I’d like to start doing an intermittent series of posts about how PHM is being implemented around the world. I think one can get tunnel vision focusing too much on how things work in their own backyard, so I figured it might be productive to zoom out a bit. PHM has so many moving parts that it’s unlikely that any one country or even region has thought of every possible angle.

Well then, what better way to begin this little odyssey than by looking at how the World Health Organization views PHM? Great idea, I thought. But, oddly enough, when you type “population health management” into the search engine on the WHO website, you get nothing.

One might think that this comes down to semantics. Maybe, one might say, PHM is an American term. But no, that’s not the case as I’ll detail below. The term “population health,” a closely related but distinct term, does return quite a few hits on the WHO site. But again, even though there is some overlap between PHM and population health, there does seem to be a consensus that they’re two distinct concepts.

Anyway, we have to delve a little deeper to find where the WHO comes down on PHM. After doing some legwork, I discovered that the organization has segmented its literature on different PHM frameworks into discreet sectors.

The sector I’ll focus on in this post is Europe. I’m sure there are plenty of good examples that demonstrate how the WHO envisions PHM rolling out across Europe, but the best breakdown I could find was in the form of a document called “The European Framework for Action on Integrated Health Services Delivery: an overview.”

The strategy outlined in this document shares many similarities with ones laid out by U.S. organizations such as the Institute for Healthcare Improvement (and their influential “Triple Aim”), the AHA, and others. Interestingly but perhaps unsurprisingly, elements such as “Stratifying health needs and risks,” “Identifying the determinants of health,” and “Linking provider payment mechanisms to performance” feature prominently in both the European and American approaches to PHM.

What seems a bit different on the surface, however, is the emphasis that European countries, along with the WHO, place on the concept of “co-locating” services. A good definition of “co-location” can be found on the website for the Rural Health Information Hub, a group supported by the Health Resources and Services Administration of the U.S. Department of Health and Human Services (HHS). They define the term like this:

Co-location refers to services that are located in the same physical space (e.g. office, building, campus), though not necessarily fully integrated with one another. Co-location can involve shared space, equipment, and staff for health and human services; coordinated care between services; or a partnership between health providers and human services providers. Co-location can streamline referrals, increase access to care, and increase communication between different providers.

Source: https://www.ruralhealthinfo.org/toolkits/services-integration/2/co-location 

This idea of co-locating services can be found in the subtext of most materials on PHM implementation in the U.S. But I’ve noticed that it seems to take on more paramount importance, and is more explicitly mentioned, in literature about the European approach to PHM. For example, the King’s Fund in the UK did a great video about the country’s vision for rolling out PHM over the next few years. At about the 29:30 mark of the video you’ll find one of the panelists discussing the key nature of co-location to the overall vision.

Another terrific example of the concept of co-location and PHM pops up in a case study, endorsed by the WHO, where PHM strategies were deployed in an Austrian municipality called Schwaz in 2016. The introduction to the case study notes the crucial importance of building community to optimize people’s health:

…co-location of services increased coordination and improved access for vulnerable groups; shared living spaces, group activities and cooperative initiatives within the House of Generations generated a sense of community, fostering a supportive and inclusive environment for vulnerable groups…

Source: https://www.integratedcare4people.org/practices/319/house-of-generations-schwaz-a-one-stop-shop-for-providing-health-and-social-services-in-austria/ 

As the case study describes, in the early 2000s “the Municipality of Schwaz observed a narrowing of family structures within its ageing population.” I take this to mean that a breakdown of the family unit was becoming worryingly obvious, especially among vulnerable groups. Adding to this already concerning situation, a “fragmented organization of local health and social services, coupled with a lack of coordination between the various actors involved, hindered the system’s ability to respond to increasing demands.”

So the city took action. 

To ensure better coordinated service delivery, by 2007 they’d built a “’one-stop shop’ for health and social services.” I won’t go into the details, but building up an infrastructure where people’s health needs can be met all in one place, and where their care is consequently more tightly coordinated, has fostered a strong sense of community spirit that has led to an overall improved quality of life for residents.

In all the talk of coordinating care in the U.S., there seems to be less of an emphasis on the proximity of that care. Part of this is no doubt due to the vastly geographically larger nature of the U.S.; indeed, it’s not unheard of for people to travel from one state to another, driving hours each way, in search of quality care. Because their populations are smaller in European countries, and because (in my estimation) mobility from one city to another isn’t as integral to the culture as it is in the U.S., organizing coordinated care around local communities to cultivate healthier lifestyles is more of a natural fit in Europe. Still, there’s something to be said for incentivizing people to forge stronger bonds in their local community, and this seems to be the guiding principal for much of the work in this space across Europe.

Stay tuned for more installments in this series. I’m excited to learn more about how other regions are handling this shift to more targeted, coordinated care, a trend which seems to be taking root across the world. 

This week I attended the virtual Health Care Payment Learning & Action Network (HCP-LAN) Summit. For those who don’t know, the HCP-LAN, or “LAN” for short, is, in their own words, “an active group of public and private health care leaders dedicated to providing thought leadership, strategic direction, and ongoing support to accelerate our care system’s adoption of alternative payment models (APMs).” 

CMS, along with a host of private payers, is interested in motivating health care entities to share in risk alongside insurers to lower healthcare costs and slow down out-of-control healthcare spending. The LAN works to make this transition to risk-sharing between insurers and providers happen.

So how does this fit into population health management?

Before going on, it might help to understand what risk-sharing means exactly, and the reasoning behind this paradigm shift. I found a good, straightforward formulation of this on a blog for a health insurance company I’ve never heard of called “Oscar.” I’m not endorsing this company, but I like their wording:

By sharing the risk, you (i.e. the patient), your insurance company, and doctors/hospitals are protected from individually bearing the costs for those who need a lot of health care or have medical emergencies during the year…From a statistical perspective, risk sharing works best when there are more people in a group – and this is especially true for health insurance.

Source: https://www.hioscar.com/blog/how-health-insurance-works-risk-sharing

For the sake of clarity, I added the “(i.e. the patient)” to the quote. And there’s more nuance to this, such as the fact that this kind of pool is optimized when not only older and less healthy people, but also younger, healthier people are included. But for our purposes, the above quote serves as a good starting place.

Suffice it to say that there is a big transition going on right now in the way medical providers are reimbursed, and the LAN is working to build a smooth glide path for that change. In the past, insurance companies (whether public or private) have paid doctors based on individual services rendered to patients under a pay structure called “fee-for-service” (FFS). But this approach led to a wasteful, inefficient system where care hasn’t been tightly integrated, costs have spiraled out of control, and quality outcomes haven’t kept pace with our peer nations. This result is down to care delivery not being viewed holistically, but rather as one-off events that don’t take the full scope of each individual patient into account.

To remedy this, CMS and a growing number of private insurance companies are transitioning to a value-based, coordinated care environment where the reimbursement structures (or payment models) are based on shared risk. In other words, providers who hit certain pre-determined cost and quality targets and who consequently save the health system money can share in these cost savings by receiving payment bonuses. This is what’s called an “upward payment adjustment.” In some payment models in this space, there are only “upside-only” or “one-sided” risk-sharing arrangements in which doctors are only rewarded if they beat expectations, but don’t get dinged financially if they don’t.

But in the future, if CMS has their way, it appears that all doctors and healthcare entities will move into two-sided risk models. The idea behind this flavor of payment model is that costs are held down and quality patient care is enhanced when providers are held accountable for patient outcomes. The most prominent kind of two-sided payment model is the accountable care organization, or ACO. “Health Affairs” has a handy definition of the two-sided risk payment model:

In two-sided risk models, providers still share in the savings but are also responsible for some of the loss if spending is above the benchmark.

Source: https://www.healthaffairs.org/do/10.1377/hblog20181011.442864/full/

That’s where the LAN comes in. They act as a bridge between the current, mostly FFS landscape and a future where healthcare providers are held accountable for cost and quality care. This group brings together payers, providers, manufacturers, policymakers, patients, and others to figure out new ways to, as they say, “lower care costs, improve patient experiences and outcomes, reduce the barriers to APM participation, and promote shared accountability.”

So this past week I attended a few LAN Summit sessions, and I noticed something that only earned a passing mention during one of the presentations: population-based payment (PBP) models. Although one of the less utilized alternative payment models, it seems to have made steady progress in the past couple of years toward adoption by a growing number of healthcare practitioners.

So what are population-based payment models? I actually didn’t know anything about them until this meeting, so I decided to look into them. Here is a helpful definition by the LAN:

The core premise of a PBP model is that providers are accountable for patient-centric care for a specific population over a fixed timeframe and across the full continuum of care.

Source: https://hcp-lan.org/2016/05/pbp-models-overcoming-barriers-accelerating-adoption/

The LAN has convened a PBP Work Group, which has identified four priority areas essential to adoption of these models:  patient attribution, financial benchmarking, data sharing, and performance measurement. One reason these models have lagged in their uptake among healthcare providers seems to be because, unlike other, more popular models, the PBP model is completely uncoupled from FFS. As the site Health Payer Intelligence notes:

Population-based payment models are models that are detached from fee-for-service reimbursement altogether, as opposed to other alternative payment models that continue to build off of fee-for-service structures…

Source: https://healthpayerintelligence.com/news/how-payers-can-support-population-based-payment-model-uptake 

Evidently the LAN has identified three types of PBP models, which are as follows: condition-specific population-based payment model, comprehensive population-based payment model, and integrated finance and delivery systems model. I won’t go into the finer points of these, which can be found in a 2017 white paper published by the LAN, but overall, as they say, “…(PBP models) represent the furthest departure from traditional FFS payments, while they simultaneously ensure that providers possess the strongest possible incentives to deliver high-quality and efficient care.”

I could easily see writing a future blog post that breaks down the details of each type of PBP model. I have a feeling that motivating providers to enter into this kind of forward-thinking model will involve developing new quality metrics that incentivize them to do so. I will keep track of this and update the blog with any new developments I come across. 

Over the past two years or so, the Centers for Medicare and Medicaid Services (CMS) has placed increasing importance on a program they’re in the process of developing which is collectively called MIPS Value Pathways, or MVPs. MVPs are seen by some as the next step in the process of incentivizing physicians and everyone else in medicine to move away from the Merit-based Incentive Payment System (MIPS) as their value-based payment model of choice, and instead begin taking on risk as part of any one of a number of advanced alternative payment models.

So what is MIPS, what are value pathways, and what the heck does any of this have to do with population health?

Let me explain it as I see it.

What is MIPS?

There have been a couple of iterations of MIPS so far, but the most concise definition I’ve seen was put forward by the American Academy of Physical Medicine and Rehabilitation:

The Merit-Based Incentive Payment System (MIPS) is the program that will determine Medicare payment adjustments. Using a composite performance score, eligible clinicians (ECs) may receive a payment bonus, a payment penalty or no payment adjustment.

Source: https://www.aapmr.org/quality-practice/quality-reporting/merit-incentive-payment-system 

MIPS is one of two tracks set up under something called the Quality Payment Program (QPP) by CMS, the other track being something called Advanced Alternative Payment Models (APMs). I won’t spend much time on APMs in this post, but they’re basically the future of physician reimbursement and probably deserve a post of their own.

Anyway, under MIPS, participating physicians, groups, or health systems report the measures and activities they collect during a performance period within four different performance categories: Quality, Improvement Activities, Promoting Interoperability, and Cost.

I won’t go into the myriad activities that count toward attesting to these requirements, but suffice it to say that participants earn a composite score in these four domains and, as the AAPM&R website says, “Your final score determines the payment adjustment applied to your Medicare Part B claims.”

Put simply: if you outperform the overall measurement expectation, you’ll earn a positive pay adjustment and receive a bonus. If, on the other hand, you underperform on the metrics, you’ll get a negative payment adjustment. On the bright side, CMS collects and calculates all cost measures, so at least that burden doesn’t fall on the physician. 

This formulation is at the heart of risk-sharing as conceived by CMS. The point of all this is to rein in out-of-control medical spending while improving the patient experience. I’ve left out plenty of other details about MIPS, such as the fact that not every physician fits into the MIPS framework and are thus exempt from participating, but you get the general idea.

What are MVPs?

I’m glad you asked, because I’m still trying to figure this one out myself. For the most part, MVPs seem like an attempt by CMS to overhaul the MIPS process because they got a lot of negative feedback from the medical establishment that participation was too challenging.

So through MVPs, CMS is attempting to streamline the reporting process while at the same time empowering patients to take on more of an active role in their care. And in another encouraging sign, CMS is actively soliciting ideas for MVPs from the field.

According to a 2019 article in Health Affairs, CMS has put forward several guiding principles which it hopes will resonate with the house of medicine. According to the article:

…value pathway design should reduce barriers to provider participation in alternative payment models (APMs) by prioritizing quality measures that align with improvement activities and cost measures. CMS proposes holding clinicians from all specialties accountable for a foundational set of health information technology and population health measures, while allowing quality, improvement activity, and cost measures to vary by clinical specialty. Creating common measures to assess performance in fee-for-service (via MIPS) and APMs may facilitate the transition to APMs.

Source: https://www.healthaffairs.org/do/10.1377/hblog20191107.686469/full/ 

To put a finer point on it, here is how CMS explains this new approach:

The MVPs framework aims to align and connect measures and activities across the quality, cost, and improvement activities performance categories of MIPS for different specialties or conditions. In addition, the MVPs framework incorporates a foundation that leverages Promoting Interoperability measures and a set of administrative claims-based quality measures that focus on population health/public health priorities and reduce reporting. 

Implementing the MVP framework honors our commitment to keeping the patient at the center of our work. In addition to achieving better health outcomes and lowering costs for patients, we anticipate that these MVPs will result in comparable performance data that helps patients make more informed healthcare decisions.

Source: https://qpp.cms.gov/mips/mips-value-pathways

So not only is patient empowerment at the heart of this program, but these measures will “focus on population health/public health priorities.”

What Role Does Population Health Play in MVPs?

This is the part I’m foggiest about. Part of the reason for my confusion is that CMS is still developing this concept, so it’s only natural that there would be gaps in the public’s understanding of it. Speaking of gaps, in 2020 CMS commissioned something called an “MDP Population Health Environmental Scan and Gap Analysis Report.” As stated above, one major goal of this report — and of MVPs in general — was to bring about “a future state of MIPS that CMS envisions to include a foundation of measures focused on population health.”¹

But when CMS highlights “population health” as important to their program, it’s not entirely clear to me what they mean by “population.”  I previously covered the flexibility of this term, so for more on that, check out this blog post.

In other words, are they taking an inclusive perspective, where “population” can mean everything from designing metrics around patient populations in clinical settings to those predicated on geographic pools of potential patients, such as a county with higher-than-average smoking rates? Or do they have a narrower definition in mind?

According to the report, “The scan was organized by six topics of population health: access, clinical outcomes, coordination of care and community services, health behaviors, preventive care and screening, and utilization of health services.” To me, this signals that CMS is applying a more inclusive definition of “population” beyond just the clinical setting. And they seem to be really keying in on underserved populations in both urban and rural settings. Much remains to be seen, though, so stay tuned.

Endnote

1. CMS Quality Measure Development Plan. “2020 Population Health Environmental Scan and Gap Analysis Report For the Quality Payment Program.” Page 1. Accessed on 12/1/21

I’ve been reading quite a bit about population health management (PHM) lately since the concept seems to be taking on increasing importance in medicine. PHM has been around for decades, but only now is it catching everyone’s attention, particularly in light of COVID which has laid bare the serious flaws with the U.S. medical system.

Bottom line: the medical system likely can’t continue in its present form without delivering worsening levels of care to large swaths of the American public. So it’s of paramount importance for people to be incentivized to live healthier lifestyles to avoid them ending up in the hospital system for as long as possible.

For me, when trying to get my mind around an enormous topic like PHM, it helps to start with first principles. In the case of PHM, in my opinion, that takes the form of asking who is the “population” in “population health management”?

Well, the answer to that question depends on who you ask. As part of their Medicaid and Public Health Partnership Learning Series, the Association of State and Territorial Health Officials (ASTHO) released a document titled “Public Health and Population Health 101” which I find helpful.

As their website states, ASTHO is the “national nonprofit organization representing public health agencies in the United States, the U.S. Territories, and the District of Columbia.” Because of the constituency they represent, their view is pretty comprehensive. To me, this document provides the best breakdown of the differences between public health and PHM — and the best parsing of the word “population” within that definition — that I’ve come across yet.

Here is what ASTHO says about what the word “population” means to different groups, and why, because of this range of views, it’s so challenging at this point to nail down a definition of PHM:

For example, public health officials, community organizations, and business leaders often use the word “population” in the geographic sense, because they work in terms of geographic units (e.g., federal, state, or county.) This usage differs from how “population” is used in the clinical healthcare system, which often more narrowly refers to patients using a facility within a specific time period, members of an insurance plan, or patients being treated for a specific diagnosis.

And then there are insurers, particularly Medicare and Medicaid, who have their own definition:

Medicaid typically uses the term “population” to refer to the total population of Medicaid beneficiaries (a subgroup of the total United States population) or a specific subgroup of beneficiaries, like pregnant women, children, seniors, individuals with disabilities, or dual eligible individuals. Medicaid serves its part of the population by constructing health plans and reimbursing providers.

So, as you can see, it may be a while before everyone has a clear picture of what the terms “population health” and “population health management” mean. At the end of the day, I suspect that the terms will go away at some point, replaced by more precise terms that better align with the various stakeholders involved.

I’ve been reading about a concept that’s been around for a while but I’m only now becoming aware of: narrative medicine. It has interesting connections to population health management, and places a premium on building empathy between physicians and patients in this era of hyper-short interactions.

According to an article on the Association of American Medical Colleges’ website, narrative medicine is a discipline taught in a handful of medical schools that helps future doctors better understand how to interact with patients on a human level.

The article connects this approach to population health management in the sense that understanding each patient’s specific life circumstances (i.e. social determinants of health) is at the heart of narrative medicine.

Those stories can illuminate how a person became ill, the tipping point that compelled them to seek help, and, perhaps most importantly, the social challenges they face in getting better. 

AAMC website

Fans of prose fiction and poetry — and creative non-fiction, for that matter — will be aware of narrative’s power to transport readers into the consciousness of characters that may, by outward appearances or circumstances, seem much different than themselves. In this way, the reader comes to strongly identify with someone to whom they might otherwise not have given the time of day. Practitioners of narrative medicine assert that by teaching future doctors to develop this same kind of empathy for their patients, overall patient health can improve.

Interest in learning the craft of writing has taken hold in medical programs like the Columbia University College of Physicians and Surgeons, Brown University’s medical school, and the University of Nevada, Reno School of Medicine. Rita Charon, MD, PhD, executive director of Columbia’s Program in Narrative Medicine and founder of the field, says, “It’s not a stretch to say we need help to look at our own processes or to see and appreciate what patients are telling us. For me, it became a way for patients to feel heard and noticed.”

That last part really gets me: “A way for patients to feel heard and noticed.” Traditionally, doctors were often considered diagnosticians, and still are to some extent. But it feels like things are changing: now, although time is still at a premium and interactions between healthcare providers and patients is too short, there seems to be a growing acknowledgement that the patient should be allowed to exercise some amount of agency over their own course of care.

Interestingly, baked into the concept of narrative medicine is “close reading,” another approach borrowed directly from literary studies. As defined by the writing center at the University of Wisconsin-Madison, close reading is a “deep analysis of how a literary text works; it is both a reading process and something you include in a literary analysis paper, though in a refined form.”

Fiction writers and poets build texts out of many central components, including subject, form, and specific word choices. Literary analysis involves examining these components, which allows us to find in small parts of the text clues to help us understand the whole.

The Writing Center at the University of Wisconsin-Madison

Understanding the constituent parts of a patient’s life can begin to help their care team to hone a care pathway that conforms to their actual life circumstances. Such an outlook goes beyond aspects that can be entered into an electronic health record, such as which zip code they live in, whether or not they have reliable transportation, and how likely they are to adhere to different care regimens. And treating the whole patient in this way can be an antidote to physician burnout as well, giving doctors permission to construct rich stories around those in their care as opposed to seeing them as a collection of data points.

Because of the opportunities it affords physicians to gain a deeper appreciation for those they treat, the discipline of narrative medicine has spread beyond the U.S. Several medical schools in Taiwan, for instance, have incorporated this new mode of understanding into their curricula. In one study of 15 participants engaged in narrative medicine or medical humanity education at various medical schools and hospitals across Taiwan, the authors “performed manual thematic analysis to identify the themes and concepts of narrative competence through a six-step theme generation process” with an eye to “exploring the dimensions and connotations of the clinical narrative competence of medical personnel.”

Unsurprisingly, the study found that “Cultivating narrative competence in medical education can complement traditional biomedical orientation” and that “(p)atients often use metaphors or symbols to explain their conditions.” In other words, treating patients as one-dimensional figures ignores a rich source of information that could help physicians deliver higher quality care and improve the health of entire populations.

I’m currently taking a great Coursera class on population health management (PHM) called “Assessing and Improving Community Health.” I find PHM to be a very interesting area. Its shares many similarities with the more well-known concept of public health, but there are also some differences. The teacher does a fantastic job of navigating the intricacies of both public health and PHM, but to me, the biggest factor that distinguishes these two concepts lies in which areas of the health spectrum they focus.

Here’s what I mean: while public health practitioners often deploy interventions on a clinical level (and initiatives under this umbrella are constrained via funding to stay within these boundaries), PHM initiatives seek to intervene further upstream, in areas that medical care doesn’t touch such as education, nutrition, exercise, safety, and overall healthy lifestyle choices.

What I like so far in my PHM journey is that its practitioners — along with those in the public health space — are very non-judgmental. Most people, I think, realize that what zip code you’re born into determines a lot about the paths your life takes. Often times just getting kids enrolled in pre-K schooling, for instance, can have cascading effects that last throughout a person’s entire lifespan.

This brings me to community needs assessments. The Coursera class explains at length how these assessments can be utilized to identify health issues in communities. According to the National Academy for State Health Policy:

Tax-exempt hospitals receive billions of dollars in tax exemptions each year. In exchange, they are required to invest in the health of their communities. But to do that, hospitals must first identify the health needs of the communities they serve.

Because of this, federal law requires tax-exempt hospitals to conduct community health needs assessments, or CHNAs. These must be done every three years, and the IRS is even involved. The NASHP goes on to say the following:

The US Internal Revenue Service (IRS) requires tax-exempt hospitals to solicit and take into account input from at least one state, local, or tribal public health department as well as from medically underserved, low-income, and minority populations in their communities. It also says that hospitals may solicit input from consumer advocates, community organizations, academics, local governments, school districts, providers, health plans, business, and labor representatives.

This approach ensures that these hospitals spend more on the communities that surround them. I’ll be sure to add more information as I get further into the class, but that’s all for now.