Value based care – reality, promise or a myth? (Part 2)
Opportunities around value based care and the Israeli angle
Part 1 of this post touched upon the shortcomings of VBC as currently practiced, in particular a) most VBC models are actually fee-for-service augmented with some element of a VBC payment model, and b) there is no consensus amongst payors and providers as to how to properly measure and demonstrate value.
In this second installment, I’ll suggest some opportunities to overcome these gaps, mainly around capitated models and unlocking risk bearing opportunities for providers. The Israeli healthcare system, which is based on a global capitated model for the past ~three decades (with extensive supporting digital infrastructure), serves as a helpful case study on how capitation can be sustainable and drive value to patients, providers and payors.
Utilization of Capitated Payment Models
The Israeli health system came into being in the early 20th century, at approximately the same time as the US system (and a couple decades before Israel was formally established as a country). Nonetheless, the context in which the US and Israeli health systems developed, and their stated goals at inception, were very different. These early differences set the two countries on different paths, and decades later, the downstream effects of these early differences remain embedded within the two systems.
In a nutshell (a tiny nutshell, for broader history I’d recommend “American Sickness” by Elisabeth Rosenthal), the modern US healthcare system started from medical facilities trying to overcome growing debts by offering the first kind of insurance. The model was simple, providers would offer “all inclusive” hospitalization for X days per year in exchange for a flat annual fee. Eventually, this crude system evolved into the notorious fee-for-service model that has defined the US healthcare system.
Around the same time, a group of Israeli labor associations realized employers were missing too many working days due to lack of access to clinical care. As working hands were scarce and each missing day was crucial, they decided to fund (provide) care for their members by collecting funds for each (hence the name “sick funds”). Quickly, the so-called “sick funds” expanded their mandate and began establishing community clinics by themselves to enable more accessible and affordable care (as there simply weren’t sufficient clinics to serve their beneficiaries).
Over the years, these sick funds grew larger, established hospitals and expanded their members poll to include most of the Israeli population under the same model – an annual fee that provides access to care with very limited co-payment. However, over time several problems emerged that drove the system to the verge of collapse. Heads up, this may sound familiar.
First, some sick funds only accepted healthy patients from high socioeconomic backgrounds. Meanwhile, sicker, poorer patients were still accepted by the more “social” sick funds, pushing them to bankruptcy. At the same, there was no enforcement of the scope of care required to be provided by each sick fund to its members. This led sick funds to restrict access to care, thus forcing the independent hospitals (some hospitals were owned by one sick funds, most weren’t), to the brink of bankruptcy. This led to a vicious cycle, where as a reaction to their shortfall in revenue, hospitals increased rates for the sick funds, prompting further restriction of access to care, and so on… (sounds familiar?)
The system was finally stabilized in 1995 when the “national health insurance” law was enacted, which led to the implementation of a sustainable capitated model.
Let’s pause for a moment.
Many elements in the pre-1995 structure of the Israeli system overlap with the current form of ACOs in the US. Interesting to note that ACO models started as the “poster boys” of VBC about a decade ago (see part 1) but now CMS is starting to back down and discontinue some of these programs as their value in cost reduction becomes questionable.
Interestingly, and in a way following the transition the Israeli system underwent, it seems like CMS is starting to double down on capitated models with the launch of direct contracting (Check out this great piece for more color and the difference between direct contract models and ACOs)
How to support a capitated model (and its benefits) – The Israeli way
For some background, I’ve already elaborated on the current (post-1995) structure of the Israeli system in another post, and made the case that the structure of Israel’s health system has been the driving force behind the success of the COVID vaccination effort. As short recap, post the 1995 legislation, each Israeli citizen became entitled to receive a rather broad “package” of medical services (e.g. treatments, services and medications). This “package” would be provided by 4 sick funds funded by the government through a global capitation formula that only considers limited socio-demographic parameters. The sick funds must accept any new member for whom they currently receive upfront payment of ~$1500 annually while offering (by law) the same type of “package”. Regardless of any member’s health status or how many times he or she utilized the services offered, the payment for the sick funds remains $1500 per year, i.e., universal coverage and global capitation. The results? Rather impressive (again, see the vaccination post).
By large, capitated models are articulated on rather fixed and general upfront payments providers need to handle with their cost structure (i.e., get the money first, then cope – prospective). Shared risk models are in essence much more focused on a particular patient, patient group or procedure (hence more specific risk calculations) and are more retrospective in nature as savings generated are measured against a target. (This is an over-simplification, as capitated models also have a varying degree of risk calculation to adjust the capitated amount, but it captures the theoretical difference between capitation and shared risk payment models).
To our point, the question is how the Israeli sick funds provide services with sub $1500 expenditures per patient (and thus being sustainable / enjoy the upside risk), while still maintaining high quality of care (on a global perspective).
Here is an attempt to breakdown the main elements that enable these results.
Focus on prevention (capitation = you want less procedures, less misuse of procedures and essentially keep people healthy)
It is well accepted that the earlier you intervene in the course of a disease, the better the outcome and the lower the cost. The real question is what structures and elements can enable early intervention and allow the system to operate efficiently and effectively for all stakeholders.
The Israeli mechanism is the following:
Low attrition rate (~1% annually). Despite no law restricting members from switching funds, they rarely do. This creates an incentive for funds to act in the long term interest of their members (i.e. worth to invest in prevention, since they will see the payoff down the road). The fact that families usually enroll in the same fund only increases this incentive.
Essentially no DRG codes, hence no incentive for extra procedures (as they won’t be reimbursed).
Strong and visible competition between sick funds, mostly on service level (one of Israel’s national pastimes is to compare…). What do patients complain about the most? Lack of access (as it is much harder to compare quality), mainly to specialists and diagnostic services, driving the sick funds to improve access to these activities. Both play a major role in preventive medicine.
Tight geographic distribution of clinics and hospitals. This can be traced to the genesis of the Israeli system, with many clinics and hospitals based in or near the original facilities established to serve workers in the early 20th century. Still today, clinics are organized and managed in clusters. This makes it more efficient to provide administrative support and to monitor patients and guide referrals to specialists, procedures and so forth.
Reliance on nurses, in particular to promote preventive care, including actively calling in members with chronic conditions (diabetic, HF and COPD for example) for follow ups, routine vaccinations (COVID…), HTN control (done in almost every clinic visit) and other procedures, such as diabetic foot encounters, chronic wounds treatments, stoma control, etc.
Focus on primary / out-of-hospital care to meet patients earlier (see below).
Focus on doing everything possible to avoid hospitalizations (see below).
Digital infrastructure to capture the entire continuum of care (see below).
Strong primary care and out-of-hospital system
Israel consecutively leads the OECD in terms of the ratio between outpatient and in-patient health expenditure.
~90% of all out-of-hospital care is owned and managed by the sick funds (compete on access…)
Constant use of data from multiple sources to study satisfaction and access to care. These surveys are frequently published, driving providers to practice on top of their licenses, and the sick funds to create better supporting venues.
GPs (called “family physicians” in Israel) get, on average, the best compensation packages (even more than surgeons).
Specialty care is provided predominantly in community settings (less cost, same results). Increasing number of fields are available, also as a first encounter (without a referral – limited gate keeping). There are an equal number of specialists and primary care physicians working in the community.
Provider-to-provider virtual consultations (GPs to specialists) are very common.
Zero to very low co-payment (less than five dollars) for primary care visits and lab work.
Moving as many procedures as possible out of the hospital (one of the last milestones was immune modulating treatment (like anti-TNF)
Multistep “shunts” to avoid hospitalizations
It’s 10pm and you have this annoying headache, and your temperature is high?
Pick up the phone and consult a nurse. Do a video encounter with a physician. Didn’t get the answer? Use a walk-in “emergency-center” (again, very limited co-payments) or a home visit by physician.
You just need to get your regular I.V. treatment but can’t come to the clinic? Continuum of care units, manned primarily by nurses, might be the answer. They are also great to maintain a proper follow up (as mentioned above) and allow the patient to cope better with his complex condition. None of that worked? Let’s talk about being hospitalized at home. Yes, not just long-term hospitalization, rehabilitation, and hospice but also acute home hospitalization for a wide array of indications.
An additional layer in the system involves highly efficient, narrowly focused care facilities, either to offer comprehensive, one-stop-shop care for a specific condition, like diabetes or child development or to offer the same small set procedures repeatedly, from diagnostics to surgeries (akin to ASCs) - “24 hours stay” hospitals devoted specifically to endoscopies, hernia repairs, or elective PCIs to name a few examples.
To emphasis, as opposed to the US system, most (>90%) of these facilities are owned by the sick funds, and the reaming 10% of private clinics are affiliated to one or more sick funds under long term contracts.
Digitalization (to enable care continuity)
The system relies on a centralized structure of four sick funds (two of which provide care for 75% of the population), and each has one major EHR system that links all community-based activities (primary care physicians, specialists, laboratories, and pharmacies) along with a nation-wide system (OFEK) that connects in-hospital and community clinics = interoperable data.
Tremendous attention is given to writing meticulous discharge letters (residents spend a big chunk of their time on this activity) + no incentive for upcoding (no fee-for-service) drives more precise documentation = more accurate data
Israeli residents have a unique I.D. Number (Israel identity) which is used in every encounter making all care interactions ID coded and geotagged= traceable data.
As there are more incentives coming from CMS to operate under capitated models and many of them (such as Direct Contracting) are granting increased flexibility on the benefits design and patient care initiatives that can be offered, there seem to be a big opportunity for more nimble players to shine. No surprise that many have applied to participate in the program (Iora, Oak, Clover, VillageMD, etc…)
In that perspective, it is interesting to follow the newly announced acquisition of Iora by One Medical, perhaps a hint on One medical’s desire to enter the Medicare market / risk on top of their current commercial population with FFS contracts.
To conclude this part, in my opinion, much can be learned from the operating structure of the Israeli sick funds and their evolution on the how new entrants and incumbents can better position themselves to seize capitated models.
Outside of capitated models there are some additional opportunities that can either enable more value-based interactions or to improve these kinds of engagements that are worth mentioning.
Enable more (efficient) risk bearing providers
Although participating in risk sharing models might be a lucrative engagement for providers, many pass of the opportunity simply because they don’t know how to calculate the risk or to reduce it. Risk-bearing is a core strength for insurance companies. They’re good at measuring risk and mitigating it (actuaries…) Most hospitals, on the other hand, lack the knowledge and resources to do it. This situation lead to a very asymmetrical relationship, where there is already great degree of mistrust. And all of that is only compounded by the additional difficulty of translating risk into cost (see first part of this post and this nice piece from McKinsey )
There are couple of large opportunities here:
“Risk bearing as a service” (assuming part of the risk / helping in calculating and managing the risk) – Attempts coming from payors / insurtech companies, like BrightHealth’s recently announced NeueHealth, are targeting this aspect (for more color there is a very comprehensive review of BrightHealth’s S-1 coming from Kevin O’Leary). Laguna health (who we are proud to have in our portfolio) is doubling down on this market opportunity, but is focusing specifically on post hospitalization rehabilitation.
Adding care provision capabilities to payors in a synergistic manner. Knowledge of how to manage risk – check. No need to justify large hospitals on their balance sheet (as opposed to IDNs) and thus opportunity to focus on community care – check. There are already many attempts by most of the large payors to add care provision elements (Optum / UHG, Anthem, Cigna, Humana…) by acquiring providers groups, adding care navigation / coordination / MSOs, telehealth… The main question is what is the right mix and how to glue together the payor side and the provider side.
Closing the “risk loop” (when you know the risk and now need to do something to reduce it) – Identifying a sub population (shall not say an individual) who is “intervenable” (currently have high risk for X and that risk can be reduced by some action Y) is helpful, but only address part of the problem. To truly reduce risk you must create an end-to-end intervention starting from identifying a true patient’s needs earlier in the care journey and then notify the relevant stakeholders, making sure he or she acts upon their directions, and tracking the impact (i.e. closing a care loop). This is a whole different challenge, even in a contained system. I’ve touched on this point on when discussing remote monitoring. Interesting example would be kidney disease management and how companies such as Somatus , RenalytixAI ,Strive and Monogram are attempting to do this.
New Ways to Measure Cost and Prove Value
Finally, the challenges in measuring costs also create some opportunities due to advancements in technology and supporting regulation (such as data interoperability and price transparency)
Overcome the network problem (lack of data and aligned incentives)
As care journeys are usually fragmented and managed by different stakeholders (primary care vs. specialists vs. hospitals, etc.) and contracts between them prevent providers from seeing the entire spectrum of care data (i.e., what happened to the patient in other siloes) it is hard to assess the real outcome of a given intervention. Seamlessly connecting data between different stakeholders (provider-to-provider and payer-to-provider) is not just a technological challenge (data harmonization, cleansing, overcoming missing data, converging claims and EHR data…) but also an incentive alignment issue (as in some cases specific stakeholders don’t want to expose the data…). Big tech (Amazon, Google, Salesforce) and many new entrants (Abacus, Innovaccer, Collective medical, SmileCDR, Diameter Health, Bridge Connector, Redox, Particle health and others) are placing big bets there.
Creating new data sources for a more holistic picture
In some events, even if you obtain all available data from all corners of the universe there still might be substantial blind spots resulting from the gaps between the different sites (what happened to the patient between visits) or from information not being documented during encounters (what happened intra visits). Remote monitoring is super relevant to eliminating these blind spots, but IMO there are very interesting opportunities in the “intra session” data capturing side as well, primarily revolving around utilization of voice and image recognition for example. We are working on couple of investments in these areas which I’ll hopefully share later on.
Proving value by different data analysis and research methods
Collecting the relevant data is one thing, but analyzing it is also required and is not a trivial task. The gold standard of RCT can be very expensive and in many cases isn’t feasible due to operational, financial or ethical constraints.
It is hard to understand where a specific intervention made a difference in a retrospective manner (observational study), especially considering that care is composed of multiple actions and thus isolating one intervention and successfully emulating RCT by creating a true counterfactual (control group) is super challenging. For instance, in trying to prove a reduction in hospitalizations, how can companies prove it was their intervention that caused the reduction? There are emerging approaches on how to tackle this issue.
Another angle is to embed prospective analysis tools into the product, similar to retail and edtech. Can you assemble any sort of A-B testing within your offering allowing you to learn and improve as you go?
To conclude, I think there are many insights that can be learned through comparative study of different health systems around the world, and other industries entirely, that shed light on opportunities to measure and create value in the US healthcare system. The better version of learning from your mistakes is learning from others mistakes 😊
Stay tuned
Want to have these posts directly to your mailbox?