II. Universe A: The Health Care Polis

(This is part of a series: start here if you please.)

The Polis conception of the health system dominates health policy today. Visionaries like Don Berwick have dedicated untold hours to preaching its virtues and nudging the system in its direction. It’s a vision to which, in a previous life, I dedicated much of my time.

If It’s Broke, Fix It

In the Polis, the flaws in the health care are systemic, with inefficiencies and overspending stemming from fundamental misalignments between how care is paid for and delivered and how it ought to be under a rational, coherent system.

As a result the delivery of health care services is severely fragmented. Health care providers operate in narrow silos that focus on specific, specialized components of a patient’s problems without connecting their work in any meaningful way to a more holistic and complete understanding of a person’s health status.  Seizing on a wide range of literature convincingly demonstrating that this discord in health care delivery is driving up spending and resulting in costly and sometimes dangerous redundancies, errors, and missed opportunities (not to mention accounting for much of the purported physician shortage we face), reformers have pushed aggressively for more coordinated care across the entire spectrum of health services.

Accordingly, in the Polis, cohesive systems of care must be consciously designed, consistent standards of care must be agreed to, collaboration and buy-in for a team-based coordinated model of service delivery are essential. Ideally, health care ought to be a continuous process, with seamless handoffs between health care providers specializing in different domains of patient care. Each provider need not be a jack-of-all-trades but rather must be contributing to her component of a coordinated, holistic care plan. And a single point of human contact with medical knowledge, likely a primary care provider, must be conducting that orchestra of medical specialists to make sure that the total medical composition is played to perfection, for the sake of the patient.

Chasing Population Health Management

If the Polis was simply about applying a better, cheaper bandage that would be laudable enough. But it serves an even broader philosophy. It seeks not only to build a cohesive working system that rationalizes the delivery of health care, but also one that supports the maintenance of health in the first place. This approach, falling under the somewhat nebulous and oft-abused term “population health management,” aspires to both better coordinate medical services and integrate them with non-medical supports (be they social or behavioral, etc) essential to an individual's well-being.

We can get a sense of this philosophy from a quote shared by the recently departed CEO of a prominent integrated health system in the Northeast, Gary Gottlieb:

I’ll compete with anyone on cost and quality, if you give me the whole patient… People with hip replacements also have dementia…Someone with heart disease not taking his medication might also be depressed. We want to compete on how well we treat that whole patient.

Attending to only the immediate problem presenting isn't enough. Really tackling costs and improving health outcomes for a person requires a comprehensive strategy. Rather than a disjointed, piecemeal approach, health care providers must build relationships inside and outside of their walls to simultaneously address everything that makes a patient unwell. The ACA is a major driver of this philosophical shift, as the American Hospital Association attested in its report on Trends in Hospital-based Population Health Infrastructure:

Myriad factors are driving hospitals and care systems to address the nonmedical determinants of population health. Most notably, the Affordable Care Act implicitly and explicitly promotes a population health management approach to care delivery. Not only does this legislation expand health insurance to a majority of the United States population, it compels hospitals to address the socioeconomic, behavioral and environmental factors that affect people before hospital admission and after discharge. The ACA is accelerating the shift of reimbursement models from fee-for-service to value-based, a structure that promotes better health outcomes, improved quality of care, illness prevention and coordination across the continuum of care. Care systems are now being held accountable for the health of their patient population and are responsible for implementing health improvement strategies to address community health needs. Adopting a population-based approach to care that encompasses the spectrum of determinants of health is essential for care systems to thrive in the ACA era.  

To improve health outside their walls, hospitals and care systems must engage in multisectoral partnerships with community-based groups, health departments and public health organizations. By bringing together stakeholders from across the health care system and local community, hospitals can collaborate to identify population health priorities and develop strategies to address the health issues unique to their specific community. The federally mandated community health needs assessment process can provide a forum for enhanced collaboration between hospitals and their partners.

We're talking about nothing short of re-aligning the health care delivery system such that health, not sheer service volume, is its bread-and-butter. That's a big shift. And I'd be remiss if I failed to note that it's a shift that aligns almost perfectly with the insurance market reforms in the ACA, which move premium determination and benefit coverage decisions to the community, not individual, level.

Put Your Money Where Your Mouth Is: The Rise of Value-based Payment

But the plot thickens. If this is indeed the end goal, we find ourselves at present in a bizarre in-between stage. We can, if we squint, see the Promised Land but we need to lay down the infrastructure (policy, financial, and technical) to get us there. And that’s tricky. For instance, if we want to manage the health of populations and share information in real time between all participants in the extended care team tending to a patient, we need to shove health care into the 21^st century (or, arguably at least the late 20^th century) and get them on electronic health records that, through the miracle of structured data, can readily extract and give meaning to trends in the health of patients. But if the financial incentives underlying the system are still focused on churning out more services and billing more health widgets then, whatever its long-term potential for improving quality and managing health, converting to EHRs may just drive up spending by making it easier to quantify and bill for discrete services. And that indeed seems to be what has happened so far.

That volume-based approach to reimbursement bakes in an inherent bias toward delivering more and more services, as exemplified by Roemer's law that "in an insured population, a hospital bed built is a filled bed." Perversely, that means that the prevailing payment policy in the health care system effectively penalizes providers who do a better job and thus end up having to do less in the aggregate. This observation from the chief quality officer at Intermountain Healthcare, an integrated health care delivery system in Utah, sums up the challenge of improving in that environment:

Quality improvement is innately a preventive strategy. It achieves most of its cost savings by improving care “upstream,” thereby avoiding “downstream” failures and their associated recovery costs. Most clinical savings stem from reduced hospitalizations, reduced emergency department visits, and reduced resource consumption within care delivery episodes. David Clark and coauthors provide specific examples of the cost savings that resulted from clinical improvement efforts at Intermountain. Such savings extend well beyond savings from administrative improvements. 

Unfortunately, health care providers today are paid for precisely those care delivery episodes that quality improvement seeks to reduce. As Intermountain teams implemented clinical management, clinical outcomes improved and costs fell. However, our payments also fell—often even further than our operating costs. For example, although improvement in Intermountain’s appropriate elective induction rates saved the citizens of Utah more than $50 million per year through reduced payments, Intermountain’s costs fell by only about $41 million. Intermountain thus lost more than $9 million per year in operating margins. Implementing better care required us to invest in education, work-flow redesign, and new data systems. As we improved, the resources to drive further change disappeared.

If population health management is a new philosophy underpinning the delivery of health care, it needs a new model of reimbursement from health payers, like private health insurers or Medicare and Medicaid. How we pay for care heavily influences how doctors and hospitals provide it. Historically we've paid for each and every health care widget, with little mindfulness of the results they achieve or the cost efficiency with which they're selected. The result has been that service volume obsessed, uncoordinated, unfocused, and myopic non-system that the Polis is dedicated to upending.

The central conceit in the Polis is that we, as a society, aren’t getting enough value—read health, broadly defined—for  our health care dollar. And while the pricing of health services is necessarily a component of this, at a fundamental level this is a conception of the volume side of the cost equation being the problem. Not long before the ACA passed, Thomson Reuters famously estimated that $800 billion of our national health dollars are wasted annually on unneeded services, mistakes we should avoid, conditions we can better treat.

Proponents eagerly exploit that argument, pointing to vast waste in the American health system

resulting from poor coordination and dangerous volume incentives: duplicated tests, unnecessary procedures, a “do more” philosophy that neatly aligns with the “pay more” defeatism of many payers. With “better” (generally understood to be some variant of “streamlined”) care delivery, we can not only meet a patient’s needs with fewer resources, we can serve a fuller conception of the person. Health care becomes more than the sum of its parts. A well-functioning system of care thus generates savings beyond what it costs to put it all together.

This philosophy, that we must first pay for better care before we can expect to receive it, is taking off as reimbursements from public and private payers are increasingly linked to quality outcomes and cost efficiency. The Centers for Medicare & Medicaid Services has announced a goal of linking 90% of Medicare payments to quality or value by the end of 2018, and a coalition of prominent private sector payers and providers has pledged to get 75% of its payments in value-based arrangements by 2020.

Industry watchers from Catalyst for Payment Reform (the creator of the handy infographic above) to The Advisory Board to Fitch Ratings have tracked the explosion of value-based payment and risk-based contracting over the past five years, driven in no small part by incentives put forth in and the Polis-oriented philosophy pushed by the ACA.

Percentage of lives covered by ACOs, by hospital referral region

This shift in how care is paid for (giving health care providers some of their savings back if they do better—to avoid the Intermountain conundrum—and in some cases asking them to pay back some of the excess if their spending goes over a certain budget) has given rise to new modes of organization/delivery in the provider world. The accountable care organization (ACO), theoretically a tightly linked network of different kinds of doctors and hospitals working together collaboratively to manage the health of their patient populations, is proliferating rapidly.

And the health policy world is pinning much of its hopes on the Polis--both the tightly integrated organization of health care providers and the payment schemes put in place to nudge it. PricewaterhouseCoopers (PwC) looks at the expected drivers (and, working in the opposite direction, deflators) of medical inflation every year. Two of the three cost deflators they identify for 2015 are related to these kinds of ACA-supported delivery reforms: (1) "systemness," PwC's word for the holistic, coordinated vision of health care delivery embraced by the Polis, and (2) the financial champion of systemness, risk-based payments (the third deflator, price shopping, is one we'll meeting in Universe B).

I Come to White-Coat Caesar, Not to Bury Him, or: How I learned to Stop Worrying and Manage the Health of Populations

One of the most important organizational shifts all this entails is a rejiggering of the institutional logic that governs health care organizations, particularly those delivering care. To steal from Institutional Change and Healthcare Organizations:

Institutional logics refer to the belief systems and associated practices that predominate in an organizational field. . . Institutional logics provide the "organizing principles" that supply practice guidelines for field participants. Logics specify what goals or values are to be pursued within a field or domain and indicate what means for pursuing them are appropriate. Thus, logics tap into both the cultural-cognitive and normative dimensions of institutional environments. The logics are only salient to the extent that they affect action within the field--sometimes being carried by established participants, sometimes by outsiders who influence behavior within the field. 

If fundamental change in the health sector is indeed taking hold, this is where it comes from: the understanding of what health care ought to be and how it ought to be delivered held by critical actors in the health system is changing, and with it the way the entire field does business. The "goals or values" that the health sector is actively pursuing seem increasingly (thanks, in part at least, to the incentives and Polis-esque worldview being pushed under the ACA) to be embracing the Triple Aim--the notion that we ought to be simultaneously improving health, care, and holding down cost growth. Goals that are now being facilitated by agreements to be paid differently and to adopt new care models.

But few communities are truly democratic at their roots. The Polis—ruled, in the thought of antiquity, by the philosopher king—must give in to Michel’s Iron Law of Oligarchy, as all forms of organization ultimately must.  “Who says organizations say oligarchy.” For all the lip service—admirable and well-intentioned as it may be—to patient-centeredness, physicians retain control in the Polis. But more important than the role it provides the individual clinician, the Polis privileges the mighty Health System (ACO or otherwise), a conglomeration of disparate physician groups, hospitals, perhaps other components. If the name of the game is integration and coordination, then a (relatively) closed system of providers that collectively take on responsibility for the health of those patients who utilize them is what we're talking about.

Inherent in this concept of coordinated care is greater vertical integration in the health care system. Played out to its logical conclusion, the Polis ultimately promotes the formation of empires: vertically integrated systems capable of servicing the full continuum of health needs of their patients. The ultimate Health Care Oligarchs. This, for instance, is what the accountable care organizations forming under the Affordable Care Act are intended to achieve; while at present they preserve patient choice, allowing their patients to seek care outside them at their discretion, this is considered a serious weakness of the current model, not a strength.

For the alternative to the Health Care Oligarchs we must proceed to Universe B.

I. Squeezing Blood from a Stone

(This is part of a series: start here if you please.)

I have in the past recommended the fantastic book Policy Paradox by Deborah Stone. Besides being excellent, it had an outsized impact on me as a young convert to public policy from the staid world of physics. It coaxed me to abandon naïve ideals of policy as some sort of quasi-science mucked up by the messiness of political decision-making. It opened my eyes to the fascinating, unsettling reality that even (apparent) facts and figures are at their heart political/philosophical questions. This was the start of a fundamental shift in my worldview from which I've never quite recovered.

Much of the book is structured around Stone’s conception of political philosophy as a dichotomy between two philosophical constructs: the Market and the Polis. She didn’t capitalize them but, for allegorical purposes, I will. 

For many reasons, this Manichean picture of political decision-making strikes me as an apt model for understanding the two pre-eminent dueling narratives about the future of the American health system. Two competing philosophical visions of what health care can (and should) be. Delightful!

One centered on the individual and driven primarily by self-interest, the other oriented around the community and motivated primarily by the public interest. A key difference between the two is that in the latter “system” is not a structure that emerges merely from the myriad interaction of individual rational actors making decisions against some structured backdrop, but rather is a deliberate and purposive creation:

It would be as much a mistake to think that the market has no concept of public interest as to believe that the polis has no room for self-interest. But there is a world of difference between public interest in a market and a polis. In theory, the public interest or general welfare in a market society is the net result of all individuals pursuing their self-interest. In economic theory, given a well-functioning market and a fair initial income distribution, whatever happens is by definition the best result for society as a whole. In a market, in short, the empty box of public interest is filled as an afterthought with the side effects of other activities. In the polis, by contrast, people fill the box intentionally, with forethought, planning, and conscious effort.

The Market and the Polis distinctions characterize so many political-policy debates and health care is no exception.

For our purposes (considering the next steps in extending and improving upon the Demi-Decade of Cost Control), each primarily attacks a difference piece of the health care cost equation. When we think of costs we’re really thinking of health care spending, both in the aggregate (national health expenditures) and in our personal lives, when it comes to how much money flows out of our wallets and paychecks and into the health sector.

But, at the risk of oversimplification, cost is the product of how high the price is and how much we need to buy: 

The Market and the Polis distinctions characterize so many political-policy debates and health care is no exception. And they correspond loosely to the prices and quantity consumed components of this not-quite-real equation, respectively. I believe they give us the best window on the two prevailing theories of how we ought to go about maximizing quality while minimizing costs.

Onward to the Polis.

The Demi-Decade of Cost Control

Happy (slightly belated) fifth birthday, Affordable Care Act!

Nearly five years ago now, in The Demi-Decade of Coverage or: The Scalpel and the Chainsaw, I predicted that the ensuing five years or so (say 2011-2015) would be a coverage story, as millions of people gained health insurance. In that respect, mission partially accomplished: 16.4 million and counting so far. But while the ACA had many components that had potential to ultimately tame rising health care costs, that wasn't something we could anticipate to start seeing in the first demi-decade. Maybe not even in the second demi-decade:  "All of this means that we can't really expect real cost curve bending to happen until the second decade of the law's existence (the 2020s)."

Wrong!

Aided in no small part by the Great Recession, health care spending growth fell. But even as the

economy recovered, it stayed historically low. The last five years have been unprecedented. More and more observers, even those reluctant to look for culprits beyond the sluggish economy, are now recognizing that structural changes in the health care sector--real, lasting transformation of the way that industry does business--may well be taking root.

This first five years of the ACA era has indeed become not just the Demi-Decade of Coverage but also, surprisingly, the first Demi-Decade of Cost Control.

And now we find ourselves at a crossroads.

We're (largely) solving our health insurance coverage problem, but how do we simultaneously tackle our cost growth and quality improvement conundrums?

I'm about to lay out my thoughts on that in some depth through this series (these will become links as I get around to posting them):

I. Squeezing Blood from a Stone
II. Universe A: The Health Care Polis
III. Universe B: The Health Care Market
Interlude: The Biggest Question
IV: Untitled Synthesis

Let's begin!

Unsettling Thought of the Day

It's interesting to consider how seriously many noted thinkers take certain sci-fi-esque existential threats. When we think of artificial intelligence, do most of us really consider the culmination of the march of technological progress resulting in a Terminator- or Matrix-style struggle for existence a likely outcome? I would guess probably not. Ditto for the Search for Extraterrestrial Intelligence, the ongoing quest to find someone else, somewhere, that uses radio telescopes (or at least signals in the radio bandwidth). Would intentionally or unintentionally making our presence known to another galactic inhabitant result in an Independence Day-style brawl instead of a Contact-esque exchange of culture and knowledge? The worryworts among us aren't so sure.

When it comes to advancing artificial intelligence, some smart folks have warned against "summoning the demon":

“I am in the camp that is concerned about super intelligence,” Gates wrote. “First, the machines will do a lot of jobs for us and not be super intelligent. That should be positive if we manage it well. A few decades after that, though, the intelligence is strong enough to be a concern.” 

Gates’ concerns about AI come in the wake of Elon Musk’s remarks last year that artificial intelligence represents “our biggest existential threat” as a species and that toying with AI is akin to “summoning the demon.” 

Stephen Hawking, meanwhile, was even more blunt about it and told the BBC that “the development of full artificial intelligence could spell the end of the human race,” especially since “humans, who are limited by slow biological evolution, couldn’t compete and would be superseded.”

And there's an active debate in the SETI community as to whether we ought to just be passively listening for signs of others or if we ought to be taking a more active approach to attracting attention.

Physicist Stephen Hawking has argued against that strategy, saying that the signals could attract the bad kind of aliens depicted in movies ranging from "War of the Worlds" to "Independence Day." He imagines that our first contact with extraterrestrials could be like the Native Americans' first contact with Europeans, "which didn't turn out very well for the Native Americans." 

In Hawking's view, it's better to lie low, and science-fiction author David Brin agrees. During sessions at this week's annual meeting of the American Association for the Advancement of Science in San Jose, Brin faced off against Vakoch and the director of the SETI Institute's Center for SETI Research, Seth Shostak. 

"We've had some disagreements lately," Brin admitted.

He said the reason we haven't been hearing radio traffic from our extraterrestrial brethren could well be because "they know something we don't know." 

If the solution to the Fermi paradox, the reason we've never found evidence of any other intelligence in the universe (despite it otherwise seeming probable there could be a number of other civilizations out there), is that folks tend to "know something we don't know"--well, that's a rather ominous thought.

The Problem is Still Choice

Please, as I was saying she stumbled upon a solution whereby nearly 99 percent of all test subjects accepted the program as long as they were given a choice, even if they were only aware of the choice at an unconscious level. . . As you adequately put, the problem is choice. -- The Architect, The Matrix Reloaded

The human mind demands choice--even the illusion of choice will do. That means if you're going to limit choice in some way, people will demand a choice for a choice.

Let's continue the Speakeasy's trend of channeling fascination with choice, human behavior, and health care. And build on that last post! We're going to talk about what market dynamics mean here. And if they can really be sustained in a multi-payer insurance market.

Choosing (How) Not to Choose

After a miraculous turnabout following the IT disaster that made exchanges a national punchline a year ago, the first open enrollment period under the Affordable Care Act turned out all right. Seems like just yesterday we were talking about how the then-forthcoming exchanges would work.

A good chunk of enrollments in December were from re-enrollments by people who first signed up last year. Recognizing how powerful inertia is--people tend not to take action--the administration made the decision to automatically re-enroll people into the plan they chose last year. That means people who don't bother to come back to the exchange website and do it themselves wind up in the same health plan this year.

That's good for keeping the number enrolled up and avoiding subjecting people to lapses in coverage due to their own laziness, but there are dangers there. Exchanges work to keep premiums in check because they're competitive marketplaces. We've seen new insurers flooding into the exchanges this year, in many cases offering 2015 premiums that undercut insurers who led the market last year. Great news!

That's all well and good right now because in these early years for the exchanges there's a continued influx of new customers picking a plan for the first time each year.

But, as things settle down, there's a danger that inertia will undermine the competitive pressures the exchanges are meant to generate. If there's a certain "stickiness" to plan selection, with people either sticking with the plan they know or being automatically re-enrolled in it by default, then an insurer that captures significant market share doesn't have to work as hard to retain that market share. Indeed, going into 2015 we're seeing that insurers who scooped up significant enrollment last year have raised premiums more than some of the newbies. Part of the reason average premium growth has been remarkably low in the exchanges is that new entrants to the market in 2015 are undercutting insurers who sold plans in the exchanges last year--competition at work!

Let's make this clear: competitive pressure is contingent upon consumers threatening to defect to competitors if they're not satisfied with the prices (premiums) or service offered by a market participant. This, however, requires action on the part of consumers. And insurance is hard. It takes work for people to evaluate competing options, provider networks, and whatever factors play in.

In a recent proposed federal rule, HHS suggested a solution to consumer inertia: people need to be given a new choice (during their initial enrollment) if they're going to cede choice later (during the re-enrollment process).

The current re-enrollment provisions ... prioritize re-enrollment with the same issuer in the same or a similar plan with the goal of maximizing continuity of coverage and care. However, because premiums may change significantly from one year to the next, the plans that are most competitively priced in one year may not continue to be the most competitively priced in subsequent years. For this reason, default enrollment in the same or similar plan may sometimes encourage consumers to remain in plans that are significantly more expensive than the lowest cost plans in the market. Because we believe that many consumers place a high value on low premiums when selecting a plan, we believe that consumers could benefit from alternative re-enrollment hierarchies. 

In particular, we are exploring implementing in the FFE an approach under which an enrollee, at the time of initial enrollment, would be offered a choice of re-enrollment hierarchies and could opt into being re-enrolled by default for the subsequent year into a low-cost plan (such as the QHP of the same metal level with the lowest premium in the enrollee’s service area, or one of the three such QHPs with the lowest premiums by random allocation), rather than his or her current plan or the plan specified in the current re-enrollment hierarchy. 

This alternative enrollment hierarchy could be triggered if the enrollee’s current plan’s premium increased from the prior year, or increased relative to the premium of other similar plans (such as plans of the same metal tier), by more than a threshold amount, such as 5 percent or 10 percent. As is the case under the existing approach, a consumer would retain the option to take action to enroll in a different plan during open enrollment if he or she wished to do so. We are considering applying an alternative hierarchy for the first time when re-enrolling consumers for the 2017 coverage year. On this timeline, consumers could opt in to the alternative hierarchy during open enrollment in 2015 (or during special enrollment periods occurring during 2016).

Stated a little more directly: when signing up for coverage, consumers would be given a choice. They would be asked how they want they want their automatic re-enrollment to be structured. They could choose, if they took no action next year, to be re-enrolled in the very same plan they're selecting this year. Just as things work now. But relative premiums change from year to year. They could also choose, if their current plan's premiums increase too much, to be automatically enrolled in the cheapest plan in their metal tier. In other words, they're given a choice as to just how they want to cede choice next year.

It's difficult to overstate how important that choice is.

If people are complacent and don't bother to return to the marketplace website to choose a plan for next year they (at present) can be re-enrolled in the same plan. Inertia keeps them where they are. Which empowers those insurers who scooped up the majority of the business in a single year: it gives them the security to jack up rates higher than competitors because they know most (?) of their enrollees won't bother to shop around and move their business elsewhere. Thus their incentive to retain competitive premiums is reduced.

On the other hand, if every enrollee is asked when they enroll how they want that re-enrollment process to work a year down the road if they forget or neglect to come back, that gives people the option to let inertia take them into cheaper plans if the default option raises premiums too much. That means people are automatically shifted to the most competitive (on a price basis) insurance offering. Which gives some incentive to dominant insurers to temper their proposed premium increases. Because they will automatically lose business if bested by their competitors. Which returns a semblance of market dynamics to a marketplace whose works may well be gummed up by the relative disinterest and inaction of its participants.

Exchanges are grappling with two mutually contradictory impulses: (1) Insurance is hard--once people choose, they're apt to not come back and choose anew, instead relying on automatic re-enrollment, and (2) price competition requires insurers to believe that their enrollees will defect to competitors if rates rise higher than justified. Auto-enrollment in the same plan dulls impulse (2). But impulse (2) is key to making well-functioning insurance markets preferable to a single-payer system (see post below!).

Yet re-enrollment in a cheaper plan, even if the consumer opted for that scenario when they first signed up a year ago, can be dangerous. They won't know what plan they're getting into (until notified) and they may not be able to keep their providers of choice, since provider directories vary by plan. And certainly in the comments submitted by the public on this proposal many advocates had feedback. The Sargent Shriver National Center on Poverty Law suggested in its public comment making the process slightly more complex, giving people a menu of factors they wanted factored into the re-enrollment determination (if they didn't bother to log back in to make a selection themselves):

We commend CMS’s willingness to consider alternatives to the current default re-enrollment position (re-enrollment with the same issuer or with the same plan). We think that over time consumers will become much more sophisticated about plan choice, and we think that discussing re-enrollment default positions with consumers is part of that educational process. We also think that, while cost is of great concern to most consumers, continuity of care, quality, and access are very important, too, and sometimes paying a little more will result in coverage that is significantly better for a particular patient. With all that said, we suggest that CMS adopt a de-fault enrollment process that includes the following:

• asking consumers to identify the factor they consider most important for them in the event of a default re-enrollment from a list of factors including cost, provider network, and benefit design; 
• asking consumers to identify their factor after they have chosen their plan (so they will be familiar with the factors they recently took into account); 
• reminding consumers of their chosen factor in their renewal materials and informing/reminding them that they have the option to change plans during open enrollment; and 
• giving consumers a window for switching plans if they are auto enrolled at renewal in a plan they do not prefer. 

Fine suggestions, though there are questions about the feasibility of folding consumer preference into an automatic renewal process to that extent. If one's preferences are that strong, perhaps one ought to take the time to log back in and comparison shop. And indeed, as the New York Times reported not too long ago in "People are Shopping for Health Insurance, Surprisingly", more people have actively re-enrolled that would be expected from previous experience in similar programs. That's a good sign but with ~70% still choosing passivity, the structure of the re-enrollment process is going to be important for ensuring the continuing existence of the market dynamics that incentivize insurers to compete hard to keep prices low and quality high.

Choosing How to Limit Choice

We'll be briefer on this one (more on this in forthcoming posts) but a significant issue that has arisen in the ACA exchanges is network adequacy--does your insurance plan have all the kinds of doctors you need to get the care your benefit package promises you? And if the answer is yes, is that enough for you?

To keep down costs (part of that surprising cost containment push by private insurers I mentioned in the previous post), insurers are cultivating smaller networks than many people, particularly those with employer-based coverage, are used to. Unlike people in employer plans, people buying plans on their own--even with the help of a federal tax credit--are, unsurprisingly, extremely cost conscious. And by cutting out some of the highest priced health care providers from their network, health insurers can keep the costs of insurance plans down. And those relatively cheaper plans are turning out to be very popular.

But a narrow network means you can't go to just any doctor or hospital, you've got to find one that's in your network. And, unlike with wide network plans, that may be a minority of doctors and hospitals. So when you seek care, you've got fewer choices. But you've in effect chosen that slate of choices in advance when you selected the insurance plan with that narrow network. You've traded a choice for a choice.

Take this bit from "Narrow Networks Enjoying a Resurgence":

Steve Wojcik, vice president of public policy at the Washington, DC-based National Business Group on Health, says several key healthcare industry stakeholders are embracing narrow networks. 

"Employers and health plans in particular have increasingly realized there is a wide variation in healthcare price and quality," and narrow networks have emerged as a market-based approach to maximizing value in care delivery, he said this week. 

"We are able to select winners and losers. It's easier for us to do it [than the government programs]. We're looking for the best we can get in terms of price and quality." 

Providers and patients also are embracing narrow networks, he says. "The last time with managed care, narrow networks were formed without the support of physicians and consumers," and the recent widespread growth of narrow networks has many healthcare providers clamoring to be included for competitive reasons. 

"Price-sensitive consumers are choosing narrow networks on the new exchanges. It's not like someone is forcing them into it… As long as the public sees value in the narrow networks, there won't be a repeat of the managed care backlash."

When placed into narrow networks without a choice (the 1990s) the experiment briefly worked. But consumers ultimately revolted, networks widened, and costs returned to large increased by the early 2000s. Today narrow networks are back but with a twist: at least in the exchanges, you're only in one if you actively chose it. Wider networks are competing alongside narrow network options, they just command a higher premium.

If it sticks this time, perhaps it's because people are being allowed to swap the choice of any provider when it comes time to seek care for the choice of a narrow network when it's time to purchase an insurance plan. The problem is always choice!

Single-Pay it Forward

As the previous post should make clear, I'm not (presently) a single-payer advocate. But I'm persuadable. So perhaps it would be helpful--at least for me--to lay out (1) why I'm not, and (2) under what circumstances I will be.

My views here are, I think, relatively non-ideological and lean toward empiricism. Bu I have some built-in biases, though I would argue that there are logical/empirical roots to these preferences.

My major guiding axioms here are the following:

1. If well-functioning markets are possible, they are preferable to the alternative.
2. If people are capable and equipped to make them responsibly, choices/options are good.
• Corollary: This is true even if choice is largely an illusion.
3. A policy that can be passed and implemented is superior to one that cannot.
• Corollary: Smaller changes are generally preferable to larger ones, and larger changes arrived at through evolution are generally preferable to those arrived at through discontinuous, disruptive change.
4. Spock's Dictum: The needs of the many outweigh the needs of the few. Or the one.

(What a fucking gut-wrenching death. Part of why TWOK comes in #1 in my ranking of Trek movies).

So this is my starting point. A healthy respect for the benefits of market dynamics, coupled with a systems-oriented perspective that doesn't unduly privilege the individual to the detriment of the overall system. Perhaps that means we're starting with an intrinsic paradox or contraction but here we are.

So here's what would make me embrace single-payer over the (ACA-compliant) status quo.

Occam's razor

I consider the new exchanges (marketplaces) under the ACA a grand experiment. Not just in policy but, deeper than that, in human behavior. They're the first large-scale, widespread chance for individuals to shop for commercial health insurance in a coherent marketplace. Previously nearly everyone had their insurance choices dictated either by their employer's HR department or the vicissitudes of the chaotic, disorganized individual market that used to exist for everyone else. How people respond to this experiment is the single most important factor in my thoughts on single-payer.

We know people aren't perfectly rational calculators, they're boundedly rational beings with limited time, resources, and cognitive bandwidth. And that's why the most important function of the exchanges (after disbursing federal tax credits, SCOTUS willing) is to organize the market. Grouping plans by relative generosity into actuarial tiers identified by metal designations, bronze to platinum, allows people to easily make apples-to-apples comparisons. 

Eliminating individual rating beyond the most basic of factors--family size, geography, age, and tobacco use--means that the sticker price of a health insurance plan is in fact its price. That means competing plans can legitimately and meaningfully be lined up next to each other for comparison on a website, something never true before. Premiums and deductibles are right there for inspection by the consumer, no strings attached.

This year the federal exchange, healthcare.gov, added a feature that allows shoppers to narrow or sort plans by "medical management program." Consumers now can quickly see which plans offer special programs to support enrollees with a variety of conditions, from asthma to heart disease to depression to high blood pressure. 

Plan quality and participating providers are additional dimensions of interest to shoppers that exchanges will get better at displaying for easy comparison as time goes on.

The point of all this being that exchanges narrow down the decision-making process for health insurance by boiling it down to a few key factors and allowing shoppers to easily compare plans along those dimensions. By providing that structure, they allow consumers to make meaningful decisions and to send clear signals to insurers selling in the marketplace. The design of exchanges not only acknowledges bounded rationality, it harnesses it to salvage market dynamics. 

But a critical assumption here is that this simplified choice architecture (an idea I've explored with delight before in "Order Out of Chaos") is enough to empower consumers to be savvy shoppers. If that's not true, the entire edifice begins to crumble. We heard it last year in scattered anecdotes that provide precious little context to determine if the problem is widespread or localized. Some people bought a plan not understanding their cost-sharing responsibilities, ending up with higher deductibles than they wanted to pay. Others bought plans without scrutinizing the provider director and found desired doctors weren't included (cases where insurer provider directories were simply wrong, primarily in California as far as I can tell, are different--that's simply fraud or negligence and ought to be pursued by local authorities). 

Those stories are all examples of people buying coverage they didn't understand and thus ending up with plans that didn't accurately reflect their preferences. If the average consumer cannot properly evaluate their risk, judge when to seek "necessary" care,  or make appropriate choices that correspond to their preferences then the marketplaces will fail. Like all ideas based on markets, exchanges are predicated on the assumption that people are equipped and able to make the best decisions for themselves. If most people end up with coverage they didn't want (e.g., because their preferred doctor or hospital isn't in-network for the plan they chose or because they didn't understand the financial commitment entailed by taking on a large deductible), then we have a problem. If people can't effectively translate what they want into what they buy, the markets won't work. The marketplaces are supposed to be the mechanism by which consumers translate their preferences into purchases and thus send clear signals to sellers. If the consumer turns out to be the weakest link here, then markets aren't the answer.

Now we need to be careful here. Anecdotes are not data and Gallup polling from last November suggests most people who bought exchange coverage are happy with it. And the new marketplaces established under the ACA remain a vanishingly small component of the total insurance market, primarily catering at present to people who are the most likely to be unfamiliar with health insurance concepts and design. There will be a learning curve. But it's entirely possible that most people in employer coverage, unaccustomed to comparison shopping for insurance in the open market, are not particularly well-equipped to buy their own insurance either. And those who hope that down the road the exchanges become a platform for the future of health insurance selection (well beyond the ~10-20 million people expected to gain coverage through a public exchange in the next few years) may yet have to grapple with that reality.

So that's the primary (empirical) thing that will turn me to single-payer. If people prove unable to competently shop for a health insurance plan that matches their preferences, then there's no reason to have competitive insurance markets. Indeed, a more paternalistic system would be demanded by such a situation. And maybe I need to surrender my moderate liberal/center left card to say so, but I really hope that's not the case.

But if we need a paternalistic system, an Occam's razor to further simplify the options available to people (to none, I suppose, since they'll get a standardized policy with likely no cost-sharing and all-providers in the U.S. in-network), then single-payer is the answer.

E Pluribus Pluribus

I argued in the previous post ("E pluribus unum") that single-payer isn't necessary if we can achieve its primary goals by aligning multiple payers toward similar ends. If we want providers to adopt a new health care delivery model that requires a different payment model to sustain it, then all payer-boats need to start rowing in the same direction to enable those providers to make that change. All health care is local and that sort of alignment needs to happen at the local/state level, though arguably such a major payer as Medicare needs to be onboard if those local changes are to succeed. And the federal government is furiously providing seed money to the states to make those local changes.

But if multi-payer alignment fails to achieve those goals then it's not going to be enough. And by failure I mean that either these coalitions fray and disintegrate, or they're simply too weak to enable or push the change they seek on the provider end. There are a lot of moving parts here and there's no guarantee that the current push to get competing payers on the same page will succeed. Payers need to maintain enough differentiation to try and best their competitors in the marketplace. Is this kind of homogenized differentiation sustainable?

I said in the previous post that a a primary argument for single-payer is that it provides unified policy direction across the health system. If that unified direction can't be achieved in a multi-payer environment then single-payer may well be the only hope we have to fix our broken payment and delivery system. This, again, is a prospect I don't relish.

Prices!

There's plenty of reason to believe that much of what's wrong with health care costs in America isn't just traced to our woefully inefficient delivery system but to our prices (see: "It’s The Prices, Stupid: Why The United States Is So Different From Other Countries" and "21 graphs that show America’s health-care prices are ludicrous").

If we don't find a way to tackle high prices on the health care provider side (a departure from this talk of health care insurance market dynamics) then we're sunk anyway. In fact, there's some reason to believe a competitive insurance market could be counterproductive:

(h/t to the Incidental Economist)

What the graph says is that if a single insurer dominates a market, they'll jack up premiums and consumers will be hurt (a great argument for competitive insurance marketplaces!). But if a consolidated health care provider organization dominates the market, a fragmented insurance marketplace may just make it easier for them to extort ever higher prices for health care services (see my now classic in the field "The Case Against Providers") by weakening the bargaining position of any individual health insurance company. In other words, insurer competition is good...to a point. And that point is where insurance market fragmentation benefits health care providers at the negotiation table. What a tenuous balance!

Ultimately if insurer competition means that hospital systems (which are quickly buying up physician practices, as well) can demand higher prices because insurers are poorly positioned to say no, then we all suffer. Which would suggest we need a single payer with the monopsony power to simply dictate reimbursement rates to provider systems.

Again, I don't believe that's ideal but if competitive insurance markets turn against the consumer in this fashion (after all, higher provider reimbursements directly translate into higher insurance premiums) then decisive action is needed. Time will tell!

Requiem for a Dream

I want markets to work. Long-time readers (j/k, I know there are none) will know that five years ago I was a huge, huge, advocate of having a robust public option in the new exchanges. On the assumption that an insurer paying (at least initially) Medicare +5% rates would put tremendous pressure on health care providers to capitulate to lower rate/price demands of newly-en-balled commercials demanding more reasonable rates to avoid consumer defections to a cheaper public option. The value of the public option to me was not that people would enroll in it, but rather that its mere existence would re-shape the provider-commercial insurer relationship to ensure that contracted rates came back to earth so that commercial insurers could reasonably compete with a public option empowered to dictate its own rates.

I have been pleasantly surprised to find that the public option seemingly wasn't necessary. Private insurers selling in the exchanges have been going crazy with cost containment. PwC found that exchange premiums were 4-20% (!) cheaper than employer-based plans of equivalent generosity. McKinsey found that narrow provider networks of the sort gaining popularity in the exchanges could lower insurance premiums by as much as 26% (!) with no appreciable difference in quality of care.

The reality is that in a world where health insurers can't find savings by shedding risk (read: dumping or avoiding sick people) they have to keep premiums down through a few strategies, including: 1) smarter benefit design, nudging shoppers toward higher-value services and providers, 2) active population health management to preserve and encourage health on the part of enrollees, coupled with support for key delivery system reforms to ensure efficiency in care delivery, and 3) more selective provider contracting or network design to attack price growth. And, amazingly, competition in the new marketplaces is pushing them to pursue these strategies--an outcome I thought only a public option competitor would finally commit them to pursuing. I was, happily, wrong. 

I hope that I'm right that multi-payer competition can work to give consumers what they want while simultaneously muting commercial health premium growth. Because if I'm wrong, Uncle Sam is our last chance.

E pluribus unum

With the recent news that Vermont's push to implement a "single-payer" health system is all but dead, it's worth taking a few minutes to reflect on what that even means.

I put single-payer in quotation marks because what was on the table there (passed in principle a few years ago through a law that left the details to be worked out later) was more single-payer lite than anything else. Part of the complexity of the American health care system is that we have so many different insurers/payers for different segments of the population. To wit, some of the major ones are:

• Medicare, primarily for the over-65 set
• Medicaid, aimed primarily at the poor and disabled
• Commercial insurance companies, which sell health insurance to individuals or to certain employers who want to buy fully insured products for their employees
• Self-insured plans, in which an employer effectively acts as its own insurance company and takes on risk for the health benefits it offers to employees

A true single-payer plan would get rid of those distinctions and replace them with a single national plan for everyone--a sort of "Medicare for all" as its proponents often describe the idea. But as a state, Vermont wasn't really in a position to do that. Self-insured plans, for instance, which the majority of privately insured people are in, aren't generally subject to state regulation. And Vermont doesn't have control over Medicare. So their approach would've been more modest than that advocated by national single-payer fans, restricted to those parts of the insurance landscape over which Vermont's government has authority.

But when viewed that way, it's fair to ask whether they really need single-payer. In my estimation, there are four primary arguments for implementing a single-payer system.

1. Universal coverage

This is, I suspect, the strongest selling point for most fans of single-payer. Everybody's in so uninsurance essentially disappears. And it's a strong argument. Cost--at the individual level--isn't a barrier to gaining insurance, nor is any kind of discriminatory effort on the part of the insurance industry since that industry has more or less been supplanted by the government.

But in all likelihood Vermont will inch toward that goal over the next 2-3 years--granted often with less generous coverage than under a single-payer system--as the coverage expansions under the Affordable Care Act continue to take root. They've expanded Medicaid and have a state-based health insurance exchange that will continue doling out subsidies for private insurance coverage regardless of what the Supreme Court does. Gallup had their uninsurance at 8.5% as of last summer and it's virtually certain to fall at least a few more percent before all is said and done. They won't get to 100% coverage but they'll get close enough for government work.

2. Administrative simplicity

The sheer number of health insurers adds layers of complexity (and expense) to the health system. For instance, a hospital or health system may have to grapple with widely varying approaches to reimbursing them from each insurer that sends patients their way (different financing arrangements, different logic to the way reimbursements are handled, different reimbursement rates, different cost-sharing for patients, etc). Vermont certainly expected savings to accrue over time based on the simplification of the billing processes in that state.

But I'm left wondering how big an impact that would really have on the state. As already mentioned, self-insured employer-based plans would persist under Vermont's proposed system so we're really talking about the fully insured commercial market where people buy plans from insurance companies.

But Vermont is a tiny state with a correspondingly tiny insurance market. 80% of their insurance market is in the hands of their local Blue Cross Blue Shield affiliate (counting a wholely owned subsidiary of BCBS that also competes), 17% belongs to MVP, and 3% to Cigna. In other words, they're not all that far from a single-payer system--albeit one in which BCBS is that payer--as it is. 

I can easily imagine that large states with dozens of insurers--e.g., California or New York, two states that have considered single-payer in some respect--could wring quite a bit of savings out of a simplification of their markets. But Vermont is a much less complex market than most places, so I have to question how much benefit will really accrue from streamlining it.

More importantly it seems to me that single-payer advocates sometimes mistakenly believe that single-payer means the private insurance industry evaporates. But the government isn't an insurance company--it generally doesn't have the know-how to administer or operate such a program. That's why Medicare is operationalized through contracts with a variety of regional private companies that handle the day-to-day business of insurance, like processing claims and making routine coverage determinations. It's why most states have privatized most of their Medicaid programs, which are actually in the hands of competing private managed care companies. So while there are certainly some administrative simplifications from transitioning to single-payer, complexities remain.

3. Unified policy direction across the health system.

In my mind, this one might be the strongest argument for single-payer. There's a very prominent school of thought that suggests real change in our system requires 1) changing the way care is paid for, and 2) leveraging those smarter health care payment approachs to re-design care delivery so as to offer better care.

But if payment is going to push and enable health care providers to re-organize the way they do business to improve, you can't have every payer sending different signals. Providers need a coherent message from the insurers reimbursing them. Single-payer would certainly achieve this. Even now, as payment and delivery reform is spreading across the country, Medicare has been a driving influence because it's such an important payer by virtue of its size. Now imagine it enrolled everyone.

Our system at present is unabashedly multi-payer. Under such circumstances, the solution to achieving (loosely) aligned policy direction is encapsulated in another bumper sticker-ready phrase: all-payer. Getting all or most insurers on the same page so that their payment approaches are directionally aligned, pushing providers in the same direction, can have substantially the same effect.

There are ongoing efforts in dozens of states to line up the efforts of health care payers--not just the various commercial insurance companies but public payers like Medicaid as well--to achieve essentially what some hope single-payer would. Vermont is no stranger to this effort, being one of the earliest states offered federal funds to promote this work. And they've not been shy about using the regulatory authority of the state to compel private insurers to align payment efforts to support certain health care delivery models.

So there's a strong case to be made that unified policy direction to push the system in a positive direction not only can be achieved in a multi-payer system, to some degree it already is being achieved in states like Vermont. 

4. Rate-setting

This last item gets at a particular danger of our system as currently constituted: better, cheaper care need not necessarily translate into lower costs for consumers and insurers. It's possible that vagaries of the intersection of a region's insurance and health care provider markets--relative market clout, pricing terms established on a payer-by-payer basis through the contract negotiation process with health care providers--could keep prices for health services artificially high.

This is what the vaunted robust public option debated in 2009 was to have attacked. And it's what single-payer would remedy. Presumably a single-payer plan would set the rates paid to providers, eliminating the possibility that an especially large or powerful health system could negotiate higher reimbursements just by virtue of its market clout.

But, yet again, similar outcomes could likely be achieved in a multi-payer system. Various academics have suggested that all-payer rate setting systems adapted from European models could be transplanted here to replace our haphazard pricing system. The state of Maryland has been using an all-payer system to set hospital reimbursement rates for decades, though their system underwent a transformation this year. If any state has the regulatory climate and a bite-sized enough market on which to cut our national teeth by testing out such a course it's Vermont.

* * *

So given the pioneering work Vermont is doing to drive multi-payer change, amid the backdrop of historic coverage expansions happening under the ACA, it's fair to ask how much would actually change in practice under a single-payer system vs. the kind of all-payer strategy for change the state is already pushing? I suspect the difference is ultimately fairly modest.

And that's because you can get most of the results folks hope for under a single-payer system by aligning all the parties in a multi-party system toward those goals. Single-payeresque outcomes--a single, unified and equitable system--in a multi-payer environment. E pluribus unum.