who are entering that game are not equal at the beginning, right. We know that safety net institutions have suffered disinvestment and gotten lower payment rates for decades relative to other institutions. Then the game is already rigged against those safety net institutions, and preexisting differences end up influencing who wins and who loses. Now, there are a couple of kinds of material ways that happen. One is, as I alluded to before, that these programs are complex. They have a lot of reporting requirements.
They have a lot of logistical requirements, and oftentimes hospitals, and you know this from being at [Boston] Children's Hospitals, health systems, and others, will need to hire consultants. They'll need to get technical expertise. They'll need data scientists and others to help them figure out what to report. So how do we be as successful as possible in these programs? How can we be compliant with the new quality reporting? That is required. Those are all the expenses that are incurred. Some institutions can afford to spend more money on these expenses than others. And those institutions that do have those resources will be more likely to be successful in those programs because those programs are inherently susceptible to gaming where picking the right metrics will give you better scores, which will give you better payments.
And so, that's one reason why having more resources at the beginning will help you become a winner in ways that just happen to be due to preexisting differences as opposed to how you actually performed in the program or how well you're taking care of your patients. Another way and there are a few others that are covered in the paper. I think the other one that I'll highlight here is that our risk adjustment methodology is still quite immature. You know, we've been at it for a while. Risk adjustment is really critical in these sorts of programs because you don't want to penalize the providers who take care of sick folks who are more likely to have worse outcomes and then penalize them for those worse outcomes, especially if you're comparing them to other folks who, compared to other providers who just take care of healthier people, are therefore more likely to have good outcomes.
Risk adjustment models are intended to account for those differences and prevent unfair penalties or bonuses. But we've long known that our risk adjustment methodology is not fully complete and is not, um, completely accurate. And part of the reason for that is that it focuses more on medical risk factors and doesn't focus much on social risk factors yet. And so, in that way, it's immature. And so, people who take care of more socially disadvantaged populations are less likely to have that aspect of the risk of their population considered when making payment adjustments or when assigning markers of how well you're performing. And so, those are some of the reasons why they have had this kind of regressive effect of transferring resources away from disadvantaged populations.
Data to impact?
Brociner: What's your perspective on ACO REACH’s mission to collect social determinants of health data but then lack of clarity on how to use that social determinants of health data to have an impact?
Dr. Gondi: 100%. Yeah, And that's a conversation that's happening. In a lot of places now, there's a push to increase screening, to ask patients about health-related social needs, and to collect that information. I think there are actually really good concerns. I have real, very reasonable concerns that if we can't impact those health-related social needs, then we shouldn't be collecting the information in the first place. I understand those concerns given how many questions and checklists are already imposed on doctors and patients. though I still do think it's important. And for a few reasons. One, those things are inherently important to me as a clinician when I'm taking care of patients. It is helpful to me when I'm trying to manage someone's diabetes. Even if I know that they suffer from food insecurity, even if I can't solve that food insecurity.
So, clinically, I think it's actually important to kind of build the physician-patient relationship as well as know kind of the environment that our patients live in and come from. I also think that there are some ways that we will be able to actually fulfill some of those health-related social needs. And so, having the data will put us in a good spot when we can make those connections. And finally, I think we should collect the data so we can do social risk adjustment right now. If we turned on a switch and we started adjusting on the basis of social risk as well as medical risk, we would be in a bad spot because we don't have the data. Right. And having that data will allow us to make those adjustments, which will, in turn, make these programs more fair.
The future of payment models in healthcare
Brociner: What do you hope the future of payment models in healthcare will be in the next 10 years?
Dr. Gondi: Yeah, so I'll be hopeful instead of cynical and paint a picture of a payment system in which we have normalized payment rates, such that these disparities between Medicaid, Medicare, and commercial are much narrower, such that the same services for patients who are all else equal, but just have different types of insurance, no longer exist. I think that is a key factor in why some people get worse care than others in experience and have worse outcomes than others. And so I think narrowing those payment disparities is really crucial. I also hope that we'll be in a place where we are actually in true value-based care, which I would define as not, you know, bonuses and penalties for arbitrary measurements of quality, but rather a true shifting of risk to primary care providers who then manage the total cost of care and clinical care across the full spectrum for their patients.
And we're seeing those fully capitated models in some places. And I think those are the models that have the most promise in terms of best-best aligning incentives. And so I hope we move away from these sorts of intermediate models, which, as we've shown, have regressive effects, and towards models with full risk sharing. and importantly, that we have a more mature risk adjustment methodology that takes into account clinical and social risk. So we are ensuring that providers are being fairly assessed and are being paid commensurate to the actual outcomes that they're delivering. And then finally, I would just say that a lot of these problems, particularly with respect to health equity, cannot be solved only through payment reform. And I think we have often been misguided in thinking that payment reform, Medicare and Medicaid regulations, are the mechanisms to solve a lot of these problems. Rather, I think a lot of these problems are far more fundamental and will require not just continuous health reform but actual new economic and social policies that strengthen the safety net, narrow income and wealth inequality, and lead to, you know, a reversal of generational disinvestment in certain communities in our country. And so, I think that last component will actually be the component that drives us further along the journey towards health equity.
Conclusion
It was really interesting to hear the many different reasons value-based care is actually not a win-win for all patients and may not help the hospitals and providers who treat patients at community hospitals. It was also interesting to read that so many people are so deeply invested in the system of value-based care that there'd be resistance to honestly questioning the fundamentals of whether it works or not. To allow for real change to make healthcare better for all patients, there needs to be flexibility to evolve, even if that means changing directions. Which, while difficult, it is necessary to focus on decreasing hospital prices. Or, at the very least, reduce the increasing pricing trends.
Although value-based care for healthcare pricing is highly complex, pricing for medications should be easier to implement because determining whether a medication was effective or not is much easier than attempting to understand the health outcome of an entire hospital stay. Pricing medication for value is something that Stacie B. Dusetzina, Ph.D. believes what the future of medication pricing should be.
I am interested in seeing the hopefully beneficial impact ACO REACH has on improving payment models for underrepresented patients. From a machine learning perspective, I would hope that future risk models will incorporate social factors because incorporating social determinants of health factors has significant impact on understanding the greater context of the patient. Though I completely agree with Dr. Gondi that in the long-term we need to continue to have broader conversations about focusing on improving the health of Americans. We need to look to broad economic and social reform to improve health in the country by addressing homelessness, food insecurity, and other critical social factors.
I’ve got a lot more coming this year from The Future As It Should Be. I always welcome ideas for interviews or topics you’d love for me to cover, so reach out anytime to @evanbrociner on Twitter.