Before Arlo became a full-stack health insurance for small and mid-sized businesses, Karthik and I set out to improve the member experience through a care navigation solution. We tried hard selling it to incumbent health plans. But even with so-called “modern” plans, it felt like walking through glue. At one point, we found ourselves on a Zoom with an incumbent carrier organization, pitching to a group of 20 VPs—each of whom wanted a say in shaping the member experience. The proposed pilot start date - 4 months out.

That’s not a model for change. It’s a model for preserving the status quo: no risk-taking, no rocking the boat, just incremental improvements at the margins.

And while carriers only are really interested in solutions that do incremental improvements, at the same time they complain about “point solution fatigue” - when really, they should call it what it is: innovation blocking. These organizations are too big, with too many stakeholders to satisfy, too many legacy business models to protect, and no organizational or technical capacity to drive real change.

But healthcare needs change. There’s so much broken—and so much opportunity to make health insurance not a completely miserable experience:

• Insurance should offer peace of mind - yet when you go to the doctor, you have no idea if it’ll cost $100 or $2,000.

• Benefit designs are confusing. (Do you really know the difference between coinsurance, deductible, and copay? Most people don’t.)

• Health plans have outsourced member service to HR teams and brokers—carrier helplines are of little use when issues arise.

So we knew that if we wanted to move beyond incremental change, we had to think bigger. That’s exactly what we did.

Over the past two years, we’ve been building the foundation for continuous innovation at Arlo. We invested in the technology and underwriting platform needed to launch new kinds of products - quickly, sustainably, and with the flexibility to evolve them over time.

It hasn’t been easy. Building mission-critical insurance infrastructure from scratch is quite different from the typical YC-style iteration loop, where you get fast, and immediate feedback from customers whether you’re on the right track or not. And getting brokers and employers to trust a new player in something as critical as health insurance takes time. After all, they are putting the livelihood of their employees in our hands. I’ll never forget standing on a golf course, mimosa in hand, trying to convince brokers about Arlo - only to watch them drive off in their cart with their BUCA carrier rep in the same foursome.

But our bet is paying off. By doing the hard things first - owning the critical parts of the stack end to end and laying the right foundation - we’ve set ourselves up to move fast when it matters. We are now not just a point solution selling to incumbents, but directly competing with United Health Care, Aetna and Blue Cross Blue Shield. We can now move at a speed, that these organizations just can’t.

Take Arlo Clear as an example. It began with a simple question: what if we could remove the financial uncertainty people face when they go to the doctor? Our members told us this was one of their biggest concerns, so we built a product around it:

• In one week, we created a prototype and completed the actuarial analysis to ensure pricing sustainability

• We brought the idea to brokers and employers to gather feedback.

• We iterated on the concept and, after another round of feedback, saw clear demand.

• Within weeks, we made the necessary operational changes, added it to our quoting portal, and launched a sales campaign.

• Just 10 weeks after the initial idea, we’re going live with our first Arlo Clear customers - faster than many software product launches.

I’m so proud of the team that made this possible. Huge shoutout to Sean, our Head of Actuary, and Tanmay, our Head of Health Plan Product. They moved mountains.

We’re not held back by outdated tech, bloated org structures, or the need to protect a legacy business model. That’s our superpower.

And Arlo Clear is just the beginning. We believe AI will fundamentally reshape how people interact with the healthcare system—and what insurance will look like. We’re ready to rapidly deploy what’s next. Stay tuned—we’ve got more exciting launches in the pipeline.

We’re building Arlo into a place where things actually get done in healthcare. If you’re tired of bureaucracy, tired of meetings about meetings, and tired of watching good ideas die in committee - Arlo is different.

You can help reshape healthcare here - and see the results in weeks, not years. Come join us. We’re just getting started.

When I am not writing, I am the CEO & Co-Founder of Arlo. We're on a mission to bring affordable health insurance to small businesses across the U.S.—and we need passionate, driven people to help us do it. We want to hear from you if you're excited about using cutting-edge technology to improve a complex health care system! We're hiring Engineers & Sales professionals. Explore our open positions here!

At Arlo, I’m often asked, "Why are your health plans priced so low?" It’s a valid question—after all, sustainable underwriting and not underpricing risk are crucial to avoiding the downfall of carriers like Bright Health and Friday Health.

In this article, I will share our secret sauce: We eliminate the hidden fees that traditional carriers bake into their plans.

A recent Health Affairs study revealed that the big three ASO (Administrative Services Only) carriers—Cigna, Aetna, and Anthem— directly charge employers around $225 per enrollee per year, much more than independent administrators would charge. This is an average number, and you can assume that smaller employers pay much higher fees than larger ones.

But these direct fees are not the whole picture. Beyond these administrative fees, large carriers also generate revenue through their pharmacy benefit managers (PBMs). Furthermore, the ASO carriers are making additional revenue by charging the employers fees categorized as claims expenses, such as visits to provider groups they own, or vendor fees, such as out-of-network repricing fees.

So, how exactly are the administrative fees broken down, and are they truly necessary for running a self-funded (or level-funded) health plan?

In this article, I’ll examine the main components of these fees, including:

• Administrative Fees

• Medical Management & Utilization Review

• Network Access Fees

• Marketing & Broker Fees

• Vendor Fees

• PBM Fees

• Fees Disguised as Claims

Understanding Direct Fees in Health Plans

Health plans have several direct fees, which are explicitly outlined and paid on a per-employee-per-month (PEPM) basis. Let’s examine the most common ones and how they impact overall plan costs.

1. Administration Fees ($5-$60 PEPM)

When a company offers a health plan to employees, they will hire a Third-Party Administrator (TPA), which handles the core operational tasks of the plan, including:

• Processing, adjudicating & paying claims

• Conducting eligibility checks

• Managing member service

• Handling compliance and reporting

TPA fees can vary widely, typically ranging from $5 to $60 PEPM, depending on their scope of service and the size of the group - large groups get substantial discounts for volume. If a TPA charges a fee below $20, it often indicates that they are cross-subsidizing their revenue through other hidden charges, such as vendor markups or claim-based fees.

2. Network Access Fees ($6-$35 PEPM)

Health insurance runs on provider networks - basically, nobody pays the list price at the doctor, but 95% of the time, they are paying some form of ‘contracted’ rate. Different entities hold these contracts with the providers, and some are renting these contracts to others to access their preferred pricing. The major players—such as Cigna and Aetna— rent their provider network to TPAs as long as they follow their rules. There are also smaller regional networks, such as Midlands Choice and Alliance/Trilogy.

Some players don’t offer a comprehensive network but still offer preferred pricing contracts for certain types of procedures, including:

• Surgical bundles (e.g., Carrum Health, Employer Direct)

• Cash-pay networks (e.g., Coral, MDsave, Sesame)

Groups must pay network access fees to utilize these contracts, typically ranging from $6 to $35 PEPM. In some cases, these fees are charged as a percentage of claims instead of a fixed PEPM rate (more on that below)

3. Medical Management & Utilization Review ($5-$15 PEPM)

Medical management plays an essential role in ensuring that healthcare services are appropriate and cost-effective. This involves:

• Reviewing proposed treatments and medications to confirm medical necessity

• Identifying cost-saving opportunities without compromising care quality

• Managing chronic conditions and preventive health programs

TPA often outsources these services to specialized vendors, costing $5 to $15 PEPM.

4. Marketing & Broker Fees ($20-$70 PEPM)

To distribute their plans and facilitate sales, health insurers include marketing-related costs. These fees are primarily allocated to:

• Broker Commissions: Brokers advise employers on plan options and assist with enrollment. Brokers typically get compensated between $20 to $50 PEPM

• General Agency Fees: General Agencies operate like wholesalers, helping carriers reach individual agents and smaller agencies. General agencies compensation usually ranges from $5 to $15 PEPM

Many traditional carriers build these fees directly into their premiums, making it challenging for employers to understand the actual cost of their broker. However, the CAA now requires brokers to disclose their compensation when their clients ask.

5. Additional Vendor Fees (Varies)

Many employers enhance their health plans with additional services that improve member experience and drive cost savings. These services include:

• Telehealth providers (e.g., virtual doctor visits)

• Specialty Care Programs like - Musculoskeletal (MSK) programs to address back pain and joint issues

• Concierge care navigation to assist employees in finding high-quality care

• Analytics tools to give advisors and employers insights into spending data

The cost of these add-ons varies widely, depending on the vendor and level of service provided.

6. Pharmacy Benefit Management (PBM) Fees (Varies)

PBMs are a major source of revenue for traditional insurers, and their fees often fall outside the standard administrative costs. Employers typically encounter PBM-related charges such as:

• PEPM administrative fees which range between $0-$30 PEPM

• Dispensing fees of $3-$10 per prescription

• Rebate retention fees, where PBMs keep a portion of manufacturer rebates

• Spread pricing, where PBMs charge the plan more than what they reimburse pharmacies

• Specialty Pharmacy Revenues: The large PBMs often own specialty pharmacies where they steer members to certain high-cost drugs.

Despite growing pressure for transparency, many large PBMs still include hidden fees that erode employer savings.

7. Stop-Loss Insurance and Risk Premiums (20-30% Load)

Most self-funded plans (including all level-funded plans) include stop-loss insurance to protect employers from catastrophic claims by capping their financial risk. However, insurers charge a risk margin, which represents an additional buffer over the expected claims cost to account for volatility.

Independent stop-loss carriers typically aim for a 70-80% loss ratio, meaning that 20-30% of premiums are retained to cover potential losses and their administrative expenses.

Stop-loss insurers owned by the large ASO carriers often operate at lower loss ratios (80-90%), but they compensate by embedding other fees elsewhere in the plan.

Understanding Indirect Fees in Health Plans

While direct fees are explicitly outlined in a health plan’s cost structure, indirect fees are often hidden within the claims, making them harder to detect. These fees can significantly impact an employer’s overall spending without clear visibility into where their dollars are going.

In many cases, vendors structure their fees in a way that allows them to retain a percentage of the savings they generate. While this can be a performance-based approach, it can also lead to misaligned incentives that ultimately drive costs higher rather than lower.

1. Claims Review & Audit Fees

Errors in medical billing are common, often favoring providers over employers and employees. To combat this, health plans use claims auditing vendors to review bills for accuracy, such as:

• Verifying that services were actually provided

• Ensuring proper coding was used

• Identifying overbilling or duplicate charges

While audits can recover significant savings, vendors typically charge a percentage of the savings identified. Conflict of interest can arise if the TPA performs the claim audit since the TPA may be incentivized to oversee errors in their claim adjudication process and only recover them later for an additional fee.

2. Out-of-Network (OON) Repricing Fees

When employees seek care outside the plan’s network, carriers often employ repricing vendors to set the reimbursement to a more reasonable rate. However, these vendors usually take a cut of the “savings”, which are pegged to inflated benchmarks that don't always align with market rates.

A recent New York Times article highlighted how some carriers use OON repricing as an opportunity to introduce hidden costs, ultimately driving expenses up for employers.

3. International Prescription Drug Sourcing Fees

To combat the rising cost of specialty drugs, some health plans are turning to international sourcing programs, which procure medications at a lower cost from countries like Canada. While this can lead to substantial savings, vendors often charge a percentage of the savings, which may result in fees that are disproportionate to the effort involved in sourcing the medication.

In some cases, employers have found that the fees paid to the vendor exceed the actual cost of the drug, significantly eroding potential savings.

4. Provider Access Fees for Bundled Services

Some organizations offer access to specialized provider networks for services such as:

• Surgical bundles

• Centers of Excellence programs

• Cash-pay provider arrangements

Rather than charging a flat PEPM fee, these vendors may bill employers a percentage of the provider’s billed charges.

5. Subrogation Fees

Subrogation occurs when a health plan pays a claim that should have been covered by another insurance policy—such as workers’ compensation or auto insurance. Specialized vendors recover these payments on behalf of the plan but retain a portion of the recovered amount as their fee. While this can bring valuable recoveries, these fees can add up.

Why Indirect Fees Matter

Indirect fees add complexity to health plan pricing, making it difficult for employers to fully understand their actual administrative costs. These fees often appear under vague categories like "shared savings" or "performance incentives," which can obscure their real financial impact.

Key considerations when evaluating indirect fees:

1. Alignment of incentives: Are vendors genuinely driving savings, or are they profiting from unnecessary utilization?

2. Transparency: Are all fees clearly outlined and accounted for in the claims fund?

3. Benchmarking: Are savings being measured against realistic, market-based benchmarks rather than inflated baselines?

Taking a Holistic Approach to Health Plan Fees

Not all fees are inherently bad—many serve essential functions that ensure a health plan operates efficiently and delivers real value to members. However, the key to sustainable and cost-effective health benefits lies in taking a holistic approach to assessing fees and understanding their true impact on the plan's overall cost.

At Arlo, we carefully evaluate each plan we power, ensuring that every dollar spent contributes to better health outcomes and a smoother experience for both employers and employees. We partner with organizations that are transparent about their fees upfront and charge fees that are aligned with the value they provide.

The healthcare industry is notorious for unnecessary complexity and hidden costs, but we believe it doesn’t have to be this way. By leveraging automation, smarter technology, and operational efficiency, plans can eliminate excessive fees and reduce administrative burdens without compromising the quality of care.

Our mission is simple: to strip out the inefficiencies, focus on what truly matters, and deliver health plans that are lean, effective, and transparent.

If you want to join us on this mission, you can find our open positions here.

It’s hard to believe how far we’ve come in just a year. Looking back at 2024, I’m reminded of a moment in Q4 when we received a note from a small business owner who had just enrolled in an Arlo health plan. They told us how switching to an Arlo-powered plan not only saved them money but also allowed them to offer comprehensive health coverage to their employees for the first time ever. Moments like these remind me why we started Arlo - to make affordable, high-quality health care accessible to more people. I finally found some time to reflect and look back at some of the things we have achieved and learned from the last year of building a health insurance startup.

This year has been a whirlwind. From launching our industry-leading online quoting platform to partnering with some of the best stop-loss carriers in the industry, we’ve come a long way - and I am proud of the entire Arlo team that got us here. We’re starting 2025 with over 200 small and medium-sized businesses covered under an Arlo-issued policy, and this is just the beginning.

Building a new, modern health insurance company hasn’t been easy - it has been full of challenges, pivots, and hard-learned lessons - but it’s also been incredibly rewarding. As we step into 2025, I wanted to take a moment to reflect on our journey and share some insights we’ve gained along the way. I deeply respect the many fellow builders in this space, and I hope some of these lessons resonate and prove helpful.

We're on a mission to bring affordable health insurance to small businesses across the U.S.—and we need passionate, driven people to help us do it. We want to hear from you if you're excited about using cutting-edge technology to solve complex challenges! We're hiring Engineers & Sales professionals. Explore our open positions here!

Lessons 1: Finding our Focus - Building an AI-based Underwriting System

From the beginning, we knew that to fundamentally change how people experience healthcare, Arlo needed to be a health insurer, not a point-solution vendor making incremental tweaks. This decision meant taking on significant responsibility (and financial risk), but we believe it is the only way to make a meaningful difference. At the same time, we recognized the danger of trying to solve every problem at once or focusing on problems that we don’t have the expertise to solve.

For Arlo, this was an important journey to truly understand where our value proposition is, which parts of the health insurance stack we need to own, and which parts we can partner with other organizations. Ultimately, we realized that we could be the best in the world at building an AI-powered underwriting system and deriving actionable insights from data. At the same time, we found invaluable partners in the form of TPAs and regionally-rooted health plan architects to jointly deliver first-class, community-based health plans. By focusing on what we do best and collaborating with others, we’ve found our unique place in the healthcare ecosystem.

One key lesson we've learned is that the U.S. healthcare system is a complex maze of various stakeholders, regulations, and incentives. It’s filled with countless problems that demand solutions. But not every problem is worth solving—many are just symptoms of deeper issues and can be highly misleading. To truly make an impact, you must identify and address the root causes. This often means expanding your scope but without overextending yourself.

Lesson 2: Counterpositioning - Transparency & Eliminating Random Fees

The health insurance industry is dominated by a handful of powerful players. In many geographies, the BUCA carriers (Blue Cross Blue Shield, UnitedHealthcare, Cigna, and Aetna) control over 75% of the market. Competing with such giants can feel daunting, but we’ve learned that the key isn’t to outdo them at their game—it’s to play an entirely different one.

At Arlo, our counterposition is transparency. The major carriers are notoriously opaque when it comes to sharing data on their population. They keep employers and members in the dark, concealing hidden fees like spread pricing on pharmacy benefits, retained rebates, and out-of-network repricing charges. They also rarely disclose the true effectiveness of their population health programs, leaving employers guessing about rate increases year after year.

We’ve taken a different approach. As a lean, technology-driven startup, we’re not burdened by the administrative bloat or shareholder pressures that force traditional carriers into secrecy. We don’t rely on hidden fees or arbitrary charges to sustain our business. Instead, we’re committed to being fully transparent about costs, performance, and how we allocate healthcare dollars.

This transparency isn’t just ethical - it’s competitive. By eliminating unnecessary fees and cutting out inefficiencies, we can offer a more affordable, higher-value product to employers and their employees. For our members, this means access to health plans that are not only cost-effective but also built on trust and clarity. In an industry where opacity is the norm, transparency is our superpower.

Lesson 3: Collaboration - Fixing the Payer & Provider Relationship

One of the biggest challenges in U.S. healthcare is the fractured relationship between payers and providers. Both sides often view each other with suspicion: providers accuse health insurance companies of denying necessary care to increase their profits, while payers argue that providers charge unreasonable prices and overtreat patients without improving health outcomes. This mistrust has created a multi-billion-dollar industry focused on fighting each other - revenue cycle management (RCM) vendors on the provider side and fraud, waste, and abuse (FWA) systems on the payer side. The result? Billions of healthcare dollars are wasted on administrative overhead rather than patient care. This trend will only worsen with more and more AI-based systems coming to market that claim to help either side streamline their process.

At Arlo, we believe this relationship can and must be repaired. Instead of working against each other, health insurers and providers should collaborate to deliver cost-effective, high-quality care. Providers know what needs to be done to achieve the best health outcomes, and health insurers have critical data and insights that can help allocate healthcare dollars more efficiently. When these two sides work together, everyone benefits: patients, employers, and the healthcare system as a whole.

There are two key paths to fixing this relationship:

1. Align financial incentives through creating “Payviders”: Unfortunately, for many health systems, the commercial employer population is a profit center. However, there are excellent examples of how health systems can profitably take on risk and provide cost-effective and high-quality care. Here is a great video by Eric Bricker on how Ochsner Health has done it.

2. Partner with independent multi-specialty direct care providers: There’s a growing movement toward direct primary care, and for good reason. These models empower providers to focus on proactive, personalized care, prioritizing prevention over reactive care. This approach keeps people healthier, reduces hospital visits, and emphasizes preventative care - ensuring healthcare dollars are spent where they make the greatest impact.

We’ve seen the power of collaboration through our partnerships with highly accountable, direct primary care organizations. By closely working together, we provide them with benchmarking data and cost-saving insights, and they find innovative clinical pathways that reduce costs without compromising care. These relationships are not only productive - they’re a glimpse of what’s possible when insurers and providers truly work together.

Lesson 4: Go-to-Market - Getting the Right Partners

If you’re selling through employers, brokers are a key part of the equation. Yet, in the world of healthcare innovation, brokers often face criticism: “They’re not open to innovation”, “They serve their own interests, not their clients” or “They’re just unnecessary middlemen pocketing commissions.” While frustration with some benefit brokers is understandable, it’s far more productive to view brokers as partners rather than adversaries.

At Arlo, we’ve found that brokers play an invaluable role as trusted advisors to their clients. Healthcare is a deeply personal and often intimidating space; employers rely on brokers to help them navigate their decisions. This makes brokers uniquely positioned to explain why an innovative solution like the Arlo plans is worth the change, easing fears about risk and uncertainty.

Brokers also bring insight. They know which clients are ready to embrace something new and which ones may need more time. For a startup, this is gold - working with early adopters saves time and builds momentum while chasing clients who aren’t ready can drain resources.

We’ve built strong relationships with the broker community because we understand their value. Brokers are not just conduits for sales; they’re allies in introducing much-needed innovation to employers and employees alike. In many situations, they are co-builders of the product we bring to the market. By working together, we’re not just delivering a product—we’re transforming how businesses think about health benefits.

Arlo in 2025

Reflecting on everything we achieved in 2024, I’m both proud and motivated for what lies ahead. This past year laid a strong product and financial foundation, but we’re just starting. In 2025, we’re doubling down on our mission to transform healthcare and make it more affordable, accessible, and transparent for businesses and their employees.

Here’s where we’re focusing our efforts:

• Advancing our technology and data infrastructure: We’re finding even smarter ways to leverage AI and data to tackle rising healthcare costs and streamline the customer experience.

• Deepening broker & partner relationships: We’ll continue to collaborate with brokers across the U.S., empowering them to bring innovative solutions to their clients.

• Expanding clinical programs: We’re investing in programs that ensure every healthcare dollar is spent where it matters most, delivering maximum value for our members.

2025 is about growth—growing our team, our partnerships, and our impact. If you’re passionate about solving big problems, applying cutting-edge technology, and improving people’s lives through better healthcare, I’d love to hear from you (just reply to this email, or if you read this online, DM me via LinkedIn). We’re actively hiring for engineering, sales, and clinical operations roles, so check out our career page and join us on this journey to reshape healthcare.

When I am not writing, I am the CEO & Co-Founder of Arlo. We're on a mission to bring affordable health insurance to small businesses across the U.S.—and we need passionate, driven people to help us do it. We want to hear from you if you're excited about using cutting-edge technology to solve complex challenges! We're hiring Engineers & Sales professionals. Explore our open positions here!

We often take it for granted how amazing the free market is. But you realize it very quickly when you are in a situation where the market fails. This year, I went to Italy on a short vacation. We booked a rental car ahead of time, but our flight was delayed by a few hours. When we arrived at the car counter, the - not so friendly - attendant told us that our reservation had expired and we would neither get a car nor our money back. We were forced to book another car right on the spot for 4X the price of our initial reservation. The company truly exploited our situation, and because there was no alternative, we had to pay.

You probably had or know of someone who had similar stories with car rental agencies. It seems to be their business model to lure customers to their counter and then hit them with additional charges when people cannot price compare. They try to limit their exposure to a fair and transparent market.

Free and transparent markets are amazing - they allow you to trust that when you pay a certain price, you are getting the best value, and you can trust that you’re not taken advantage of. However, abuse of market power exists in many other industries other than car rentals. Probably the most critical is health care services in the US. Prices for health care vary significantly for the same services and are disconnected from the quality of care you’re getting. Today, I want to dive into some of the reasons why the free market is broken in health care and talk about some potential fixes.

Why are prices so arbitrary in health care?

Let’s first talk about how prices for healthcare services are set! And I say prices - plural - for a reason, as there is no single price in health care. Every person pays something different when they show up at the doctor's office, even if they get the same services. There are two main mechanisms for how prices for healthcare services are set in the US:

• Government Set Prices: A large portion of services are provided to people covered by traditional Medicare (not Medicare Advantage) and Medicaid programs. CMS and state agencies set the rates for each service by reviewing lots of data and consulting with different stakeholders. Providers who want to participate in these programs have to accept government-set rates. In other countries, this is the standard approach for all health care services, i.e., provider representatives, payer representatives, and government agencies get together every year to set a universal sustainable price.

• Payer Negotiated Rates: Besides government-set prices, rates are usually the result of negotiations between payers and providers. This includes Medicare Advantage plans (privately administered Medicare) and health plans from the individual marketplace or employer-sponsored group health plans. While Medicare Advantage rates are often relatively similar to the CMS set rates, commercial rates vary wildly, sometimes up to 358% of Medicare rates. Rates also vary greatly between payers; some payers are able to negotiate better prices than others.

• Provider-set prices: A final but very small number of prices are set by providers. These include cash-pay prices or the so-called charge-master rates, i.e., the official price lists that hospitals charge for care. However, these rates are usually never charged as most people either use their insurance.

Free market proponents would argue that the government is not good at setting the right prices - they might have a point, but it seems like private companies are not doing any better. This graph shows an example of payer-negotiated hip-replacement rates in the Los Angeles Metro area, which are all over the place in terms of cost and not correlated to quality metrics. Note: If we had a functioning market, “provider-set” prices would be much more common - and market forces would punish or reward over-and-underpriced institutions.

Why Negotiations are broken

Let’s talk more about these payer-provider negotiations. Quality of care plays a role, but the hard factor in the negotiations is all about patients and payment volume. This is what represents negotiation power.

Thinking back to the car rental example, if someone can choose between different agencies, each agency will need to present itself with attractive prices or better services. But when there is only one agency around, they can charge whatever they want. Likewise, if there are many booking sites, car agencies will not list themselves on the most expensive platforms. But when there is only one booking site, the booking site can demand much higher referral fees from the rental agency. When it comes to rate negotiations, both payers and providers are doing their best to strengthen their bargaining power:

• Payer consolidation: “UHC acquired Bind”, “Centene buys Wellcare”, and lately, “Cigna and Humana in Merger talks”. Payer M&A is ruling the headlines. There has been so much consolidation going on that we are now at the point where one or two health insurance players now dominate most markets in the US. In the large group market, for example, the average market share of the largest insurer is 63%, and the top three payers make up more than 80% in most markets. Payers are using this market power to negotiate preferred prices with healthcare systems. Because they control where patients can go through their network directory, they can push providers to give them lower rates. This is particularly true when providers don’t have much market power, i.e., smaller, independent clinics, which are what economists call price takers. These practices have to get used to the annual “rate update” letter by the large carriers. Payers also use price-setting power to create an economic moat. When they pay lower prices than their competition, they can offer lower premiums and gain more market share. Providers who are willing to give lower rates to larger carriers should bear in mind that they are making their counterparts’ negotiation power only stronger in the future.

• Provider Consolidation: To keep their position at the negotiation table, providers developed other strategies to react. It is a common playbook for PE shops operating in health care: Buy a bunch of dermatology clinics in an area that used to be price takers from the insurance companies. Once you have bought enough providers in an area, they force the insurance company to the negotiation table to negotiate better rates and then sell the group with a profit. But this strategy is not just employed by private-equity firms but also by larger health systems. They employ highly paid M&A teams who have been buying out hospitals and practices, small and large, to ensure they become “too big to be out-of-network”. In 15 states, health systems are now the largest employer and, in many others, the second or third largest (behind Walmart).

• Branding & Marketing: Another way to strengthen a health systems negotiation position and avoid being dropped from a payer’s network is to invest heavily in branding and marketing. This is to make sure there will be an outcry by insured members when a payer dares to remove a health system from their network. If you have ever driven on a Houston highway and wondered why not-for-profit organizations spend so much money on outdoor billboard advertisements or why they drop millions of dollars on Airport wallpapers, you now know why.

This dynamic between payers and providers has led to price competition drying up in many markets and is one of the reasons why healthcare prices behave so strangely. And we all pay for this in the form of rising costs in the form of…:

• …Waste: When there is limited competition, there is no need to innovate, no need to become efficient - money will just come in, no matter what. This is why we still see antiquated technology in most health systems, bloated management organizations, and why most processes are still incredibly costly and manual.

• …Payer Profits: Regulators decided to cap the profits that health insurance carriers are allowed to make by forcing them to pay out at minimum 80-85% of the premiums they collect. While this cap is probably well-intentioned, it creates a perverse incentive for payers - the main way health insurance can increase their absolute profits is not by reducing the cost of care but by increasing the total premiums they collect. So when payers negotiate with health systems, they might not negotiate as hard as they could, also because payers with market dominance know they can just push down the higher prices in the form of higher insurance premiums.

The lack of a free market is truly a root cause for many problems in the US health care system. But not all hope is lost…

Creating competition

Because health care costs are rising and somebody has to pay the bill eventually, more and more employers are looking for solutions to how they can maintain affordable benefits for their employees. Some employers and their third-party administrator partners are trying to break this unhealthy negotiation dynamic between large payers and behemoth health systems. Here is what they try to bring price setting away from the payer-provider backroom deals:

• Reference-based pricing: The idea behind reference-based pricing is to tie a rate back to the state-set prices (which, in general, are deemed reasonable). Instead of negotiating rates, a health plan will pay up to a certain amount, and a provider can decide whether they accept the price or not. As this is a fairly uncommon way to get paid and reference-based pricing rates are usually lower than commercial rates, this approach creates some friction for members when accessing care. However, there are several health plans and vendors out there helping with setting up provider reimbursements and handling the friction.

• Narrow networks & Direct Contracts: Some daredevil health plans are trying to break the “too big to be out-of-network” narrative and are trimming down their network to a few hand-picked providers. They find a set of providers they deem high-quality and then negotiate preferable rates with them. Providers sometimes are willing to agree to these rates as the plan will guarantee that they are the “exclusive” provider for their plans and other competing providers are not covered. This approach is very interesting. However, it requires that at least some competition is left in the market.

• Procedure Carveouts: Certain high-cost procedures are highly profitable for many health systems. These procedures include cancer care, many types of surgery, as well as organ transplants. However, there are more and more facilities that specialize in this type of care, and that can offer these services at a dramatically lower price (and often better quality of care). Employers with enough power with large health systems sometimes carve out these services to certain specialty network providers to ensure their employees get the most cost-effective care for these conditions.

• Dynamic Copays & Tiered Networks: Another way to increase competition is to align incentives between health plan members through different cost-share tiers. Instead of giving members carte blanche and the same copay for every provider, members will have to pay different copays depending on the provider they choose. Recent price transparency regulations have given rise to several vendors and plans that try to make costs transparent to the members. Unfortunately, many provider contracts prevent health plans from differentiating copays across providers within a health system or against other systems.

• Cash Prices & Upfront Payments: Last but not least, some health plans are trying to leverage cash prices and upfront payments to obtain lower rates (and to avoid contract negotiations). The idea is that in some cases, providers may be willing to offer lower cash prices because they can collect the money upfront, thus having lower default rates and less billing overhead. Some health plans are built entirely around this approach; others have it as one feature of their plan.

Better infrastructure is needed

While many of the strategies above have been shown to reduce healthcare costs, there is still a long way to go to break the pattern of backroom deals between large payers and large providers and bring more competition to the healthcare market. However, operational challenges also need to be solved to make more competition possible.

Authorization Infrastructure

These days, when tapping our credit card, we don’t think twice. Neither does the merchant, as they trust the credit card system that once a payment is authorized, they will receive it. However, this has not always been the case. When the first credit cards emerged, the merchant had to call the card issuer to confirm whether the card number was still valid and if they could complete the transaction — an operational nightmare. So, back in the day, credit cards were only used for larger purchases, greatly limiting their use cases. Only when Visa and Mastercard built technology to automate transaction authorizations did credit cards start to displace cash. Doing authorizations and verifications via phone calls? Sounds familiar! Currently, most doctor offices are not set up to accept patients with “alternative access models” like reference-based pricing or direct contracts, even if they, in principle, would agree to the payment terms. From my experience at Arlo, I can confirm that practices may have a contract with a network, but because this network is not on the post-it note of the front desk attendant, they won’t accept the patient.

Health Care Service Marketplace

Negotiating narrow network contracts and direct contracts with providers is quite a feat. Fee schedules and side agreements can be very complicated and take time. Thus, providers are only willing to entertain contracting conversations when they think the number of patients is large enough. This makes it harder for smaller, innovative players to break in. The healthcare industry needs a standard contracting language that allows for easier network creation and payment for healthcare services.

Beyond Price Transparency - Network Freedom

I am a great fan of the recent regulations regarding price transparency in health care - however, they won’t achieve much unless we lift restrictions on putting this data into use. For example, many network contracts prevent health plans from adding copay differentials for different providers or allowing for network carveouts. For example, many health system contracts include inclusivity clauses, which means that a health plan’s network has to include all affiliated doctors. These clauses prevent employers from creating competition between providers and using quality and price data to steer members to the most cost-effective care.

Lets create more competition

Free markets are beautiful - they create vast value. The internet has created many new markets we never thought could exist: Uber for car sharing, Airbnb for spare rooms, and UpWork for international freelancing. We should work on bringing more competition to health care where providers compete to give the best care, and patients can be sure they are getting care worth their money. Feel free to reach out if you are building in the provider contracting or network space. Would love to have a conversation!

The individual marketplace is experiencing a surge in popularity, with over 16 million people in the US securing coverage through the ACA individual marketplace in 2023 - a 38% increase from just five years ago. With this trend comes a growing concern among brokers, whose traditional business model may be disrupted by ICHRAs (Individual coverage health reimbursement arrangements).

ICHRAs allow employers to give their employees tax-free reimbursement for coverage they can obtain through the ACA individual marketplace. This arrangement often eliminates the need for brokers to scout the market for group health plans, negotiate with payers, and run RFPs for their clients. Many benefit advisors are worried about the future of their practice. But how much of this worry is warranted?

In this article, I’ll explore why employers are turning to ICHRAs, which companies will benefit most from ICHRAs, and evaluate whether ICHRAs will really take over the group benefits space.

Why ICHRAs are appealing!

There are several good reasons why employers are choosing to offer ICHRAs over of fully-insured or self-funded group health plans:

• Stop worrying about rising health care costs: ICHRAs provide employers with a way to stop worrying about rising premiums and shopping around for carrier plans yearly. They allow employers to set a reimbursement amount for their employees, and it is up to the employees to shop for healthcare. Premium increases for individual marketplace insurance plans have been relatively modest in the last few years.

• Ability to customize benefits & contributions: In an ICHRA, employees can choose a health plan that matches their specific situation. For example, a younger and healthier person may choose a high-deductible bronze plan with a lower premium, while a family with high medical needs can choose a richer, more expensive plan. Employers can also vary the amount they pay out to employees based on age (to some extent), which can make health benefits more fair, as older adults usually have higher healthcare costs.

• Easy compliance: Employers with more than 50 employees are required to offer “affordable health coverage” to their employees - also called the employer mandate. For some companies, it is difficult to find reasonably priced group health plans, and they might opt to offer an ICHRA that helps them comply with this requirement.

What are the limitations of ICHRAs

At first sight, ICHRAs and individual marketplace plans seem very appealing to employers, and some people are predicting ICHRAs to be the end of group health plans as we know them. However, the success of ICHRAs greatly relies on the success of the individual marketplace. If the individual marketplace cannot offer attractive plans, ICHRAs will have limited adoption. Unfortunately, there are several fundamental flaws with the individual marketplace:

• Adverse selection and high costs: Individual plans often offer unattractive rates due to adverse selection. Marketplace plans cannot consider pre-existing medical conditions when underwriting a quote. As a result, they must set rates at the average expected cost for the entire population - called community rating. This cost averaging means that groups with relatively healthy populations may choose to find coverage elsewhere (more on this below) and leave the marketplace population, making it, on average more unhealthy and thus driving up costs for those who remain.

• Lack of Options: While ICHRAs theoretically provide employees with choices, in many locations, the options available on the ACA marketplace are pretty limited. In many geographies, there are only one or two insurance carriers to choose from, and the selection of plan types is typically limited. Most marketplace plans are HMO plans. ICHRA plans are thus not a great option for companies that rely on competitive benefits to attract employees.

• Restrictive plans: ACA marketplace enrollees tend to come from lower-income households. A slight increase in premiums often induces people to switch their carrier to a better-priced one. While this is great to put pressure on insurance premiums overall, it also means that insurance carriers don’t benefit from longer-term population health initiatives. Instead of investing in the health of their members and reaping the benefits down the road, they choose to control costs by restricting benefits. Most marketplace plans are HMOs and narrow network plans. For most employers, marketplace plans don’t offer attractive benefits for retaining talent.

• Driven by subsidies: The main reason for marketplace enrollment growth seems to be that households that make less than 400% of the federal poverty level, i.e., below $58,000 annual income, can get premium tax credits. 75% of people on ACA individual plans receive government subsidies for their premiums. However, employees cannot use tax credits and ICHRA reimbursements at the same time, and often, it makes more sense to use premium tax credits instead of making use of the ICHRA reimbursement. Given the current debate on government spending cuts, subsidy-driven growth of the individual marketplace does not seem to be sustainable.

Taking all these, it is clear that ICHRA plans won’t be a fit for many employers. However, they offer a great option for employers who cannot afford the rising healthcare costs of their group health plans. Usually, this is true for relatively unhealthy groups. ICHRAs can also be a great option for groups that don’t rely on health benefits to be a differentiator when it comes to attracting talent or who want to use ICHRAs as a cheaper and easy way to comply with the employer mandate.

What will actually disrupt group health plans?

Some people predict that ICHRAs will march through traditional group health plans, but there is another trend that is more likely to disrupt traditional fully-insured group plans: level-funded plans - a form of self-funded health plans. Here are a few reasons why level-funded plans are on the rise:

• Medical underwriting: Unlike fully-insured and individual marketplace plans, level-funded plans can consider pre-existing medical conditions when setting premiums. Employers with relatively healthy populations can get dramatically lower rates compared to community-rated individual plans or fully-insured group plans (for companies with fewer than 50 employees).

• ICHRAs as fall-back: Ironically, ICHRA plans may enable smaller groups to take on more risk and move to a level-funded arrangement. Individual marketplace plans and small group health plans are guaranteed issue, i.e., payers cannot refuse coverage based on prior medical conditions. This means employers always have the option to move to an ICHRA if their level-funded or fully-insured health plan becomes too expensive. It can act as a “fail-safe”. In a way, the guaranteed issue of community-rated plans allows employers to be more risk-taking and consider level-funded options.

• Pooling risk: Insurance relies on the law of large numbers. When you pool individual risks together, they become predictable and manageable. Employers who pool risks from several employees into a group plan can often save significantly compared to insuring each individual separately. This is the main reason why PEOs and group plans can offer much lower rates than individual insurance. This is a true benefit a company can provide over just offering health insurance reimbursement for individual coverage.

• Longer-term cost containment: This point is somewhat open to debate, but overall, group health plans have less turnover than individual marketplace plans! This incentivizes group health plans to invest in long-term cost-containment solutions, such as better primary and preventative care. Additionally, the regulatory framework for self-funded plans is generally more flexible than that of ACA plans, which allows for more creative cost-containment solutions that are not possible with ACA plans.

• More innovation in the level-funded plan space: Many associate level-funding with products from large carriers like Cigna level-funding or United Allsavers, but there are more and more independent vendors (including Arlo) that offer level-funded health plans for small and medium-sized businesses. These smaller plans are often faster to adopt innovative cost-containment solutions and can often offer lower premiums. ACA marketplace plans suffer from a lot of regulation, making it hard to launch a marketplace plan. This reduces the competitiveness of these plans. For every startup launching a fully-insured/ACA-compliant plan, there are 5 startups launching innovative level-funded products.

• Better benefits: Because self-funded plans often come at a lower cost, they can also offer richer benefits through lower deductibles or a broader network. Level-funded health plans are also more flexible in including additional benefits, such as virtual physical therapy and wellness benefits, that can be customized for a group. Marketplace plans need to be registered and reviewed by the regulators and are generally slower to adopt more innovative benefits.

The new role of brokers in ICHRA-times

ICHRAs are here to stay, and they continue to play an important role in groups to manage their healthcare costs. However, they don’t represent the unraveling of the group health plan.

Brokers and employers need to know how to operate in this new world. Here are a few of my thoughts:

• Become an expert in ICHRAs AND level-funding: Knowing exactly which groups benefit from which arrangement will help benefit advisors differentiate themselves. Also, each arrangement will require different skills to serve their clients: In an ICHRA, the role of the broker will require helping employees select the most appropriate health plan based on the individual marketplace offers; for level-funded plans, brokers need to evaluate whether the plan will fit the needs of the group as a whole.

• Advances in underwriting technology: Medical underwriting is becoming easier and easier, and it will be more and more effortless for brokers to evaluate whether a group would benefit from being in the community-rated risk pool or in their own underwritten arrangement. We will see more employers switch from fully insured to level-funded arrangements in the future once they realize they can get better quotes through medical underwriting. Brokers should be aware of this trend and find partners that allow them to do thorough risk assessments without needing individual health questionnaires and claims histories.

• Switching as competitive advantage: Brokers traditionally don’t like to switch their clients from one plan to another. It requires a lot of effort, employee education, and process change. However, in this new world of ICHRA and level-funded plans, switching employer groups between arrangements can be a competitive advantage for brokers to win and retain business. Evaluating opportunities to switch from ICHRAs to level-funding and vice versa should be done every year, and the broker who can leverage technology to make the switch as easy as possible can realize incredible value for their clients.

ICHRAs will have their place in employee benefits, but they won’t be the unraveling of the group health plan. Unless regulations around the individual marketplace are altered, I don’t see this changing. However, brokers should know how to leverage ICHRAs and level-funded plans to offer the most cost-effective coverage to their clients. If you want to chat more about this topic - please reach out or leave a comment!

When I am not writing, I am the CEO & Co-Founder of Arlo. We're on a mission to bring affordable health insurance to small businesses across the U.S.—and we need passionate, driven people to help us do it. We want to hear from you if you're excited about using cutting-edge technology to solve complex challenges! We're hiring Engineers & Sales professionals. Explore our open positions here!

I’ve been excited about this post for a while now! Lately, I have been getting more and more questions about - Jan-Felix; what are you actually doing? You left Palantir now a while ago - what are you cooking up? It's time to open the curtains and give you a sneak peek into what we are building.

But before I get to that, I would like to share a bit more about what motivated me on this journey:

After undergrad, I got to experience the world of management consulting at McKinsey - a crucial part of their business is to help organizations streamline their processes and reduce unnecessary costs. I remember working on a study at an automobile manufacturer, where we sifted through every expense line to find opportunities to reduce costs. The manufacturer was struggling to keep up with their recent growth in business, and their unit economics were deep in the red. Nothing was safe, even the coffee machines on the factory floor (no worries - they kept them). We made sure that workers would practice on top of their license for example that line workers with higher hourly pay would not have to drive a forklift to get materials, but this task is done by forklift drivers who need less specialized training. We replaced parts from high-cost suppliers with lower-cost suppliers with similar quality. We reduced the inventory to free up capital,... - you get the gist. One thing I realized: In order to turn this company around, no silver bullet existed - not one initiative that turned a failing factory into a profitable business, but the aggregate of all the above yielded success.

Fast forward a few years, I got to work with some of the largest healthcare payers in the US. To me, they looked very similar to a poorly run factory - claims were not being paid-out to providers in time, millions of dollars in penalty interests were incurred, claims were adjudicated manually with inadequate accuracy, and there was no quality control when paying providers and facilities.

If a factory is poorly run, it will go out of business - and another better-run factory will take its place. Unfortunately, this dynamic does not exist in the US healthcare system. Poorly run payers found a way to make employers and patients pay higher and higher prices for their services, instead of having to improve their operations and increase their value. Read here if you want to know why.

And this is hurting people. I don't have to tell you about the devastating effects medical bills, high deductibles, and a lack of medical coverage bring to families across the country. 66% of bankruptcies are caused by medical debt, and many people are uninsured or underinsured because they don't think the premiums are affordable. We need a better way - something where payers are incentivized to create value. That's why we are building Arlo - a value-based health plan for small and medium-sized businesses.

Capitation in commercial insurance? How is that going to work?

While value-based care and capitated provider payments are prevalent in Medicare and Medicaid, they are still looking for adoption in the commercial space. There are several reasons for this:

• The commercial insurance population is younger and healthier than the Medicare population. This means that, in general, a large quantity of healthcare costs is incurred by just a few high-cost claims, such as cancer, organ transplants, and premature babies. A lot of the playbooks from Medicare ACOs, such as better chronic disease management and home health for better access to primary care, have limited effects on commercial health care spent.

• Health insurance coverage is tied to employers, and people change employers frequently. This means that health insurance carriers are not incentivized to invest in population health initiatives that only show effects after 1-2 years. They might not reap the benefits of investing in better preventative and primary care.

• A lot of procedures in the commercial sector are profit centers for health systems. There is little incentive for them to cannibalize these profitable procedures through capitated payments - with the commercial rates often being a multiple of Medicare rates, this is a more significant issue than in the Medicare world.

While we are far away from the broad adoption of capitated provider payments in the commercial sector, we already have one form of capitated payments. We just call them differently: insurance premiums.

The role of health plans

At its core, the payer is the primary "risk-bearing" organization - it is their duty to offer access to high-quality care and ensure that premiums are spent wisely. By wisely, I don't mean that the health plan should decide on treatments and medical actions a member should take. That is and should stay the task of physicians. However, a good health plan should ensure that any money they pay out is maximizing the value for their members and the premium dollars are not being wasted on unnecessary care or unfairly-priced services. Together with a robust primary care team, the health plan should be the quarterback to ensure financially sound decisions are being made. A value-based health plan can achieve this without burdensome restrictions and unnecessary tedious utilization management. Instead, they should serve as a partner to the member and make transparent the financial consequences of their actions. The physician will decide on the course of treatment, and the health plan should open up different options for the member, explain their financial consequences, and make these options easily accessible. Here are a few examples:

• Help members find out-patient facilities for IV-administered drugs, which often cost 3-5 times less than an inpatient hospital setting. Support them to easily get an appointment at these facilities and organize transportation.

• Negotiate bundled payments for specific procedures and provide incentives to providers for high-quality outcomes.

• Offer concierge care navigation to triage members to the right place of care and to avoid overutilization of specialists.

• Make sure members have a primary care follow-up appointment after a hospital discharge.

• ... and 50 more things a health plan can do to reduce cost and improve outcomes.

Value-based health plans already exist. In fact, many large employer groups are running quite efficient value-focused health plans - their benefit levels are often much higher than small-group health plans. They employ data analytics teams, and medical management teams (for example, Morgan Stanley has their own Chief Medical Officer) and bring targeted population health solutions to their employees. Because these large groups are usually self-insured, they have the incentive to utilize their healthcare dollars most cost-effectively. It directly impacts their bottom line.

However, many people currently don't have access to these value-based health plans. Those are the small and medium-sized businesses in America. They either have to go with a community-rated health plan or are stuck with incumbent health insurance players, that even two years of work-from-home forgot that a lot of care could be delivered virtually, that have 60-minute customer service wait times, and that provider directories are as accurate as a phone book from 1998. These health plans often suffer from "vendor fatigue", which to me sounds very much like an excuse for not innovating and trying new approaches. Over the last decades, these health plans have not taken the necessary measures to offer better premiums: the most effective way to get better premiums was to reduce coverage by introducing high-deductible health plans and shifting costs from premiums to out-of-pocket cost sharing. It is time for this to change.

Announcing: Arlo - a value-based health plan for SMBs

I am excited to present Arlo - a value-based health plan for SMBs

We will keep a few details to our chest, but here are a few principles we are following with building out our plan:

1) Aligned incentives and performance bonuses

The key for a value-based health plan to work is to align incentives. Align incentives between the plan and the member, as well as the plan sponsor. The first element for us is to remove the perverse incentives of most fully-insured carriers by unbundling their services. When you utilize services from United Health Group, you most certainly have to use their owned PBM as well as stop-loss carrier and contract with UHG-owned providers. The ASO plan certainly doesn’t have the incentive to squeeze their owned entities for better prices or switch them out for a more competitive solution they don’t own. At Arlo, we don’t have these restrictions and are choosing the best-in-class vendors for our health plans.

Second, while most health plans are shared savings plans, fully-insured carriers are limited on how much they benefit from shared savings. Fully-insured carriers have to have a medical-loss ratio (MLR) of at least 85%, which means that 85% of the premium needs to go toward medical care. If a group has fewer expenses than that amount, the carrier needs to refund those premiums. This setup does not incentivize prudent financial behavior. For the health plan, it does not matter if a group has 60% or 70% in medical expenses. They will always make the same profit. Yes, they don't want a group to exceed 85% or even 100% - which would mean they will run a loss - but they usually address these high MLRs by increasing the renewal rates for this group or dropping them altogether.

Arlo is operating on a performance bonus with our groups. Our health plan will thrive if we can cost-effectively manage healthcare spending.

2) Unified Member Experience

Many employer-sponsored health plans suffer from a mesh of solutions that are bolted together rather poorly, leading to a disjointed member experience. While great care navigation solutions, convenient pharmacy apps, and good virtual urgent care tools exist, they must be optimized to work well together. I had to make this experience myself when I tried to access my care navigation app via my health plan portal and ran into an “App not found - 404” error. No wonder patient engagement is a challenge - nobody wants to juggle eight different apps to manage their health benefits. As the health plan, Arlo is in the prime spot to integrate different solutions into a coherent member experience. We built a modern digital member experience that integrates care navigation with virtual primary & urgent care. Our unified experience makes it easy for our members to make informed decisions about their care.

3) High-value providers & data-driven cost containment

I've written many of my thoughts on the opportunities for cost containment here, and most of them will find their way into Arlo in some form or another. I would like to highlight two elements here, though:

First, our health plan is built on identifying high-value providers and helping members find appointments with these providers. Prices and quality can vary a lot between providers, and we are using different data sources to identify the providers with high-quality that also offer a reasonable price. Our integrated care navigation team will then help members find the most appropriate and cost-effective care for their condition. We will write more about how we achieved this on our company blog - make sure you're signing up.

Second, we developed a robust, modern data infrastructure that lets us detect cost savings opportunities by monitoring claims, appointment bookings, prior authorizations, and member interactions with our plan. This intelligence layer ensures that nothing falls "through the cracks", and we identify cost savings opportunities in real-time.

4) Administrative Ease

To win with small businesses, you must make health plan administration as easy as possible. I was staggered when I learned in my conversations with HR leaders that many health plans only allow member enrollments manually via online forms without bulk uploads available. Imagine doing this every year for 200 employees...

Small business owners have a business to run, and it should not be their task to spend hours on bill reconciliations and payroll system syncs. Also, the HR teams at most companies already have enough to do with hiring, payroll administration, labor rule compliance, and DEI initiatives. They should not be their health plan's outsourced customer service department (which, in most companies, they actually are).

Aligned Incentives, a unified member experience, identifying high-value providers, and administrative ease are the four pillars of our plan, and I am very excited to announce that we are launching for enrollment in mid-year 2023. Check out the quotes we can offer compared to existing options for small businesses!

What this means for the blog

Building a health plan takes a lot of time, but I will continue publishing things and keep you with me on the journey - I will also start publishing more health plan-specific content on the Arlo website.

At this point, thank you all for being such great readers and supporting me on the way. I have met so many great people through these articles, including my awesome co-founder: Karthik Bhaskara. It's been a true pleasure working with you and building this company together.

I will share some more updates here moving forward - and if you want to continue to support me (and Arlo) in its mission to bring affordable health benefits to US workers, please share this post with a CFO at a small and medium-sized employer or with a broker you trust - I would love to talk. You can reach me at team@joinarlo.com. I am happy to show them around. And if you are an engineer or nurse/ care coordinator who wants to bring affordable health care to the US - we are hiring. Check out our open positions here.

Arlo is hiring - check out our job openings here.

When I wrote the previous article, one question did not leave me rest: Why is it not done? If it is all there and there are all these opportunities to reduce cost and close efficiency gaps, why are they not being implemented? Basic economic theory tells us that if one company finds a better way of producing a particular output, it will start to dominate the industry through better pricing or quality of service. This will create follower companies that imitate the new method and drives down prices for all. According to this dynamic, health plans should have more broadly adopted cost containment measures such as surgery bundles, more accessible primary care, and innovative provider contracting methods.

However, this is not the case in the healthcare industry in the US. Broader adoption of cost containment methods still has plenty of opportunity to grow and costs keep rising. Why is this the case?

Today I want to share my thoughts on this $1 trillion question. I cover the most prominent reasons, but there are probably plenty of smaller ones, too. Here is what I will cover:

• Delegation problems

• Information asymmetry

• Matching frictions

• Risk aversion & denial

Let's get to it!

Incentives: Principal Agent problems

The first set of barriers to broader adoption lies in misaligned incentives. Many stakeholders benefit from the current "wasteful" way care is financed; for these stakeholders, this is not waste but profit.

In economic theory, these misaligned incentives are called principal-agent problems. A principal delegates a specific task to an agent and hopes the agent completes the task with the principal interest in mind. However, the agent also has their own interest; instead of maximizing the principal's interest, they will maximize their own benefit. There are plenty of principal-agent problems regarding employer-sponsored health plans, and here are a few.

1. Employer - Employee: Delegating coverage level

It starts with the employee delegating the health plan purchasing decision to the employer. The employee usually should know what is best for their health care, but the employer will choose which coverage level they will offer - they will decide on the cost-sharing amount, access restrictions, and carrier network. This delegation of the decision becomes especially tricky as the employer is optimizing for retention and cost rather than for a plan that maximizes each individual's needs. Especially for larger populations, there will always be some level of compromise the employer needs to make. In addition, an employer might also over-purchase insurance. Employers might be afraid that their benefits are not competitive and thus might buy higher coverage than the employee would buy on their own. However, an individual employee may want to choose a higher salary over a better benefits plan.

This delegation of health insurance buying also goes the other way around. After the employer has picked a set of plans to offer their employees, the employee usually has to choose which plan to enroll. Let's assume the employer covers most of the plan premium, and there are several plans to choose from with different premiums. Why not choose the most expensive plan, if it basically comes for free? I assume many people just enroll in the most expensive plan, even though they do not need as much coverage.

2. Broker: Delegating benefit decisions

As discussed in Part I, it is inefficient for most employers to run their own health plan. They often do not have the resources to oversee their health plan administration and design, so they heavily rely on outsourced vendors to help them with that. In fact, most small and medium-sized businesses outsource the selection and administration of benefits to a broker/ advisor firm to help them select the right plan.

Unfortunately, there are a lot of adverse incentives at play here:

• Don't get fired: A primary goal for an advisor is to keep their clients happy and to continue the relationship with them. Because of that, they are better off recommending an established carrier that works vs. pushing for more modern and cost-effective health plan options. (I will cover later why the established carriers don't really have the incentive to reduce costs). Also, they are not directly incentivized to minimize cost, as most advisors are paid a PEPM fee. Thus, they only have to keep expenses low enough to keep their clients happy.

• Kickbacks: Brokers do not only get compensated through a fee from their clients. They often get significant bonuses from the carriers and point-solution vendors (PBM, Stop-loss carriers, etc.) they recommend to their clients. These financial arrangements can become very complex. For example, a firm working with a particular carrier might get a bonus if they sign a certain number of lives with that carrier in a year. In another arrangement with a PBM, an advisor/ broker might get a bonus for every prescription the PBM processes.

The complexity of the health plan landscape makes it very hard for employers to oversee what the broker recommends and understand the genuine reasons for the recommendations. To protect employers more from this information asymmetry, the Consolidated Appropriations Act passed in 2021 included a provision that requires brokers to disclose their third-party fees. However, it is up to the employers to ask their brokers/ advisors to disclose their fees.

One last note: similar to financial advisors, brokers fulfill a critical role and there are plenty of great individuals delivering great service. However, employers should do their homework when selecting the right advisor.

3. Vendor: Delegating admin tasks

For self-funded and level-funded plans, brokers will leverage various third-party vendors to deliver the health plan. These vendors include a TPA, a PBM, and point-solution vendors such as case managers and digital health tools. A key thing to consider here is how these different vendors are paid and how they make their margin. Here are three common payment methods:

• Per-member-per-month (PEPM): Under this arrangement, the vendor will get a monthly fee for every enrolled member. A challenge with this model is that the vendor is not paid based on outcomes. The primary way for the vendor to increase their margin is by streamlining their processes and reducing their operating costs. Sometimes a conflict can arise, though. Consider this: A TPA or ASO carrier can improve their profit by reducing their operational staff - instead of reviewing every claim, they often rely heavily on auto-adjudication systems. While this is nothing bad, there is no real incentive for the TPA to improve the accuracy of these systems. They would rather automatically pay out a claim than spend manual effort reviewing it and getting it right.

• Volume-based: Another way to get paid is by getting a percentage of the claims volume. You can probably see already where the problem lies. If a vendor receives a percentage of money paid out, there is minimal incentive to reduce the overall cost. This type of reimbursement also affects "shared savings", i.e., if a vendor gets paid a cut of the savings they achieve. While these incentives can work, the vendor must not control the benchmark. PBMs and hospitals have long gamed this system by artificially inflating list prices instead of realizing savings.

• Utilization-based: This method is pretty straightforward and can be very fair. For example, an obesity management program will only get compensated if it can effectively engage and enroll employees in its program. This gives incentives to build better care. However, it can also give vendors incentives for over-utilization and to bill unnecessary procedures and treatments.

As you can see, every payment method has its advantages and disadvantages. There is no silver bullet in paying vendors. Still, it is essential to know the adverse incentives that exist and account for these, for example, through reporting and performance bonus payments.

Another thing to consider when delegating tasks to a vendor is hidden fees impacting incentives. Especially on the prescription drug side, organizations are masters of hiding how they make money. The major PBMs, ExpressScrips, OptumRx, and CVS Caremark are not charging any PEPM fee - but of course, that does not mean they are free. You can bet they are making a healthy margin on the backend in ways that are not obvious to the employer customer. A particular case of hiding fees is the strategy of vertical integration. The three PBMs I just mentioned are each owned by one of the major health insurance carriers. These carriers have done a great job in the last few years in buying all types of companies in the health plan value chain, from clearing houses to health care providers. This is a great way for them to expand their margin, as a cost center (such as paid-out claims) suddenly becomes a revenue stream for the group. Optimizing these cost centers is not a priority. 

4. Carrier networks: Delegating rate negotiations

One type of vendor is a source of unique delegation problems: the provider contracting network. These networks negotiate the rates with the providers the employer eventually has to pay. Carriers like Cigna and Blue Cross Blue Shield are allowing independent TPAs to rent their network and give them access to their negotiated terms.

To understand this delegation problem better, we must look at the dynamic between payers and providers. While many health systems are non-profit organizations, they still have to keep an eye on their profits: given their significant overhead and often inefficient operations, they need to optimize for profitability, or they risk going out of business. Better rates with commercial payers is a crucial way of achieving this, and to improve their negotiation leverage, they have employed several strategies:

• M&A: Most health systems have been incredibly active over the last decades in buying up smaller practices, specialties, and primary care and adding them to their organization. Because of that, many markets are now ruled by a few large hospital systems, giving them much power in payer contracting negotiations. Many payers are forced to work with the dominant health system for a specific market to maintain network adequacy.

• Branding: The traditional way of increasing price elasticity is to get people to believe that you are the best. Creating a brand lets Apple sell its devices for twice as much as the competition, even though one can argue whether they are twice as good from a pure performance point of view. Health systems are achieving this branding through PR campaigns (have you also gotten LinkedIn and YouTube ads from your local non-profit health systems lately?), and they are investing in marketing so that employees expect the system to be in-network.

Carrier contracting networks now find themselves in a difficult position. On the one hand, they need to offer a competitive network regarding the breadth of coverage, as their customers expect certain anchor providers to be in-network. On the other hand, they need to have a competitive network regarding negotiated rates. However, the carrier networks have decided to choose breadth over competitiveness in price. They can afford to loose a few health plan customers, but they cannot afford to loose an important health network.

Some of the restrictions that contract networks put on how health plans can use their networks indicate that provider relationships matter more to carrier networks than the economic value. For example, they may not allow TPAs to "tier" their network, i.e., actively try to steer members to cheaper providers in their networks. Often a network can only be used as a "whole", i.e., one cannot pick a certain provider from the network, but everyone has to be in-network. They also may restrict the number of claim audits a TPA can make to reduce the burden on providers.

5. Payer-Provider: Delegating care delivery choices

Payers don't deliver care. Doctors do. This is the most classic task delegation that happens in health care. However, the fee-for-service reimbursement structure does not give a whole lot of incentives to the provider to become more cost-efficient. They benefit from more utilization, not from less. And while I don't want to throw doctors under the bus - after all, we all rely on them - they often are unaware of the financial impact their choices can have on their patients. For example, ordering the image "down the hall" in the hospital-owned imaging center may be a convenient option for the patient. Still, it can cost a multiple of an image done by an independent facility. Many health systems instruct their doctors only to refer patients within their health network to avoid "revenue leakage".

CMS has pushed alternative provider reimbursement models in the government programs to align incentives between providers and payers. By reimbursing for outcomes, providers are incentivized to be more economical with their resources, optimize for long-term outcomes instead of short-term procedures and improve their operational processes. But there is still a long way to go, and these arrangements have yet to take hold in the commercial space.

Incentives are vital factors to consider when trying to reduce healthcare costs - but there are other challenges to delivering more cost-effective health benefits. 

Lack of data access - a form of information asymmetry.

People do many things when they know nobody is watching. This is a key reason delegation problems exist - "I don't know what you're doing over there". A meaningful way to keep agents accountable is through data transparency and asking: where did my money go? This includes getting information about any third-party commissions but also - even more critical - a thorough review of claims data. Claims information can reveal a lot about how resourceful a payer or administrator has spent their claims, whether a health plan is performing at the benchmark or whether it has missed opportunities to realize savings.

However, it is pretty challenging for plan sponsors and their advisors to obtain access to the claims data of their population. Usually, fully-insured health plans don't provide access to claims reports and thus don't allow the employer to review their spending behavior critically.

Companies have found ways around this by using web scraping technology to obtain claims information from their payer, but this has its own challenges. Scraper technology can be brittle, you need customer consent, and it might violate the terms and conditions of the payer web portals.

"Whack a mole" - matching frictions for narrow point solutions

Let us say a plan sponsor gets access to their claims data; there is still a challenge with what to do with it. The cost of the same health condition can vary significantly from person to person and is often very well justified. Identifying the right actions to take to realize cost savings without compromising the quality of care can be quite challenging.

Thanks to the investment of CMS into value-based care arrangements, many vendors have popped up that target different saving opportunities: from importing drugs from Canada to pinging primary care doctors when their patients get hospitalized. However, many of these solutions are often very narrow in scope and only apply to a small number of people. While this can be millions in the US on aggregate, for a single employer, this could mean that only 3-4 employees would qualify for this solution.

There is a significant matching problem here. Every population is different and requires different point solutions. The current friction in evaluating and implementing new vendors makes health plan administrators often focus only on the top 3-4 conditions. The smaller the employer gets, the more difficult the ROI case gets, as the fixed implementation costs can vastly exceed the solution's benefits.

Inertia - for employees & employers alike

One has to be careful when blaming slow innovation on user inertia. It is an easy way to explain away a problem without looking into the underlying reasons. For example, a lack of great doctor-facing software is often blamed on the unwillingness of doctors to adopt new technology. In my opinion, this is not the reason. There are many structural reasons why Epic is the dominant software rather than more modern tools.

But still, there is quite some inertia that needs to be overcome to drive change.

Patient Engagement

A key challenge for digital health and effectiveness often depends on their ability to enroll people in their program. I always have to get back to Brendan Keeler's chart here.

Any solution that depends on the member becoming active by themselves is having a much harder time than a solution that makes it frictionless and inserts itself into existing user journeys. A lot of people only actively engage if there is immediate pain, the rest, you have to meet them where they are. Through better design, better immediate benefits, and other means.

A second challenge is a need for choice. Americans hate restrictions imposed by their health plans: prior authorizations, narrow networks, referrals, etc.! These utilization restrictions can be very effective in reducing healthcare costs, and they can even improve outcomes as they protect patients from over-utilization and unnecessary procedures. However, people dislike the burden associated with them. While this is a valid preference, they probably over-index the importance here because they often don't feel the cost of a non-restrictive health plan. A way to address this is to have people pay the price for a less restrictive health plan.

Buyer Engagement

For most companies, health benefits are a top-three line item in their annual budget. But it is rarely treated with as much scrutiny as other vendors, such as real estate or raw material suppliers. Health benefit design choices are often 100% outsourced, but the employer has to foot the bill.

There needs to be more understanding of how the healthcare supply chain works. There should be more education for CFOs and HR managers alike on how PBMs, provider network contracting, and plan design work. Another essential concept is risk premiums, i.e., understand that anytime you hand over risk to someone else, you will pay a premium. Rightsizing the risk, for example, through a level-funded plan, can be a practical first step to reign in costs.

However, making changes to benefits is a topic loaded with risk aversion. Usually, employers only make significant changes if something has gone very wrong: a critically denied service or a 40% renewal rate. HR leaders are often afraid of the challenges that a change in health plans brings with them, to how this could impact their ability to attract and retain talent.

My thoughts

With all these challenges ahead, I hope you have not given up all hope. There are ways to address those issues:

• Align incentives: The economy lives on the division of labor and specialization, but this only works if there are clear performance criteria and transparency. Health plan sponsors should be aware of these adverse incentives and push for measures to address them. This includes asking advisors to reveal compensation and commissions from third parties, putting the right reporting metrics in place, and not only looking at the monthly PEPM price but also considering the total cost of ownership. 

• Better integration of point solutions: Standards around who the point solutions are for either by ICD-10 codes or other inclusion criteria, as well as clear standards on the effectiveness/ expected outcomes. This will help organizations decide which MSK solution works best for their population and significantly reduce the friction for adopting modern solutions. There is also an opportunity for an aggregator platform that allows health plans to bring in the right solutions given their claims data. 

• Unbundling the behemoths: The increased consolidation for health care providers and carriers is a major source of rising costs. Hospital-owned primary care doctors refer to health system specialists rather than independent practices; carrier ASOs require their customers to only work with their owned PBMs and stop-loss carriers. While I let the FTC figure out how to break up these organizations sensibly, there is also another way for employers: choose unbundled health plans. With the right partners, they can build a plan that leverages best-in-class solutions for each vendor category. 

• Open contracting: Another way to improve competition is to reduce the contracting friction between payers and providers. Allowing smaller health plans and employers to build direct contracts with providers and let them pick and choose providers from a health system vs. contracting with the entire organization can significantly improve competition between providers and create pressure to improve operational efficiency and prices.

Many of my above thoughts are anecdotal, derived from conversations with experts in the field, and based on theoretical principles from economics - if you find data points and examples that prove me wrong, please don't hesitate to reach out or comment.

We live in exciting times - all these challenges are not insurmountable, and I am always interested in chatting with people having a passion and thoughts on this space, so please reach out.

So after we looked in Part I at the different ways companies can finance their benefits, let's look at all the tools available to employers to reduce their overall benefit costs. 

In this article, want to avoid the prevalent trap of anecdotal savings. "Measure X has saved $1 million here, activity Y saved $500k there". There are plenty of anecdotes where a single initiative has yielded considerable absolute savings. But how generalizable are these strategies? I think it is worth taking a step back and looking at benefits costs from a high level and how they are breaking down.

Non-claims costs

• Administrative expenses include everything required to set up and administer a benefits plan. Depending on the type of funding, they are sometimes more and sometimes less obvious. For example, in a self-funding arrangement, the administrative expenses include the TPA fee, the benefits consultant fee, and the PBM fee. In a fully-insured model, the administrative costs are included in the insurance premium. My take on this is that even though 25% of healthcare costs are administrative, this is not a huge opportunity to optimize. The TPA and benefits consultant business is pretty competitive. In fact, the focus on administrative sticker prices has potentially eroded the quality of administrative services.

• Risk premiums: As mentioned in the last part, handing over risk to someone else usually comes at a cost. Here I am not talking about the expected claims cost - it is evident that a less healthy population has to pay a higher premium than a healthier population. But a group pays an additional premium if their risk is less predictable or measurable. Insurers will always assume the worst about you if you cannot prove them otherwise. 

Claims-related costs

• Price: This is the average cost for a specific procedure or an episode. There are major differences in the rates for a certain procedure, and provider rates are often not correlated to quality. A key strategy for employers is to obtain better rates.

• Quantity/ Utilization: Many cost containment efforts have focussed on unnecessary or avoidable utilization, like ER visits and elective surgeries. However, it's worth looking at the data to determine what is actually impacting rising costs. This interesting report shows how much of health care cost increases are driven by higher prices, not more utilization (see chart below).

  

• Share of claims paid: The last important lever for costs is how much of the claim costs are even paid by the health plan. Several factors are going into this: the amount of cost-sharing with the employee, services that are included or not included (for example, fertility benefits, and acupuncture), and the number of rejected claims (for example, in case of fraud or lack of medical necessity).

Let's go through these buckets and see how payers are trying to address each of these.

Risk Premiums: Get a grasp on your risk

As discussed earlier, healthcare costs consist of the expected cost for claims and a risk premium if you decide to move some or all of the risk to someone else. For fully-insured companies, this is usually a 25% markup on expected claims (assuming an 80% medical-loss-ratio/ MLR, and for stop-loss, this is a 30-40% markup (assuming an MLR of 60-70%). Basic economic theory tells us that the more volatile your expected risk is, the higher your risk premium. Therefore, employers can try to reduce the volatility of expected risk.

• More accurate predictions through better data: The first way to reduce the risk premium is to improve your expected claims prediction accuracy. A more precise prediction minimizes the uncertainty of the risk an insurer has to underwrite. Suppose an insurer gets proof that all group employees are healthy and don't have cancer or other costly conditions. In that case, they will offer a cheaper risk premium than if they only get limited information. The most important source of this proof is by reviewing claims data. However, claims data is often unavailable - especially for fully-insured employers; the employer usually only gets aggregated reports. A typical hack is to web-scrape claims data from insurance portals or to use health questionnaires. However, you will need consent and collaboration from each employee to hand over their data.

• Pooling risk (reducing volatility) - Captives, MEWAs, PEOs: The basic idea of insurance is to leverage the law of large numbers - the more people you cover, the more predictable the risk will become. An insurer would be much happier to take on the risk for a group of 1,000 people, where they get $10m in premiums a year and have a 95% certainty of a $1m claim vs. insuring a group of 10 people, where they get $100k premiums in a year, but have a 5% risk for a $1m claim. Thus small employers are trying to pool their risk with other employers to save on risk premiums. Several options exist: In a captive arrangement, multiple employers contribute to a shared pool of funds, which will act as a buffer between the self-funded claim expenses and when the stop loss carrier kicks in. For example, an employer will pay for any claims up to $50k, from $50k to $250k the captive will cover the claim and after $250k the stop-loss carrier will take over. Similar arrangements are MEWAs (multi-employer welfare arrangements), where an employer association creates a health plan with a joint risk pool that a stop-loss carrier will insure. Another advantage of risk pools is the better negotiating power for both reinsurance rates and provider contracts. Owning more patient volume can improve costs. This fact is also used by professional employment organizations (PEO), a third way for employers to pool their risk. PEOs are co-employment arrangements where the PEO will take care of everything related to payroll, HR, and benefits administration. The PEO combines the workforce from several smaller employers and purchases group insurance for all their customers. Through their larger size, they can offer the insurance at lower rates, and often are the only way for small organizations to avoid insuring their employees through the small-group ACA marketplace.

Price: Avoid paying inflated prices

There are two common strategies to address provider rates. 1) Exploit price differences between providers and send members to cheaper doctors and 2) send them to the same doctors but try to get more favorable rates.

Improve provider competition through member steering

One of the most shocking facts about the US health system is how much variance in price and quality there is and how uncorrelated they are. Prices for very standard procedures can vary a lot. For example, a colonoscopy can range from $1,000 to $4,000, and an EKG can be $200 or $3,000 depending on the facility. Health plans have long tried various methods to exploit these price differences. Here are some common strategies to steer members to cost-effective providers:

• Exclude coverage: The most effective way to avoid paying for costly services are hard coverage restrictions. Health plans can decide not to cover costs for particular services or providers. Narrow networks have proven to yield significant savings. While this is an effective strategy, many people dislike these restrictions, as they can disrupt the continuity of care if employees change health plans. Also, most people get their doctor recommendations through their friends or primary care doctors, which can be frustrating if their insurance network does not cover them.

• Utilization management: Because hard restrictions are so unpopular, payers came up with many soft restrictions: prior authorizations, mandatory second opinions, primary care referrals, etc.. These are all ways to double-check medical necessity and quality of the approach before a member can go ahead with a specific drug or procedure! But be aware - providers and pharma companies are fighting back against these restrictions, and they have become an outright bureaucratic war with the member caught in the middle.

• Virtual-first HMOs: This approach became more popular with the rise of virtual care - in a virtual care HMO, a patient will need to obtain a referral from a virtual doctor before seeing a specialist. The idea is that a virtual care visit is not as much of a burden as an in-person primary care visit. The virtual care team can quickly determine whether a specialist visit is necessary and refer to a cost-effective provider.

• Financial incentives: Since price transparency rules kicked into effect, payers are now trying to make the price difference more evident to their members. Health plans waive copays to members for visiting preferred providers or give them lower or higher copays based on the cost-effectiveness of providers.

• Soft nudges & concierge care navigation: A prevalent approach over the last years has been concierge care navigation, i.e., assisting members with finding a doctor and setting up appointments. Examples include text-based care navigation services, nurse hotlines, and free virtual care visits. Many employers have adopted this approach as care navigation: a) does not require many changes to the plan, b) does not add any restrictions, and c) can even be sold to employees as an extra service. People like to follow the path of least resistance, and there is so much friction in making appointments that care navigators can nudge people in the right direction. However, the onus is on the employee to find and make use of these services. As many people find their doctors online or through their friends and family, members often don't engage with this service and go directly to the doctor they want to see.

Get better rates

The second approach is to obtain better rates with provider groups. Often this depends on the willingness of the provider group to negotiate. Since many providers have built regional market power, they are not always willing to do that. However, here are some strategies:

• Single case agreements: Single case agreements are one-off negotiations if a member needs to see an out-of-network doctor before their visit. The price transparency rules have given payers more tools to prepare for these negotiations and obtain a better bargaining position.

• Bundled payments & direct contracts: In these arrangements, employers negotiate special rates upfront with their preferred providers for certain procedures or procedure bundles. 

• Reference-based pricing: A new trend to pay providers is to only reimburse providers based on the Medicare benchmark rate, for example, 170% of the amount that Medicare would pay. However, this puts patients at risk for balance billing, i.e., the healthcare provider might charge the patient the amount not paid by the carrier.

• Medical tourism: If you cannot find reasonable rates in the US, many other countries around the world offer similar, if not sometimes better, quality of care at a much lower price. For example, an IVF treatment in Germany costs $3,000 - $5,000 vs. the average price of $15,000 - $30,000 in the US. Some employers now include medical holidays in their benefit plans to get lower rates.

• Import drugs: Sometimes people don't want to travel abroad, but they can have the drug come to them. In 2020 the FDA and HHS relaxed rules around importing drugs from Canada. Those drugs are essentially the same as those purchased in the US and manufactured by the same companies, but they are often much cheaper. Companies like Sharx help employers procure drugs outside the traditional PBM route and realize cost savings.

… and the 1001 one-off things

There are several other, more minor ways to reduce prices and waste: i.e., use mail-order pharmacies, choose the right setting for physician-administered drugs (they are much cheaper if they are done at home rather than in a facility), etc., etc. A key challenge for these solutions it that they are usually only effective on a case-by-case basis. Health plans have to identify these cost saving opportunities and make a timely intervention, but doing these smaller, one-off opportunities can add up.

Reduce share of paid claims: Avoid paying in the first place

This is a very popular way of handling benefit costs - just don't pay for it or find somebody else to take on the risk. Here are a few popular options:

• Shift risk to member: Shifting some of the risks to the member makes sense from a principal-agent perspective and aligns incentives. Making members pay a copay will incentivize them not to abuse their plan by seeking unnecessary care. Employers hoped that high deductible health plans (HDHPs) would lead to more shopping behavior in their members. However, this has not proven to be the case. They have only led to people postponing much-needed preventative care and, in the end, just shifted more of the costs to the members. In most cases, a member does not meet the deductible and has to foot the bill for their medical expenses alone. Basically, HDHPs are a reduction of health care benefits. Sweetening these plans with tax-favorable HSA accounts does not change this fact. The graph below shows how employers have embraced HDHPs and slowly shifted costs to the member (instead of increasing premiums, which would be a more obvious shift of costs).

• Shift risk to provider: Another way of dealing with risk is by pushing it to the provider. This could be a win-win situation. The employer sets a fixed price for a specific episode, and the provider will get that amount regardless of the resources required to care for the patient. The more efficiently the provider treats the patient, the more money the provider can make. The provider is in the best position to do so, as doctors should decide what is clinically necessary and not payers. However, capitated models and payment bundles have yet to take off in the commercial sector. The main reason for this: providers tend to make more money billing fee-for-service than taking on the risk for outcomes. Why would you want to cannibalize your profitable business? There are also plenty of other obstacles, such as risk aversion, admin challenges and benchmarks (discussed further at the end).

• Shift risk to another payer: If you cannot push the risk to the member or provider, you might find another payer to foot the bill. There are a few - not always very ethical - practices to shift costs to other payers. The first example is members with chronic kidney disease. After 30 months in dialysis, they qualify for Medicare, and Medicare will be responsible for their care - getting them to Medicare right at the 30 months mark instead of waiting longer can yield significant cost savings. Another strategy addresses high-cost employees that have left the company and qualify for COBRA. If they are on a self-funded plan, it might be beneficial for the employer (and the employee) to move them onto an individual marketplace plan. Even the premiums for a rich platinum plan can be much cheaper than the employer paying for medical care. The third method is to make your benefits unattractive for certain high-utilizing members so they will self-select to seek coverage elsewhere. This strategy includes removing certain high-cost specialty drugs from the formulary so the beneficiary will opt to get covered by their spouse's plan or find a more suitable individual plan. Last, there are cases where the payer should not pay out a claim because they are the secondary payer. Common examples are medical claims related to work accidents, which should be covered by workers comp insurance, or car accidents, which could be covered by liability insurance of the liable party. It is important for the plans to catch these situations and reject paying the claim.

• Reject claims: It is estimated that 25% of all medical bills contain errors, and in most cases, these errors favor the hospital system that sent the bills. Payers must implement appropriate systems to catch errors and dispute them in time to avoid overpaying. This area of claims processing, called FWA (fraud, waste & abuse), tries to catch anything from upcoding (i.e., charging for a higher level of service that the provider did not deliver) to duplicate claims and unnecessary medical and billing inconsistencies.

Utilization: Address avoidable events and remediate them if they occur

Avoiding unnecessary care and improving the overall health of their population is an important area for health plans. It is also a weired one: shouldn’t this be the task of the physician? In theory yes, but adverse incentives and differences in quality push health plans into action. Here are three levers for reducing costs through improving outcomes:

• Better primary care & improved access to care: The site of care matters significantly when it comes to cost: a primary care office visit is cheaper than an urgent care, which greatly beats the emergency room. In addition, primary care visits cost much less than specialist visits. Good primary care can take more tasks off the specialists' plates and should offer easy access to care to avoid urgent and emergency care. But changing primary care from a referral mill into more holistic care comes at a cost. I recommend reading Kevin O'Leary's article on the economics of more comprehensive primary care.

• Chronic care management: Unmanaged chronic conditions such as diabetes and hypertension lead to more expensive acute care down the road. Thus it is vital to ensure members are equipped with all the drugs, services, and devices necessary to manage their condition. There are plenty of digital-first vendors out there addressing a range of chronic conditions via continuous engagement and small but frequent interventions.

• Navigate costly episodes better: 80% of health care costs are caused by 20% of members. Some episodes and conditions will always be expensive - think kidney care (check out Zach's blog on it), cancer, or spinal surgery. If the care team does not properly manage these episodes, they can easliy become a multiple of the initial estimated costs. This can be due to a miscommunication between the different specialists, a poor quality provider choice, or unnecessary procedures and tests.

There are plenty of other mechanisms for how better care can yield cost savings. If you are building something in this space - feel free to reach out. I would love to hear your approach. I am planning another article just on this space.

JF's Thoughts

As always, here are some final thoughts on all these cost-containment strategies:

• Point-solution fatigue: You’ve seen that there are plenty of ways to address rising healthcare costs. There is no silver bullet. Implementing the right point solutions for the right population can feel like whack-a-mole. Depending on the specific employer population and geography, some strategies work better; others are less effective. Orchestrating all these solutions can be a daunting task, and it is not getting easier that there are now hundreds if not thousands of solutions trying to sell to employers, all with their very nuanced offerings: For example, some MSK solutions specialize in treating shoulder problems others focus only on lower back pain. There is definitely a need for a better “matching” solution that helps employers identify the right solutions for their population and connects them to their health plan, particularly for smaller health plans. The current consulting-heavy approach only makes financial sense for health plans/ employers with a large population.

• Point solution sales: The previous point has ramifications for the pricing model of many digital health and points solutions. Because there are so many providers now, they will need to consider their pricing model carefully. A PEPM model often does not make financial sense for many small to medium size groups if they only have 2-3 cases in a year. These groups would be much more willing to pay and adopt these solutions if they come with a case-based/ value-based pricing model.

• The role of cash pay: In my article, I have left out an important trend: direct-to-consumer cash pay. Several startups are building D2C medical brands and offering cash-pay doctor networks with transparent prices. GoodRx managed to establish an interesting business here for drugs. However, it is crucial to consider who cash pay is most useful for: people who are generally healthy and who have a high-cost-share plan. If a person does not expect to hit their deductible, cash pay is often cheaper (also for the employer, as they do not have to pay for the benefit). But this model very quickly crumbles for people with higher utilization, which usually causes the most costs. There are interesting opportunities to integrate cash pay into the payment model for the plan, i.e. have the health plan “cover” the cash pay. However, this may add additional complexity for the member that needs to be handled. Also, there is still the perception of people that health plans should cover their benefits and they resist paying cash.

• Obstacles to value-based care: One last thought on VBC, which is finding more and more adoption in the Medicare world. However, we have yet to see many value-based care arrangements for commercial health plans. A major obstacle, besides provider hesitation, is the issue with benchmarks. Whether something is deemed valuable is usually measured against some benchmark. There are two main issues with benchmarks, though: 1) The "value-based" organization can influence the benchmark, and instead of creating value, they will just play a benchmark game. Great examples are PBM drug price rebates or provider chargemaster prices. Provider network carriers and PBMs optimized for percentage discounts but, at the same time, inflated the base price, so there were almost no total savings. 2) The second issue are benchmarks that are catching up too fast. Let's say a value-based organization realized savings in one year. Where should the benchmark be set for next year? At the old benchmark, the new spending level, or somewhere in between? It is a delicate balance because benchmarks catching up too fast can lead to adverse incentives where value-based organizations might only realize some of the value they could at a time.

Above, I've started to touch on some of the challenges with all these cost containment solutions, but there is more. Many health plans are not implementing everything they can do to reduce costs. But this is a topic for the next part of this series. Don't forget to subscribe and share if you have liked it.

Continue reading here for Part III.

I am always excited to chat with thought leaders in the benefits space. If you are a CFO, HR leader, benefits consultant, or broker, don’t hesitate to reach out - I would love to chat.

Will you get cancer? Will you get into a car accident? Will your newborn need to be in the NICU? You can live a very healthy lifestyle and do all your annual preventative health exams, but there is always the chance you might need expensive treatment. When talking about healthcare finance, we will always need to consider the risk of unforeseen healthcare costs.

For many, risk management in health care can be a daunting topic: it involves statistics and abstract economic concepts, but I hope you bear with me here as it is also very fascinating. In particular, employer-sponsored health plans are unique in how they take and handle risk - this affects more than 170m people in the US, and a lot of innovation is happening in the space.

Today, I am kicking off a three-part series on risk-taking in employer-sponsored health plans: Part I will cover the different forms of risk-taking and how risk-taking can yield cost savings; part II will talk about approaches to manage and reduce risk, and part III will be about why many of the current systems to manage risk are failing.

Let's dive into it…

Health plans: financing plan or insurance plan?

On a high level, health plans fulfill two roles: financing "predictable" care and insuring against unforeseen healthcare costs. In many other areas, we usually purchase these two functions separately. For example, when you own a house, you typically pay for expected maintenance costs out-of-pocket. Even if maintenance needs vary from year to year, responsible homeowners know that they need to put aside some money to upkeep their property. However, there are also high risks, such as storm damage or a fire that can destroy the whole property. For these high-risk "catastrophic" events, a homeowner must buy insurance, as the mortgage companies usually require it. 

In healthcare, we don't differentiate between these two functions. Most regular and predictable care falls under the financing category; this includes annual wellness exams, vaccinations, cold visits, preventative exams, etc. In addition, a lot of chronic care also falls under predictable healthcare finance - if a chronic condition is well-managed, one can predict the costs quite well. However, the level of health care finance required differs for every individual. The average healthcare needs depend on a variety of factors: the apparent factors are age and preexisting chronic conditions. But there are lots of other influences. For example, on average, restaurant workers are more likely to be diabetic or suffer from substance use disorders, and truck drivers and office workers suffer more from MSK-related issues.

However, health care also has true, less predictable financial risks. Here are some high-cost examples that are particularly relevant to the commercial sector:

• High-risk pregnancies and premature babies

• Surgeries (organ transplants, back surgeries, etc.)

• Cancer treatments

Of course, several "risk factors" contribute to the likelihood of these events occurring in a particular population, but these events are generally relatively infrequent.

It is worth differentiating between health financing and health insurance because there are different ways to address the associated costs. Payers can often optimize predictable healthcare spending through better primary and preventative care. However, this won't always help with unpredictable high-cost events like a car accident or certain types of cancer. Primary care can definitely reduce the overall likelihood of a catastrophic event occurring but won't prevent these from ever happening.

But I am getting ahead of myself. Before discussing how to manage and reduce risks, let's look at the different types of health (finance & risk) plans available in the commercial space.

The industry supporting this & the newer plans here

For most Americans, their employers are the primary risk bearers of their healthcare costs. Employers are obligated to offer health plans to their employees once they reach a certain size. They also cannot charge more than 9.12% of an employee’s wages in premiums for the lowest-cost plan. So employers have to bear at least some of the cost in most cases.

Employers have different options on how they can handle their responsibility to finance and manage risk:

• Individual & small group marketplace: Employers with 50 or fewer employees are not required to offer health coverage. They are considered small groups, and their employees can obtain health coverage via the health exchanges. These health plans are "guaranteed issue, " meaning they cannot reject anyone from coverage. In addition, these health plans are also restricted in the risk factors they are allowed to use for quoting their insurance premium, i.e., only age and smoking/ non-smoking are permitted. Employers can either pay the premiums for an individual plan or purchase small-group insurance directly for their employees. Another option for employers are Professional Employer Organizations (PEOs), which co-employ the employees and offer them health insurance through their larger group plan. Employers, in general, can get cheaper rates through a PEO than through the individual marketplace. 

• Fully-insured group health plans: At 50+ employees, an employer can purchase a group health plan that is not part of the ACA exchanges. These large group plans allow insurers to consider the group's specific risk profile, i.e., look at preexisting conditions, lifestyle, and the employees' occupation. Because the insurer can better predict the health care costs of the group using these factors, many employers can get lower rates than on the exchanges and get better coverage by becoming fully-insured. In these arrangements, employers still move the entire risk of payments to the insurance plan.

• Self-insured & Level-funded: Pushing risk to someone else always comes at a price - so if you're able to bear the risk financially, it's always cheaper not to pay for insurance. (yes - this is also true for individuals. That's why you usually get insurance for catastrophic risks like fire and not for general maintenance issues such as a broken gutter). Employers with enough employees can decide to self-insure and take responsibility for financing medical bills themselves. However, there is still the risk of catastrophic claims, as discussed earlier. That's why self-insured or self-funded employers usually purchase stop-loss insurance, which covers very high-cost claims. Companies have different risk tolerance levels, and they decide how much risk they actually want to take and how much they want to offload. In general, self-funded arrangements only make sense for employers of a certain size; only for a larger population will, stop-loss carriers be able to underwrite the group at a reasonable premium. However, there are ways also for smaller employers to become self-insured. They often opt for so-called level-funded plans, which are a special type of self-funded plans that have very tight stop-loss levels. In level-funded plans, the stop loss kicks in after the employer has paid the expected amount of claims during a year. This gives the employer more piece of mind in case a large claim hits.

Recently self-funding has become more and more popular among smaller employers. In particular, employers with attractive risk pools, for example, white-collar groups, are opting to bear their own risk. Going self-funded has various advantages: health plans are not bound by state regulations, and ERISA exempts self-funded plans from most state regulations. Also, they are not subject to state insurance tax premiums, and employers have much greater flexibility in their benefit design.

The industry behind self-funding: build your health plan

Self-funding basically means you have to run your own health insurance company. It is inefficient if every company has to figure this out for itself. Thus, a whole industry helps self-funded employers design and run their health plans. Every health plan needs a mix of different core functions. Here are some of the main activities:

• Claim administration & Payments: The most apparent activity of a health plan is to adjudicate and pay out claims whenever a provider renders a service. This function is usually performed by a third-party administrator (TPA), and they generally get a $20-$40 PEPM fee and a fee based on the volume of claims processed.

• Member Services: In addition to handling claims, a health plan also has to deal with member requests and needs - these can range from finding an in-network doctor to handling denied claims, sending out member cards, and answering coverage questions. This task is usually also performed by the TPA.

• Stop-loss: As discussed earlier, health plans usually buy reinsurance in case catastrophic claims occur. Depending on how much risk is offloaded to the reinsurer, stop loss can make up 10-40% of the overall health plan cost - depending on the population's risk profile and size.

• PBM: Drugs make up about 30% of medical spending in the US, and special companies handle drug coverage called PBMs. They help with formulary design, i.e., which drugs are covered, utilization management, pharmacy claim administration, and negotiating drug prices with the drug manufacturers- for better or worse.

• Contracted Rates: Most health plans leverage a network of providers where they have preferred contracted rates. As provider contracting is a huge undertaking, this is often not feasible to do for employers. Several large carriers, however, rent out their provider contracts so that self-funded employers can access their rates for a monthly fee. Examples here include Cigna, Multiplan, and FirstHealth. The rates here can range between $5-$20 PEPM (per employee per month). There are a lot of exciting trends here in this space - I will cover more of this in my next article.

• Point-solutions: If you have ever attended HLTH, you will know what I am talking about. Hundreds of vendors are addressing smaller and larger buckets of healthcare spending and offering different approaches to reduce or avoid healthcare spending. These solutions range from online physical therapy to travel agencies that organize elective surgeries in Mexico and Europe.

Each of these health plan components has an established ecosystem of vendors, but combing those and picking the right solutions can still be daunting for an employer. Every population is different and has different needs. But don't worry - American health care has another set of useful middlemen that can help out: benefits brokers and consultants.

HR leaders rely heavily on the network and expertise of brokers, and benefit consultants leaders lean when they make decisions around health plan structure and implementation.

On a high level, benefits brokers usually recommend one of two things: (1) go with a large ASO carrier or (2) build a customized health plan using independent modules. United Health Group, Cigna, Aetna, and the Blues offer employers to administer their self-funded health plan using their infrastructure. These ASOs (administrative services only) often come as packaged deals where employers can choose from a few plan packages. The carriers will select their preferred stop-loss partners and PBMs (which they often own). The second option is to choose a more customized approach and pick and choose different vendors. For example, use an independent TPA, send out an RFP to get the most favorable stop-loss rate, and pick and choose a set of point solutions that exactly match the needs of the employer's population.

Innovators disrupting traditional self-funded health plans

The current model of delivering self-funded health plans has plenty of opportunity to be disrupted: It is loaded with misaligned incentives (I will write more about this in part III), ASOs and TPAs do not always efficiently utilize cost containment solutions, and customization of plans can be very consulting intense.

There are a few innovative solutions that are trying to change this and offer a new way how self-insured health plans are being constructed and delivered:

• Modern TPAs & Point solution navigators: Building customized self-funded health plans is a labor and consulting-intensive process. Traditionally only large employers would choose this option, as the additional cost of customizing the health plan can yield significant cost savings by integrating specific point solutions such as surgery bundles or onsite primary care clinics into the plan. For smaller employers, this often requires expensive administrative effort to set up. However, modern TPAs and plenty of point solutions now target smaller employer groups to customize their health plan and point solutions. Examples here include Bennie, Flume, Collective Health, and Aither.

• Virtual-first & PCP-first health plans: Since COVID, probably everyone has tried virtual care - and even the large carriers have understood the potential of virtual care. Leveraging virtual care visits as the first touch point in a person's care journey can significantly reduce utilization (avoid urgent care and ER visits) and the overall cost of care (refer to cost-efficient providers). Thus more and more virtual-first health plans are serving the employer market. United Health Group, for example, partnered with Galileo to deliver their virtual-first health plans, and Firefly Health is now offering virtual-first plans (under the Firefly brand powered by Flume or in partnership with BCBS in Massachusetts). Crossover Health, a direct primary care group selling to employers, has already made the move to offer a health plan. I expect more provider groups, such as Hint Health, Eden Health, Carbon Health, or Everside will take a similar step.

• New health-plan platforms: A whole array of younger health plan platforms are building new-age health plans for self-insured employers. Some are based on new provider payment models (Arlo Health), some leverage dynamic co-pays that offer price transparency for every visit (Bind, now Surest after United Health Group acquired them), or they are starting by building strong and cost-efficient local provider networks (Centivo). Other players include Angle Health and Gravie - a lot of exciting innovation is happening here. It remains to be seen if any of these players can reach escape velocity. These players will most likely get more competition in the next few years. I predict we will see more companies entering the self-funded health plan space from the care navigation side. Transcarent could be in an interesting position to launch a plan, and other care navigators, such as Included Health or Rightway are probably thinking about this as well.

JF’s thoughts

As always, here are some additional of my thoughts on this topic:

• How much are employers actually insured?: Fully-insured health plans, in general, mean that employers can offload their risk to an insurance carrier, but not their responsibility of health care financing, and in most cases, if a bad event happens, the carrier will claw back the money through higher renewal rates. Let's say an employer has a bad year, and two of their employees incur several hundred thousand dollars in claims because they have cancer; this will impact the MLR of the health insurer. The health insurer will then use this as an argument to increase the health premiums for the employer in the following year. The same is true for stop-loss carriers. If the cancer episode is going on, they might even choose to "laser" these two high-cost individuals, i.e., exclude them from the stop-loss coverage, and the employer ultimately has to foot the bill. The moral of this story is that in many cases, insurance can smooth out payments for high-cost bills, but in the end, the employer has to bear the cost anyways in the long run. Employers can offload the volatility risk of healthcare costs, but they cannot avoid financing them.

• State of denial: From my conversations with brokers, I learned that many employers don’t understand that they are finally responsible for financing their employee’s healthcare costs. But many employees are not thinking about it that way, and many are unwilling to make small changes to their benefit plans that could yield in significant cost savings. Most probably spend more time on optimizing their travel expense policies than designing more efficient benefits. Pundits have long predicted the upcoming “breaking point”, when employers can no longer bear the 10-20% annual rate increases, and they will be forced to make more significant changes to health benefit finance. But so far, employers have found ways to move costs around (reducing formulary coverage or increasing deductible for employees, etc.). But maybe, in a post-COVID world combined with a bad economic climate, we might finally see some more fundamental shifts.

• Lack of risk transparency: A key piece for taking on more responsibility for efficiently covering health care expenses, employers will need more visibility into their employee's risk pool. This is particularly true for smaller employers that are fully insured or insure their employees via the marketplace. To accurately predict the claim costs for a group, you will need to get as much data about the employees as possible. This is an essential condition to make an informed decision about whether self-funding would make sense, and it is vital information for getting favorable stop-loss rates. However, the fully-insured carriers often don't share claims data with the employers, so they often fly blind. In many cases, they don't even share information about high-cost claim events, and employers need more information about why their rates got increased by 20-40%. Because the carriers don’t have incentives to share the data, this might be an issue where the regulator needs to become active.

This concludes my high-level overview of employers' different options for handling healthcare expenses and benefits. But there is so much more to talk about. Subscribe if you want to get Part II (How employers manage risk & cost) and Part III (Why risk and cost management is hard) directly to your inbox.

If you made it until here and you liked the article maybe you are or know a benefits broker/ consultant, a new health plan or an employer who is doing innovative things in the space. Please don’t hesitate to reach out - I would love to talk.

In 2022, 28 million people were enrolled in a Medicare Advantage plan, 48% of all Medicare beneficiaries. This share has grown from 35% only five years ago. But why are more and more people choosing Medicare Advantage plans over traditional Medicare? Medicare Advantage makes sense for a lot of people, especially people that tend to be healthier or that have lower incomes. Medicare Advantage (MA), with its $0-premium plans, is often a more affordable option for many people. But there is another reason - MA growth is fueled by a well-oiled broker machine that pushes MA plans out into the world. There are over 130,000 health insurance agents in the US, and probably a good chunk of them are agents selling Medicare.

The Medicare broker industry is fascinating, and today, I want to dive into the ins and outs of this space and look at some interesting trends. And I’ll give my opinion on how this industry will develop. Stay tuned…

The basics of Medicare brokering

Medicare is complex - when you turn 65, you will receive a 128-page manual that describes the different coverage options you can choose. Overall, people can choose between either A) traditional “fee-for-service” Medicare (the government pays 80% of all eligible services) or B) a Medicare Advantage plan, which is a bundled plan including out-of-pocket maximums, drug benefits, and often additional benefits such as vision and dental coverage. However, this is where the simple choices stop. There are special needs plans, plans for dual eligible (i.e., people eligible for Medicaid), people who choose traditional Medicare can opt-in to buy a Medigap insurance plan to cover the 20% coinsurance, and Medicare Advantage plans offer a variety of supplemental benefits! It is a confusing process. Most people are just overwhelmed, and there is definitely a need to help people choose the right plan. In most cases, brokers fill this important role.

Medicare brokerage offers quite an attractive business model. Brokers are not only paid a commission when they enroll a member into a plan, they also get an additional commission every year the member renews the plan. Most brokers receive about $600 when a member enrolls into a MA plan and then $300 every year they reenroll in the plan. These numbers are the maximum commission that health plans are allowed by CMS to pay to brokers.

But it wouldn’t be health care if health plans did not find a way to add additional incentivize for brokers. In addition to commissions, they pay broker firms - also called field-marketing organizations (FMOs, more on this in a second) - something that is called an override, basically reimbursement for any overhead costs of about $200-$300. Larger broker firms then also can get marketing dollars from health plans so that they can push more generic marketing materials into a certain market.

To understand the industry structure better, let’s have a brief look at what the main players are:

• Independent Agents: To sell Medicare Advantage plans, one must obtain a state-broker license and follow all CMS rules regarding Medicare marketing and communications. These agents work for themselves, i.e., they don’t have an employer and live off the commissions they bring in. This can be a lucrative business if you can build a substantial book of business - remember, the broker will get commissions as long as their clients stay enrolled into the plan they enrolled with the brokers.

  Brokers will be responsible for generating leads for their health plan contracts - a common way is to buy “Turning 65” lists and send out mailers. There are pretty tight rules around what brokers can and can’t do. For example, they cannot cold-call leads or solicit them in public spaces (like a supermarket or mall).

• Field marketing organizations (FMOs): In most cases, independent agents are affiliated with an FMO. These firms build contracts with insurance carriers and offer tools for agents to enroll members and get rates from health plans. They also help brokers with compliant marketing materials and sometimes offer lead support. It is free for brokers to join an FMO. The FMOs make their money from the previously discussed overrides, i.e., they receive $100-$200 from the health plan for every member that one of their affiliated brokers enrolls. In addition to that, they get “marketing dollars” from plans to cover certain marketing initiatives.

• Medicare broker agencies: Unlike FMOs, Medicare agencies employ agents as W-2 workers. The commission from the health plan is paid out to the agency, which then pays the brokers a salary (and most likely a performance bonus). They can also collect an override and marketing dollars from the health plan.

Recent Trends in the brokerage industry

In the last year, we have seen some drastic developments in the broker space: The two public MA brokerage firms got crushed in the public markets - which might initially seem surprising, given the macro trends of an aging population and the growing demand for MA plans.

But there are a few immediate trends that are causing their abysmal stock performance, besides the general market drawback:

• Retention rates and short-term disenrollment rates (basically, a member can leave a new plan within 90 days of coverage start) have increased, which hurt the company's expected LTV. For those publicly traded brokers, LTV is measured at the time when they enroll a member.

• At the same time, the CAC has increased substantially. For eHealth, the CAC has increased from $685 in 2019 to $906 in 2021 - a 32% increase. GoHealth does not list these numbers in its annual report, but they are probably quite similar.

• If both things happen simultaneously, this is the perfect storm that leads to substantial cash-flow problems. CACs occur before the member enrolls in the plan, and there can be quite a lag between paying for marketing and other expenses. In addition, if you have unexpectedly high-disenrollments, payers will claw back the paid commission. Disenrollment can be hard to predict, which makes it hard to predict how much these firms should invest in customer acquisition. It looks like in the last few years, the CAC investment has not paid off.

What is happening here?

But if MA membership growth is a fixture of US Medicare and more digitally adept people are aging into Medicare, why are LTVs and CACs so out of balance? Here are some of my guesses:

• Fierce health plan competition elicits shady sales tactics: more and more plans offer $0 premiums, and health benefits are becoming less differentiated between plans. However, plans are starting to entice people by adding ancillary benefits to their plans. Some are quite standard, such as dental or vision benefits, but others are more creative, such as OTC drug cards, transportation services to doctor appointments, or even debit cards to buy healthy food. However, to access these benefits, you often need to read the fine print, as in many cases, only certain people with certain health conditions are eligible for these benefits. This did not stop broker firms, however, from praising the new benefit offers to their prospective clients and encouraging them to switch plans. CMS is quite aware of this practice of “malicious” marketing and now requires brokers to record their phone conversations with prospective clients. Still, competition between health plans will most likely increase plan-switching behavior.

• Online CACs are rising: This trend is true for any DTC brand selling through online acquisition channels. More and more firms are bidding for the same ad spots. The recent changes by Apple to protect users’ privacy have made targeting even harder, and conversion rates of online ads have fallen. This trend is particularly impacting those MA brokers that acquire most of their customers online (eHealth, GoHealth, etc.). But also, other “scalable methods” for lead acquisitions, like mailers and TV ads, don’t have fantastic conversion rates. A certain level of fatigue sets in if a person receives the 10th MA mailer in that week. Or do you ever open all these credit card offer letters?

• Health plans direct sales: Traditionally, the national health plans (Blues, United, Centene, Aetna, Humana, etc.) have been leaning heavily on regional brokers who own relationships in the local communities. As COVID upended a lot of the personal sales and these brokers reverted more to online and telephonic channels, the argument for these national health plans to work with door-to-door agents got less strong. This, in conjunction with the spreading misinformation about plan benefits and thus declining lead qualities, has prompted these plans to start building up their own DTC sales teams. In fact, on their latest investor day, Humana announced that they would start to invest heavily into their own online and agent sales channels after disappointing results of their broker channels. This is putting direct pressure on MA broker firms.

While the incumbents are being crushed, new players are ready to eat their meal

The sheer size of the MA broker market and overall attractive business model (= annuities) has encouraged quite a few new startup companies to enter this field and try to innovate the rather old-school way of selling health plans to the elderly. Here are some of the strategies these players are pursuing:

• Navigate the evolving complexity of plan benefits: Many old brokers don’t cover all health plans but only a selection of carriers they contract with. A range of new firms is offering a complete set of plan choices. In addition, they are adding more decision support to make it easier to match a health plan to a beneficiaries preferences. Members have different needs regarding their provider network, the drugs they take, the ancillary benefits they value, and their financial situation. These players focus on aggregating all the plan information and help with better, personalized decision support.

• Engage members throughout the year: Another strategy of the new players is to stay connected with their members throughout the year so that even if a member wants to change their health plan, they will be there to advise them on this decision. Some brokers are layering care navigation services and better client support tools to build more long-term relationships with their clients.

• Neo-FMOs: FMOs are pretty old-school organizations, and several startups are trying to build better FMOs. This includes better-quoting tools, more transparency about plan information, and offering white-label engagement tools (such as a website or a mobile app). This is a competitive space, though, as FMOs typically are free to the participating independent brokers, and they can freely switch between their affiliated FMOs.

• Offer additional services to health plans: Brokers are in quite an interesting position as they are the first person in the member’s journey when they enter a MA plan. They collect valuable data about a person's medical and financial needs, which can be very valuable for health plans. They also have a “warm” relationship that can be leveraged to offer the member a better onboarding relationship with health plans. A few brokers are trying to use this relationship to sell additional services to health plans, like benefit activation (which can increase member retention), health risk assessments, and even setting up and conducting home health visits based on the data collected during the sales process.

• Offer additional services to providers: In the same way, brokers are the first step in the health plan member journey, they can also be the first step in the member journey for value-based care providers (primary or specialty care). Brokers can help their clients take advantage of the right VBC programs or provider groups in their new plan’s network. This is a new line of business that people don’t like to talk too much about, as one has to be careful about anti-kickback laws. These prohibit any provider organization that takes Medicare patients from paying for patient referrals. However, it seems that there are ways how these arrangements can be structured in a compliant way.

• Find new acquisition strategies: Being a broker is about finding leads and converting those leads into sales. New entrants are becoming quite creative here in leveraging different channels to get in front of potential clients. For example, MA brokers leverage financial advisors, employers, and other organizations to help them find new leads. If you find a less competitive channel that works, this alone can be the foundation of a good brokerage business.

The winner takes it all?

I find a lot of these new broker approaches quite exciting. Nevertheless, it is a very competitive market, and I do not believe there will be one winner emerging. Here are a few of my thoughts on why:

• Can brokers build a “distinct brand” from the health plan: To increase the broker LTV retention will be vital, and thus their transactional sales business has to transform into a more relationship-based brand. But building this lasting relationship is not an easy feat. First, the decision of a member to stay with a plan depends on the plan’s quality, and if the plan decides to make any change to its plan, the member might very well leave. Likely, a bad experience with the plan will also affect trust in the broker who recommended it. Second, in a local community and face-to-face setting, it is probably easier to build these lasting relationships. I wonder whether the new broker firms can build scalable, lasting relationships across markets. A key challenge here is that many players compete to be the most complete, informative, easy-to-understand, helpful resource, including the government, which runs Medicare.gov, a very informative website. However, there might be an opportunity here for dedicated concierge agents that build a relationship with their clients.

• To aggregate or not to aggregate? In the end, a broker is a salesperson that needs to hit certain quotas. Most of the new broker firms are trying to change the character of this business, though, and they want to be less of a sales agent than a marketplace - an aggregator of demand and supply that matches every client with the right health plan. It remains to be seen, though, whether any new entrants can achieve scale to be truly a “neutral” marketplace or will fall back into old habits of pushing only the plans they get a commission for. In addition, a Medicare marketplace seems not very defensible. While there are many MA plans overall, in each specific market, there are usually only a few dozen. Just collecting information about these plans is not a competitive advantage. This is very different from other online aggregators in the travel or other insurance industries. Thus, the differentiation of a marketplace might lay less in the aggregation of plans than in the user experience of matching with the right plan. But will this be enough to win over the fierce competition?

• Catastrophic Risks: Running a business is inherently risky. However, some businesses are facing catastrophic risks, i.e., risks that, if they materialize, will drive the company out of business. Medicare brokers face quite a few of them. For example, being a marketplace is quite challenging for Medicare plans, as health plans are unwilling to work with any broker firm. If they realize a broker always recommends their competitors’ plans, they might pull their contract, and losing one of the big brand carriers can be fatal for a marketplace. Furthermore, CMS increased their oversight of Medicare brokers quite a bit - if a new broker business is built on a loophole of CMS regulations, it can quickly happen that CMS closes the gap and thus steals the basis of the innovation.

A final thought: Even with all these challenges, we will continue to see new broker businesses enter this market. While it might not be a winner-takes-it-all market, we will see many solid businesses going to be built on the need for navigating MA plans and coverage. If you are building in the space, don’t hesitate to reach out for a chat.

I hope you liked today’s article! I plan to write a second part on the topic, focussing on brokers in the commercial health insurance space (small group, large group, self-funded employers). If you want to support me and know a great broker or benefits consultant, please reach out!

Times are challenging for the industry: understaffed, plagued by pent-up demand and operational hick-ups, tons of people are stranded or have to accept delayed services. I am not talking about health care here for once; I am talking about air travel this summer.

In many ways, health care and airlines are surprisingly similar:

• They are highly regulated

• Safety is a major concern, and a lot of things can go wrong

• Complex operations with lots of parties have to work together

• High fixed operating costs

However, when you look at the price development in these sectors, there are some significant differences:

• Airlines are operating much more efficiently today than a few decades ago - average domestic airfares in the US dropped from $600 in 1990 to $314 in 2021 (adjusted for inflation). More people than ever can afford to hop on a plane.

• Meanwhile, the average price for an office visit has risen from $60 to $109 from 2003 to 2016 (outpacing inflation), and while many people don't see these costs because their health plan pays for it, we all bear the costs through increased insurance premiums.

While at Palantir, I spent quite some time in both sectors and I saw firsthand some of the contrasts in how they operate. Today, I want to share my thoughts on what health care could learn from airlines to operate more efficiently and potentially bend the cost curve. Obviously, comparing those industries also has a lot of shortcomings, and they differ significantly in their "product" and processes. I have tremendous respect for professionals in both industries (doctors, pilots, technicians, nurses, etc.), and this article is more about inspiring a different viewpoint than drawing the most accurate comparisons. But several common challenges exist in both industries, and each sector has to address these. I will return to the limitations of my comparisons at the end of the article.

Subscribe now

Optimizing Utilization

Let’s first have a look at the cost structure in which airlines and health care operate. Both have very high fixed and semi-fixed expenditures. A commercial aircraft costs tens of millions of dollars, with many wide-body aircraft starting in the hundreds of millions. The same is true for hospitals, imaging devices, and many other medical facilities cost millions to build - also, labor costs make up a significant proportion of expenses for both industries. These fixed costs require them to make sure they optimally utilize their available resources.

This can be broken down into two areas:

• Supply Optimization: Optimize the availability of your asset/ services, given any operational constraints and disruptions - Think of this as: how can we maximize the number of flights/surgery slots per day?

• Demand Optimization: Optimize the utilization of services by filling all available slots - think: how can we maximize the number of passengers we take for each flight/ number of patients we can see during a day?

Supply Optimization - efficiently opening up availabilities

70% of all (!) aircraft are usually in the air - this became especially obvious when there was not enough parking space for airplanes during the COVID pandemic when air travel demand dramatically dropped. Airlines need to keep their aircraft productive, so they spend a lot of effort optimizing their operations. However, planes need regular shorter and longer maintenance checks, which must be woven into an airplane’s operations schedule. These maintenance checks require differently certified technical personnel, facilities, and/ or spare parts, and during each check, there are usually more maintenance tasks that could be completed than there is time available. Airlines have developed sophisticated optimization tools to prioritize the right maintenance issues at the right time and location to limit their planes' ground time. In addition to this overall plan, their planning must also account for unforeseen interruptions like weather or unplanned maintenance issues.

The analogy for health care operations is evident. Many procedures require an interplay of different staff, facilities, and equipment. For surgery, you will need to organize an anesthesiologist, an operating room, a surgeon, special equipment, and clinical assistants. And even standard procedures often vary in the time they take, and unforeseen complications can change the operational plan. However, from my experience in both industries, I've seen a lot of airlines employing much more advanced and sophisticated planning tools than health care organizations.

Demand Optimization - efficiently fill availabilities

Once the organization has figured out which slots they should open, they must ensure that all these slots are booked and fully utilized. On average, airlines are able to fill 85-90% of their seats. The utilization metrics are much lower for medical facilities. Let's have a look at some of the challenges in filling their availabilities:

• Handling no-shows: If passengers do not show up and their seat stays empty, the aircraft is not fully utilized. Airlines have developed sophisticated methods to avoid this from happening: They usually sell more tickets than seats on a plane, and they use sophisticated data analytics to estimate how many people will show up for a particular flight. For example, business travelers on Monday and Thursday tend to be much flakier than leisure travelers on Saturday. Additionally, they allow passengers (and crew members) to be put on the standby list so that they can fill any unused seats. The killer app to avoid no-shows, though, are non-refundable fares: we are so accustomed to this that we don't even think about it anymore. If we book a flight and don't take it, we will forfeit our ticket price. No-shows are a massive issue for health care organizations as well. Doctors report no-show rates of 20-40%!!! I think they can learn quite a lot from airlines here handling no-shows. Double-booking using statistical analysis, standby lists for short-notice appointments, and upfront co-pay collection can and should be more broadly employed by health care organizations.

• Demand fluctuations: certain times of the day/ week are more desirable for airline travel than others, and demand peaks during Christmas, Thanksgiving, and the summer holidays. Health care is no different - ERs have peak hours, and certain appointment times are more desirable for patients to book (late afternoon vs. lunchtime). Airlines can modulate and distribute demand through dynamic prices, i.e., making less desirable times cheaper so people will shift to these times. However, demand fluctuations are a much more significant challenge for healthcare organizations - as, in general, they cannot move acute visits to other times, and I don't think we will see "surge pricing" for health care soon. However, some interesting trends are happening around virtual care. Moving less acute care from urgent care clinics and ERs to an online setting for triage can make addressing demand fluctuations more feasible. Another interesting way health care can handle this is by filling low-demand slots more efficiently: airlines are using thousands of programs and sophisticated analytics to build loyalty with their customers and get them to travel with them. Health care can also employ more of these tactics to fill open slots with preventative care screenings by activating patients.

• Efficient distribution: Another critical piece to filling all appointment slots is efficiently distributing your capacity to anyone in need. The travel industry has developed very efficient distribution channels for their tickets: Airlines not only show their available tickets on their website and to their own booking agents, but they also give them to third-party distributors who will make their inventory available to travel websites and other agents. Health care is severely lacking this: I've written about the friction in the scheduling process, which trickles down to making availabilities more broadly available to anyone who needs an appointment.

Managing "Fare Classes"

A passenger is not a passenger. Airlines know that quite well! A passenger traveling for business has quite a different willingness to pay than a student booking their flight back home to their parents. Airlines developed various strategies so passengers self-select into different fare classes, and the airline can maximize the amount of money they get from each customer.

People don't like to talk about it, but fare classes also exist in health care. However, it is much more subtle - because a patient usually does not know whether they are a "business class" or an "economy class" patient, and health care generally does not differentiate the level of service they provide to each patient. However, patients generate different amounts of revenue for doctors in two main ways:

• Insurance rates: Health plans pay vastly different rates for the same procedure at the same hospital. The rates depend a lot on the negotiating power the health plan has with the health system and the type of coverage. Rates for commercial insurance is usually much higher than for Medicare and Medicaid patients.

• Procedures: Certain treatments and procedures are much more profitable than others for a doctor. GI doctors can bill much more for endoscopies than for spending time on patient consultations. A patient's condition can be more procedure or more consultation-heavy and thus generate different amounts of revenue for the practice.

Like airlines, health care organizations need to balance their "patient mix" so they can operate without a loss. They often say that their commercial population "subsidizes" their Medicare and Medicaid population and that they operate these at a loss. This is a bit too easy to say as you have to look at their operations as a whole. With their high fixed cost, any additional patient increases their revenue, same as any additional economy ticket is a value add to an airline. While it is true that airlines make a bulk of their income from business class tickets, they would not be able to operate profitably without their economy class passengers.

An interesting thought is what these fare/ patient classes mean for utilization metrics. Airlines are pretty fixed in how they configure their aircraft and decide on how many business class and economy class seats they add. However, they can hold business class capacity basically until boarding and then fill it with economy class passengers later. For many doctors, this is a bit more nuanced. Instead of seeing a patient once (selling a one-time ticket), they often establish a longer-term relationship with their patients. If they have seen the patient once, it is harder for them to "drop" the patient than for an airline that "upgraded" an economy class passenger. They have to optimize their "patient mix" more long-term, and they do that whenever they accept a new patient to their practice. They basically don't have great mechanisms to shift capacity between these classes in the same way airlines can move passengers. This means that there is potentially some capacity that is not being fully utilized.

Another interesting observation is that airlines have differentiated themselves to focus on certain passenger types; for example, low-cost carriers focus on economy class passengers, and regional carriers concentrate on certain types of routes! We are starting to see more and more differentiation by procedures and patient populations for doctors. For a while now, outpatient surgery & imaging centers (low-cost carriers) have tried to unbundle hospitals (legacy carriers) by focussing on streamlined processes and facilities for certain procedures. Also, value-based payment models are promoting more differentiation of providers by population groups. Cityblock, for example, focuses solely on Medicaid patients, and provider groups like Cano, ChenMed, and Oak Street focus on Medicare populations.

Passenger Operations: Booking, Check-in, and Boarding

The razor-thin operating margins for airlines have spurred a lot of creativity to streamline every operational process. Examples are: replacing paper manuals with iPads (= less weight), switching faster from aircraft power to ground power at the gate (= less fuel), etc., etc.! In addition to aircraft operations, airlines became very good at shifting manual labor from the airline onto their passengers. Here are a few examples:

• Book tickets online

• Online check-in

• Automated security background checks (TSA Pre-check, FBI no-fly lists, etc.)

• Check-in Kiosks & Bag self-tagging

• Facial recognition security checks

You can argue that a lot of this means less service (as we have to do a lot ourselves), on the other hand, they made it very easy for us to do these tasks and I usually hate it if I have to call the airline for something I could do in their app.

Many of these processes directly map to similar activities for health care organizations. However, there is much less pressure to optimize and digitize these processes. Most doctor appointments still have to be booked via phone, patient registration is done on paper, and insurance checks and prior authorizations often require the practice to pick up the phone to complete the registration.

While we see changes here, adoption is very slow. There is just much less pressure to change…

Safety & Quality: Standard operating procedures

Commercial airlines in the US have not had a fatal crash since 2009. Flying is the safest means of transport by far, and the reasons for this can be found in how airlines handle safety issues. Over the last decades, every accident or near-accident has been meticulously analyzed by investigators. They publish detailed reports about the different immediate and secondary reasons for what led to the accident, and they include recommendations on how standard operating procedures need to be adopted to avoid any future accidents. These reports are not to blame but to learn from mistakes. Even if an accident was caused by human error, they would look into how better system design could have prevented the mistake. Over the decades, this has made air travel incredibly safe. Airlines have sophisticated checklists and safety procedures to catch and mitigate errors early.

In “The Checklist Manifesto” Atul Gawande describes how adopting more checklists in the medical field can reduce the number of fatalities due to medical errors, which is a much higher number than in air travel. I don’t want to go into too much detail here, as I am not a doctor. Still, I guess there is quite some opportunity for health care organizations to better codify and share best practices and have industry-wide post-mortems to improve standards of care.

Differences - why are they not the same?

Okay, so let's examine why many of these comparisons are probably much more nuanced than what I have written above and also what some of the dynamics are why health care is less eager to address some of the issues:

• Demand exceeds supply: In the summer of 2022, airlines operate probably the most similar to health care than ever. After two years of sitting at home, pandemic-fatigued travelers just want to travel at whatever price. Also, layoffs and retirements during the pandemic put airlines in a tough staffing spot. In health care, this problem is much worse and much more chronic: demand for certain doctors and specialties often exceeds supply, and medical staff burnout is real. As long as the number of doctors are artificially constrained by the number of available residencies (vs. by a candidate’s aptitude) and nurses and nurse practitioners are restricted in their permitted activities, we will continue to see this dynamic.

• Price sensitivity: Probably the most significant difference between airline travel and health care is: patients don't shop for prices when going to the doctor. While a $10 ticket price difference can make passengers switch airlines, we don't see this dynamic for patients. They don't see the prices and generally don't have to pay them directly. Even high deductible plans have changed little about this. Also, even if they see the prices, people care more about brand and quality when they look for health care. Interestingly, patients behave quite similarly to passengers traveling on a company’s dime. They won’t choose the most affordable ticket but book based on their loyalty program and buy tickets that maximize their collected miles. Hello, adverse incentives!

• What is the baseline quality? From a safety perspective - all airlines will guarantee a minimum of quality. Even when flying a low-low-low cost carrier, without seat assignment, your knees in the back of your front passenger, and $10 on-board water bottles, you can be assured that the plane is very, very unlikely to crash. Defining the minimum level of quality for health care is much harder to determine. There are differences in quality and outcomes depending on the doctor and facility you go to. Still, it is hard to tie them back to the doctor's capability as health care outcomes are inherently variable. Due to the complex nature of health care, it is much harder to impose and regulate standards for everything.

• Regional diversity and diversity of needs: Health care is a regional business - the competitive dynamics in a certain market depend a lot on the specific mix of health systems and independent providers as well as a number of payers in the market. In many markets health systems form a quasi-monopoly for care. Airlines, however, compete much more on a national scale, as in principle, any airline can start and close a route at any airport, thus increasing competition. Also, entry barriers for airlines are much lower than for health. In addition to the regional dynamics, health care organizations serve much more diverse needs than airlines. Air transportation is about safely getting from A to B, with comfort being another dimension, but not much else. While aircraft maintenance crews have to handle thousands of checklists for different systems and aircraft types, the human body is unlikely more complicated. There are many specialties, many types of populations to serve, and thousands of different treatments and procedures. Health care is a much less standardized "product", and thus processes are much harder to be streamlined. 

These differences explain why health care is behind in solving many of the challenges above. There is much less pressure to change than in the highly competitive airline industry. But maybe the shift to value-based care and the financial pressure of the COVID pandemic will spur the adoption of innovations in health care - and thus creates massive opportunities for new products!

I would love to hear from you about what else health care could learn from other industries. Let me know in your comments, or reach out to let me know your thoughts!

Happy Price Transparency Month! Since July 1st, 2022, hospitals and health plans are required to publish their negotiated rates in machine-readable files on their websites. Payers and health systems have long well-guarded their networks and contracted rates and saw them as an essential competitive asset. They cannot protect it anymore. This is a massive win for increased competition and to bring to light some of the craziness of provider-payer contracts.

While increased price transparency is a huge step, procedure prices are only one part when evaluating how much an episode will cost with a specific provider. The other parts are utilization and quality. Even if a provider offers lower per-service rates, the provider might not be the best choice from a cost perspective. They might tend to order more labs and images per episode of care or opt for more expensive treatment options on average. Another factor is quality of care - if a procedure, test, or image must be redone because of poor quality, a lower price will not matter.

With prices being more in the open, network builders will need to focus more on these other metrics, which are much harder to measure. Not only price transparency but also other trends are driving organizations to invest more into their network analytics and building out provider networks. In this article, I will take a closer look at everything related to network building!

Who is building networks and why?

Network building has long been a major focus for payers - it is a significant lever for them to use their volume to negotiate cheaper rates with provider groups and thus lower their total spend on health care services. However, they are not the only ones anymore thinking of running provider analytics - here are some other examples:

• Self-funded Employers: Self-funded employers decide to bear the health care cost for their employees and thus are financially incentivized to optimize the health care costs of their employees. An important priority for them is to avoid large cost outliers. These high-cost areas include cancer care, transplants, orthopedic & bariatric surgery, fertility care, and maternity care. A single complex case in any of these conditions can easily double their yearly health expenses. Thus, they have started carefully selecting the right provider partners for these conditions to avoid adverse outcomes and exploding costs. Direct primary care, care navigators, and virtual care companies selling to employers have a similar incentive here and evaluate specialty providers based on cost and quality. This is one of their main sales arguments.

• Value-based care organizations: With ACO Reach and the move to more and more fully-capitated models, risk-bearing providers will need to think more and more like health plans. A costly MRI or colonoscopy now directly affects their bottom line. Thus these organizations are starting to build their provider analytics stack to identify high-value specialists. Actually, legacy risk-bearing provider groups (for example California primary care groups) have been developing their own high-value networks for decades.

• Virtual & digital health companies: Good patient experience is critical for digital health to retain patients. But the scalable virtual care model often clashes with the highly localized in-person model. Virtual care companies need to find the right partners for in-person care, so they will be able to handle more acute and severe cases. Identifying the right providers at a national scale and building relationships that support seamless handovers between virtual and in-person care is quite a challenge.

• Payviders & Neo-health plans: Narrow network design has been touted as a new great way to reduce overall health care costs. This narrow-network strategy is pursued by several Neo-health plans and emerging hospital-affiliated "Payvider" health plans. However, these plans must carefully balance their preferred narrow network with adequacy requirements. (more on this below)

• Cash pay platforms: Some people think the solution to many of the problems of US health care comes through cash pay networks. Cut out the intermediaries, make prices transparent and let patients decide. These new platforms are also building provider networks willing to take cash for their services. 

What does network building mean?

Building provider networks is easier said than done and involves several steps. Here is a quick overview of things that you will need to do to create a performing provider network:

• Adequacy: The first step for every provider network is to provide sufficient coverage for the health needs of its members/ patients. This means having enough cardiologists, interventional cardiologists, cardiac electrophysiologists and nuclear cardiologists etc. , that participate in the network. However, network adequacy is quite poorly defined. CMS has tightened its adequacy requirements for ACA, Medicare & Medicaid plans starting 2023 by adding more specialties to the adequacy criteria. One way to achieve basic adequacy is to borrow a network from an existing health plan, like Blue Cross Blue Shield or Aetna. There are also companies like Quest Analytics and (just recently founded) J2 Health that help organizations achieve adequacy and comply with regulations.

• Refine your network for your need: After establishing basic adequacy, the organization will need to start refining its network. Several criteria can be taken into account: First, different populations have very different needs. For example, a health plan for construction workers will need much more orthopedic surgeons in their networks. A health plan focusing on Asian-American or Latinx populations might want more providers speaking their language and having cultural competencies. Second, there are other factors such as cost and the ominous "quality" factor. For many organizations, the network-building journey ends at this step - there are many consulting firms that help with one-off network optimization projects.

• Collect Data on providers: More sophisticated organizations will start to build their own database around provider performance. There are several sources for doing this: 1) Claims data - this is often the richest internal data source a health plan/ employer or value-based care organization has access to. 2) External data vendors - companies such as Clarify, Ribbon, CareJourney, Turquoise, and Definitive Healthcare offer datasets on provider information ranging from simple things like address & phone numbers to more complex information such as episode quality metrics, cost metrics, and clinical specialization. Simple data points do not often mean easy: running my own analysis, I found that over 20% of the phone numbers in a payer’s provider directory are wrong - and this is probably a low estimate. 3) Organizations often also have access to several internal data sources, like patient surveys or even tribal knowledge from their providers, who often know pretty well which providers in the community one should work with and which ones to avoid.

• Go out and build relationships with providers for contracts: Organizations might opt to create a one-sided network, meaning that providers don't know that they are part of the network and are selected to be preferred. However, in most cases, building networks means establishing a contractual relationship between the network organization and provider. These contracts could entail preferred rates, volume guarantees, or technology integrations. Establishing these contracts and relationships, in most cases, requires "boots on the ground" and can take quite a substantial amount of time to develop. For health plans, this contracting process is also called credentialing, and there are several companies in this space trying to make this easier.

• Network enforcement: Once the network is established, the work is not done. To reap the benefit of the carefully constructed and negotiated network, one must also convince patients to choose the in-network network providers. This is easier said than done. Restricting network access through primary care providers didn't fly well with freedom-loving Americans when payers tried to establish restrictive Health Management Organization (HMO) plans. These days network enforcement has to be much more subtle and involves nudges, steering through assistance, and offering incentives.

• Learning: Network building is not a static process - the world is constantly changing. The member needs are changing, provider's performance is changing. That's why network building is an ongoing activity.

Those are some of the basic building blocks for provider network building! Here is an overview of companies, helping with different parts of the stack!

Let's now have a look at some unique approaches to network building.

1) Networks addressing total cost of care

As already discussed, in the commercial insurance population, a few conditions make up the majority of health care spending. Most of the focus for payers and self-insured employers is to control these episodes' costs. One way of handling this is through bundled payment programs, i.e., shifting the risk for a specific condition to another organization and giving them incentives to properly manage their spending. Here are a few different flavors of these bundled payment networks:

• Center of Excellence: This wonderfully sounding concept is not very well defined, but it often means that payers or employers partner with a select network of providers to care for specific, high-cost conditions. For example, Walmart partners with a dedicated network of clinics to take care of the following services: weight loss surgery, cancer care, hip replacements, cardiac surgery, organ transplants, and spinal surgery. As the biggest employer in the country, Walmart has enough volume to negotiate preferable rates and build its own preferred provider network for high-cost conditions.

• Specialty-care networks: Center of excellence networks can also be "borrowed" from vendors aggregating these provider groups. If an employer does not want to go out and set up direct relationships, they have options. Many health plans offer center of excellence services with bundled payment rates. Companies such as Carrum or Transcarent's Bridge Health provide access to a network of high-quality providers and bundled payments.

• Cash-pay networks: Cash payment networks are a newer trend in the same vein. The idea is to cut out the health plan intermediaries and create direct cash payment relationships between employers and providers. Instead of having a health plan negotiate rates with providers and resell them to employers, cash-pay networks usually let providers set their own prices for their episode bundles and act as a broker between employers and providers.

• Specialty care navigators: Certain specialty conditions like cancer or musculoskeletal conditions (MSK) can be quite confusing for a patient, and making a "costly mistake" is easy. Specialty care navigators are helping patients to make the "right" decisions, understand their options, and allow them to access high-quality (and usually cheaper) provider groups. In many cases, they come with quite deep analytics on the quality and cost of their providers.

• Imaging Networks: Medical images have high variability in cost and quality, and it is often a major focus for payers and employers to control their spending for these services. A few interesting companies in this space include Covera, which is building a network of high-quality imaging centers, and Medmo, which makes finding high-value imaging centers easier.

2) Networks filling unmet service gaps

Not every organization wants to build its own network. It takes time, lots of data analysis, and building contractual and integrations with provider organizations. So several organizations are building provider networks for other organizations to borrow, in particular in areas where access to care is often challenging and provider demand greatly exceeds their supply.

The line between network and health systems/ integrated practice groups is blurring here. Some of these network builders have quite loose relationships with their providers and act more like provider directories; others are fully integrated IPAs that come with a tech stack and unified branding. Here are a few examples for different focus groups:

• Behavioral Health: Access to behavioral health providers has been notoriously tricky - due to stigma, the special relationship between patient and practitioner, and the lack of behavioral health specialists. Many digital health companies are trying to improve access to behavioral health. Examples here are Alma, Headway, and SonderMind. 

• Post-Partum Care: Similar to behavioral health, maternity care and in particular post-partum care have been challenging to find. Many health plans often don't have doulas and lactation consultants in their network, making it hard for patients to find providers and get these services reimbursed. Zaya Care is trying to change this by creating a network of maternity care providers that are in-network with several carriers.

• Social Work: Social determinants of health are becoming more and more relevant and health equity got a much higher focus through the new ACO Reach program. Several vendors are making it easier for large organizations to connect with local social work initiatives. Unite Us is an example here, aggregating social work organizations across the country.

• Culturally competent care: More and more health plans care care about serving historically underserved conditions and populations and they see providers with cultural competencies as a critical factors here. Companies such as Violet are training and certifying cultural competencies for providers and provider groups with a specific population focus are emerging, for example Zocalo for LatinX population and Rendr for Asian-American population.

3) Networks bridging virtual to physical

You can do only so much via video consults or asynchronous texting. Some exams and procedures just have to happen in person, especially for higher acuity patients.

I've spoken to several digital health startups, and many are thinking about how they should set up relationships with in-person providers. Digital care companies often want their in-person provider to follow a specific treatment protocol. For example, they often would like to establish new care protocols that allow their patients to move directly to procedure vs. having initial consult meetings.

Bridging virtual to in-person care is a nascent area, and I have not seen many players in this space yet, but I believe we will see more growth here. A great example is Solv health, helping digital care companies gain access to urgent care appointments. Their network of urgent care clinics can support many services needed by digital primary care. Another company in this space is Rezilient health, which is lending its in-person clinics to other digital care providers. This model is especially exciting as modern in-person clinics are probably much more flexible and innovative in supporting the specific needs of digital care companies than established health systems.

Thoughts

The fragmented nature of the US health care system with its thousands of payers and provider organizations makes network building an essential activity. Here are some further thoughts on this topic:

• Provider networks offer a competitive edge: Even with price transparency rules making negotiated rates public and revealing some of the competitive advantages of provider networks and relationships, they will remain an essential competitive asset. A good understanding of utilization, quality, and negotiating preferred arrangements will still be critical for more and more health care organizations. It's interesting to look at which roles provider data companies will play in this new world. I believe that many provider data providers will become a commodity as they can be purchased by every provider organization, i.e., you have to buy them to stay competitive, but it won't provide you a competitive edge. Health care organizations must run analytics to gain an advantage and put their own spin on the data. A good analogy here is credit agencies - credit card companies have to buy your credit score but to really understand your creditworthiness and gain an edge over the competition, they will need to combine this information with their own data about you.

• Network enforcement & engagement: Having had many conversations with value-based care organizations and employee-facing care solutions, I got the impression that many organizations are thinking very actively about provider analytics but don't invest a lot of resources into enforcing their network and nudging their members and patients to their preferred network. While it might seem a logical step first to find your targets and then think about how you hit the bullseye, it is also risky. All this investment into a preferred network might not pay off if you're unable to steer your members/ patients to the right providers. I find this problem so exciting that I will write another article about this topic.

• Network contract market place: The complexity of provider networks will only increase in the next few years, and this will bring a whole set of new challenges. Insurance networks are pretty easy to understand, but a lot of organizations are starting to break the boundaries of these networks: they establish carveouts for certain conditions, they tier their networks, they develop direct-cash relationships with providers, etc., etc.! Many lawyers will have a blast billing their hours to establish and maintain all these agreements. I think there is a genuine opportunity to simplify network access for provider groups and reduce the friction contracting can take in the network building process. Maybe companies such as Medallion and Verifiable will enter this space. Their credentialing services greatly position them with provider groups to offer them more "networks and arrangements" they could join.

Network building is complex, and there is a lot of whitespace for products to be captured! Organizations are just starting to invest more resources into selecting their preferred network and analyzing how changing their network impacts their revenue and health outcomes.

If you're a healthcare organization thinking about network building & patient engagement - please reach out. I would love to talk!

That is easier said than done. Especially in the reality that many value-based care organizations operate in. Let's have a look, for example, at Medicare Shared Savings Program ACOs. In many cases, they are a loose collective of physicians and care providers that agree on implementing certain programs or technology investments to realize savings and better outcomes.

However, there are quite a few challenges that make this goal hard to achieve:

• Lack of incentives: How much revenue do saving initiatives really bring for each ACO member? ACOs usually do not have much executive power over their members in what things to do or not to do. They must convince their members that a particular savings initiative is worth implementing. If a practice only has 10% Fee-For-Service Medicare patients (vs. Medicare Advantage patients and commercial patients) they are most likely not willing to make any far-reaching investments into new care protocols, retraining staff, or adopting new technologies. If the overall incentive for realizing savings is small, i.e., maybe $30k per physician per year, they won't make significant changes.

• Lack of resources: Even for successful ACOs who realize annual savings and pay out a bonus, there is not a lot of wiggle room. Investments pay off only after a while and are uncertain; therefore, solutions need to have quite some clear ROI and scope. There is not a lot of experimentation budget, especially with the current condition of the capital markets (June 2022).

• Competing interests with their members: As always in health care, many competing incentives make decisions more complex. These competing incentives are especially prevalent in hospital-led ACOs - for example. They might choose not to implement specific savings programs, as they would cannibalize revenue from their other divisions. For example, the hospital CFO would not want their ACO to send patients to providers outside their health system's network even if they would be more cost effective.

To succeed, the value-based care movement will need to overcome these challenges. A significant driver here is new products that help value-based care organizations look beyond the low hanging fruit and that help them realize savings, providing them with a clear ROI. Several vendors are trying to enable value-based care organizations to realize savings easily. Today, I want to look at them in the following categories:

• Provider enablement - help the provider do what the data suggests

• Patient engagement - help the patient do what the data suggests

• Population programs - take care of a specific sub-population

Provider Enablement: help provider change their workflows

As I mentioned in the first part, analytics provides the foundation for value-based care. Analyzing EHR and claims data can identify populations and specific patients that need certain actions. But analytics needs to be actioned, and insights need to be leveraged by providers in their day-to-day clinical operations. Here is how different solutions are trying to solve this problem and enable physicians:

• Point-of-care solutions: Several companies integrate with a provider's EHR to produce actionable insights during a visit. These tools surface the "most valuable" action a physician can take for a particular patient, given their medical record. These actions can include ordering a preventative screening, ensuring risk factors are appropriately captured, or, if the patient qualifies, they are assigned to special subpopulation programs, like chronic-kidney disease or hypertension programs. Examples of point-of-care solutions are Vim and Affirm Health. Modern EHR systems like Elation or Canvas can integrate this logic into the physician's interface without the physician having to change tabs, which further reduces the friction.

• Incentive-based solutions: Some point-of-care solutions are taking it one step further. They not only surface the most valuable next-action but also reward the (attending) physician with a financial incentive. Quantifying the impact of a physician's decisions in a value-based care context can help drive awareness of the economic consequences their actions can take. Leading solutions include Stellar health and the Clover assistant, who reimburse providers based on their value-generating actions.

• E-Consults: Primary care is often not much more than a triage in which the doctor decides which specialist they should send the patient to. They are taking on a much more prominent role in the value-based care paradigm. As specialist visits are expensive, value-based care providers are trying to equip their primary care doctors with the knowledge to take care of increasingly complex patients. E-Consult platforms, such as Sitka or RubiconMD, are an excellent way for primary care doctors to connect with specialists and get their advice on the next treatment steps or whether a specialist referral is actually necessary.

• Referral workflow solutions: There are many non-trivial workflows in the value-based care paradigm where practices could use additional support. Great examples are outbound referrals or care transitions from one facility to another. Several companies are trying to make the hand-off between organizations easier. They allow physicians or front desk staff to triage the appropriate & cost-effective provider for their patient, transfer the patient with all the required medical records to the specialist, track the scheduling of the visit and close the referral loop. Examples here include Par8o, ReferWell, ReferralMD, and Preferwell (acquired by AristaMD).

• Training & Education: Last but not least - provider buy-in and understanding of how value-based payment models affect their clinical outcomes and practice's bottom line is vital for any value-based arrangement to work. If providers are not changing their behavior, either out of lack of incentives or lack of knowledge and understanding, nothing will change. I have not seen any provider education companies here focussing on education and training, but I am sure there is a significant need for those.

One last thought on provider enablement: It's worth noting the different business models - some players are getting paid by the health plan (Stellar), and others are getting paid by the provider group (Affirm). As payers and providers have, in general, aligned incentives in a value-based model, however, given the patient mix of the provider, there can be differences in terms of adoption.

Patient Engagement: get the patient to do things

Doctors can only do so much to make their patients healthy. The patient's actions (or in-actions) play an enormous role in their treatment success and overall health. Engaging patients is an important activity for many value-based care organizations. Not only to detect unmanaged chronic conditions early but also to get their annual risk adjustment done. There are different channels for providers to engage with their patients:

• Digital Patient experience: Digital patient outreach via e-mail, apps, text, or web portals is critical for providers to activate their population. Many people dismiss the effectiveness of digital channels for Medicare beneficiaries, who comprise most of the value-based care population. When listening to the Twitter crowd, one could think that 65+ are unable to use a smartphone or the internet. I believe this is a great myth! The share of people over 65 regularly using smartphones is increasing every month. 61% own a smartphone, and 45% are on social media. However, digital patient engagement tools need to be designed appropriately. For example, ensure that larger fonts in your mobile app do not result in endless scrolling. Lena, for instance, is successfully activating their Medicare members through a text-based chatbot, helps them set up appointments, and gives them helpful information about their patient journey.

• In-person Community engagement: With COVID cases falling, in-person engagement is making a comeback - to engage with patients, you have to meet them where they are, which is often in public, in their community. Cano health, for example, a value-based care provider in Florida, strategically selected their provider office locations in shopping malls and community centers frequented by their target population. These locations make it much easier for them to stop by their practice when needed. Another interesting approach is to engage patients at their barber shop or wherever they are in their community (churches, bingo events, etc.). Life Chair is a new Medicaid-focused company in this space.

• Home visits: The patient's home is another essential space for patient engagement. If patients are too frail or it is inconvenient to come into their doctor's office, the value-based care organization might choose to visit their patient at their home for particular exams and treatment. A big player in the value-based home health space is Signify Health. Although recently, there has been some criticism around the practice of home health providers around risk adjustment - they would visit the patient to get a diagnosis code, but they won't provide much value add beyond that for the patient.

• Remove barriers: Last but not least - you can engage your patient as much as you want, but as long as certain barriers keep them from scheduling a visit or following their treatment plan, you won't help them get better. That's why value-based care organizations focus on removing some of these barriers. This includes transport services to and from doctor appointments, child care during hospitalizations, access to food, etc. United Us is a great player here that addresses social needs and connects patients with social organizations in their community to improve their health and access to care.

Population Programs: Target specific patients

Several vendors specialize in initiatives for a specific subpopulation and offer this out-of-the-box service to risk-bearing organizations. They would use analytics (as mentioned in part one) to identify a population for which they can provide specific interventions. Here are a few great examples:

• Post-Acute-Follow-up care: A "low-hanging" fruit for cost reductions are avoidable readmissions. Strong evidence supports that a follow-up visit with a primary care doctor within 14 days of hospitalization can significantly reduce the likelihood of a readmission. This is a common use case many ACOs are focussing their resources on. Ready Responders and Signify health have products that send a primary care provider to the patient's home as a follow-up visit.

• Medication Management: Proper, holistic medication management is another compelling use case that can lead to value. Companies like Upstream and Lumi help value-based care providers conduct in-depth medication reviews for their patients and adjust their medications to the proper levels. 

• Subcapitation for specific populations: Certain costly patients warrant special attention by dedicated care programs. Some VBC organizations are developing their population health programs, while others partner with specialist providers to care for a specific subpopulation. Examples are Cricket & Strive for patients with kidney disease or Cityblock for patients with multiple chronic conditions. Aledade also recently launched its Care Solutions business, which helps its physician members take care of sick patients.

These "subpopulation" vendors are not always selling to the care provider but can also work with the health plans. This makes sense for ACOs with fewer resources that don't have the time to assign their members to the dedicated program.

Food for Thought

Value-based care is not an easy feat, and even with all these vendors helping, it requires excellent management and people to orchestrate all these solutions successfully. Here are some of my thoughts:

• Don't two-class doctors or software: From most of my conversations with physicians, I learned that doctors don't like to have different treatment plans for different health plan populations. When they purchase a new software tool, they want it to work across all of their patients. Adjusting their toolkit to each person based on their payer is another thing that occupies their already busy brain space. You can solve this problem in two ways: either you provide software and tools that are payer/ population agnostic and that provides value for any patient (or a sufficiently large group of patients). Alternatively, you have physicians/ practices focus on specific populations - that's why I firmly believe Medicare-only ACOs will succeed in their quest to improve patient outcomes and lower costs. They will have the focus required to tackle all the challenges mentioned.

• Where are the microservices?: Reviewing so many vendors in this space, it is interesting how much overlap there is in the marketing messages and features of the different solutions. The software engineer in me is complaining: "Everyone is trying to be a monolith. Where are the microservices?" For those of my readers who don't know what that means, here is a quick definition:

Microservices - also known as the microservice architecture - is an architectural style that structures an application (= care delivery) as a collection of services (= software tools) that are

• Highly maintainable and testable

• Loosely coupled

• Independently deployable

The microservice architecture enables the rapid, frequent, and reliable delivery of large, complex applications (= patient journey). It also allows an organization to evolve its technology stack.

• I wish more vendors would focus on what they are good at, really focus on their strengths and solve the problem, and build robust APIs or easy-to-deploy integrations with other services. This would allow providers to mix and match their care stack to their needs. Given the lack of engineering resources in many provider organizations, I understand that there is a market pull for monolithic, one-vendor solutions, but this leaves us with many jack-of-all trades and few jackknives.

• Adjudication & Subcapitation: Every vendor in the value-based care space seems to pursue the business model of shared risk (vs. a SaaS fee or enterprise contract). From a business point of view, this makes sense, as this is a great way to capture a part of the value they are creating. However, this makes attribution a nightmare when more than 2 or 3 solutions are deployed simultaneously. For example, what if a person has kidney disease and diabetes and is enrolled to Strive and Omada at the same time? How will you figure out capitation and saving attributions? And what if that person then has a hospital visit where a third vendor is providing follow-up care? This becomes quite a complex problem without strict separation on a population level. Even if you can separate populations, managing several subpopulation providers as a provider organization can be quite daunting and organizations might not be ready for it.

With the growth of value-based care payment models new problems and opportunities for new products are arising, and even existing solutions are ripe for disruption with well focused products. If you are a risk bearing entity evaluating new vendors or a startup building in this space - let’s chat!

Thanks to Rik Renard & Morgan Blumberg for their invaluable input!

This post is a joint production brought to you by the Health API Guy (Brendan) and Health Tech Stack (Jan-Felix) - you should definitely be subscribed to both! Tremendous thanks to Julie Yoo, Joe Mercardo, Erica Jain, Krish Maypole, Morgan Blumberg, Samir Unni, Shashin Chokshi, and Colin Keeler for their invaluable input.

For almost all of humanity, our primary healthcare goal is to never be a patient. We seek to live our normal lives working and playing and doing whatever pursuits might fulfill us in our time on this planet. However, the wears and tears of our activities, the bad luck of chance, and other factors inevitably mean that we do become patients - we seek a checkup, we need urgent care, we identify a need for a medication, a procedure or a diagnostic.

At that precise moment (paraphrasing the words of Eric Andre) we all just want to be let in. We want to be able to see the exact person who can fix our problems or at least make us as close to whole again as possible. We want that to happen as quickly, easily, and cheaply as possible.

But...this is easier said than done. Appointment scheduling is the epitome of the dysfunction of the American healthcare experience, the purest distillation of an unending onslaught of frictions, missed handoffs, and varying failures. In particular direct scheduling, i.e. the ability for everyone to see when a doctor is available and book an appointment, is rarely adopted.

Fixing this is a big, hairy, audacious goal, so it’s no wonder that many have tried and many will try. The past month alone (detailed here in Brendan’s thread) has seen an incredible amount of change on the scheduling front. Brendan and I got together to talk a bit about the why (root causes and incentives), the who (existing players), and the how (where we think things will go and businesses will be built).

Why should schedules be open?

Direct scheduling is common practice in almost every industry: you can book airline tickets online, you can make restaurant reservations, and you can schedule workout classes. Calendly even brought this to practice to person-to-person schedules.

It’s not rocket science to imagine that patients want the same ease of access and consumer experience brought to bear for doctors’ appointments. There’s an inherent urgency to health problems - finding the intersection of speedy resolution and quality care is what we all want as patients. Extrapolating this further, there are myriad reasons (and even built-in incentives) for us to expect this to already exist in healthcare:

For Patients

The value proposition for patients is clear: convenience! We all have busy lives - we want to fit the doctor’s appointment into our schedule. Also, Gen Z (and maybe even millennials) simply avoid making calls on their phone like the plague, strongly biasing towards apps, websites, and text engagements. There’s no better way to forcibly eject younger generations from your onboarding and acquisition process than by putting them on hold. Without digital interactions that they can access 24/7, businesses are distinctly disadvantaging themselves against their competition.

For Physicians

Pure play forces of our grand capitalistic economy would suggest that doctors would want direct scheduling. Simply put (or oversimplifying tremendously), a practice can earn more revenue by treating more patients - and treating more patients requires more optimized utilization. This is true for both the fee-for-service world (more patients = more procedures = more cash) and in a value-based care world (better utilization = bigger patient roster = more capitation payments). Opening up their availability to a broader audience would presumably help doctors fill open slots that might otherwise go unused.

In addition, direct scheduling hypothetically reduces admin burden on the staff needed for making and taking phone calls and shifting appointments around.

As an aside - Short waitlists and easy scheduling are literally some of the core benefits most people list if you ask them why single-payer or other alternative healthcare strategies won’t work in America, so it’s actually quite ironic how shitty the current situation is with that in mind.

From a more altruistic perspective (those do exist in healthcare!), optimally utilized physicians could actively provide more and potentially better care. Automatic routing of care would enable physicians to focus on “specialized, high-acuity care” and appropriately route lower-acuity care to navigators, coaches, or providers more appropriately credentialed, theoretically without human interaction required to book.

For Referring Physicians

Direct scheduling would also make it much easier for primary care physicians to get their patients to see specialists for preventative visits. If the primary care doctor can schedule a follow-up appointment directly at the visit, the care continuity is maintained for the patients. The friction of scheduling is work for the patient that few enjoy or relish. Currently, scheduling a follow-up appointment is a barrier (the numbers don’t lie). The higher the barrier for the patient, the less likely they will be to actually show up at necessary screening appointments. The more engaged primary care providers are currently handling this for their patients, but only with more manual overhead and boots on the ground: calling specialists to see whether they have availability, handing off the patients to care navigators, or staffing a team of referral administrators. The less engaged, however, just hand out a bunch of business cards (with QR codes, even, if they’re fancy). Whether the specialist takes new patients and is available in a reasonable time frame, is up to the patient to find out.

For Payers (selectively) & Care Navigators:

More and more payers try to actively steer their patients to cost-effective providers. They have invested deeply in analytics solutions to identify those preferred providers in their network and now they are trying to steer their patients towards those providers. Direct scheduling is a logical way to make it easier for their members to access their preferred providers through the payer portal (a thesis that seems to be shared by Kyruus in their acquisition of HealthSparq). Employer-facing care coordinators such as Accolade, Rightway, Transcarent, or Collective Health also offer these concierge services for finding and scheduling appropriate care, and direct scheduling would reduce their manual tasks. Today, care navigation is heavy analog work requiring phone calls and faxes ad nauseum, solved simply by increasing staffing. Direct scheduling largely turns this ongoing services investment into more automation and allows care navigators to focus simply on finding the best possible providers.

Well - so why is direct scheduling not an industry standard?

“Wow, Jan-Felix and Brendan” you might say, “so many good points here. You’re so smart and talented. It feels objectively and abundantly clear that direct scheduling is in everyone’s best interest. So why isn’t that the case?”

Great question, valiant reader. The fact of the matter is that despite the weight of the incentives above, there are equal or greater headwinds historically to ubiquitous accessibility on a national scale.

Problem 1: Triage - “Risk of seeing the wrong patient”

While some restaurants may limit their clientele based on dress code or other factors, generally speaking, any given customer is equivalent as long as they have a willingness to pay. In this regard, scheduling a doctor's appointment is very different from making a restaurant reservation - not all customers are equally appropriate.

First of all, a missed appointment is much more costly in healthcare than for a restaurant, as there are usually not enough walk-ins to fill the space. Therefore, there’s a strong prerogative for providers to be 100% sure that when they have scheduled a patient, they actually will see that patient. There are more than a few factors that need to be cleared before the visit:

• Insurance Coverage: Is the provider part of the insurance network of the patient?

• Prior Authorization: Is a prior authorization needed and has it been approved yet?

• Referral: Is a referral needed and has it been received yet?

• Specialty Match: What is the reason the patient is coming in? Is the doctor the right person to see the patient? This is especially important for sub-specialties. For example, an electrophysiologist is technically a cardiologist, but will not see patients for medication management. These subspecialists are especially protective of their calendars.

• Experience Match: The body of work (and volume of work) of the physician correlates to the outcomes of that care. How does a patient get matched to the provider with the right experience for their acuity? The example Julie outlined years ago still resonates here.

• SDOH and Intangibles: Do they speak the same language as the patient? Is the patient able to get to the provider?

Most of these checks are done manually and laboriously today. Scheduling process optimization via technology isn’t necessarily a core competency of providers - they are comfortable doing them this way. Change inherently has some risk of missed income and wasted time.

Problem 2: Visit Type Matching

A restaurant typically doesn’t have a variety of different types of experiences to choose from. A restaurant booking may have sub-divisions for table size, indoors/outdoors, or some small features. Likewise, although hotels may have different room types, there isn’t a ton of nuance beyond that - a better room is roughly correlated with increased price, without other dependencies or sub-classifications.

The opposite is true in healthcare. Each visit is a complex web of matching patients with the visit type appropriate to the exact moment in the patient’s journey. Even if the patient is correctly matched with a provider, they will now need to be matched with a certain appointment type.

The easiest appointment type is an initial consult, as it is quite standardized and many providers actually allow direct scheduling for this visit type. However, it only gets harder from here. For other visits, like procedures and treatments, the doctor will need to allocate different amounts of time, and will need to make sure facilities & equipment are available and they might need to order drugs and tests before the visit. Imaging centers provide a great example - even though ordering an image may seem simple, there are quite a few specifications on what type of image type needs to be performed and sub-divisions of those image types, such as with or without contrast.

What makes this even harder: every organization has its own flavor of appointment types. This means that it is quite hard for an outsider (even for a medical professional) to discern what the right visit type is actually correct to book. For this reason, most practices gate their appointments via the front desk, a shadowy priest class present at every healthcare institution who have memorized and studied the tribal knowledge and who alone can divine an appropriate time to see the professionals they represent.

Problem 3: Fluid calendar - patients and operations are unpredictable

It’s like clockwork. A practice starts the week with a perfectly planned calendar, a beautiful work of art balancing patient types and conditions, with the right breaks for charting and lunch. But by Monday afternoon, everything has changed - the schedule is a minefield geared into hyperdrive, long days laden with overlapping appointments, interruptions, and no breathing space. Doctors’ calendars need to allow for quite a bit of flexibility for several reasons:

• No-shows & Rebookings: You’d think the urgency and acuity of healthcare might mean people fulfill their promises and show up when they say they will. You’d also be wrong. Patient’s fear of needles, logistics of getting to the provider’s location, and other factors all combine into relatively frequent no-shows and rebookings. This is especially prevalent when coming from online scheduling services (like ZocDoc) and/or when no prior relationship exists. As a result, some practices copied the practice from airlines to double book or overbook their day or allow for walk-ins and waitlists to fill the open slots.

• Appointment lengths: Appointment lengths are also not always predictable - certain procedures can take much longer than expected if acuity is not judged properly or an emergent condition is uncovered. Like a butterfly flapping its wings, this has a ripple effect cascading into a hurricane throughout the whole calendar. This can mean that some appointments need to be rebooked, decreasing patients’ trust in the provider.

• Acuity: Some doctors keep capacity open for acute patients. While many procedures can be planned well in advance, often patients need care on the same day or week. Handling these types of patients requires quite some appointment juggling.

Problem 4: Control & Doctor Preferences

Besides the hard business factors, there are also soft factors for doctors on what constitutes an optimal schedule. They range from easy preferences like “I don’t want to see patients on Friday afternoon so me and the boys can hit the range” to complex scheduling logic “I would like to schedule patients only consecutively starting backwards from 2 pm” to hyperspecific logic predicated on incredibly nuanced patient attributes, like “I would like to not see two highly emotional patients in a row”. A lot of these preferences live in the schedulers’ brains and are internalized over a long period of time (the aforementioned tribal knowledge above). As a result, it is quite hard (and perhaps impossible) to codify those intangible preferences into machine computable logic.

Problem 5: Why see more patients?

Many doctors have more patients than they are able to handle and they have long waitlists. This is especially true for doctors with strong established relationships in their community or that have a good reputation and/or low coverage of their specialty in their region. These providers are obviously fine with the current way appointments are scheduled and lack an incentive to move to more modern processes. In fact, gatekeepers and scheduling barriers might even be desired to control the flow of new patients!

Even for PE-owned and hospital-owned practices, the incentives for taking on more patients and fully utilizing the time is not a given. Often compensation schemes are salaried and not based on the number of patients a doctor sees, further degrading the incentive to squeeze in more appointments.

Problem 6: Relationship & patient preferences

There are many people in this world with different preferences; thus, a phone call is a communication medium that some populations of patients (and their providers) actually still prefer. The highest cost patients are usually in their 70s and older. This population distinctly lacks the adverse attitudes toward phone calls that millennials and Gen Z have. They are not savvy with text or typing, they don’t download applications as frequently, and they default to conversation to solve problems. For providers serving Medicare or other groups of patients, relying on phone calls is actually the optimal way to maintain the relationship with the patient and keep them engaged.

Beyond that, there are other preferences that complicate scheduling, such as cultural factors. Few of these outreach solutions have multi-language support, which is especially important among physicians serving communities in which English is a second language.

Problem 7: Integration sucks

Last but not least - setting up this particular workflow with EHRs is really quite tough, at least until recently. You need availability synced to you, but few EHRs supported a query model to pull this in real-time historically. Synchronizing appointment templates is one option, but what if Dr. Smith gets sick or plays hooky to go hit golf balls? EHRs also have historically pushed back on multiple sources of truth for scheduling, so while writing appointments back into the EHR has had broader support technically, there have often been hoops to jump through. Redox literally has a whole white paper dedicated to this topic, suggesting it’s problematic enough to warrant marketing their solution.

Market Map - how providers are trying to address this

Disregarding all of these challenges, several fearless vendors are trying to address this problem and open up doctor’s schedules for you!

Here are the different approaches:

• Consumer-Facing: The main value proposition of these consumer-facing companies is offering new patient leads for the health care providers. Aside from some aggregators that rely on advertising, they usually charge the provider a fee for each patient that enters the doctor’s office through their platform. One fundamental challenge of these tools is that patient acquisition costs are usually a one-time cost vs. a recurring line item. As described earlier, doctors have to do marketing until they don’t have to do it anymore. Building a strong provider value proposition beyond that is often something these consumer-facing companies are not well-equipped to do.

• EHR & Practice Management Systems: More and more EHRs and Practice Management systems added features to their platforms that allow doctors to open up their schedules and generate links that can be embedded into websites or other applications. For the vendors, this is a way to charge the provider for another feature (as always). For the providers, this is often the easiest way to get more patients into the door without having to add another tool to their stack.

• Patient Engagement Platforms: Most EHR direct scheduling solutions are quite primitive and don’t address all the challenges described earlier. That’s why many point solutions emerged that try to offer a better scheduling experience for patients while electronifying the gatekeeping functions of a front desk scheduler. These SaaS or license-based tools are very common for practices that care about their patient experience and brand, such as digital-first health companies and concierge-like care providers.

• Health System Platforms: While similar to patient engagement platforms, organizing the schedules of a health system in order is another beast. These companies specialize in the overwhelming complexity of large health systems and integrated delivery networks. The value proposition of these platforms is (1) improving the utilization by standardizing appointment types across organizations and (2) reducing patient leakage by making schedules available within the organization.

Opportunities and Outlook - our “Hot Takes”

The recap and restatement of the status quo are all fine and good, but what does this mean for the future? Whether by the will of the government or the relentless but capricious forces of the free market, will we see easier appointments we all desire? What businesses will be built and where will legacy parts of the industry crumble to dust? Don’t worry, fam, we got you.

With the current trajectory, the quest for direct scheduling seems on the precipice of massive second-order regulatory tailwinds. The CMS Prior Authorization API Rule and accompanying ONC regulation seem to be coming at some point (detailed in Brendan’s article). Barring massive changes, none of this actually addresses scheduling head-on, but electronic prior authorization would mean one of the biggest barriers to easy appointments may fall. Additionally, EHR API requirements in Cures do not call out scheduling, but the shift to the FHIR economy has already resulted in some of the top EHRs creating availability APIs, a feature never before available in the HL7 world. We’d like to see the industry converge their approaches here, though, as many EHRs are creating different FHIR-based direct scheduling implementations.

This will enable quite a few new business opportunities which new entrants and existing companies are racing to capture. A first observation is that health system platforms and patient engagement tools will pivot into acting as “First Mile Aggregators”. Health system platforms and patient engagement tools already do the hard, thankless groundwork of getting a practice or health system’s schedule in order and automating their intake, and triage process. We expect most of these players to shift and develop or open up their APIs for other companies to integrate with. This will allow them to not only monetize through a license fee with their providers as they do today but to resell capacity/availability to other businesses (more on that in a second).

This trend is already happening today, as we see NexHealth, whose initial wedge was delivering patient engagement tools to dentists and small practices, but has created an API program and repositioned itself as an integration solution for dental applications like SmileDirectClub and Quip. Kyruus seemingly also is in the midst of this transformation, opening up the schedules of their customers to aggregators such as Google. This is a great way for these companies (where Business Agreements allow) to upsell the work they’ve already done, so we predict that more will be making the move towards becoming a first-mile aggregator and exposing their scheduling capabilities.

We think this shift will be challenging for a lot of vendors. It’s a truly different core competency to create an API and the surrounding developer ecosystem than it is to create SaaS tooling for providers. It’s also an entirely different sales muscle - will the existing sales team with their Rolodex filled with health systems and practices be equipped to actually sell to digital health vendors, payers, or other businesses? Regardless, the health system platforms and engagement tools that can pull this off will skyrocket in their value to the overall ecosystem.

So if more schedules are accessible, who will be picking up this capacity? Let’s first look at the consumer-facing companies. The first, of course, is Google. After so many failed attempts at breaking into health care in areas that were firmly outside their core competencies, have they finally found a wedge into the field? We think this fits the model of success they’ve found in other industries and they have a playbook called the Google Squeeze. In their first stage, given that they are uniquely positioned in their large pre-existing monetization of attention through ads, they will be willing to pay providers & first-mile aggregators to list their schedules on their platform. Once established as the to-go place for finding doctors and the first touchpoint of the patient relationship, their playbook entails that they may look to cut out the first mile aggregators. Eventually, they could start charging doctors for patient leads (or at least be in a position to drive up provider advertising prices). That possible future entails true patient steering power that most healthcare payers can only dream of.

ZocDoc is another candidate to aggregate the first-mile aggregators for consumers, with a huge head start as one of the first digital health companies and a large pre-existing provider footprint. However, their current business model (charging providers pay-per-lead) might not work as well with Google as a competitor that sits at their top-of-funnel on the demand side. The strange and twisted beauty of ZocDoc is that it tries to live in many worlds: building first-mile scheduling integrations, existing as a consumer-facing brand, and now selling to other businesses via API. Can one company actually be everything everywhere all at once to all people?

There are also many interesting B2B use cases for integrating with these first-mile aggregators:

1. Provider data aggregators already sit closely in the workflow preceding scheduling. It seems logical that they might take that next step and attempt to upsell their existing customer base by buying or developing this capability.

2. As noted earlier, payers and care navigators are quite interested in surfacing scheduling and availability information through their patient portals. This could improve patient steerage to the providers they deem more cost-effective while reducing their ongoing linear investment of new schedulers down to the less expensive, fixed cost of an electronic solution.

3. Digital health companies are another logical candidate, as they want to deliver a great consumer/patient experience. To that extent, direct scheduling and availability of information provide a more seamless experience when they have to refer patients out to in-person doctors.

A world built on the foundation of direct scheduling has a variety of exciting derivative business opportunities:

• End-to-end referral collaboration: True interoperability is not just about patient data transfer and chucking a CDA over the fence to your neighboring physician. It’s the radical enablement of doctors in different organizations to work together on a patient’s care in novel ways unthinkable in the pre-digital era. However, we have thrown in so many stones in form of the administrative burden of prior authorizations, fax machines, and call centers that this path of collaboration is nigh unimaginable. Thus, we sit in the era of Digital Health 1.0 digitizing legacy patterns and processes, but direct scheduling is a critical piece in the leapfrog moment to our next era: one of effortless provider collaboration. It may take a witch’s stew of provider directory vendors, direct scheduling solutions, patient data interoperability tools, task management platforms, prior-auth & eligibility APIs, payment rails, and patient communication solutions, but we believe this frothy mix will converge into an end-to-end referral tool that will cause us to question the simple software we’ve built in previous eras.

• Digital Care Navigators: If direct scheduling will be a thing, care navigators that are selling to employers could be in a tough spot. Their current operating model is pretty much based on call centers that research providers and schedule appointments for their members. Direct scheduling will give rise to fully digital and automated care navigators. Maybe that is what Comcast had in mind when dumping Accolade earlier this month... or maybe there was another reason, but this is one sector of the industry that faces massive, irrefutable disruption as nationwide availability comes into being.

Conclusion

Simply put, this is an exciting time to be alive. Market and regulatory trends point to a rosier future than the decades that have passed and digital health is ascendant. If you’re working in this space or found this article interesting, we’d love to chat with you and hear more - you can reach us on Twitter or by replying to this email (if you’re subscribed). And if you’re a provider, EHR vendor, or government regulator, we all have just one simple ask of you:

Let us in.

If you want to learn more about this topic, try:

Sandy Varatharaja’s take on scheduling challenges

Mark Olschesky’s overview of scheduling as part of the interoperability problems

Julie Yoo’s talk about the fragmentation of health care schedules

Here are some logistics: 

• To be able to answer, you need to be subscribed to this newsletter - just reply with your answer to the original newsletter or the signup email! 

• Please let me know in your response whether you want to be mentioned or stay anonymous (and share any Twitter/ LinkedIn for reference)

So let's get into the topic: 

How can independent doctors stay independent? And should they?

Let's first look at some facts. In the last 20 years, more and more doctors have given up their private practice and sold their businesses:

• According to the Physicians Advocacy Institute, almost 70% of all physicians are employed by hospitals or corporate entities. Corporate entities include investor/private-equity-backed organizations as well as payer organizations.

• This trend has been going on for a while - with every year about 2% of all doctors (!) switching from small practices to being employed by hospitals or corporate entities. The COVID pandemic only accelerated this trend.

• In 2020 the AMA provided an interesting breakdown of independent physicians by specialty, and one can see that there are quite some differences.

• There are also some regional differences. The Southern US has a greater proportion of independent doctors than the Midwest or North-East. However, the South has seen the fastest acceleration in the consolidation of private practices.

It is striking how fast the landscape of private practice ownership is changing. Several factors are driving the growth of hospital and corporate employed physicians.

Reasons why doctors shift from private to corporate

New market entrants

Post pandemic, there are more employment opportunities for doctors than ever before. It is not just private practices vs. hospital systems. Doctors can now work for publicly-traded primary care chains (Oak Street, Iora, etc.), choose one of the many retail clinics like MinuteClinic from CVS or Walmart or work from home doing telehealth. All these organizations compete for doctors, which is even more challenging during a time when medical professionals are very hard to hire. Many newly graduated doctors may find these settings more attractive than the traditional private practice route.

Acquisitions

Another big driver of the shift away from private practice ownwership is the pace at which hospitals & corporate entities are aggressively making acquisition offers. Often, this is not a hard sell - here are a few (anecdotal) reasons why private practice owners are ready to give up their independence:

• Increasing overhead: The overhead of running a private practice has increased by 40% over the last 20 years. One great example is the burden utilization management for certain drugs put on doctors' offices. You can read more about it here. Billing insurance has become more complex, and there is more and more admin work involved with prior authorization requirements, claims billing, and claims follow-up management. Additionally, practice management technology is quite expensive. An Epic implementation can cost $350k (unverified number from Twitter) for a small practice. Managing these non-clinical activities and expenditures often pushed doctors to hand these responsbilities off to a third party - they may take a pay cut but they will be able to focus more on clinical work vs. admin.

• Patient flow is drying up: The lifeblood of every doctor's office is patients coming in for visits and procedures. The primary patient acquisition channels for doctors are direct marketing and referrals from other doctors. In recent years, especially the latter channel has changed quite a lot. 1) Hospitals have been actively buying up independent practices to steer members to the hospital facilities, and they implemented sophisticated referral management solutions to keep the patients in their system. 2) Insurers have been more aggressive in their network design, and "narrow networks" have become very common. Smaller practices might have a more challenging time becoming part of them. 3) Retail clinics and new upstart care providers (virtual and hybrid) can potentially provide a better patient experience. They allow patients to combine a grocery store visit with their annual wellness exam (Walmart, CVS) or even come to the patient's house (Amazon Care). You have to contrast these trends with the fact that there are still long wait times for some specialty doctors and primary care doctors. Let me know which factors I am missing here...

• Tempting buyout offers: Doctors sell their practice because it can often be a really good deal. If you've run a business for 30 years, would like to take a step back, and don't have someone to hand over your practice (and patients) to, why say no to a $2-5 million buyout offer? It could be the ticket for early retirement or at least a life as a (worse-paid but less stressed) employee.

Who is buying & Why?

Acquestition decisions are not only driven by doctor preferences, but there are even stronger incentives for the buyers to make the private practice part of their own organization. They can offer these high pay-out offers because they think they can create even higher economic benefits from the deal. Three main buyer groups exist.

Payers

Value-based care is seen as an effective way to reduce the total cost of care, as it aligns incentives between the physician and the payer. However, value-based care arrangements are complicated: benchmarks have to be calculated, providers have to be convinced that they can realize savings, and new technology has to be rolled out.

So why go through all that hassle if you could make them part of one organization? That is what the large payers are focusing on these days. Optum (United Health Group's vehicle for everything non-insurance related) now employs 50,000 physicians, which is 5% of all doctors in the US. And they are planning to add 10,000 more to their organization. Check out Kevin & Ryan's take on the UHG acquisition strategy (paywalled).

It is an interesting approach, as payers can become an integrated delivery network like Intermountain or Kaiser, the poster-children for cost-efficient health care. On the one hand, it reduces the number of intermediaries. It aligns incentives, but I am also afraid of the market power that the big payer groups exhibit already, and the efficiency gains might not be distributed to the premium payers but to the stock holders.

Hospitals

The Stark Law & Anti-kickback statute prevents any provider from sending Medicare beneficiaries to a provider they have a financial relationship with. The idea behind these laws is that doctors should not follow financial interests but what is best for their patients when making a referral. However, hospitals found a simple way to circumvent these laws: Instead of painfully convincing providers to send their patients, make them part of your organization by buying their practice. Then make sure they only refer within the hospital system. Over the last years, health systems across the US have bought primary care and specialty care practices and instructed the new employees to refer their patients only to hospital-owned facilities.

Besides gaining access to referrals, hospital systems gained more negotiating power when setting rates with insurance companies by increasing their market share. In many markets, plans are forced to contract with large health systems to get adequate coverage for complex care (like oncology or trauma care). However, they would also need to accept other hospital-owned providers to be in-network.

Investor-backed provider groups (Private Equity and others)

Investor-backed provider groups are on the rise as well. The motivation from investors (public and private equity-backed) firms to acquire practices and create practice chains is similar to the hospitals' - get negotiating power for better rates with insurance through a larger footprint. However, there are also some additional motivations:

• Efficiency gains: A lot of processes in health care a broken (mainly on the admin side), and through applying the right process frameworks, technology stack, and use of data, corporate chains can offer better patient experience, better patient outcomes, and more cost-efficient care. The rise of outpatient surgery centers is a great example. In general, they offer significantly lower prices for standard surgical procedures. Another example are investor-backed value-based care organizations, which are promising to better care for chronically ill patients, such as Oak Street or Cano.

• Consolidated Overhead: A lot of the overhead for doctor practices can be centralized and thus made cheaper per patient, for example, implement a modern billing and payment stack, improve patient marketing strategies or invest in a centralized patient communication platform.

• Utilization: However, there is also a darker side to investor-backed provider groups, who may put profit over patient and payer interest. There are several examples where PE-backed companies pressured doctors to steer patients to higher-cost procedures and drugs (i.e., in oncology). Also costs cutting might go to far and get to a level where patient safety can no longer be guaranteed.

To learn more about private equity moving into private practices, definitely check out Olivia's takes on PE and Ambulances and PE and Fertility Care and this Bloomberg article about Dermatology.

So, how can independent physicians stay independent? And should they?

Let’s first look at what are the actionable steps an independent doctors needs to take to stay competitive:

• Learn from investor-backed provider groups, but stay ethical: The last part of this sentence is probably not necessary because one of the main arguments for many doctors to NOT sell their practice is that they care about their patients and don't want their quality of care go down. However, there are many things that practices can learn from the PE playbook. Adopt modern technology for more efficient processes on the admin side, invest in the proper marketing and patient acquisition channels, and modernize the patient engagement stack.

• Find new channels to acquire patients: If old referral paths change, practices need to react. Some are finding new interesting ways to stay afloat. One example are direct primary care provider and cash-pay specialists. Avoiding health insurance can dramatically reduce overhead as no billing staff or certified (= expensive) EHR systems are required. Other practices join ACOs to reap the benefits of value-based care arrangements. Here are a few examples of companies offering solutions. Another way for practices to find new patients is to partner with digital care companies - they often need in-person services, and I think there is potential for fruitful collaborations. Last but not least, digital channels for acquiring patients are becoming more critical. Setting up a good experience for patients to find a doctor online and get in touch is expected these days. 

• Opportunity for new products to enable independent practice owners: I think there is a lot of white space in helping independent doctors with the challenges of running a practice business. However, changing systems and processes is not easy, and I am wondering how many independents will give in and accept a buyout offer. Nevertheless, I predict we will see several new entrants in this space over the next 1-2 years that will offer better, cheaper, and more modern technology solutions. If you hear of any innovative and exciting companies in this space, please let me know. I might cover the modern independent practice tech stack in a future post.

I will not go into the “should they?” question for now. I think there is a lot to be said about the effects of market consolidation on overall health care costs and long-term quality. Also, I heard many doctors are unhappy about their decision selling their practice. But I will leave room here for other people to share their experiences and thoughts. I am looking forward to your replies & publishing them in a next article!

But realizing this potential is not an easy feat. Data is spread across different source systems and organizations, and raw data is usually not actionable. Thus, a whole ecosystem of companies has emerged to make health care data usable, and today I will look at the business archetypes that exist in this space. I will discuss their general business model, how they build economic moats, the weaknesses of their business models, and which growth strategies might work for them.

Here are a few examples of companies in these different archetypes:

Data Source Systems: Owning the Inputs

In the beginning, there was a data point - every entry in a database will need to be produced somewhere. Either by a sensor, by a manual entry, or by a system writing into another system. In most organizations, data is entered and produced by operational systems such as an ERP system or the EHR in the healthcare context.

The best examples of healthcare data source systems are EHR vendors and practice management systems. Device companies also generate raw data. Examples are WHOOP & Levels, on the lifestyle product side, and medical equipment manufacturers, like oxymeters or other diagnostic devices. Claims and financial systems are the most prevalent data sources on the payer side.

Business Model & Challenges

Data source systems usually have a straightforward business model; either they pursue a SaaS strategy or an enterprise license model. They strive to become the system of record, which will make them very sticky, and it will become almost impossible to migrate away from the solution (as it will most likely take five years and will burn through at least two CIOs).

Two main challenges exist for data source system vendors: First, their solutions depend significantly on their sales force. Because migration cycles are long, there is no perfect competition between all solutions at any time. Knowing when a customer is looking for a specific solution and being present with the right relationships and arguments is critical for the success of these enterprise deals. Second, though, many data source systems are not very differentiated. At its core, many operating systems are just databases with forms attached to them. Of course, they differentiate on how well they match their user interface to the workflow needs, but in the end, many of these systems are quite interchangeable (in theory). 

Growth Strategies

Once you are the system of record, a logical expansion opportunity is to upsell new modules on top of your data model. As the data system provider knows the software the best, it is not too hard for them to build modules on top of their data model that seamlessly integrates with the existing system of record. They can also open up their source system and allow third-party providers to build applications on top of their platform. This offers two advantages: making them more sticky with their customers & financial upside via revenue shares.

Data source systems usually have some level of control over accessing the data. This control could be direct (they control access to the database or IP rights to the data schema and decide who can use the data or not) or indirect (they own the knowledge on how to make the data usable). Depending on their control over the data, they can sell access to the data and make it available to other organizations. This control of the data generation process is so essential that other data business archetypes build out their own data capture systems. The best example is Flatiron acquiring an EMR vendor - but more on that later.

Data Routers: Make Data Available for Use Cases

No organization exists in a vacuum, and organizations need to interact and transact with each other. Sharing data here is critical, and a whole group of companies allows organizations to collaborate by enabling data exchange. Some of these "data routers" just help with secure data transmission; others allow more complex workflows that include authorization or settlement mechanisms.

In general, data routers are agnostic about the data content - they apply some normalization, but usually, they don't apply any business logic to the data. Great non-healthcare examples include (the always mentioned) Plaid, which helps consumer FinTech Apps link to financial institutions, or Sabre and Amadeus, which have access to hotel and airline inventory and connect this to travel agents and booking platforms. In health care, various data routing businesses exist: Clearinghouses such as Change Healthcare or Komodo, interoperability providers such as Particle and Health Gorilla, and Payer-to-Provider data exchanges like Flexpa and 1UpHealth.

Business Model & Challenges

Data routers can elicit considerable network effects - if they are connected to a majority of players in a market, they can become the de-facto standard for how data is exchanged and how organizations work together on a specific use case. A great example here is Surescripts, which owns the prescription workflow. 95% of all pharmacies and 70% of physicians use Surescripts to send and receive prescriptions. However, becoming this standard can also have adverse side effects, as once established, the data router might have little incentives to innovate. They might even be disincentivized to innovate as it might break certain connections. This dynamic raises the opportunity for "on-ramp" businesses that allow for easier access to these older standards and basically "innovate the user experience". If you want to learn more about these businesses in health care, definitely read this post from Brendan Keeler.

Data routers also have several weaknesses. First, their standard is subject to disruption. If the industry comes together and establishes criteria for peer-to-peer data exchange, data transmitters become obsolete (or much less powerful). Unfortunately, these standards have not prevailed in US health care or are driven by private/ semi-private organizations with their own interests (like Epic, Direct Trust, and The Sequoia Project). Regulators could probably play a huge role in forcing people to adopt a single standard. Another dynamic could be innovations from Web3. A significant promise of Web3 is that crypto will replace a lot of the data transmission networks by using trusted protocols and cutting out the intermediaries.

Growth Strategies

The first goal for data routers should be achieving network dominance, which means having a sufficiently broad coverage among all organizations. This is easier said than done in health care, as the first 50-60% is often easy, as some standards exist in source systems. But it gets incredibly tough when we get to the long tail of home-grown data sources. 

A second strategy can be to enable more complex collaboration workflows. Currently, many interoperability companies are pretty simple in their use case: send one file from one party to another. However, similar to payment networks, they have the opportunities to build more complex workflows in real-time. Examples here are claims pre-adjudication, automated prior authorizations, or referral workflows.

Another growth route is adding an analytics layer on top of the routed data. By definition, data routers are very agnostic about their data content. Still, because they touch a lot of data going through their network, they are in a great position to add analytics layers on top of their services, like flagging outliers, transforming and normalizing certain clinical concepts, and calculating benchmarks. Adding an analytics layer on top would push them into the next bucket of data businesses.

Data Refiners: Turn Data into Insights and Actions!

Raw data is often not actionable, so organizations need to turn it into valuable insights and build operational workflows around it. This usually involves several steps:

• Integrating data from different sources

• Building out a data model

• Calculating metrics and predictive measures

• Serving it to frontend applications

I am not going into detail here, as there are whole books about the proper data architecture organizations should employ and which software vendors they should use for each function. Well-known players in this space in health care are Innovaccer, Clarify, Arcadia, and Health Catalyst.

Business model & weakness

Most data refiners have a license model, i.e., they provide access to their software for a license fee (which can be per user seat or an enterprise license). These vendors will be more sticky the more central they are to the core operational workflows of an organization or when they are embedded with the IT architecture of the organization. This is very similar to the data source system vendors. But there are other strategies they can employ to get an edge: data refiners often deliver encoded domain knowledge. For example, some vendors provide rules engines that can detect uncoded risk scores or care gaps. Other examples include pre-trained AI models or specific workflow frontends that codify industry best practices. This is why there are so many healthcare-specific data platforms. Data refiners are usually competing with two groups: point solution providers, that solve one specific problem really well (for example Arcadia’s risk adjustment module competes with Apixio’s risk adjustment AI) and in-house IT departments, who want to build workflows themselves using generic builder tools such as AWS and Google Cloud.

Growth Strategy

Similar to the data source systems, the foremost expansion opportunity for these companies is to expand the number of data use cases their platform supports. Ideally, they would like to extend from data consumption workflows, i.e., where users look at a dashboard or get a particular metric, to operational workflows, i.e., where users not only read data but also write data back into the platform. Achieving this allows the data refiner to become a data source system, which makes them more sticky than if they are just an analytics solution. 

Another strategy for data refiners is to become a data aggregator and not only reuse insights and best practices, but build an aggregated data asset. Let's look at this next.

Data Aggregators: Bring Data Together across Organizations

Data aggregators are a particular case of data refiners. While most data refiners work within the boundaries of one organization, data aggregators bring data together from different sources across organizations. Usually, data aggregators don't own the data sources, but they can be super powerful if they do.

Once the data aggregators obtain the data from different parties, they prepare it, integrate it, and sell it to other organizations. They can sell access to the aggregated dataset on a record per-record basis (which can be identified or deidentified), or they can only sell the analytics, like benchmarks or trained models. There are many prominent examples from the non-healthcare world, including credit agencies, payroll benchmark providers like Pave or Levels.io, and consumer research companies like Nielsen (here is a great article describing data aggregator businesses in more detail). Data aggregators are also prevalent in health care, and several high-valued companies are built on this concept. Examples are Ribbon Health (for provider data), Flatiron (for cancer patient data), UpToDate (for medical guidelines), and Apple Health (for personal health and wellness data). 

Business Model & Weakness

Data aggregators don't work in all markets. They usually work well if there are high data acquisition barriers:

• Distributed data: Data needs to be distributed among different organizations for data aggregators to work. The mechanism here is that the more difficult it is for a single organization to collect and integrate the data, the more defensible the business will be. 

• Data is changing relatively frequently: Data does not only need to be distributed among different organizations but can also be distributed over time. Data aggregators will work well if the information needs to be constantly refreshed and recollected.

• Data value decays over time: This recollection dynamic is further supported if data loses value over time. For example, my credit score from 10 years might still have some meaning, but a recent score is much better for a lender.

• Incentives for data owners to share the data: Data aggregators can only exist if they can access the source system. Sometimes the source systems are "open," which means data aggregators can scrape or download the data without any agreements with the data owners. They can also be "proprietary", which means the data aggregator needs to establish an agreement with the data owner. Crafting the right incentives for an organization to share the data is critical here. Incentives can range from monetary compensation, access to the combined data asset/benchmark, access to specific software tools, and avoiding regulatory fines.

• Contracting friction: For proprietary data sources, data aggregators usually need to set up a data-sharing agreement with the data owner. Negotiating the terms around what and how data is shared takes time and effort. The larger the organization and the more sensitive the data, the longer the contracting cycles.

If data aggregators reach enough scale, they can become quite defensible. The number of agreements with data source organizations becomes challenging to replicate, especially if the barriers to recreating the aggregated data asset are high. Replicating the data connections and sharing agreements might become too burdensome for a single data user or a potential competitor.

The pinnacle of defensibility is reached if a company can become the source of truth for specific data points. If they become the de-facto registry for data points, data owners will want to update and provide accurate data to the registry. Google Maps is an excellent non-healthcare example: business owners want to make sure their address and business information is correct and recent so that people can find them. Google does not need to lift a finger for this or provide any compensation (they can even charge for it). 

Another interesting approach to data aggregators is open source repositories, where organizations or individuals voluntarily share their knowledge or data. Nikhil from Out-of-pocket started an interesting thread on this topic: 

Examples of open-source data aggregators could include a list of digital health builder tools or customers for value-based care enablement services (It is not so easy to find the right contact person for an ACOs). If everyone would benefit from the combined data, organizing some of these resources as open source makes sense. However, there are also "common good" dynamics at play. Why should I do the work and contribute when I could just wait and let others do the job. 

A significant weakness of data aggregator businesses is losing access to their data source. As they are often not the data owner, they rely on data source partners to build their services, or if they use web scrapers, they rely on a UI that risks constant modification. If there is a concentration of data sources among just a few organizations, this can become a risk. Another challenge is that the aggregators do not control data quality and how data is captured without owning the data source. That's one of the main reasons Flatiron went into the EHR business and acquired Altos Solutions. The last weakness is that data acquisition barriers could fall. For example, the ONC rules about FHIR interoperability or Payer-to-Patient data will require payers to make patient data available via a modern API. This will make it easier to pull data from certain organizations, which can undermine the business of companies that rely on these exclusive relationships and data acquisition barriers. 

Growth Strategy

Data aggregators can create a precious data asset, which might be uniquely positioned to solve particular business problems. Instead of just selling access to the data, they could opt into offering the solution that is powered by their data asset. For example, Google never sells consumer profiles, but they let people run targeted ads using their data. 

Thoughts on Data businesses

Here are some of my thoughts on these business models:

• The power of data ownership: In all of the business models, the data source owner is probably the most powerful stakeholder in the value chain. You can compare this to the oil market dynamic. If oil prices go up, the production companies benefit the most, not the pipeline or refinery owners. Because the data source owner can limit the data flow, they have better negotiating power to squeeze the benefits from the intermediaries (data routers & data aggregators). Data routers and data aggregators can avoid this squeeze by diversifying their data sources.

• Privacy: No article about data businesses should not mention privacy. Especially when handling patient data, privacy and data security need to be taken very seriously. Luckily we have the correct legal frameworks to share data in a secure and compliant way (The P in HIPAA stands for "Portability"). Unfortunately, many organizations are too afraid and use HIPAA more as an information blocking tool than to enable collaboration. 

• Data business vs. workflow business: A common theme among all these businesses is whether they should "just do the data work" or expand into higher-level workflows. Both approaches have their advantages and disadvantages. Staying close to the data and spending time making the data capture, routing, refinement, and aggregation better give a lot of focus to a business, and it can be a winning strategy. However, workflow businesses can capture more value and thus warrant higher prices. 

• Blending the models: The archetypes are often not as clear-cut as I present them here. Data refiners are trying to become data aggregators; data transmitters are trying to become refiners; data source owners are becoming refiners. If you're able to combine the different business models into one, you will build the next "Google of health care", however, I don't see this happening soon, as there is too much competition in each field - and this is probably a good thing.

Overall I believe data businesses will have a bright future in health care. Collaboration within and across organizations is rising, and data is crucial to enabling better outcomes and more efficient care. In particular, I see more opportunities in the payer-provider data exchange for value-based care organizations and more opinionated workflows regarding care coordination & specialty referrals. Let me know if you're building something interesting in this space (my Twitter DM is open!).

But when looking a bit deeper into these numbers, I had some doubts: The a16z article included Oscar and Devoted health, who are health plans, not actual care providers. Also, if you check out all startups tagged with health care on Crunchbase, you will get closer to the 3000 number. However, those companies also include biotech, wellness, and other health IT startups. So I was wondering: how many digital care providers are actually out there. In this article, I will take you on my research tour and share some of the results with you. You can also find the complete list of startups I found on my newly launched website Digital Care Directory.

What is a digital care provider?

To start, I pulled together all my prior research and made a list of every company I have come across. This list was probably highly biased towards my network and media consumption, so I decided to list all data sources containing digital health providers. I collected VC portfolio company lists, filtered Crunchbase, scrubbed the YC list, obtained customer lists of my friend's infrastructure startups, and went through slack channel postings. 

It wasn't easy to distinguish actual providers from lifestyle and other health care adjacent companies. So I put forth the following criteria to limit my search and to define a digital care provider:

• Patient-facing brand: The company needs to hold a relationship with the patient. I excluded pure white-label solutions but included marketplaces and provider networks that act as the digital front door for the patients.

• Virtual-first: The provider needs to pursue a virtual-first approach, and utilize digital channels to support the whole patient journey. I also accepted Hybrid approaches that combine in-person care with virtual care. However, I did not include legacy providers that tacked on telehealth to their practice without changing their care model. 

• Clinical benefits & operations: The company needs to show clinical value and improve health outcomes. I excluded pure lifestyle products or products with unproven clinical value. While the provider does not necessarily need to employ clinical staff, their product must be embedded in a clinical process and work alongside certified clinical care providers.  

So after sifting through all these databases and tagging the providers - here are some of my insights!

Breakdown of the numbers

Overall I found about 350 companies that would qualify as digital care providers. I was able to tag 150 so far, so the data is still quite biased, but here are some preliminary results:

Breakdown by specialties

Primary Care, Mental Health, and Women's Health make up the largest three categories. Note that Mental health, Women's Health, and Substance Use Disorder are specialties that traditionally have inadequate patient access. These categories offer great opportunities for virtual care to fill a gap in the market. Primary care is in the top position (as expected), as many primary care services can easily be delivered online. But care is not only happening online...

Breakdown by Delivery Type

Most digital care companies adopt a virtual-only approach - about 75% of the sample. However, a sizable chunk is doing hybrid care, and some are even doing most of their care in-person. The in-person companies here are mainly delivering care at home.

Breakdown by Reimbursement Model

Interestingly, only about a third of all providers in the sample are cash pay only. Many providers are also taking insurance or are reimbursed in some other form (as employer benefit, Medicare reimbursable, or provider-sponsored). I would expect the cash-only part of the pie to shrink over time, as early stage businesses will probably move beyond cash pay as they look to expand their revenue models.

Again - please take these results with a grain or better a pile of salt. My tagging is not the most accurate, and there is still a substantial amount of providers left to tag. I will write up my updated results once this analysis is done.

A closer look at the provider categories

After looking at hundreds of digital care websites, it was interesting to see several repeat patterns and business models across the different digital care providers. Here are some common types of digital care providers:

• Access to Prescription Drugs: This model is designed to increase access to prescription drugs. A prominent example here is Ro and Hims&Hers with their skincare, ED, and hair loss prescription drugs. Prescription drug models have also expanded in other areas such as psychiatry and substance abuse disorder. One has to be very careful, though, about the incentives for the digital care provider. Do they make money on the drugs and just act as "pill mills," i.e., they will prescribe as many drugs as possible? Or do they care for the patient more holistically and are outcome-focused? The latter providers will also offer other delivery models and choose prescription models only when appropriate. These providers will also supplement any medication therapy with other forms of care.

• Primary Care through virtual care: Primary care is ripe for a virtual care model, especially for low acuity visits. For many situations, a 5 min virtual visit and mail-order prescription might be sufficient to help the patient. I really like how texting and video blend in these models, and how data can be used to more efficiently triage patients and coordinate the proper care. The more data these virtual primary care providers can gather, the better they will become in triage and personalized medical experiences.

• Well-scoped specialty care for a specific condition: This is probably the type of business I am most excited about - these digital care providers usually have a very narrow diagnosis focus and a well-scoped treatment model. Examples are sublingual allergy care programs or virtual treatment for polycystic ovary syndrome.

• Chronic care management programs: These are usually not staffed by doctors but by health coaches and other trained staff. Those types of companies work as a supplement to existing care and make it more efficient. Some include education materials delivered through a virtual program or app, a patient community, and regular check-ins with a health coach or care coordinator.

• Virtual Provider Networks: Certain specialties have traditionally been challenging to access. Either because the market is very fragmented or these specialties are generally not covered by insurance (often a result of a fragmented market). Two examples here are maternity / post-partem care and mental health therapists. Several companies are building platforms that allow patients to access providers and schedule appointments with ease. In many cases, these services can also be delivered online.

These are just some examples that caught my attention, and there are many more business models. I probably also missed quite a few original and creative approaches here. Let me know what else should be included here.

Making it a Digital Care Directory

After doing this research (primarily for myself, out of curiosity), I realized that this might be beneficial for the digital health community, so I decided to put this on a website. You can now find the first 150 providers at Digital care.Directory! Here are a few more reasons why I decided to put it on a website:

• Digital health is pushing into new applications & they become more differentiated: More and more differentiated specialist solutions are entering the digital health market, focusing on specific conditions and treatments. Digital care is not only virtual consults but spans remote monitoring, digital programs, and at-home visits. When a patient or a doctor is seeking a particular program for enrollment, it is easy to get lost. Thus there should probably be a space where doctors and their patients can narrow down the number of providers to select the most appropriate program. I don't expect this directory to be a "match-maker" that helps patients find the most appropriate care - that can only decide a doctor with all the relevant context, but it might open up new options...

• Population & identify specific focus: Digital health providers not only differ in their diagnosis and treatment scope, but they also focus on different populations. Research shows that taking into account the identity and social background of a patient can lead to a stronger provider-patient relationship and improve health outcomes. Again, knowing about these options can help patients & doctors get better outcomes.

• Rising acquisition costs: While more differentiation opens up new options for patients, the flip side of more differentiation is higher customer acquisition costs for each provider as they compete for the attention of these patients. A centralized directory might solve this at least a bit, as a virtual cardiologist focusing on the elderly is probably not competing with a hormone therapy care provider targeting LGBTQ+ patients. However, both might bid for the same marketing spots and drive up ad prices. Additionally, digital care providers are likely competing less against each other and more against traditional providers. This directory could drive general awareness about the possibilities of digital care.

The digital care directory is still very much a prototype, and I am planning to add a few features and more providers to the website (there are still 200 left to be added). Please share the provider directory and ask your digital care provider founder friends to add themselves to the website if you want to support this project! Also, don't shy back from any feedback. I am always interested to learn what you think...

Value-based care requires a complete rethinking of how we deliver care. And because rethinking is never easy, many provider organizations are still hesitant to take on risk. They often fear they do not have the tools and capabilities in place to reduce cost while maintaining high-quality outcomes. But luckily, there is a whole industry of companies emerging, helping provider organizations adopt value-based care models.

It is a confusing space, though! The scope of vendors are not clear cut, and their websites are full of marketing promises - technology blends with consulting, companies saying they are an end-to-end solution but then also partner with other vendors... It's pretty convoluted. So in today's post, I want to structure this market a bit and look at the different areas where vendors are trying to help providers adopt value-based payment models. As a rough structure, I will be looking at the following areas:

• Admin & Contracting support: How to get started?

• Data & Analytics: What to do?

• Provider Enablement: Get the provider to do it!

• Patient Engagement: Get the patient to do it!

Here is an overview of players in this market. 

This map should not be a complete representation of all players in the market (there are so many) but should list a few example companies for each category. Most of the companies here also blend into other parts of the value-based care stack, but I assigned them to their strongest category. If I missed a company, please reach out, and I will include them in the picture.

Getting into Contract: Admin & Contracting Entities

Value-based payment models are a prolific breed, so it makes sense to first look at the different types of models. This is probably the best report on the current state of VBPM adoption, and they provide a great categorization of value-based payment models:

• Quality Bonuses: This is basically a payout to the provider if they hit certain quality metrics, such as readmission rates or percentage of patients that came in for an annual wellness exam.

• Claim-based (upside, upside & downside): in these models, every provider will be measured against a cost & utilization benchmark, usually calculated using historical claims data. They can be designed as upside-only, which means that the provider will get a share of the savings they have realized but won't have to pay out any money if they miss the benchmark. For upside & downside models, the provider will have to pay out money if they did not meet the benchmark.

• Capitation-based: Capitation models take value-based care one step further, and providers will get a lump sum of money every month for every member attributed to their organization. These capitation rates are usually risk-adjusted, meaning the provider will get a higher monthly rate for sicker members than for healthy members.

Each of payment program will most likely only cover a subset of the population the provider sees. So the provider will need to choose which programs are worth joining. CMS alone lists more than 11 value-based care programs for Medicare. For Medicaid, the situation becomes even more complex as it is administered on a state level, and every state defines different value-based payment programs. As value-based care is not a "one-man-show", providers will need to collaborate in order to realize better outcomes. Therefore value-based payment models often require providers to set up a dedicated contracting entity, e.g., an Accountable Care Organization, where a group of providers can collaborate and share their savings. A whole group of companies is supporting providers to become part of these VBPMs and help with the legal and administrative work involved to get into contract. One can categorize these companies by the programs they are targeting. Here are a few examples:

• Medicare Direct Contracting: Agilon & Pearl health

• Medicare Shared Savings Program: Lumeris, Evolent, Aledade & Caravan

• Medicaid VBPM Programs: Yuvo Health

These admin & contracting enablers primarily target smaller practices and clinics, as these providers often don't have the legal resources. Interestingly, some of these providers are reusing the established contracting entities to participate in different value-based care contracts. For example, an Aledade ACO might participate in the Medicare Shared Savings Program and also get in a value-based care program with a Medicare Advantage plan and potentially even a commercial plan. Although admittedly, VBPM adoption in the commercial space has been relatively sluggish. 

What should I do now? Data, Analytics & Opportunities

Once a provider is under a contract, the provider will need to work on realizing more cost-efficient care. A major factor here is good data analytics that guides concrete actions a provider can take to improve care quality by reducing costs. Here are a few examples of how data is used to inform steps that can improve outcomes:

• Reduce utilization through care coordination & integration: This is probably the most crucial area to improve care. This includes ensuring chronic diseases are well managed, avoiding duplicative care, and appropriate drug utilization. Another example is the post-acute care workflow. To avoid readmissions and fast recovery, the VBC provider needs to address any underlying reasons for hospital admissions. In addition, follow-up care needs to be delivered, and patient context needs to be shared between the hospital, specialists, nursing facilities, and the primary care doctor.

• Preventative care & quality metrics: Avoiding costly acute episodes (think emergency room & hospital) and focusing on preventive care can reduce costs in the long term. To also provide short-term incentives for preventative care, many VBPMs come with quality metrics that will give a bonus to the provider if they can meet these requirements. These quality metrics are also crucial for health plans, as they affect their star rating. Medicare Advantage plans and individual market plans use these ratings as marketing tools to attract more members. Thus, they are even more willing to incentivize providers to improve their quality scores. [At least once incentives are aligned...]

• Cost-efficient referrals: Routing patients to the most cost-efficient specialists and facilities is another area for realizing value. Unfortunately, cost and quality are usually not correlated in US health care, so making wise decisions in the referral process can significantly affect overall costs. For example, sending a patient to a high-quality imaging center instead of a hospital can bring their radiology cost down from $4000 down to $800.

• Risk Coding & Reporting: This is less on the cost side than the revenue side. If a provider is in a capitation program, they will need to report the risk scores of their members. Being diligent in this process has a significant impact on revenue. You can read more about how this works in the risk adjustment article.

All these areas will require robust data sources and a strong data infrastructure to handle the analytics.

Data Sources & Acquisition

Bulls*** in, bulls*** out! This is a truism for any data analysis. So the first step for successful VBC analytics is getting access to the right data sources. Traditionally health care data has been a bit of a nightmare when it comes to accessibility. Even within an organization, not even talking about getting external data sources. However, several vendors are now helping to get this step done for the provider organization. Let's go through the main data sources:

• Internal EHR data: This is the apparent data source any VBC provider should start with. Vendors such as Redox can make this data available so providers can push it into population analytics platforms.

• External EHR data: However, in order to get a complete view of a patient, it is important to obtain data from other provider organizations as well. This can either be done via health data exchanges, but there are also vendors like Datavant (who acquired Ciox) that allow for chart retrieval from various organizations.

• Real-time admission & discharge data: Vendor chart data can have a time lag. Thus, some vendors specialized in providing near real-time data for specific events using ADT (admission, discharge, transfer) data feeds from hospital EHRs. Bamboo health (who acquired PatientPing) is an example here. This data is essential for timely care coordination.

• Payer data: Payers have a lot of interesting data, too. Claims are probably the most prominent data source, but health plans can also provide interesting analytics around suspected diagnosis, risk gaps, and population health. While claims are don't offer as much "depth" as medical records, they can provide more "breadth" as they can give a complete historical picture of what happened across care organizations. Claims might also indicate which data from other provider organizations need to be obtained. 1Up.health is a vendor here, facilitating this data exchange.

• Provider benchmarks: Provider benchmark data is even broader than payer data, as a good benchmark draws data from different payers. This data gives indications about which providers are cost-effective and have a high quality of care. Several organizations have specialized in providing this type of information and aggregating it across payers. Examples here include CareJourney and Ribbon Health.

• Other sources: New data sources are emerging for value-based care use cases. A great example here is data sources around social determinants of health - information about educational background, food security, access to transportation, etc. - will be much more important moving forward. The adoption of social determinant metrics is also driven by the new CDC regulations around the direct contracting program now called REACH (a value-based payment model in Medicare).

Data integration & preparation

Getting access to all the data sources is just the first step in turning data into actionable insights. The data needs to be turned into a usable format in a second step. Several challenges in preparing the data exist here - some examples include:

• Technical Barriers: Data might come in very cryptic formats. I've heard of one provider organization that would get some payer data in a corrupted 2001 Excel file that could not be opened on any computer they had. The plan did not have the resources to update the file, so the provider had to build an extractor that could handle this file.

• Logic Variation: Even if data comes in formats that can be processed, data comes in different data schemas. Examples here are patient identifiers that don't match across records or business logic that varies across EHR files. The FHIR data model should alleviate some of these paint points here. Still, even in a world of a single data schema, organizations might fill the schema with different data assumptions.

• Time Lag: Data can have a significant time lag. This is especially true for claims data, which can take several months to settle. Therefore time gaps might need to be closed using projections or extrapolations from incomplete data. 

• Noise: Most health care data is high noise and low signal data. A lot of the information in medical records is boilerplate code, and there are instances where the EHR can contain 20,000 (!) record pages for a single patient. Filtering out the signal from the noise is a challenging task.

Because of these challenges, value-based providers need to invest significant resources into their data preparation and integration stack. Some vendors helping with data integration are Innovaccer, Arcadia, Ursa and Edifacs.

Analysis & Continuous Learning

Once the data is integrated into a usable data schema, the interesting part can begin: Turning data into action. For some use cases, the data can be used directly for decision-making. Having a clear metric can drive a lot of decisions. For example, a risk-adjusted cost metric could decide which providers should be preferred for referrals. Other use cases require more complex modeling and analysis. Here are two examples:

• Population stratification: In this use case, the provider creates a risk model for their patients to predict specific health outcomes. They can then use this model to assign patients into different risk groups. These risk groups can then help assign them to certain special programs to prevent negative health outcomes.

• Intervention evaluation: We still early in value-based care, and there are a lot of interesting approaches providers can take to improve the health of their patients, like home visits, remote patient monitoring programs, and leveraging digital health providers. However, these interventions need to be constantly evaluated, and using the available data is a critical use case here.

A whole group of software vendors, consulting firms, and actuarial firms help providers with their risk modeling and data analytics. Examples include Clarify, Milliman, and many of the larger consulting firms. Also, the major EHR vendors have their population health analytics tools that give providers actionable insights. 

My Thoughts on this space...

Actually, I will keep them for the next part. Part two will cover how different companies help turn the insights into actions by supporting providers on the operation side. You can read it here.

Two years into the pandemic, Zoom fatigue is real. That's why I am even more impressed if so many people show up for an online "mini-conference" and bring their unique perspectives to a fruitful discussion. After the specialist referral discussion last month, we had another online session with the health tech nerds community about risk adjustment in value-based care". The participants had a wide range of backgrounds. Value-based care providers, health plans, actuaries, and a former regulator joined the discussion and shared their thoughts on the topic. In today's post, I will summarize some of the main insights from our conversation. 

As these "mini-conferences" are so insightful and exciting, I plan to organize them more regularly. If you're interested in participating and getting the invites, please sign up here. Before jumping into the article below, if you're new to risk adjustment, you can read some of the basics in this article covering the basics of risk adjustment.

Introduction to Risk Adjustment

Daena Bell and Erica Rode from Milliman kicked off our discussion with an introduction to risk modeling - Milliman is a global actuarial and consulting firm with quite some expertise in risk adjustment. It's worth looking a bit at the terminology - at its core, a risk model is a predictive model that assigns a risk score to each individual. The risk model can be used to predict various factors, such as utilization, the likelihood of acute events, or the total cost of care. Risk Adjustment is the process of using these risk scores for a specific use case. Here are some of the most common applications:

• Payment arrangements: Risk scores are used to compensate organizations based on their members' risks. This finds application in Medicare Advantage Organizations, ACA Marketplace plans, ACOs performance payments, and other commercial value-based payment models.

• Care management: Risk scores can be used to identify members that are likely to experience adverse events - providers and health plans can use these prediction models to design specific interventions for these groups. This is also known as risk stratification.

• Quality measurement: To accurately measure a provider's performance, risk scores can be used to remove some of the bias in the patient data. Certain providers see sicker patients, so this needs to be considered when evaluating their outcomes.

Given the advancements in data science over the last decade, the world of risk models is also subject to change. The workhorse for actuaries and risk modelers remains to be the linear regression. The advantage of this model is that it can be easily interpreted and works well on smaller datasets. However, risk adjusters are exploring more complex models such as decision trees and neural networks. These models might be better suited to detect non-linear relationships between input variables and predict outcomes more accurately. But these complex models can also be computationally intense to train and need large data as input. Furthermore, the parameters of these models are harder to interpret.

An important thing to consider for risk model calculation is biased data. If the population used to train the risk model is substantially different from those used for the prediction, the predictions would be biased and less accurate. We discussed the bias in the next section.

Critical voices about the current risk adjustment system

The risk adjustment system put forward by CMS for Medicare Advantage and Direct Contracting has been under quite some criticism lately. Read this article in health affairs if you want to learn more about the controversy. We also had a few critical voices in the group:

• CMS-HCC model disease prevalence: There is a potential considerable statistical bias in the risk model that CMS applies, which is only trained on data from fee-for-service Medicare but does not include encounter data from Medicare Advantage. Because fee-for-service Medicare providers don't have incentives to be very careful in their risk coding, the model might underestimate the prevalence of a particular condition. This bias is sometimes called coding intensity bias. A solution for this could be to include MA encounter data into the CMS risk model.

• Models can underestimates the risk for underrepresented groups: Risk models can only be as good as the underlying data - groups without access to health care or with little health data overall might not be reflected properly in the government risk model. This means that the compensation for conditions in certain marginalized groups might be too low. This can in turn strengthen the bias in the data, if Medicare Advantage Organizations are less financially incentivized to care for these groups and capture their data. An idea to fix this is to take non-medical data into account. We talked more about this during the social determinants of health part.

• Adverse incentives: Risk scores for different diseases may encourage providers to do diagnostic tests that are not necessary. Doctors might be inclined to do a diagnostic test if a particular diagnosis is higher compensated than others, even if the clinical guidelines do not deem it necessary. [I guess we will never get incentives right in health care...]

Some interesting insights related to this criticism came from a former CMS regulator. CMS is aware of some of the issues related to risk adjustment, and they are trying to improve the methodology for risk adjustment. When setting these reimbursement policies, they need to balance the reporting burden with the reporting accuracy. Also, when deciding their risk model, it is not just about predicting costs but also about pushing certain policies.

Social determinants of health in the risk model

Moving on, we discussed how social determinants of health could play a role in risk adjustment or at least in the risk stratification process. Quite a few people are currently working on Social Determinants of Health (SDoH) in health care these days. SDoHs consider the conditions in the environments where people live that affect their health. Factors include economic stability, education, health care access such as access to transportation and availability of doctors, social factors, and environmental pollution. In theory, providers and health plans could use these data points to predict better whether a patient is at risk and design targeted interventions. The group brought up a few challenges here:

• Data Collection: There are no established methodologies to collect SDoH data. A common source is surveys via phone, mail, or at the doctor's office. However, these surveys are rarely standardized and are difficult to connect to other survey results. Another potential source of data is publicly available data on a community level, for example, ZIP-level income information, air quality data, school district performance data, or occupational information. For example, an asthma patient in Montana might be less at risk than in downtown Los Angeles. But all this data is not easy to handle at scale, as it comes in different file formats and has varying scope and methods for calculating their metrics.

• Turning data into action: Even if data is available and tied back to a patient, it is still unclear how providers can use it for their clinical decision-making. Doctors are generally not trained to take SDoH into account for treatment decisions. Just giving them the raw data will not achieve much, even if it can predict certain conditions. Therefore, SDoH data needs to be analyzed, and specific interventions must be designed, such as access to care programs (like paying for a taxi to and from the doctor's appointment) or food security programs. A great example brought up during the discussion was to work with community institutions such as churches to drive awareness about medication adherence.

• Include SDoH into the Risk Adjustment Reimbursement Model: There are thoughts to include the SDoH into the CMS reimbursement models. However, it isn't easy to accomplish. In particular, there are no standards (yet) for capturing SDoH data, and thus they are difficult to codify and include in a nationwide risk model. Also, the administrative burden of requiring physicians to collect SDoH data for their practices might not outweigh the benefits of more accurate risk adjustment. However, it might be a possible backdoor to push an industry-wide standard for SDoH data capture, expanding their use. 

Operational challenges with risk gaps

As risk scores play an essential role in value-based care compensation, we continued our discussion by covering the operational challenges of capturing and reporting risk factors. Jay Srivastava from Cityblock shared his views on what a successful risk documentation program would look like. 

A critical effort of many stakeholders in the risk adjustment process is to surface risk gaps and suggestions at the point of care. A risk gap means that the patient is likely to fall into a particular risk category but currently is missing the compliant documentation required by CMS. A great example for a risk gap, is if the patient had a chronic condition last year, but it has not been coded this year. Managed care organizations would ideally like to inform the providers about open risk gaps at the point of care so they can provide the appropriate diagnosis documentation. This in turn gets the managed care organization or value-based organization the full risk-adjusted reimbursement for their member, which can be shared with the providers or reinvested if savings are generated. However, this puts an added burden on the provider, and to make this program successful, the following points should be considered:

• Motivation and Incentives: As risk adjustment may put some coding burden on the practice, they need the right incentives and motivation to implement the proper reporting processes. For example, if a practice only has 10% of their patients in value-based payment arrangements, there is little incentive to pay much attention to risk reporting. However, suppose the share is closer to 90% or there exist incentive programs from payers that pay bonuses on reporting accuracy, the practice will be much more inclined to pay attention to their risk adjustment program. A critical factor for success is buy-in from the practice management and leadership. They need to understand the mechanics of their risk adjustment program and impact on revenue, so they will sponsor implementing the right processes and commit resources.

• Select the right tools & integrate them into the clinicians' workflows: There are a variety of vendors providing tools to help doctors identify and close risk gaps. The idea is to bring risk model outputs to the provider at the point of care so that the provider can act on the data. We discussed a few features that make these tools successful. First, the data needs to be well embedded into the clinical workflow, i.e., the EMR, and it should not come as a stand-alone Excel file sent monthly via email. The provider should be able to act on the data in as few clicks as possible, and it's imperative that they can trust the data. Some discussion participants shared that instead of giving doctors black-box risk scores and prompts, doctors would be much more likely to act on data if they get specific supporting evidence, such as references to past claims or medications.

• Education & Training: Even if there is motivation at the practice management level, they must communicate the relevance of risk coding to the attending physicians. Health plans and value-based care providers need to invest in good risk adjustment training. People shared that dedicated training & practice sessions on properly using risk adjustment support tools can greatly improve their adoption.

• Clinical Value: In the end, the risk adjustment model should not just be used for reimbursement purposes but also to guide patient outcomes. A good risk model can identify patients who need dedicated outreach, are at risk of acute episodes or are skipping their care. While the risk gap and the care gap, i.e., the gap between needed care and actual care, are not always overlapping, they are often related, and risk gaps can help identify gaps of care.

If you wanted to sell stuff online in 2006, you had to bring quite some technical knowledge to the table. You had to know how to set up a scalable web server, build a front-end, and cobble together different e-commerce & payment solutions tools. A group of friends running a snowboard shop in Canada didn't like any of the tools out there for their online shop, so they started to build their own shop system. 15 years later, their e-commerce platform is a $100bn company with 1m+ businesses using their platform, and yeah you guessed it, it's called Shopify.

In 2022, digital health is similar - setting up a digital health practice requires quite some technical knowledge. The situation is not as dire as for e-commerce 15 years ago, as many great stand-alone solutions are emerging and embraced by digital health startups. I've written previously about some of the infrastructure solutions out there:

• APIs for insurance billing and payment processing,

• telehealth platforms and white-labeled provider networks,

• on-demand services for various parts of the care delivery chain,

• cloud-based EHRs, digital front doors, etc., etc.

But with these ever-expanding options of point solutions, there is a need to combine these point solutions into a seamless patient and doctor experience. Digital health startups often have to hire developer teams to do the plumbing between solutions, or they fall back to write data from one system to another manually. This integration step might slow startups down significantly in a world where time to market is critical. And if they ignore the integration task, the manual swivel-chairing might prevent their business from scaling to more users than their manual processes can support. So, who will build the platform that will enable digital health startups to go to market faster and scale more easily with their number of patients?

What does Shopify do?  

When I asked Twitter, "Who is building Shopify for digital health?" people had many different opinions on which companies would fall into that category.

To set the stage, let's clarify the different dimensions that a "Shopify for digital health" would need to cover:

• Setting up a front-end for patients and providers: Shopify is probably best known for allowing non-technical users to set up a nicely designed online store. They can upload photos of their products, set prices, and quickly set up a checkout experience that accepts most payment types. The equivalent for health care would be a platform that provides an easy way to set up provider scheduling, patient intake, and other patient-facing journeys. The key here is that the digital provider could brand their patient experience without hiring a full-stack developer team.

• Taking care of operational & clinical workflows: But Shopify is not only a no-code front-end builder. It also takes care of many backend services, such as order fulfillment and invoicing automation. Some of these tools are Shopify's own services and building blocks, but Shopify also partners with a range of third-party integrations. For health care, many of these third-party infrastructure (insurance billing, credentialing, etc.) and on-demand service (labs, home visits, etc.) service providers already exist and would need to have easy integration into the platform.

• Integrate point solutions into workflows: Quite a few workflows need to span multiple point solutions in health care. For example, setting up proper referral management would be a combination of e-consult services, making a referral decision, transferring patient data, and closing the loop with the referred provider after the visit. Point solutions exist for each step (i.e., RubiconMD, Ribbon Health, Health Gorilla), but they will need to be put together into an end-to-end workflow. The same can be said about revenue cycle management (i.e., insurance billing), which often involves working with several point solutions and clinical workflows.

• Enable workflows between organizations: The holy grail of health care is reaching true interoperability and building cross-organizational workflows, like discharge management and other care coordination tasks. For this to work, organizations must speak the same language, i.e., use the same data schema and exchange protocols. Achieving this is not an easy task. This usually succeeds if an established organization or the government pushes a certain standard. A great (non-medical) example is Skywise, an aviation data platform that Airbus developed and promoted throughout the airline industry. With 100+ airlines on the platform, Skywise became the de-facto standard for data exchange between airlines, Airbus, and manufacturers.

With these requirements in mind, we can look at the current market of companies playing in the "Shopify for digital health" category. There were a few debates about where to put each company on the chart, and I have to admit that the lines are not as clear cut as they seem in this chart, but this is my current assessment:

In my opinion, no market player is currently offering a solution for all functions Shopify provides, but many are seeking to. But health care is also different than e-commerce, so here are a few potential futures on how these integration platforms will evolve.

Future 1: (Closed) Full-Stack Solutions

The first scenario is less of a future than a description of the present. Traditional health care technology providers have chosen a very closed approach. Companies like Epic or Kareo provide a suite of tools that integrate well with their primary system. These legacy players also make it easy for organizations to collaborate if they use the same system. However, there are quite some restrictions on what type of integrations they allow. Also, these vendors usually don't serve the requirements of modern health care organizations and digital health startups. Still, Epic & co are so ingrained in the hospital sector that I don't see this change anytime soon.

I don't think this future is very desirable, as innovators are at the will of the closed platforms. Startups will have to give up a significant revenue share, and they have to accept long time-to-market when their offers have to go through inefficient approval processes with the legacy vendors. I heard that integration with App Orchard from Epic could take up to 6 months.

Future 2: Frontend RPA & Backend RPA

Because most legacy systems are closed and don't have a great way to integrate via API, many organizations fell back to a very clunky but effective technology called Robotic Process Automation. RPA is a click-bot that can help automate processes from one front-end into another. The advantage of this solution to bring together systems is that it almost works with any interface. Still, the significant disadvantage is that it is not scalable at all - anytime there is a slight change in the UI, the bot might break and needs to be updated. Olive has identified that generalist RPA companies such as Automation Anywhere and UiPath don't have much health care integration knowledge, so they became the market leader for health care RPA.

Automation gets more interesting when applied in the backend, so it is not subject to frequent front-end changes. A great non-healthcare example here is Zapier. It allows users to automate tasks between different systems. Zapier does not currently sign a BAA, and thus, it is not HIPAA compliant and can't be used for patient data. Also, clinical workflows need to be 100% reliable, and there should be alerts if backend automation is failing - something that Zapier currently does not focus on. Examples of companies trying to be the Zapier for digital health are Tellescope and Phase Zero. Another exciting player here is Awell from Europe, offering a visual interface to orchestrate different health care APIs. Olive is also planning to enter this space and migrate RPA-Bots to more seamless integration solutions that are more maintainable and can be stand-alone products.

Future 3: EHR as the integration platform

The EHR is traditionally the point where all clinical information is captured, and it is the workhorse for the provider. However, the EHR does not need to be restricted just for a subset of clinical and reimbursement workflows. The EHR could expand its functionality beyond the traditional system functions and include CRM functionalities and more operational workflows like supply chain and staffing. Brendan wrote a great take on this future with his headless EHR piece.

Modern EHRs are morphing into practice management systems. We could already observe this happening with athenahealth. Canvas is an exciting new player in this market, allowing easy backend workflow integration with their EHR platform. For example, when a provider orders a lab, Canvas enables them to easily connect their order via API to a home-health service provider like Axle or Workpath. It is easy to imagine that EHR platforms can include CRM functions that track non-clinical touchpoints with the patients or take care of other practice management functions. 

Jonathan Bush's Zus is expanding this EHR platform vision. The Zus EHR platform not only enables workflows within provider organizations but also wants to allow workflows across organizations. 

Future 4: Orchestrators & Wrapped Solutions

Moving from a legacy EHR to a more modern EHR platform that expands its functionality might still not work for every provider. They might want to choose something more flexible and not be locked in with one vendor that takes care of all their workflows. These providers might look for modular workflow solutions and then choose a platform to orchestrate all these modules. In this world, the EHR will be one module among others and not the integration platform.

Combining infrastructure pieces into wrapped solutions

This future opens up the possibility for wrapped solutions, which I define as modules that provide utilities into plug-and-play end-to-end solutions. An example of an wrapped solution would be an end-to-end insurance billing solution, which would cover insurance enrollment, claims coding, claims submission, and payment processing. On the back end, this solution would probably leverage a few other API utilities, such as Change Healthcare, Eligible, and Stripe. Wrapped solutions open some exciting opportunities for a range of new analytics products, which have faced many barriers for easy adoption in health care.

Orchestrators

Wrapped solutions are one way to combine different infrastructure services into plug-and-play solutions. But then these solutions also need to be connected with other solutions. This is where solution agnostic orchestrators come into play. A good orchestration framework provides a way to mix-and-match different solutions into one framework to seamlessly manage a digital health practice. Orchestration frameworks differ from back-end automation. They impose a data schema and connect applications more tightly - they are not point-to-point automation solutions but work like a bus everything connects to. The closest company building this out is Capable Health. They provide a framework and data model that enables providers to combine different out-of-the-box solutions. For example, they could use IntakeQ as a patient intake form, Square as payment solution, and Elation as EHR, and then put it all together using the Capable platform.

Closing thoughts

As always, here are some closing thoughts:

• Who would use the Shopify for digital health?: The main value proposition of Shopify is to enable SMBs to launch their own e-commerce shop. I am not sure whether we will see this anytime soon for brick-and-mortar practices as, unlike retailers, they don't struggle with falling demand. It's quite the opposite, as staffing shortages and physician burnout strain doctors' supply. On the other hand, about 7000 digital health startups are pushing into the market, and all of them will need infrastructure to build their business on.

• Is integration on top of mind for digital health providers?: When digital health companies start, they probably do not care much about integration. In a world where speed-to-market and iteration velocity is critical, they might be perfectly fine with juggling several point solutions. Other more pressing issues, like getting labs to work or becoming part of a health plan network, are more important for digital health to solve right now. Once digital health companies reach maturity, this will become more relevant, as integration and automation are essential for scaling.

• Is building on the Shopify for digital health a defensible strategy?: Keep in mind that many digital health companies strive for tech valuations - for this to be true, they will need to develop defensible and scalable tech and migrate away from any platform solutions. If they don't, they will essentially be a tech-enabled services business that scales linearly with the number of physicians they employ. If the Shopify for digital health becomes real, digital health providers might have the same fate as many e-commerce shops: they might be good businesses but rarely unicorns. Let me know in the comments or reach out if you agree or disagree with this assessment!

• Wrapped Solutions - the future?: Instead of integrating everything, the next evolution step for digital health infrastructure might be wrapped solutions, i.e., end-to-end workflows that help the provider solve a specific problem. We will probably see more M&A between companies that offer point solutions, as we have already seen with Ro buying Kit.com & Workpath. However, one concern with wrapped solutions is that they might not be very defensible. If they rely a lot on other providers, they have a lot of points of failure, and if they work, one of their suppliers might move up the stack and starts competing with them.