Flip on CNBC right now and you’ll hear pundits cheering on an absolutely frothtastic capital market. As I write this, Dell (!) is up 35% on the day, Anthropic just raised one of the largest private investment rounds in human history, and SpaceX is poised to deliver a payday so large it could transform the entire venture capital industry.

At the same time, consumer loan delinquency rates are hitting their highest levels since the ‘08 crisis, new grads face an impregnable job market, and Americans are so stretched that CPG executives are warning that customers are “literally running out of money.”

Maybe you’ve heard… we’re deep in the K-hole. Elbow-patched academics have pulled off a Patagonian Toothfish-level rebrand of the “wealth gap.”

This divergence between high- and low-income consumers is reverberating across industries - healthcare included.

For a perfect example, consider two competing headlines from this week.

Axios published an analysis showing that most seniors cannot afford long-term care (assisted living, nursing home care), even though an estimated 70% will need it at some point.

Contrast that with an article from the Wall Street Journal on the emerging “med-à-terre” trend: wealthy seniors buying a second home near their preferred doctors.

Parisa Afkhami, a New York-based real-estate agent at Coldwell Banker Warburg, recently helped a client buy a Manhattan med-à-terre for close to $5.5 million. The international buyer has homes in Dubai and London but plans to visit the city for checkups twice a year.

Source: WSJ

Affluent individuals are increasingly becoming healthcare “super-consumers,” deploying discretionary income to escape the traditional system. A glossy ecosystem has emerged to serve this demand - premium longevity and concierge clinics, Prenuvo scans, Function Health panels, high-end aesthetics treatments.

Meanwhile, low-income individuals are being increasingly squeezed out of the traditional healthcare system. Medicaid cuts, ACA subsidy reductions, and the steady creep of higher premiums and deductibles are forcing consumers out of insurance-intermediated care. Some early innovation has emerged to offer low-cost access, like discount pharmacy offerings and D2C telehealth. But it hasn’t organized as deliberately to satisfy the growing demand. Instead, individuals pushed out of the system are more likely to engage in care avoidance rather than care direction. Facing an outlook of chronic clinical undersupply, it’s hard to imagine care access improving until regulators adopt a more flexible posture towards AI-based prescribing platforms.

Earlier this month, Utah released initial YTD data from its pilot with Doctronic, where patients could renew existing prescriptions with Doctronic’s AI. The early results are promising - when the AI recommended prescription renewal for a patient, the reviewing physician agreed 91% of the time.

As care deserts expand across the country, it’s hard to ignore the promise of AI to provide some access to care when the alternative is no care at all.

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

“Then in my heart I wanted to embrace the spirit of my mother. She was dead, and I did not know how. Three times I tried, longing to touch her. But three times her ghost flew from my arms, like shadows or like dreams. Sharp pain pierced deeper in me as I cried…” - Homer, The Odyssey (tr. Emily Wilson)

Last year, a dead guy testified against his killer.

Chris Pelkey was shot in a 2021 road rage incident in Chandler, AZ. At the time of the shooter’s sentencing, Pelkey’s family presented a digital twin of Chris generated with AI to provide an impact statement.

For the first time, the deceased had the last word. The judge loved it.

“I loved that AI, thank you for that. As angry as you are, as justifiably angry as the family is, I heard the forgiveness,” Judge Lang said. “I feel that that was genuine.”

Source: BBC

The shooter was sentenced to ten years for manslaughter.

A couple of years ago, I wrote about emerging innovations in end-of-life care: advance care planning platforms, technology-enabled hospice, marketplaces to help families navigate the logistical burdens of bereavement.

I was optimistic then that these advances could elevate the standard of care. But I never imagined that technology would soon fundamentally alter not just how we die, but what it means to be gone.

A new wave of “griefbots” is emerging. AI-powered digital twins that promise to satisfy one of the most primal, consistent pains of life - the desire of the living to access the dead. It’s fascinating, alarming, and probably unstoppable.

Digital Necromancy

Griefbots are a derivative of the emerging category of digital twins - virtual chatbots or avatars trained on a real person to closely approximate them in live conversation.

You may recall when Reid Hoffman famously interviewed “himself” in early 2024, representing an energizing, unnerving new frontier for generative AI.

Within months, people would move on from using AI to create facsimiles of the living to reanimating the dead.

There is the case of Joaquin Oliver, a victim of the 2018 mass shooting at Marjory Stoneman Douglas High School in Parkland, FL. Oliver’s parents created a digital twin of their son, conducting interviews with journalists to offer his perspective on gun violence.

There’s Shotline, a website using AI to recreate the voices of gun violence victims. Users can send these calls to their Congressional representatives.

And soon, the griefbots showed up. Apps like StoryFile, HereAfter AI, and You, Only Virtual let users create customized bots to preserve and converse with the deceased. A vivid illustration is offered by a trailer from an app called 2Wai founded by *checks notes* former Austin & Ally star Calum Worthy.

We’re getting a glimpse into a future where dying doesn’t mean you’re “gone.” That could have real implications on the healthcare economy.

A Virtual Afterlife

If we can mute the pain of loss, could it change how we approach end-of-life care?

It is well documented that as individuals approach the end of their lives, their chief concern is not living longer. It’s comfort - for themselves and for the ones they leave behind.

Despite this, many patients undergo extensive interventions designed to prolong their lives. For example, we know that 10% of Medicare recipients had surgery in the final week of life. And we know that hospice is underutilized, particularly among non-White patients.

So if people say that they want to prioritize comfort, why are they empirically not doing that? One major obstacle to enrolling in hospice is the idea of “being ready.”

Some patients and family members remarked that they didn’t enroll because the patient was “not ready” for hospice… family members spoke of how enrolling in hospice meant acknowledging their loved one was dying. One widow commented, “[Hospice meant] admitting we were dying; we were trying to live.”

Source: “Why Don’t Patients Enroll in Hospice? Can We Do Anything About It?” Journal of General Internal Medicine.

The emotional gravity of navigating palliative care can be suffocating for loved ones. What it means to “be ready” lives at the intersection of medical, spiritual and personal considerations. If it’s possible to reduce the grief burden on those we leave behind, would it be easier to “be ready?” I have to think the answer, for many, is yes.

Then if griefbots help more patients and their families confidently choose palliative care pathways, what would the implications be?

There’s some data that suggests employers lose billions in labor productivity from grieving employees. Could we see employers offer griefbots to support their workforce through bereavement? See you at work on Monday. Terminal end-of-life care can cost health plans hundreds of thousands. Imagine a health plan calling to say “hey instead of this expensive surgery for your loved one, what about a digital twin?” Maybe we’ve just discovered the next great prior authorization gate.

Black Mirror of Erised

Let’s assume griefbots live up to their grand promise of diminishing the pain of loss. That could be the most dangerous outcome of all.

We already have cautionary tales of self-destructive relationships with chatbots. A man who was convinced by Gemini to commit suicide to transcend and join it as a digital being. Another who died on the way to “meet” his AI girlfriend in NYC. A VC fund manager unraveling into what observers called “AI psychosis”.

There’s a fascinating study from Anthropic drawing on its own conversation data on the incidence of “disempowerment” - cases where users exhibit signs of unhealthy reliance or distorted thinking in their interactions with Claude. The global average rate is small: severe disempowerment appeared in approximately 1 in 1,300 conversations. Still concerning given the absolute number implied by Claude’s scale. But that global average is diluted by domain - personal conversations carry significantly higher risk of disempowerment than say writing code.

But the study goes on to examine several “amplifying factors” - authority projection, reliance & dependency, vulnerability, and attachment. These amplifying factors map alarmingly well against grief. Individuals experiencing grief are likely experiencing severe levels of each of these amplifying factors all at once.

Users exhibiting these amplifying factors have dramatically higher rates of disempowerment. For example, individuals with moderate to severe authority projection - ranging from “you know better than me” to “I submit to your wisdom completely” - have distortion rates approaching 50%.

In short, grief is like jet fuel for the amplifying factors of AI psychosis. If people are forming unhealthy or even dangerous relationships from zero with “stranger” chatbots, what would happen when it’s not Claude, but Mom? Dad? A lost spouse? The proliferation of griefbots risks putting a powerful technology with an addicting form factor into the hands of some of the most vulnerable users.

Like Shadows, Like Dreams

There’s a moving scene in The Pitt where a pair of siblings struggles with the decision to honor their father’s wishes to terminate treatment. At a loss for what to do next, Dr. Robby (king 😮‍💨) takes them through a forgiveness ritual to ease them through the initial stage of grief.

It illustrates how grief is among the most personal experiences one can have, profoundly unique in how it manifests. And yet, it is also among the most universal.

The griefbots will get better - this is the worst the technology will ever be. As that happens, it will be hard for the grieving to avoid their call. How do you stop someone from trying to satisfy the primordial longing for those we’ve lost?

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

Indeed, one of the most fascinating little pockets of the healthcare economy is the market for non-human primates (NHPs) for clinical research. NHP prices are sky rocketing, and upon close examination, we can see that this market is a microcosm of multiple macrotrends reshaping the US economy:

1. The fragility of a globalized supply chain reliant on foreign imports

2. A shifting global balance of power in critical innovation sectors

3. A race to leverage technology (and yes, AI!) as a strategic off-ramp from foreign dependency

To accompany this post, I’ve created an interactive timeline tracing the evolution of NHP usage in clinical research - check it out here.

In Vivo Veritas

Before commencing clinical trials on humans, pharmaceutical companies typically conduct tests in two animal species to assess safety and toxicity.

Preclinical safety testing typically includes studies in one rodent and one non-rodent species. The choice of species often depends on how closely the animal’s biology mirrors our own. For example, NHPs are commonly used in biologics research where a close approximation to humans on genetics, anatomy, physiology, and immunology is vital.

Biologics are large, complex, and difficult to characterize molecules created through multistep biologic processes... These new treatments for cancer, autoimmune disease, heart disease, infectious disease, post-transplant organ rejection, and genetic disorders depend upon animals closely related to [humans] … Therefore, it is only currently possible to assess the safety and efficacy for biologics intended for human patients in other (non-human) primates.

Source: Charles River Laboratories

NHP research spans diverse therapeutic areas, each category utilizing various subsets of primates (with cynomolgus macaques and rhesus macaques being the most common).

At first glance, the scale of NHPs in clinical research seems small - of the 14 million vertebrate animals used in research in the US, NHPs represent just 0.5%, of which private industry accounts for ~50% of the usage. But NHPs have an outsized impact on clinical development - of the >17,000 individual products in the US R&D pipeline, an estimated 56% of them have used or will require the use of NHPs during their development.

NHP markets have been tightly indexed to overall investment in scientific research and biopharma development. When industry funding and research activity increases, demand for NHPs climbs. When funding dries up, demand falls. However, supply can't pivot in lockstep with demand. Breeding colonies take 5-10 years to establish, reproductive cycles take 12-18 months, there is typically 1 offspring per cycle, and it can take years for NHPs to reach sufficient maturity for research use. Demand moves in financial quarters; supply is stuck on biological time.

This structural mismatch creates significant baseline vulnerability to volatile price swings. But over several decades, a death by a thousand cuts would make those swings increasingly extreme.

The Blurst of Times

Monkeys have been used in clinical research for over a century. By the turn of the 20th century, they were already becoming central to polio research. In 1908, Karl Landsteiner and Erwin Popper demonstrated that monkeys could model human polio transmission via spinal cord injection from a child who died of the disease. The following year, several groups successfully demonstrated monkey-to-monkey transmission of polio. From the onset, research with NHPs exhibited both the promise and the burdens of the practice:

Simon Flexner had infected monkeys with polio, but researchers in the 1920s were unsuccessful infecting rabbits, mice, guinea pigs, dogs, cats, pigs, chickens, sheep, goats, and calves. Until Charles Armstrong succeeded in infecting cotton rats in the late 1930s, monkeys were the only laboratory animals that polio researchers could use. Researchers’ reliance on monkeys to conduct polio research substantially increased the cost; monkeys averaged about $8 each through the 1930s, and each monkey could be used only once.” John R. Paul observed that “the great majority of universities and private research institutions were unable to furnish the expensive facilities, technical assistance, and the supply of monkeys which were so necessary at that time for poliomyelitis research.

Source: “Basil O’Connor, the National Foundation for Infantile Paralysis and the Reorganization of Polio Research in the United States, 1935–41.” Journal of the History of Medicine and Allied Sciences

Researchers today might lol at that price point (~$150 today adjusted for inflation), but the cost burden of NHP research was clearly an obstacle even then. Nonetheless, without an alternative, imports climbed steadily through the 1920s. By 1931, the trade had grown large enough to appear in annual government reports for the first time. Within a decade, the number of rhesus monkeys imported for research would triple.

Persistent demand for Indian rhesus monkeys was driven largely by a significant cost advantage (~$10/monkey from India vs. say $75/monkey from Puerto Rico). However, supply was precarious. Monkeys hold deep religious significance for Hindus in India, so any perceived mistreatment could threaten to choke off supply - a fragility that would soon surface. As the 1947 Partition of India unfolded, American researchers found themselves contending with an odd predicament:

[via State Department inquiries]: “There is a possibility of an interruption in the supply of monkeys from the United Provinces because the trappers are all Moslems and there has been a large movement of Moslems out of the province in consequence of the communal riots in and around Delhi. The influx of Hindus into the area from Pakistan has contributed to the confusion and has served to make the Moslem trappers nervous about the continuation of their activities because of the Hindu veneration for the monkey.”

Source: “Between Simians and Cell Lines: Rhesus Monkeys, Polio Research, and the Geopolitics of Tissue Culture” by Tara Suri

It wouldn’t be the last time that religious concerns threatened the supply of research primates.

Still, demand kept climbing. In the late 1940s, a large-scale polio typing effort required some 20,000 monkeys to confirm that there were three strains of the virus. Jonas Salk’s vaccine breakthrough further accelerated demand, as monkey kidney tissue was essential to its production. By 1955, the US moved to formalize its import/export relationship with India to secure access to a supply chain that American science had come to depend on.

“T. T. Krishnamachari, Commerce and Industry Minister, told Parliament last week that the United States was likely to release polio vaccine by the middle of this year and for this purpose 100,000 monkeys were required … the monkeys cost $35 each… the monkeys are sacrificed after twenty-eight days.”

Source: New York Times - April 14, 1955

The agreement had an interesting caveat: the Surgeon General would have to certify that imported monkeys would be used solely for medical research or polio vaccine production - explicitly prohibiting their use in atomic research.

Oops! It turns out they were.

When reporting in the 1970s revealed that the Air Force had been testing the effects of radiation exposure on monkeys over the preceding decades, activists like Shirley McGreal brought this to the attention of Indian authorities. Exports of monkeys to the US were swiftly banned in 1978.

Meanwhile, domestic production of NHPs was still in its infancy. Congressional appropriations established what would become the National Primate Research Centers in the early 1960s, but funding was directed mostly to research activities rather than large-scale breeding. So when India closed the door, the US had no meaningful domestic alternative to fall back on.

At this point, the NHP market bifurcated. Academics continued to rely on rhesus monkeys but struggled to source supply as NIH-funded breeding centers slowly ramped up production. Meanwhile, the biopharma industry pivoted to cynomolgus macaques from Indonesia, the Philippines, and Mauritius. These nations filled the vacuum left by India, but only temporarily. They would soon be eclipsed by a behemoth.

Crouching Tiger, Hidden Monkey

In 1985, China established breeding programs for cynomolgus monkeys and began exporting in 1990. Within a decade, China was producing tens of thousands of macaques for export per year.

China also emerged as the default logistics hub for exports of NHPs bred elsewhere. In 2001, approx. 7,000 macaques of Indonesian and Vietnamese origin were shipped to the US via China, as conservationist pressure had forced most airlines to stop transporting primates; Chinese carriers stepped into the gap.

With inherent cost advantages and superior logistics, China soon became the dominant exporter of research macaques.

As it consolidated its grip on NHP supply, China’s goals around clinical research were far more ambitious. From the 2000s onward, China made a deliberate push to become a global player in biotechnology innovation. But 2015 marked a genuine inflection point. China adopted a sweeping set of reforms to supercharge its domestic biotech industry, increasing the number of regulatory reviewers and slashing timelines to achieve clinical trial approval from two years to 60 days. In the following years, domestic clinical trials doubled and drug approvals skyrocketed:

As US researchers grew increasingly reliant on China, which had grown to represent two-thirds of NHP imports, Chinese domestic demand was surging, increasing from 7,000-8,000 NHPs in 2013 to 30,000 in 2019.

Then, in 2020, COVID nuked the NHP market.

Bored Ape Research Club

Ironically, as the pandemic took shape, the hyper-volatile market for digital monkey NFTs turned out to be a surprisingly apt metaphor for the real thing.

At the onset of COVID, China moved to ban the export of macaque monkeys.

US researchers scrambled to find other suppliers. At first, Cambodia filled the market gap left by China and quickly became the new number one supplier to the US. But in 2022, scandal struck as Cambodia was identified as the hub of an international primate smuggling ring, in which wild cynomolgus macaques were being falsely labeled as captive-bred and illegally exported. Cambodian macaque supply shut off almost as quickly as it had rushed on to the scene.

That left Mauritius as the lead horse. But Mauritius has limited breeding supply; to fulfill demand, wild monkeys are trapped at scale for export. This has spawned a cottage industry:

For some locals, capturing monkeys is a way to remove animals that can bite and tear up yards; villagers can also earn up to $200 for allowing Bioculture to set up a trap on their property.

Source: Science (AAAS)

While researchers and industry tout Mauritius as a lifeline for scientific progress, this supply remains fragile. Like India, monkeys hold deep religious significance in Mauritius. Public sentiment could close the door to exports just as quickly.

Where does this leave us today? The supply of NHPs is on tenuous footing. The US has not shown meaningful appetite to build NHP breeding capacity. In fact, the Trump administration has signaled a strong interest in reducing NHP testing - recently, the CDC announced it would end its internal monkey research program. At the same time, China has demonstrated that it intends to reserve its NHP supply for its domestic biotech industry.

According to our checks, China’s move to shut down its NHP export business was done in an attempt to gain competitive advantage over the supply of NHPs and to drive increased drug development business in China.

Source: Mizuho Capital Markets - CRL Initiation Report

Subscale domestic production and limited access to foreign supply mean that the US is one supply shock away from a preclinical research emergency.

The US was never going to out-breed China on NHPs - China’s cost advantages and ethical calculus are insurmountable. Rather, the strategic off-ramp from reliance on this brittle supply chain will require doubling down on areas where the US does have structural advantages - world class technical talent and capital markets that reward technological innovation. And now, a vision for credibly diminishing dependence on NHP testing is coming into focus.

No More Monkeys Jumping on the Bed!

The FDA recently published a roadmap of promising alternatives to animal models - “New Approach Methodologies” (NAMs) - spanning organ-on-chip models, AI/ML toxicity platforms, and other advanced computational modeling approaches. Each approach tackles different pieces of the preclinical puzzle, from modeling how drugs move through the body to predicting immune responses before molecules ever reach a living subject.

Taken together, these approaches represent a potential reinvention of preclinical research - faster, cheaper, more predictive of human outcomes, and not dependent on a supply chain that other countries can turn off.

For its part, the FDA is outlining an aggressive path for adoption of these new methodologies:

In the long-term (3-5 years), FDA will aim to make animal studies the exception rather than the norm for preclinical safety/toxicity testing… Ultimately, the vision is that no conventional animal testing will be required … instead, a comprehensive integrated NAM toolbox (human cell models + computational models) will be the new standard.

Source: FDA New Approach Methodologies Report

Taken literally, this feels like an unrealistic timeline. While these technologies have shown promise, there are legitimate scientific obstacles that will constrain adoption. Many proposed methodologies struggle to capture the full complexity of physiology and how biological systems interact over time.

A key issue is that current NAMs typically offer insights into single cells or organs, failing to capture the complex interactions among multiple organs or the systemic effects of a drug across the whole body. For instance, organoids, which are 3D cell cultures that mimic key organ functions, do not have a vascular system that can replicate whole-organ physiology. While progress is being made, replicating the full complexity of human physiology—encompassing systemic drug distribution, metabolism, and immune responses—continues to be a substantial scientific and engineering hurdle.

Source: McKinsey

There is also the harder problem of the “unknown unknowns.” NHP studies are routinely credited with catching things researchers were not looking for - safety signals that current NAMs may not be designed to detect. Garnering institutional trust could take a long time.

The most likely path forward is NAMs gradually incorporated alongside, not replacing, animal testing. If these technologies prove out to be reliable proxies at scale, the sunsetting of NHP testing will eventually move from aspiration to reality.

The insiders are taking the over. Charles River Laboratories, which takes painstaking effort to emphasize that it is investing millions of dollars in alternatives to NHP research, announced a $510mm deal in January to vertically integrate a Cambodian NHP producer.

Monkey Business

I find myself reflecting on two recent anecdotes.

Late last year, a truck transporting rhesus monkeys from Tulane overturned in Mississippi. The local Sheriff’s Department put out a notice that the primates were infected with hepatitis, herpes and COVID. They were not (the Sheriff later blamed this misinformation on the truck driver).

Nonetheless, this led to a weekslong primate hunt. Hearing that disease-riddled lab animals were set loose in their community, the citizens of Mississippi literally took up arms.

Then there is the saga of Punch, a baby Japanese macaque at the Ichikawa Zoo in Japan, which captivated the internet (and me) this month. Little Punch-kun was abandoned by his mother and struggled to socialize with the other macaques. Footage of the other monkeys bullying Punch went viral. The zookeepers gave Punch a stuffed orangutan (specifically a DJUNGELSKOG from IKEA) which he ran back to and snuggled for comfort.

The staggering difference between these two stories feels like an extreme example of the Identifiable Victim Effect - thousands of macaques used in research are a statistic, Punchy is a tragedy.

I’m convinced that continued advancements around NAMs will eventually make testing on live non-human primates obsolete. Probably not on the FDA’s five year shot clock, but eventually.

When the day comes that biopharma companies and other researchers can fully transition to these alternatives, it will bookmark a chapter in history in which we subjected non-human primates to testing that we will almost certainly look back on as barbaric.

We’ll say it was necessary - the advancements enabled by NHP research have saved millions of lives, and will probably go on to save millions more. But it will be hard not to wince.

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

Q4 2025 earnings calls for managed care stocks offered a bevy of insights for early stage operators and investors. Here are some of the key takeaways if you don’t have dozens of hours to listen to the likes of Stephen Helmsley say the word “execution” 1000 times.

For me the big clear reads are:

• Medicaid is getting crushed. There looks to be a real opportunity for solutions to bend the cost curve in areas like behavioral health, high-cost drug spend, and complex care management.

• Payors seem to be ready for enterprise scale AI. Management teams are talking about investing billions into technology with expectations to realize multiples of that in long-term cost savings. I think we will start to see a rapid dispersion among early stage AI vendors. In 2025, a clear theme was seeing a lot of vendors doing similar things all lingering around the pilot phase with large enterprises; this year, I think the solutions proving early ROI will have tailwinds around wallet expansion which could fuel highly capital efficient growth for the early-movers.

• Government funding pressure has to relent at some point. Across categories, we’re seeing more members pushed out of coverage as health insurance becomes unaffordable for patients (e.g. now expired ACA subsidies) or the plans themselves (e.g. MA plans exiting markets with poor unit economics). Something has to give as more people get pushed out of the margins.

Vickspicks Hc Payor Earnings Roundup Feb 11 2026

4.07MB ∙ PDF file

Download

Download

This year, I asked ~30 people (down from last year, recession indicator?): if you could only pick ONE thing you discovered this year to recommend to others, what would you pick?

The recommendation could be in any category - a book, travel destination, restaurant, product, movie, activity, idea. BUT you can only pick one thing.

Last year’s picks were amazing. I was inspired to try making fruit and crème sandwiches by Kevin T., tune into Caleb Hearon’s So True podcast by Katie O., and watch the Penguin by Juan R. All bangers!

This year’s picks seemed to swing philosophical. This was a year to be inspired by concepts, not objects. More self-care, less retail therapy. Less travel, more introspection (recession indicator?).

As for 2026? I’m far from qualified to provide a definitive list of IN’s and OUT’s, but I’m feeling:

• Bearish on: coffeeshops, the All In podcast (please), vaping, long form written Linkedin posts (ironic), espresso martinis, Federal Reserve independence, sports betting being cool, America’s measles elimination status, cold plunging as a status symbol, Grand Theft Auto being released in 2026, people calling everything “agentic AI,” GenZ being viewed as a monoculture, extra protein in everything

• Bullish on: the Glue Guys podcast, negronis, Casey’s General Stores, the Chicago Bears, Atrioc, Chris Fleming, budget hacks as a status symbol, Waymo, Olivia Dean, Miami, sending voice notes, prioritizing fiber, recession indicators

Below are the picks (edited a bit for clarity). Thanks everyone for participating (p.s. if you have a highly recommends to add to this, email me!!) Love you all and here is to a great near year ahead!

The Single Best Thing I Discovered in 2025:

• Sänger Rubber Hot Water Bottle

  • Meagan W. strongly recommends this for the folks feeling a little chilly this winter. In her own words: “I’m obsessed with this hot water bottle. We are in love. I carry it everywhere like a stuffed animal. I would be frozen if it weren’t for her.”

    

• Blozzom Pizza in Miami Beach

  • Karen J. makes a very strong claim that Blozzom is home to THE best pizza in all of Miami. Mushrooms, zucchini, potatoes, honey - whatever the topping, Blozzom delivers!

• Amla Oil

  • Nakita P. (my sister :) ) says Amla Oil has been a game changer for her. It’s great for scalp health and eczema!

• Lovable

  • Michael M. had a lot of fun vibe coding this year. Even though he didn’t build anything serious, he found it incredible that he could easily develop working front ends without actually knowing how to code. After messing around with a few different platforms (Cursor, V0, etc.) he declares Lovable to be the most user friendly as a non-technical person

• Breville Coffee Machine

  • Juliette C. says she never buys coffee out now that she has a Breville at home. It takes a little practice but she swears, “once you get it, you get to make lattes better than the boujee NY coffee stores every morning.” As a recent purchaser of a Bambino Plus, I wholeheartedly sign off on this pick!!

• Lurker

  • Cinephile Mike G. picks this as his film of the year follow up to last year’s pick of The Substance. From Highly Recommends resident Roger Ebert: “No moment is wasted in this directorial debut about para-social relationships and the diminishing barrier between fans and celebrities. It’ll make you squirm, drop your jaw to the floor, and make you wonder: what’s the difference between love and obsession?”

• The true meaning of “get out of your own way”

  • Andrea M. channels a little Fleetwood Mac with her pick: “If you’ve ever been part of a large institution and told to “get out of your own way”, what they really mean is “we don’t have the resources, skills, or culture to support your ambition.” This is usually a good time to become your own boss and see what you can achieve without them in your way.”

• Mowe Napa Valley

  • Joe S. says the seven mile run from Yountville to St. Helena is home to the best wine America has to offer. But he singles out this particular winery located about a mile or so from downtown St. Helena if you’re looking for great wine, incredible views and great food nearby (especially Bouchon and Charter Oak).

• Saariselkä Ski & Sport Resort

  • Cat G. says that skiing under the northern lights is a magical experience. “It’s bitterly cold but the magic outweighs the cold. Find the trails that are lit with strung Christmas lights!” Dreamy!

• The New York Times Super Mega Crossword

  • Per Ellie B., “Every year, NYT releases a Super Mega Crossword in their December issue that takes up a whole spread. It’s got 700+ clues which may sound daunting but can end up being more fun than the regular weekly’s since there are more low hanging fruit. You can really get into a flow and it’s fun to do with a friend or a partner—especially at a bar over an Old Fashioned!”

    

• Mel Robbins’s podcast episode with Jim Doty, “#1 Neurosurgeon and how to manifest anything you want”

  • Gemma B. says this is a must-listen about mindfulness, positivity and goal-setting (hello, new year new me!!). “His story is incredible and this has been one of the most impactful podcast episodes I’ve ever listened to. Leaves you feeling positive and uplifted but he also provides tools on how to visualize/manifest things you want (from little things to big dreams).”

• Getting a pet cat

  • Andrew L. found his convictions shaken this year. “As a lifelong dog guy, I thought I would never appreciate having a cat. Living in NYC, Juliette and I had been wanting a pet but did want not to deal with having a dog in a cramped apartment / no backyard space. We adopted a pair of cats earlier this year and it was the best decision we made in 2025. They have wildly funny personalities, are extremely low maintenance, and are way more affectionate than I ever thought they would be. To all the non cat guys - highly recommend getting a cat!” I’m actually allergic so I won’t be doing this but I did care for several Neopets from 2001-2004.

• The Poison Squad by Deborah Blum

  • Carl M. makes this book his pick of the year which details the history of how food companies have been adulterating food for decades and how the industry leverages its power to lobby government to protect these practices. A fun (?) fact I didn’t know that Carl shared: “in the 1900s, milk used to be watered down, had chalk added to whiten it, blended calf brains to replace the cream at the top, and formaldehyde to preserve it. And that was common practice.”

• McDonald’s in Paris

  • Kevin T. is back with another culinary pick and says you simply must try M*******s in Paris. According to Kevin, “sometimes French cuisine can be so heavy and rich that a McCrispy, 6 nuggets, fries, and a Diet Coke is the most refreshing meal you’ve had in days. Also, the quality of the food is just better abroad than it is in the States. Besides the Vietnamese food, McDonald’s were some of the best meals that we had in France. We went three times.”

• Unwanted by Christina Keim

  • Distinguished equestrian Sara O. says that this book is great for anyone with an interest animal welfare or nonprofit work. “This book discusses the unfortunate outcomes of horses that are “unwanted” - old, injured, neglected, etc. Unlike cats and dogs, horses walk a fine line between livestock and pets, so they are subject to different rules and regulations, not to mention very expensive and time consuming to care for. This combination of factors means any horse is one sale away from an uncertain future. Despite the depressing description, this book is actually really hopeful. The majority of the book is spent describing various nonprofits that are tackling this issue, how they operate, and how they work together. As an owner of sport horses, it made me think more deeply about planning for my horses’ retirements after their athletic careers are over. I also Chair the board of an equine nonprofit, so seeing how other groups operate was very helpful. You may not own horses or work on horse nonprofits, but I think you will find this read fascinating and encouraging nonetheless!”

• Tesla Full Self Driving

  • Chance R. keeps the explanation simple: “If you are getting a car, why not get the car that can drive itself?”

• Physical Therapy

  • Andrew T. says PT helped him a TON with his lingering back pain. As a current PT patient I’m hoping he’s right about the life changing impact of a little targeted stretching!!

• Adding peel-and-stick wallpaper to your home

  • Emily C. says that peel-and-stick wallpaper TRANSFORMED her apartment making it feel “like an actual home vs. a four-white-walls box.”

• Prime Target

  • Juan R. is putting us on to another TV pick this year and says Prime Target (ironically on Apple TV) is “a smart, fast-paced conspiracy thriller series that blends high-stakes espionage with clever twists, keeping you guessing every episode.”

• Andor

  • Lincoln Z. says Andor is “a timely epic about the true costs of rebellion. Though it’s Star Wars, is passes for Sunday night HBO Prestige TV.” This pick gets a strong second from me - FREEDOM IS A PURE IDEA!!

    

• Eye Drops!

  • Jenn M. discovered that it wasn’t glasses she needed, but enhanced ocular lubrication! “Turns out my eyes were just dry and causing my vision to strain. The doctor said that as we age our eyes become dry and that everyone should be using eye drops multiple times a day.” So yes I’m not crying, my eyes are just performing their daily prescribed water aerobics.

• Metamucil

  • Phil E. has pick for all the high cholesterol bros (including yours truly 🫡). This reliable fiber supplement has a secret superpower of lowering cholesterol! Let this pick be the signal that the Year of Fiber is truly upon us!

• Figs

  • Christine J. says a highlight of moving to California this year has been indulging in fresh figs. If your experience of the fig has been limited to the Newton form factor, here’s your sign to bite into a fresh one this year!

• Propagating Plants

  • Chrissy B. encourages us to get in touch with our agricultural roots. “The best part of plant propagation is its simplicity, just a pair of scissors, a willing plant, and a bit of patience. Being able to take cuttings from my mom’s and grandma’s gardens, grow them in my own apartment, and even gift them to other loved ones makes the process even more meaningful. It’s a way of sharing living pieces of family history, and now you don’t need to spend another cent at the garden center!”

• Electric salt and pepper grinders

  • Sean E. cements his status as the king of seasoning with this one. Electric grinders make seasoning while cooking much easier because “you only need one hand and you feel super cool using them.”

• Booking a Solo Trip!

  • Katie O. says you should treat yourself to a solo trip for some primo R&R. “I took my first solo vaca this year to the Fairmont Resort in Mayakoba and I highly recommend everyone try traveling alone at least once. It was so enjoyable to just relax and do exactly what I wanted, when I wanted. It also pushed me to socialize outside of my comfort zone and gave me the opportunity to meet new people when going out to eat or exploring the resort. Spending 2-3 days to just relax and reflect is my new favorite form of self care!”

• Jaya at the Setai Brunch

  • Zeev G. is letting you in on some inside baseball. “Here’s the secret, the brunch at Jaya is unreal. Unlimited beverages and a food selection that goes as far as the eye can see. Sure you have roughly a 2 hour time limit but trust me, you won’t need more. Book massages at the spa before and go at 11:30am for brunch. The cost $110 per person, drinks are included and no, that is not a typo. Enjoy!”

• Handwritten notes, letters, and cards… just because!

  • Kenzie S. suggests that the best reason to get sentimental is actually no reason at all! “Although it’s easier to send a quick text, something about a handwritten message to friends, family, and loved ones feels much more meaningful to write and to receive. Just something simple to say have a great week, thinking of you, or thank you. Grab a stack of cards at the store at the beginning of the year and send them to your loved ones throughout the year to let them know you’re thinking of them. A nice gesture with a personal touch goes a long way these days.”

• Tomato-tinis

  • Chelsea V. sums it up: “It’s time to evolve from our 2022 ways and cast the espresso martinis aside - it’s now the time for tomato-tinis. More fun than a salad, classier than a Bloody Mary. Don’t forget some basil garnish and a splash of balsamic on top. Maybe it’s the Italian in me, but the one at Avra nyc is the best drink I’ve ever had and you bet I will try to convince Peter F. to let me have this signature cocktail at our future wedding.”

• Olive Oil Martinis

  • Joe B. continues the campaign of martini variations with this pick. Joe thinks that these are easy enough to make and far superior to a traditional martini!

• Heated Rivalry

  • Parisa D. says you should watch Heated Rivalry on HBO. If you haven’t heard, it’s “an incredibly unique, sexy, intimate, tear-jerking, thought-provoking show!”

• Investing in Sleep

  • Steven N. says “Seriously, half of your inconveniences in life are due to not getting enough sleep. Invest money, sure. But more so invest your bandwidth into purposefully getting better sleep. Future you will thank todays you for having made the investment.” As a new father, Steven’s advice on catching your ZZZ’s carries some extra weight :)

• The Best Comic Books Ever Written: Akira and Watchmen

  • Technically this is two picks but I’ve made an exception since Peter F. wrote a college thesis extolling the virtues of these great works in the form:

    I don’t read graphic novels, but the two I read this year after seeing the movies, Akira and Watchmen, happen to be the uncontested #1 and #2 greatest of all time (I have incredible taste). But don’t take my word for it: when my girlfriend asked the mouthbreathing proprietors of Midtown comics, “do you have anything better than Akira?” They replied, “NOTHING is better than Akira.”

    Katsuhiro Otomo is both the director of the movie and author of the comic book “Akira”. He released both at the same time in 1988 making the film the first and last book of the comic series. The movie was the first widely released rated R cartoon and is widely regarded as marking the birth of anime as a genre. It consists of 2,212 shots and 160,000 single pictures, which is 2-3 times more than normal. It also used 327 different colors, which is a record for animated films. 50 of those colors were exclusively created for the film. With a record budget for the time of $10m they essentially had multiple warehouse of Japanese laborers drawing extra frames because it’s fucking sick. Beyond its technical achievements, the story is utterly brain melting. It is about life, death, and the meaning of the universe, told through a story observing the classic jeering trope of male friendship. It is the definition of a masterpiece. Reading the comic was the single greatest novel-film combination because of the rare case of the author and director being the same person and planning both to sync with each other. The comic book gives so much more detail behind small moments in the film and is truly beautiful.

    6 months later I watched the Snyder cut of “Watchmen”, and the proceeded to watch it several more times in the same week (it is VERY important to watch the Snyder cut, and he theatrical release misses things making the plot confusing. I then immediately visited the mouthbreathers and was brought to the Watchmen dedicated section of Midtown comics. The story is of course unbelievable, it is a superhero movie that is also a complete deconstruction of the superhero concept. The historical references are incredible and the lore is as rich as Mansa Musa’s net worth even after adjusting for inflation. The graphic novel similar to Akira made the movie even more satisfying and the color illustrations gave me genuine pause as if I was in an art museum.

    You don’t need to read any other comic books in your life, but you must at least see the Akira and Watchmen (remember, SNYDER CUT) and then read one of them. Bonus points if you watch Akira in Japanese with subtitles, imo the English rips you from the world of Neo-Tokyo. I once tried to get [REDACTED] to watch it at my apartment and he left abruptly 30 minutes after walking through the door. It wasn’t because he was bored, it was because his worldview was challenged such that he was afraid.

• The General Concept of Everyone Chilling Out

  • Evan M. is feeling a little stressed out these days and says “whether it’s launching your AI-powered toaster oven startup, lowkey being cool with fascism so you can knock a percentage point off your income taxes, or telling me I should know who Sombr is…everyone, please just relax. This is all a bit much for me, so let’s all take a deep breath and think about how your choices affect my mood before you do anything else I don’t like. Additionally, I enjoyed the new Knives Out movie.”

• Loggerhead Marinelife Center in Jupiter, FL

  • And last but not least, MY pick of the year is this incredible marine life sanctuary in South Florida. In late summer to early fall, they host turtle hatchling releases at night. After an informational session on the secret life of turtles (did you know the sex of hatchlings is determined by the temperature of the sand??), you get a pair of nightvision scopes and head down to the beach to watch baby turtles waddle out into the sea. It’s like watching a turtle Mission Impossible and it is absolutely incredible. A must-see experience IMO.

    

  Happy new year!!!

I love an ambitious attempt to draw out the contours of major market trends, so I figured I’d take a shot at it this year.

Enclosed is an analysis of healthcare performance across asset classes, capital flows, underlying supply-and-demand drivers, company formation, startup fundraising, and the broader ~vibes~ that defined the healthcare technology industry in 2025.

When we look at the data and underlying industry signals, a clear picture comes through: the foundations of the system are hitting a breaking point. How we pay for care is under destabilizing pressure, and constraints to access are intensifying.

In response, early-stage healthtech activity is clustering around two vectors: leveraging AI to “fix the system” and building parallel models that help (largely affluent) consumers bypass the traditional system and take ownership of their health.

Healthcare Trends 2025

4.54MB ∙ PDF file

Download

Healthcare Trends 2025 by Vickram Pradhan

Download

As always, I welcome your questions, comments, and feedback! Thanks for reading!

Titles can have a magical impact on the perception of a job.

Banks have armies of “vice presidents.” Venture capital firms hire recent college grads as “partners.” I, for one, folded clothes and rang up customers as a lowly JCPenney “customer service associate” when my peers across the mall were doing the same thing as “models” for Hollister and Abercrombie. Now I’m forced to toil in the thinkpiece mines, groveling for Substack likes to feed the crater this left in my ego 😔😔.

But perhaps the most striking recent example of job title alchemy comes from Palantir and their “forward-deployed engineers” (FDEs). Here is the origin story, straight from Palantir CTO Shyam Sankar:

In 2006, Alex Karp asked me if I knew why French restaurants were so good. I had no idea. He told me that at a French restaurant, the wait staff is actually part of the kitchen staff. They intimately understand the food, the methodology, and the technique. They are not merely carrying the food from the kitchen to the table, but are instead part of a subtle and complex system that affects kitchen operations. He wanted me to build that, but for engineering.

I did just that, and in 2007, I gave it the name “forward deployed engineering” (FDE) in homage to our customers.

FDEs embed alongside our customers and work to ensure our software solves their problem and not some proxy for their problem...

Just like French restaurants don’t blindly hand off food from an uninformed waiter to the diner, we didn’t believe in throwing our software over the wall in the hopes the customer would divine the correct meaning from it.

Source: “The Primacy of Winning” by Shyam Sankar for Pirate Wires

Gergely Orosz does a nice job of demystifying the concept:

Job adverts for FDEs describe a few main characteristics:

• Software engineering basics…

• Collaborate with Sales to close customers… Basically, when a customer is undecided on whether they can effectively use a product, Sales offers to provide an FDE to help them successfully integrate it.

• Embed into customers’ teams… It’s pretty standard that the job involves travelling to customers to sit alongside them a few times. Palantir expects around 25% of FDEs’ time to be spent onsite with customers, and healthcare AI company Commure estimates up to 50%…

• Contribute to the core product, and embed with core product engineering teams

• Help customers succeed. The core of the role is to make customers successful by building on top of a company’s product offering

Source: The Pragmatic Engineer

The FDE role can be a source of significant consternation for investors - particularly for Palantir bears who see the company as little more than a zhuzhed up consulting firm. This high-touch, embedded services layer contradicts the dogma for “pure-play” software companies, where the goal is to minimize labor inputs to the delivery model. In their defense, what exactly is the difference between “help customers succeed” and “customer success?”

Palantir, for its part, points to FDEs as one element of a complex ecosystem of internal infrastructure, workflows, and processes that together form a “platform.”

It’s hard to argue with the results - Palantir seems to have outrun the “glorified consulting firm” allegations. While leading public technology consulting firms trade around 2x revenue, Palantir, backed by superior fundamentals and perhaps a touch of memey-ness, trades at ~100x revenue.

Meanwhile, FDE roles are exploding in popularity as they become woven into the fabric of prevailing technology delivery models.

This trend has legs. The idea of using a high-touch approach to implementation and value-creation seems like it will be critical in the next wave of AI adoption. Nestled in the recent, highly publicized “95% of AI is failing to deliver any ROI” MIT report is an important caveat:

Organizations stuck on the wrong side continue investing in static tools that can't adapt to their workflows, while those crossing the divide focus on learning-capable systems.

Source: MIT NANDA

The idea of “learning-capable” solutions mapped to existing customer workflows is practically a bat signal for FDEs.

It’s worth mentioning that AI businesses have (for now) inherently weaker economics than traditional software companies - inference compute costs represent a meaningful variable COGS item. The trade-off for lower gross margins is that AI tools should be able to capture more of the “work” happening across enterprises today. Said differently, applied AI companies are chasing a bigger pie than traditional SaaS firms that should yield more overall gross profit dollars. But to get that extra share of work, they have to do more hard stuff - and to do more hard stuff, they need a more flexible delivery model. As such, the demand for FDEs probably increases as AI applications permeate the market.

Valuation Considerations

The trick for founders is structuring FDEs in a way that reinforces a platform story rather than a services one. Here are some areas worth scrutinizing.

Contracting Discipline

What’s actually in the contract language? For SOWs and purchase orders, is spend delineated between software licenses and human resources? If so, then a probing buyer might ascribe a sum-of-the-parts valuation credit to that book of business. In contrast, if customers are paying for an all-inclusive package or specific outcomes and deliverables, it’s easier to justify that they view the offering as a platform. Are the contracts long term (e.g. 12-months or longer with defined renewal criteria)? Or are they short-term projects? The latter is going to be priced at a significant discount by investors, even if they are “recurring” in nature.

In a frothy market, investors and buyers tend to squint and call everything “ARR.” If the luster on AI companies fades in future years, we can expect investors and buyers to evaluate revenue quality more tightly, with contract language serving as the source of truth.

Technology Leverage

How exactly are FDEs using the core technology platform to deliver more value to customers? Often, this looks like faster speed-to-implementation and quicker time-to-value realized through the platform.

How much custom development by FDEs is reused and resold to other customers versus delivered as one-off, bespoke customization? For great platform companies, a track record often emerges quickly, showing that modules have been resold across multiple logos.

Complexity as a TAM Unlock

Perhaps the biggest advantage of the FDE model is that it allows a vendor to absorb more complexity on behalf of the customer. Good platforms will have a clear story around how FDEs expand TAM. For example, there are some customers that lack the ability, bandwidth, or appetite to consume software on a self-serve basis (e.g. biopharma companies notoriously prefer hands-on vendors). Moreover, amid a massive technology shift, customers are increasingly looking for vendors to serve as trusted thought partners rather than merely selling something off-the-shelf.

In other categories (e.g., government end markets), the complexity is so daunting that it’s impossible to serve customers without significant custom buildout. In these cases, there’s a clear story about using high-touch labor resources to configure the “platform” at the edge and meet customers where they are.

Show Me the Money

Ultimately, if FDEs deliver clear advantages, it will show up in the financials. The path looks something like this:

Applied AI companies and other FDE-led businesses will face the same kind of scrutiny SaaS companies have endured over the last decade. Well after the peak of AI-bubble mania fades, I expect that the market will continue to reward companies for retention, unit economics, and scalability.

Ambitious AI companies that build for long-term financial viability from inception will outlast hype cycles and command premium valuations over the long haul.

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

Today, there are roughly 40 approved CGTs and hundreds more in the pipeline, promising major advances for a range of conditions. Based on this pipeline, the number of patients receiving CGTs is expected to >10x over the next decade.

At the same time, these promising clinical advancements represent a brewing storm for pharmacy costs in the U.S.

For the promise of CGTs to be fully realized, healthcare stakeholders will need to create and adopt novel payment frameworks. The natural friction points that emerge from these frameworks could create opportunities for innovative point solutions in areas like risk underwriting, utilization management, and outcomes measurement.

The proliferation of CGTs will be a key component of the ongoing shift to more personalized, precision care. But it may also force a reassessment of the roles of insurance and the U.S. biomedical innovation apparatus from first principles.

Biotech Innovations → Financial Dislocations

CGTs are expensive. List prices for CGTs can reach $3-5mm per treatment. Why is that?

First, there is a tremendous amount of baseline R&D expense to cover. The average cost to bring a pharma product to market is >$1bn. For CGTs, estimates suggest that R&D costs are >2x that amount, driven by complexity (e.g. smaller trial sizes, trial execution costs).

CGTs often target rare disease categories. Smaller addressable markets of eligible patients support more aggressive pricing on a per-treatment basis.

Further, the manufacturing and delivery of CGTs is significantly more complex than traditional medications, increasing variable costs.

Most commercially approved cell therapy products are classified as autologous, meaning that the cells used to make the therapy for a patient are donated or collected from the same patient… Mass producing autologous therapies can be especially challenging, as the material from each patient must be manufactured separately. This would be like if each person in the world required a specialized car to be built unique to them, with the steering wheel in a different place, the pedals a custom length from the driver, and the windshield its own shape.

Source: Charles River Laboratories

But probably the biggest driver of the eye-watering list prices for CGTs is simply the way they work.

CGTs are generally curative and durable, meaning that a single treatment should meaningfully resolve the target medical condition and that its effects should persist over time (rather than requiring ongoing treatments). This means that biopharma companies need to generate all of their profit in a single transaction.

Interestingly, CGTs are typically priced to deliver a financial ROI for the medication relative to the status quo.

However, the upfront timing of the payment distorts how ROI is realized. Instead of recouping ROI alongside ongoing payments, the payor absorbs a large upfront charge and hopes to realize the ROI over subsequent years.

This creates a problem for payors. For example, a member who receives a CGT might later switch coverage. The member gets the lifetime benefit, but one payor shoulders the costs while another payor reaps the long-term savings.

Struck by Lightning

How much does this actually matter to health insurance plans? It depends on scale.

By the law of large numbers, payors with more members are more insulated from the risk of any given “lightning-strike” claim. For these groups (e.g. Medicare/Medicaid-scale payors, large commercial plans, or very large employers) navigating ultra-high-cost CGT claims is more about medical cost optimization than survival.

That said, even larger payors are feeling the impact of high-cost drugs, including CGTs. For example, Centene has a managed Medicaid book exceeding 10mm members.

Despite this scale, Centene recently attributed a meaningful share of higher-than-expected medical costs in its Medicaid business to high cost drugs - specifically naming gene therapies as a culprit:

Turning to Medicaid. Our Medicaid portfolio also fell short of expectations in the second quarter, producing an unanticipated and unacceptable health benefits ratio of 94.9%. Driving this underperformance was a step-up in medical cost trend in 3 areas: [~50%] behavioral health; [~30%] home health; and [~20%] high-cost drugs… with cancer drugs and gene therapies among the major categories contributing pressure.

Source: Centene Q2 2025 Earnings Call

For smaller-scale payors, particularly the long tail of self-insured employers across the U.S., the risk from ultra-high-cost-claims can be existential.

A single seven-figure “lightning-strike” claim could push a self-insured plan underwater. As a result, most small-to-mid sized self-insured plans use stop-loss to blunt the impact of catastrophic claims.

While stop-loss coverage provides some peace of mind, it does not shield self-insured plans from the inflationary burden of CGTs. Coverage for these drugs typically comes with higher deductibles and premiums. Stop-loss premiums are already increasing at a ~10% CAGR, and that’s with only 50-60% of stop-loss plans covering the few dozen approved CGTs. Premium inflation could accelerate if 1) CGT coverage expands across carriers and/or 2) CGT approvals proliferate.

Precision Finance for Precision Pharma

The urgency to manage the time-based financial dislocations from CGT claims is spurring the development of novel reimbursement frameworks.

Three key risk vectors define the scope for potential financing solutions: payment timing risk (spreading out the upfront payment), performance risk (ensuring clinical value for the treatment dollars), and actuarial risk (managing overall uncertainty in a given payor’s population).

With these risks in mind, here are some of the interesting alternative payment models that I have seen floating around:

Performance Based Annuity Model: First, we could spread the upfront payment as a multi-year annuity.

This approach addresses payment-timing risk, but has a zero-sum trade off - improving payment timing for the payor makes it worse for the drugmaker.

A simple annuity model does not address performance, which is why a natural extension is to add a performance mechanism. A performance layer drives alignment between patients, payors and drugmakers by tying payments to clinical outcomes.

In a milestone-based payment model where there are future payments from the payor to the pharma company, the outcomes measurement burden likely falls on the drugmakers, since they want to get paid in later years. A variation to this would be a milestone-based rebate model, where the drugmaker is paid upfront, but issues rebates to the payor if milestones are not met. Here, the outcomes measurement burden likely falls on the payor, since they want to get rebates in the later years.

The concept of paying for CGT performance is gaining traction. For example, CMS recently launched the Cell and Gene Therapy Access Model for Medicaid.

The CGT Access Model is a voluntary model for states and manufacturers that tests whether a CMS-led approach to developing and administering outcomes-based agreements (OBAs) for cell and gene therapies improves Medicaid beneficiaries’ access to innovative treatment, improves their health outcomes, and reduces health care costs and burdens to state Medicaid programs.

Source: CMS

An open question is how performance will be adjudicated. In theory, both sides have incentives to shape the analysis of patient outcomes. Moreover, neither drugmakers nor payors may want to build and manage this adjudication function in-house. A performance framework might require an independent party to play referee.

Mortgage / Therapeutic Health Credit Model: Adding a financial engineering layer on top of the annuitized payment stream could also be compelling. This structure would allow drugmakers to get paid upfront, while enabling payors to spread costs over time.

The “Financed” Performance Based Annuity Model: This sets up a kind of a “holy grail” model combining the benefits of financialization and cost-spreading from the aforementioned models.

The idea is a “NewCo” that serves as an outsourced adjudicator of performance milestones. NewCo could also financialize the process with a loan product. Further, NewCo would likely provide care navigation to help ensure patients are hitting key follow-up care milestones, critical to achieving target clinical outcomes.

This model is really cool, but could be prohibitively complex.

Standing up NewCo would require combining biotech expertise, institutional financial/risk-underwriting acumen AND care-navigation capabilities within one company. A more achievable version might entail NewCo building the “pipes” to connect this system, while leveraging partners for key aspects of the value chain (e.g. a financial institution capitalizes the loan, a care navigation company handles last-mile patient support).

Aligning all incentives and building the right execution capabilities is a herculean task. But history demonstrates that innovators that simplify complexity for stakeholders can create and capture outsized enterprise value. With CGT spend set to accelerate significantly over the coming decade, the juice might well be worth the squeeze.

Cure Orchestration

Regardless of which payment models gain adoption, stakeholders will need extensions of existing underwriting and cost‑containment tools to manage CGT expenses. To fully unlock the potential of CGTs, they’ll also need solutions for developing and delivering these therapies at scale.

Enhanced Risk Underwriting

Historical claims data underpins much of the work done by health‑insurance underwriters.

A challenge for CGTs is that, because the category is so novel, they rarely appear in historical claims data. As a result, underwriting that relies on historical claims data alone understates the risk of future CGT claims.

In the absence of underwriting precision, carriers default to raising premiums to create a margin of safety, driving higher costs for self-insured plans.

This seems like an opportunity for a solution that incorporates rich, alternative data sources (perhaps even synthetic data?) into underwriting decisions to deliver cheaper coverage.

Utilization Management

Based on discussions with decision-makers, from those running self-insured plans directly to the “gatekeepers” of employer insurance (e.g. TPAs and brokers), there is a strong preference for solutions that go beyond quantifying risk to acting on it.

One widely used cost-containment tool in healthcare is utilization management, which seeks to avoid inappropriate medical spend by creating checks on physician orders (e.g. prior authorizations). These checks are typically focused on ensuring medical actions are clinically appropriate.

Utilization management of CGTs will be especially challenging, even compared to other specialty drugs. Determining clinical appropriateness for CGTs is very complex, requiring patient education, genetic testing, and completion of first-line therapies. Given the convoluted patient journey, as illustrated in the below graphic, carriers have numerous potential checkpoints to test clinical appropriateness.

A generous read of utilization management would be that it ensures that patients are getting the exact right type of care for their condition. On the other hand, overly aggressive utilization management can spark ire, as these roadblocks often feel adversarial to patients and their physicians. Making it harder for people to access life-saving care will inevitably invite scrutiny.

That said, I can’t see a world in which payors don’t default to utilization management as a lever to control CGT spend. We may even see adoption driven by reinsurers and stop-loss carriers, who could offer cheaper coverage if cost-avoidance mechanisms like utilization management are in place.

My sense is that utilization management solutions that build trust with physicians by focusing on a collaborative approach to clinical appropriateness rather than acting as punitive roadblocks to patient care will be stickier and be more successful long-term.

Development, Manufacturing, and Delivery

Meaningful adoption of CGTs will require innovation across manufacturing, care delivery, and R&D execution as well.

For example, developing high-trust centers of excellence for specific treatments could be lucrative, particularly given the significant geographic constraints many patients face. For example, per Milliman, less than half of patients with sickle cell disease have a treatment center in the region where they live.

As mentioned earlier, care navigation is also a compelling opportunity given the significant dollars at risk with these treatments. Patients would be more likely to meet clinical targets if they have support across the chaotic CGT journey.

More commercial success for CGTs could create a flywheel for capital formation around CGT R&D, serving as a tailwind for the ecosystem of clinical trial technology and services companies, which can help in the difficult aspects of identifying and recruiting oncology and rare disease trial participants.

A glass-half-full view of the coming wave of CGT complexity is that high prices create margin dollars to invest in innovative solutions, potentially unlocking an ecosystem of enabling companies.

The Price of Healthspan

In closing, it’s worth reflecting on what this wave of critical but expensive innovation reveals about the U.S. healthcare system.

On one hand, the pharma industry functions as an innovation lab, generating scientific breakthroughs for the benefit of future generations. The time-boxed monopolies granted to new products via patent protection create a dynamic in which taxpayers effectively subsidize a big pot of gold to incentivize talented people to dedicate decades of their lives to advancing human health. It’s hard to imagine this dynamic changing without degrading one of the greatest innovation engines at work across the U.S. economy.

If the system is set up to incentivize leaps forward in healthspan, it’s worth asking what role insurance companies should play. For example, employers across the U.S. are effectively paying a tax to stop-loss carriers in the form of premiums to cover high-cost drug claims. Yet even when aggregated through stop-loss carriers or other pools, employers have less leverage due to fragmentation.

One alternative would be to carve out all CGT risk into a separate pool. At the risk of sounding like a sOcIaLiSt (never forget Mitt did Obamacare first), the premium currently flowing to stop-loss carriers could instead fund a government-sponsored risk pool.

The widest risk pool can manage “lightning-strike” claims most cost-effectively, while maintaining the greatest leverage in negotiating pricing and performance obligations. I like this idea because it feels pro-science, pro-business, and pro-people getting breakthrough, life-saving care.

Perhaps one day we’ll have this kind of centralized Life-Saving Care For All or Universal Catastrophic Coverage program. Until then, it’s clear that in order to maximize the benefit and reach of the scientific breakthroughs poised to emerge on this frontier over the coming decades, the industry needs to develop the commercialization and orchestration layers necessary to power CGT delivery at scale.

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

The short version: Family offices and other evergreen vehicles have a few commonly understood advantages in venture strategies. They’re more flexible in how they can pick and structure investments, they focus more on investing than on fundraising/reporting, and they’re better aligned because “eating your own cooking” emphasizes returns, not getting fat on fees.

But the biggest edge is duration, which is often misunderstood. People seem to conflate long-term orientation with infinite patience. While family offices can be more patient with respect to capital deployment, portfolio maturation and exits, that patience has limits. Rather, the real unlock of duration is the ability to compound via re-invested proceeds, which enables an entirely distinct approach to portfolio construction.

This is not to say traditional VC funds are “bad.” As Bill Gurley recently put it: “There are people and firms taking action that change the state of the field and I think they're all acting reasonably and in their best interest. The aggregate effect may not be positive for the world, but I'm not ascribing malintent...”

Ultimately, VC as an asset class is far from one-size-fits-all. The key for entrepreneurs navigating early stage capital markets is clarity. Know what you’re trying to achieve and seek capital that specifically aligns with it.

The Unbearable Lightness of Having No LPs

Shunning outside capital gives family offices several strategic/qualitative advantages in direct venture capital strategies, including flexibility, focus, and incentive alignment.

Flexibility

Family office VCs are not constrained by a fixed mandate or portfolio construction.

In a closed-end fund, LPs need to understand exactly what they are signing up for. This forces GPs to lock in a strategy and structure upfront. Once set, they can’t easily deviate from it.

In contrast, family offices generally don’t need to commit to a specific mandate. This means that they can stretch across asset classes and investment structures. John Zito, Co-President of Apollo, explains the advantage of this flexibility (structured credit in his case, but the logic carries):

The directive is: do the best risk / return.

Most firms are set up by fund, they're set up by okay, go find a company or a preferred or a debt instrument that's going to make 15% rates of return. And we're set up, let's assess the company, let's assess the best solution for that company.

It could be a buyout, it could be an investment grade solution, regular way bond deal. But understand risk, reward, or cost of capital structure per unit of risk. It's a completely different framing because we have no walls.

Source: John Zito on Invest like the Best Podcast

This flexibility means that family office VCs can be better capital partners to entrepreneurs. When the mandate is just “make money,” investors can offer bespoke capital solutions that fit the needs of the company, rather than something off-the-shelf (e.g. VCs who have rigid ownership targets).

Focus

Family office VCs don’t have to fundraise. Instead of churning through 1,000 point questionnaires from prospective LPs, family office investors can dedicate bandwidth to sourcing, execution, and supporting investments. Reporting still exists, but it’s far less onerous.

Every minute not spent on the “annoying stuff” is more time spent helping companies create enterprise value.

Incentive Alignment

Family office investors don’t live off of management fees. The incentives are simple. They only make money if investments do well. That drives tighter alignment with entrepreneurs, while avoiding many of the distortions that come from the pressure of managing outside capital, like:

• Pushing companies to take more capital than they need, to put more fee-carrying capital to work

• Keeping companies on the fundraising treadmill so they can show paper markups

• Performing cap table gymnastics to inflate headline valuations

For many venture capital firms, the product is the next fund. With this backdrop, it’s no wonder that valuation distortions are endemic across the venture ecosystem today. As Bill Gurley put it:

The thing that most people may not believe, but I guarantee you is true — no one has an incentive to get the marks right. For those that don't know this world — private investing, both on the PE side and the VC side — is this weird world where the GPs, the people responsible for the investments, report the price to the LPs. They get to pick it… But the thing people may not realize is, the managers of the VC group at the large endowments have no incentive to try and right-size this number. In fact, many of them are bonused on paper marks. So if anything, they have the reverse incentives to get them right.

Source: Bill Gurley on Invest like the Best Podcast

When you’re “eating your own cooking,” the incentive structure is brutally simple1: make money.

Here for a Good Time AND a Long Time

The most profound advantage of family offices (and other evergreen capital vehicles) comes down to the simple math of duration. But the benefits of long-term orientation are often misunderstood.

The prevailing view is that family offices are just more patient with respect to capital deployment and timing exits.

It’s true that evergreen investors are not forced to invest on a timeline or sell early due to a fund life cycle. Instead, they can wait for favorable market conditions and ride out asset level volatility.

Traditional funds, by contrast, face mounting pressure to return capital on schedule. You can see this playing out in rise of PE continuation vehicles and the growing popularity of PE and VC secondaries (e.g. Yale Endowment’s “Project Gatsby” to offload $2.5bn in PE/VC secondaries).

But the idea that family offices can wait indefinitely for liquidity has limits.

Some family offices benefit from alternate income streams. Take Steve Ballmer, who seems to be clipping (lol) ~$1bn in Microsoft Dividends each year.

You’re just trying to shovel the money that is coming in the door, out the door.

Source: Acquired Podcast Interview with Steve Ballmer

Groups like Ballmer’s would have far less pressure to generate liquidity from their investments2.

But many family offices don’t have that kind of external cash flow. Those groups need to return capital in order to fund new deals.

Many believe that the lack of a “sell by” date means that evergreen investors can exhibit more patience around startup growth, particularly in early years when it takes time for product-market fit to materialize. That is sort of true. Ultimately, all investment outcomes are governed by rates of change. No rational investor, evergreen or otherwise, is going to sit around content with flat growth indefinitely.

Double It and Give It to the Next Person

The real advantage of duration is compounding. This unlocks a different style of portfolio construction, which often aligns better with founders than the standard spray-and-pray VC playbook.

A recent paper from Michael Mauboussin and Dan Callahan at Morgan Stanley lays out the math behind this idea.

An essential point is that mean/variance optimization generally assumes you are deciding based on one period. But the approach is different if you consider multiple time periods and your goal is to maximize the likelihood that you will have the most money at a date far in the future.

This insight came from John Kelly, a physicist who used information theory to develop a strategy for optimal betting over the long term. Kelly noted that if a gambler made one bet of one dollar per week but could not reinvest his winnings, he should maximize expected value. This is Markowitz.

But the math changes if the winnings are reinvested from one period to the next.

Instead of seeking the outcome with the best arithmetic mean, the objective is to find the opportunity with the highest geometric mean. This is called the Kelly criterion, or Kelly strategy.

The Kelly criterion can’t be directly applied to venture capital because the probabilities/risks/payouts are not known in advance, but the concept is still instructive:

The probabilities and payoffs allow us to calculate the arithmetic and geometric means. We can see that of the three opportunities, the expected value, or arithmetic mean, is lowest for A, in the middle for B, and the highest for C. If you bet the same amount every time you should maximize expected value. Markowitz would say that the best choice is a function of an individual’s preference. But opportunity C is the most attractive, all else being equal.

If your profits or losses in the prior period are reinvested into your bankroll, you should use the Kelly criterion and maximize the geometric mean. In this case, A is the most attractive opportunity and C is second best…

The main takeaway from this discussion is that there is a crucial difference between selecting the opportunity with the highest expected value for one period and reinvesting returns in opportunities over multiple periods. Mean/variance optimization is built for the former and geometric mean maximization works for the latter.

Source: Morgan Stanley

Closed-end funds don’t get to reinvest proceeds. They are playing a one-shot game. Thus, it is rational for them to optimize around expected value per fund. This is why the spray-and-pray approach of having 1-2 massive wins in a fund is so common.

But this reliance on power law outcomes becomes a trap. Everything gets underwritten to be a potential billion dollar exit, whether it deserves to be or not. “Perfect” is prioritized while sacrificing a multitude of “good” to “great” outcomes.

Family office investors can play a different game because they can roll proceeds into future investments. Modest outcomes still matter because they become fuel for the next investment.

This is not to say that family office VC investors aren’t looking for big outcomes. But they can use a more measured approach to achieving them. The power law is ultimately a statistical phenomenon, not a set of operating instructions. The traditional Silicon Valley obsession of chasing home-run outcomes crowds out meaningful, durable wins that could be life-changing outcomes for entrepreneurs.

That said, there are real constraints for family office VCs too. The biggest challenge is scale. Big firms with billions in AUM have the infrastructure to power robust sourcing, whether via content or brute-force. This can be critical since those strategies are incumbent on intercepting the rare category-defining company.

Ultimately, I think raising capital really is a choose-your-own-adventure for founders. If you want to shoot for a bazillion dollar outcome with haste, you should raise from a traditional VC. If you want a more measured approach, seek alternative capital. The ongoing maturation of the VC industry means that it’s not a one-size-fits-all product.

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

In healthcare, I see a few distinct categories of innovation, each with a unique risk/return profile that informs everything from investment sourcing to portfolio construction.

The Standard of Care

At the risk of oversimplifying for brevity, I will define the healthcare market as the goods and services of the standard of care, where the “standard of care” is how medical issues are treated.

In this context, we can say that the contours of the healthcare market are shaped by:

• What the standard of care is (clinical)

• How the standard of care is delivered (operational)

• How the standard of care is paid for (commercial)

These buckets create a framework to classify innovations. The chart below illustrates how I think about risk and return across these categories.

It’s not perfect, but this framework helps me screen and assess investment opportunities. My current investment mandate favors a high-conviction, concentrated portfolio of assets with asymmetric upside on low capital intensity (i.e. doesn’t require a massive amount of capital to scale). It also favors flexible liquidity, prioritizing exit optionality and avoiding write-offs over pure “unicorn” potential (i.e. not every company needs to be a “grand-slam”).

With this mandate, I’m drawn to commercial innovations and niche operational innovations, for reasons I’ll explain below.

As always, I would love to hear your feedback because this framework is something I plan to calibrate over time.

Clinical Innovation

What is it: Clinical innovation changes what the standard of care is. Clinical innovations advance the prevailing set of methods by which medical conditions are treated. In short, they change the “medicine” itself.

Examples: In this context, “medicine” is the combination of diagnostics and clinical interventions to treat issues. This would include diagnostics, biopharma products, medical devices, surgical interventions, and other direct treatment modalities. I would also include technologies that support the direct delivery of those diagnostics and interventions.

Risk Profile: First, clinical innovations have a baseline scientific “does it work?” risk. Companies must show that clinical innovations “work” via hard research. Getting through this research phase (e.g. FDA approval process for biopharma products) is hard. Most attempts fail — e.g. the average success rate for new pharma products is <10%.

This expensive, low-hit-rate research makes clinical innovations extremely capital intensive. Importantly, this capital intensity comes long before there's any certainty of commercial viability. Regulatory outcomes are generally binary - clinical innovations either outperform the status quo or they don’t.

These dynamics explain why the cost to bring a new drug to market exceeds $1bn.

Theoretically, once you have proven that the “new thing” is sufficiently better than the “old thing,” commercializing the new thing should be easy. There should be no reason for a doctor to prescribe the inferior, old thing.

Further, because clinical innovations typically have robust patent protections, theoretically there’s no competition for your new thing for a definite period of time. This should enable clinical innovations to capture massive market value unencumbered.

Sure enough, these market dynamics have enabled multiple blockbuster drugs to achieve billions in lifetime revenues.

In practice, however, there are many hurdles to the pot of gold at the end of the regulatory rainbow.

First, the payment model for biopharma products and medical devices is highly convoluted. This creates reimbursement risk. For example, pharma companies have to convince insurance companies that paying for their innovations will deliver a compelling ROI.

Also, doctors and patients need to be educated about the product. There is real commercial execution risk involved in things like engaging key opinion leaders and establishing distribution channels.

Still, in broad strokes, we can say that if a clinical innovation “hits”, the financial outcome is massive.

Portfolio Implications: This results in a boom-or-bust risk/return profile for clinical innovations.

Successful investing is based on deploying capital with a margin of safety. In venture capital, you can either build that margin of safety into the disciplined pricing/structure at the asset level (value investing, at the extreme) or diversification at the portfolio level (“spray-and-pray,” at the extreme).

But there is no real way to de-risk a clinical innovation at the asset level; even with the perfect team, the fundamental “does it work” risk can only be resolved by the research.

At the same time, these companies offer “return the fund” outcome potential. Given this dynamic, successful clinical innovation investing requires a diversified portfolio rather than a concentrated portfolio.

I avoid clinical innovations because of the capital intensity and binary regulatory risk, which doesn’t fit with the concentrated portfolio construction of the strategy I’m on. And also because I am just a dumb excel monkey who gets queasy around blood and chemistry.

Still, I deeply admire investors and operators in this category — clinical innovations are ultimately the biggest drivers of progress in human health.

Operational Innovation

What is it: Operational innovation changes the way that the standard of care is delivered and administered.

Examples: This is the broadest category, encompassing most services and software businesses that attack the massive administrative layer in healthcare.

This category includes provider-focused innovations (e.g. documentation tools, clinical decision support, prior authorization automation) and payor-focused innovations (e.g. population health analytics, utilization management).

There might be some patient-focused innovations too, but because patients are not usually the “buyer” for healthcare innovations, the scope for creating and capturing value from patients directly is limited.

Risk Profile: Operational innovations have significantly less “does it work” risk. Companies can quickly assess whether a new software or services model is going to work in a binary sense, even before generating revenue. The remaining “does it work” risk is based on how much ROI/impact the innovation delivers, which can typically be measured once early customers adopt the offering.

Unfortunately, having less “does it work” risk is a double-edged sword. It lowers the barriers of entry, which means that you can get to market faster, but there is also more competition.

A central challenge with operational innovations in healthcare is that the problem areas with the big TAM opportunities are generally well-known — e.g. documentation, billing, coding, prior authorizations, etc.

A common market pattern when there is a new technology is that a lot of people think — hey, let’s take the new technology and apply it to the known problem with a big TAM. When there are low barriers to entry and limited points of differentiation, this leads to rapid saturation of new categories.

We’re seeing this now with the proliferation of AI companies attacking the same opportunity sets — like the >50 “ambient AI scribe” companies in the market.

The low barriers to entry create go-to-market complexity, which drives capital intensity. Companies must spend more to sell in a crowded market and invest heavily in R&D to stay ahead of competitors. As a result, operational innovation categories can easily turn into arms races where players are incentivized to raise massive war chests in a bid to “win” the market.

Complexity is a lever for operational innovations to build a competitive moat. Absorbing complexity from the customer enhances stickiness and differentiation. This often drives services and software models to converge. Services companies often need to invest in infrastructure and technology in order to absorb more complexity at scale. Software companies often need to invest in a services layer in order to ensure customers rapidly and sustainably realize ROI.

An important consideration in assessing the commercial viability of an operational innovation is to look at where on the customer’s P&L the innovation delivers an impact. Specifically, does the innovation a) increase revenues or b) reduce costs?

Operational innovations that increase revenue are almost always more compelling. There’s less friction involved in making the revenue printer go BRRR. Cost reductions, which typically involve reducing headcount or obviating vendor spend, elicit more resistance through the sales cycle.

Portfolio Implications: Operational innovations can be really compelling for a more concentrated investment strategy if there is a meaningful, durable differentiation — either differentiation built into the offering (e.g. complexity approach) or in the market focus (e.g. overlooked niche positioning).

Operational innovations are highly dependent on execution. That means investors in this category need to be comfortable leaning in to help entrepreneurs scale. For example, in an operational innovation business, things like making the first head of sales hire and developing the right sales commission structure are super important. Investors who have the ability and appetite to assist founders in thinking through execution at that tactical level are better positioned for success. This dynamic fits well with a concentrated portfolio because there’s less dilution of time/resources/bandwidth.

So while I am scared of the “arms race” categories with low barriers to entry and limited differentiation, companies that build differentiation through a niche focus or complexity can be very interesting for high-conviction portfolios.

Commercial Innovation

What is it: Commercial innovation changes the way that the standard of care is paid for. Often, these are business model innovations that work to align stakeholders in the value chain, while capturing a significant financial spread or revenue opportunity.

Examples: Value-based or risk-based business models. Most value-based care companies boil down to doing some clinical intervention or activity really well and using the improved outcomes to create and capture a financial spread.

Another category I would lump in here is marketplace businesses that transform how specific aspects of healthcare are transacted (e.g. rentable clinician networks). I’d also include companies here that have developed unique go-to-market approaches wherein they absorb risk to make it easier for a customer to say “yes.” As an illustration, TapestryHealth (Sopris portfolio company) takes on some financial risk of installing hardware for skilled-nursing facilities in order to reduce the friction for adoption to customers.

Risk Profile: The major risk for commercial innovations is making sure the rails of the “mousetrap” are set up correctly and that the Company can actually deliver on the risk it assumes. Once you know it “works,” then scaling via more volume/risk is relatively straightforward. Because these models often absorb risk from customers, there is a clear value proposition and less sales friction. This means less overall go-to-market execution risk.

Portfolio Implications: Companies with a novel commercial innovation tend to be the most rare. Creating a new way to transact requires a diverse skill set, more so than other categories of healthcare. For example, a fee-for-service care delivery company needs to excel at care delivery. Meanwhile, a risk-based model requires actuarial and contract design expertise. With each additional layer of complexity in a commercial innovation, the founding team needs a wider range of experience to be successful. The scarcity of these assets is an important portfolio construction consideration — it would be difficult to build a large portfolio constructed solely of commercial innovations. But for a concentrated investment mandate, the scarcity is less concerning.

From an outcomes standpoint, this category offers robust exposure to asymmetric upside. Once you know it “works,” commercial innovations can scale in a rapid, continuous motion. While clinical innovations also offer step-function value upside, they require orders of magnitude more capital. Commercial innovations are more attractive from a dollars-in to dollars-out standpoint. One caveat: models that function like quasi-insurance companies may face capitalization complexity if there is a need for an insurance-like balance sheet reserve (e.g. if they are subject to downside risk).

🧨🧨 Short-Fuse Asymmetric Upside Investing

Ultimately, I find the categories of niche operational innovations and risk-based commercial innovations to be the most compelling. They offer faster feedback loops to commercial viability and lower commercial intensity (less competition). They’re hard to find, but they offer really attractive asymmetric upside with optionality that fits well with a concentrated, durable, returns-oriented early-stage portfolio.

So if you’re working on something like this, hit my line asap!!

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

II. Executives

Mark Zuckerberg

Fascinating because: Zuckerberg is the prototypical tech founder of the internet age.

Mark Zuckerberg is THE archetypal founder of the consumer internet age that I grew up in. Meta’s transformation from quirky dorm room side project into a digital narcotics cartel advertising cash machine is a creation myth for the modern technology industry. A lot of tech culture is built on Meta/Zuckerberg’s foundation - the product focused founder CEO, “move fast and break things,” having a “growth” team.

The most interesting aspect of Mark Zuckerberg as an executive is the control that he has over Facebook because of the dual-class stock structure. His superpower is that, unlike basically every other public company CEO, he cannot be fired.

This enables an unparalleled decisiveness - illustrated in situations like the Yahoo/Microsoft buyout offers and the transformative M&A deals in the Company’s history. The biggest advantage of this dynamic is that Zuckerberg can underwrite to time horizons that far exceed the typical CEO; other CEOs have to be fixated on the next quarter, Zuckerberg can plan for the next decade.

This job security has enabled Zuckerberg to mature through multiple phases of his tenure at the helm. There are Eras of Zuckerberg. Melting down servers at Harvard Zuckerberg, turning down $1bn to bet on himself Zuckerberg, getting yelled at by Congress Zuckerberg, and current edition UFC/wakeboarding/startups-need-to-be-on-TRT Zuckerberg.

And he is still so young. It’s crazy to think that his best days might yet be ahead.

Personally I am here for the Amiri collab, Porsche pimpin’ era of Zuckerberg. I have been wondering why he’s become much more prolific with his public persona, including the recent shift to “tech-right” political narratives (oh nooo DEI is killing my company, it was only able to add $1 trillion in market cap over the past 5 years😔😔).

In this era, I see strategic decisions from a noted strategy game enthusiast to position Meta to survive the next phase of the Company’s life. If I had to guess, Zuckerberg is looking at Elon Musk and seeing that his larger-than-life cult of personality serves as a foundational support to the his companies’ stock prices and ability to fundraise. He probably wants some of that action to bolster his own position on some of the Company’s longer duration bets. The political movements track with how he has operated in response to government/regulatory pressure in the past. We had the very demure version of Mark Zuckerberg and the Company during the high-scrutiny Cambridge Analytica fallout phase because that was necessary for survival. Similarly, I think he now sees an administration that is transparently open-for-business and one where perceived loyalty will forgive all manner of corporate harm by consumer-tech companies (please ban tik tok).

I am sure that Mark Zuckerberg’s tenure as CEO of Meta will continue to fascinate me. That being said, I do think he should be arrested for putting this into the world:

Barry Diller

Fascinating because: Diller’s management style of cultivating talent by pushing responsibility down resonates with me.

Barry Diller is a media powerhouse. His career spans ABC to Paramount, founding FOX with Rupert Murdoch, Building QVC and then InterActiveCorp.

He is well known, in particular, for his style of management. Frederik Gieschen captures the "Diller System” really well:

Diller cultivated so many talented executives with this approach - the “Killer Dillers” (side note, “Diller’s Killers” was right there🤦) include Jefferey Katzenberg (Dreamworks … and Quibi RIP), Michael Eisner (Disney), Dara Khosrowshahi (Uber), Strauss Zelnick (Take Two / CBS). The story of Diller mafia member Anjali Sud (Vimeo) synthesizes the method:

[Anjali Sud is] a great example of that. I was running Vimeo on an interim basis for close to a year. The company divided into two pieces. One was Anjali’s side, tools for video creators, and the other was using the Vimeo platform to be a destination where we were creating our own content for audiences. At some point, she said to me, “I can make this business much bigger under the strategy that I’m learning.” And I said, “Great, do it.” And she did. We scrapped [the content creation] and moved everybody onto the other when we promoted Anjali to CEO.

Source: Wharton (🤮) Magazine

The sink-or-swim approach to management and development is one that resonates with me - it’s the kind of system I feel most comfortable in and the type of system I’m most comfortable deploying. Thus, I find it sort of inspiring to see the cadre of talented operators who were forged through Diller’s system.

Also, Diller’s “Little Island” off the WSH in Manhattan is easily a top 5 billionaire pet project.

Jamie Dimon

Fascinating because: Dimon is the ultimate statesman CEO whose scope as an executive is unparalleled.

The reach of JP Morgan Chase is absurd. It banks consumers and small businesses. It banks multinational corporations and nation states. It has >300,000 employees. And all of that breadth is under the purview of Jamie Dimon.

Dimon commands this veritable empire with extreme charisma. In many ways, his role at the helm of JPM is to be this grand statesman across the myriad counterparties of the firm. In this role, he has a track record of deftly navigating ultra high-stakes flashpoints. Basically at gunpoint he led the buyout of WaMu and Bear Stearns in ‘08 and then orchestrated the rescue of First Republic in 2023. His muted commentary on the 2024 election is an example of the measured statesmanship required to navigate uncertain political waters. For all the criticism of wanting his cake and eating it too, his stance on crypto is also a great example of this:

Dimon has remained a fiercely outspoken bitcoin and crypto critic even as Wall Street began to warm up the technology, calling it a "fraud" and a "Ponzi scheme" in April. Dimon has said that if he was in government, he'd "shut it down," and branded bitcoin a "waste of time," calling it a "pet rock" that "does nothing."

However, Dimon has led JPMorgan in developing and deploying its own blockchain and cryptocurrency, JPM coin, to speed up and reduce the cost of transactions.

Source: Forbes

His statesmanship provides a great model to learn from. Bill Ackman said it best - Dimon for President baby.

Michael Ovitz

Fascinating because: Ovitz’s work is a masterclass in tactical execution, winning the key moments “in the trenches” that compound to build success.

There are so many fascinating Michael Ovitz anecdotes that highlight why I think he is such an expert tactician. I just finished his memoir “Who is Michael Ovitz” which is full of situations where Ovitz and his team at CAA had a one-shot opportunity to dazzle a new client. They all have this common theme - wherein Ovitz hyper-analyzes how to win a single pitch that moves his business a step further.

One of my favorite anecdotes that encapsulates this approach to winning every moment is when Ovitz and CAA pitched for the Coca-Cola advertising business against its longstanding partner McCann Erickson:

We put together a team of in-house experts in movies, television, music, and books and we combed every Coca-Cola print ad and commercial since McCann Erickson got the account in 1955. Two weeks later we met in Atlanta with ten division heads, and then we broke into small groups for no-holds-barred discussion. The meetings lasted twelve hours, the best one-day education I ever had.

Source: Who is Michael Ovitz

Michael: [Coca-Cola] said, will you come do what's called a shootout? A shootout is where you come in and all the ad agencies pitch to the executives….

They set the date... for 11:00 AM on a weekday, and I flew our team of five people in the night before. We checked out the room we were going to do it in. We rehearsed what we were going to pitch, and we went for a very nice dinner. I bought everybody new Armani clothes and we got up the next morning. I'll never forget this. We went to the gym. We had a nice breakfast, we walked in at a quarter to 11:00, sat down, and we looked the part.

Ben: One would say it's a show of strength.

Michael: At 5 to 11:00 AM, the McCann people, there were 35 of them, drifted in having just gotten off a 6:00 AM flight from New York. I'll tell you they looked like they were all in a blender. They kind of looked like live smoothies. They were wrinkled, ratty, and all screwed up, and nice people too, but whoever sent them to fly in the morning made a critical error.

Anyway, they flipped a coin and they said they won. We were hopeful that… they'd go second because we wanted to go first and blow their doors off, and we did… We went in and we presented 35 ideas and we acted them all out. Each one ran for 30 seconds. We did it crisp and clean with music, with expression, with enthusiasm.

… We had these guys. I will never forget, I was sitting next to Bill Haber but next to me was John Dooner from McCann-Erickson and then a guy who had been friends with Don Keele for 35 years… This guy wrote Dooner a note. I saw it by accident and it said, we're screwed.

Sure enough, we finished, and then they came up. They were so bad that after seven or eight minutes of presentation, they sat down. We got awarded the account. Of the 35 commercials we presented, they approved every single one.

Source: Acquired Podcast

A recurring theme I see in exceptional commercial teams at early stage companies is that they treat every pitch like it is the Superbowl with a clear goal of how they want to make the customer feel. I’m not sure there is a better example of someone building a business brick-by-brick, client-by-client with that approach than Michael Ovitz.

Disclaimer: This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

I posed this question to ~40 of my wonderful friends. The recommendation could be in any category - a book, travel destination, restaurant, product, movie, activity, idea. BUT you can only pick one thing.

Below are the picks (edited a bit for clarity). This list is great inspo for your 2025 vision boarding. Thanks everyone for participating (p.s. if you have a highly recommends to add to this, email me!!)

Love you all and here is to a great 2025!

The Single Best Thing I Discovered in 2024:

• “The Highest Calling: Conversations on the American Presidency” by David Rubenstein

  • Sara O. says this book highlights aspects of the presidents that they didn’t teach in school, like their personalities and personal lives. It’s best as an audio book because of the interview format

• Japanese fruit and crème sandwiches on white bread

  • Kevin T. loves these little guys because they are so fresh, so fruity and so creamy 🤤

• The Outdoor Boys YouTube Channel

  • Per Peter F., “A Mormon father quits his tax attorney practice to make YouTube videos about survival in the wilderness. As dead internet theory pervades our algorithm at an accelerating rate, Outdoor Boys is so exceptional that for me, it is a sign of hope for humanity.”

• Sign up for a race! Walk/Run/bike/anything!

  • Steven N. insists that the race itself doesn’t really matter. “The weeks and months of training you’ll put into it will do wonders for your physical and mental health. We are goal oriented people [(Steven works in tech sales can you tell?)], and signing up for a race creates some self generated benchmarks to look forward to.”

• Full blood panel and biomarker testing by Function Health

  • Through the diagnostic tests offered by Function, Andrea M. learned that she had insanely high mercury levels from eating too much sushi, which other docs never suggested to check!

• “I Who Have Never Known Men” by Jacqueline Harpman

  • Cat G. says while this book sounds like it is going to explore a feminist dystopia, it focuses more on the universal need for connection

• Mexico!

  • Per Julia E., “Mexico it’s a huge country with diverse landscapes and rich culture. There is incredible natural beauty (Beaches! Desert! Jungle!) AND lively cities (Mexico City! Mérida! San Miguel Allende!). And so many more lesser known gems to explore (Puebla! Puerto Escondido! Valle de Guadalupe!). Mexico offers a chance to explore the world right next door to us. And also.. tacos.”

• Oaxaca, Mexico

  • Heather K. says Oaxaca has an amazing food and art scene and it is very walkable. Apparently there’s a rule that houses next door to each other can’t be the same color and there’s not a McDonalds in sight

• Carajillos

  • Continuing with the Mexican theme, Andrew T. describes carajillos as the Mexican take on an espresso martini. The Carajillo was the drink of the year in 2024 and should be a staple for years to come. I’ll co-sign this one in a heartbeat

• Bringing pickles to parties

  • According to Hayley G., “There is not a dinner party, housewarming, tv season finale watch party, etc that would not be improved by a container of fresh pickles. Everyone else is bringing wine/ cheese/ flowers, but you can really light up the room with some half sours (the best version is purchased at Zabar’s, but any non-vlasic option will do if the uws is not easily accessible). Plus, by bringing pickles to the party you also bring pickle juice which is a key ingredient for the best drink, pickle back shots!”

• Applying sunscreen every morning and retinol every night, no days off!

  • Emily C. says this is the best way to maintain your youthful skin. I'd like to co-sign this from the Sunshine State, I’m basically an Elta brand ambassador now ☀️

• The Coach Makers Arms in London

  • Mitch R. says that this pub in Marlyebone is one of his favorite places in London. Fantastic upscale pub food and welcoming whether you’re drinking a pint in a three-piece suit or a sweatsuit!

• Capsule Pharmacy

  • New father (!!! 😊😊) Chance R. is bullish on Capsule because with your prescriptions delivered right to your door you’ll never have to go to the pharmacy again. Co-signing this because I think my local CVS is haunted

• The “So True with Caleb Hearon” Podcast

  • Katie O. actually recommended this to me earlier this year and I can corroborate her opinion that this podcast is actually LOL-funny. Also Katie adopted a dog this year and would highly recommend that because the unconditional love is life changing

• The Substance

  • Mike G. says this campy horror soon-to-be-classic is a MUST WATCH because of the stellar performances

• “Smaller Acts” by Zach Bryan

  • This was Sean B.’s most played song of 2024. He says this song brings you the beauty in the benign and the romanticism of simplicity - all in 3 minutes!

• Fly Fishing

  • Jake G. says it’s an easy way to spend some zen time outdoors and you don’t have to worry about gross fish handling because nobody ever catches anything!

• The Perez Art Musuem in Miami

  • Zeev G. says this museum is great because it offers Miami views, cultural connection as well as plenty to do, including a restaurant and outdoor entertainment including, live music and movie nights. A Miami must for locals and tourists that stands apart from the rest. I personally hard agree with this pick, it is stunning

• The Penguin on MAX (HBO)

  • Juan R. thought the plot and acting were insanely good and it’s one of Colin Farrell’s best roles

• Za’atar

  • Sean E. can’t stop putting this stuff on popcorn and roasted cauliflower. It’s a sneaky-great, super-versatile spice mix often missing at home

• Sleepy Girl Mocktails

  • Nakita P. (my sister 🥹🥹) recommends combinations of pomegranate/tart cherry juice + an olipop or iced chamomile tea in a cocktail glass because it will put you to sleep before you know it. Bonus cozy points for pairing with a matching PJ set and a face mask

• At Home Sleep Study (w/ a BONUS pick of: Mandibular Advancement Device)

  • In Salem C.’s own words, “After nearly three decades, I can no longer pass off my snoring as a charming personality trait—because an at-home sleep study revealed I've been essentially waterboarding my brain with its own carbon dioxide every night. Now that I'm finally getting proper air flow, I'm transforming into what I imagine will be an unstoppable force of well-rested excellence.”

• The Purple Pillow

  • Lincoln Z. insists that the pillow is worth the price, he’s never slept as well in his life as he has using this cloud-like pillow. As a bonus pick since he couldn’t decide, he also insisted that you know that San Sebastian in Spain is a must visit because it is beautiful, has an amazing/vibrant culture, and is home to the most Michelin stars per capita in the world.

• Eight Sleep

  • Will P. thinks the Eight Sleep significantly improved his sleep quality. It cools your mattress in the summer and warms it in the winter. It automatically adjusts the temperature throughout the night to optimize for deep and REM sleep

• The Best Stovetop Rice Recipe

  • Abby N. used to hate making rice because it always got too dry, the bottom burned or it was flavorless! But then she discovered this recipe:

    • Rinse 1 cup of jasmine rice in a fine mesh sieve until the water runs clear.

    • Boil 2 cups of water. When water is boiling, add rice.

    • Once water comes back to a boil, add a teaspoon of salt and half a white onion (the onion is important)!

    • Turn heat to low, cover, and let simmer for 20 minutes.

    • After 20 mins, uncover and add 1 tablespoon of butter. Cover again until it melts, then uncover and fluff with a fork. Done!

• Visit Ireland

  • Parisa D. thinks Ireland is an incredibly underrated travel destination. It has beautiful nature/wildlife, hiking, small cute towns, lots of history, and she loved the inclusive/unique pub culture. The food/drinks scene is phenomenal and the people are the nicest she’s every encountered abroad

• Visit Iceland

  • Chrissy B. says it’s hard to narrow down a list of the best things to do in Iceland, there’s something for everyone in the Land of Fire and Ice!

• Slow Horses on Apple TV

  • Kashmira P. (my mom 🥹🥹) made this her pick because the show really demonstrates how people who might fail at something still have a lot of potential to offer under the right leadership. I strongly endorse my mom and her pick, all four seasons are 🔥

• Grace Bay Club in Turks and Caicos

  • Joe S. says that this resort in the Providenciales region of Turks and Caicos is the best resort he has ever visited. Beautiful beaches with the softest sand he’s ever felt, great food on site, and a quick walk to great restaurants and gambling at neighboring resorts on the island

• American history in two books - “How to Hide an Empire” by Daniel Immerwahr AND “American Nations” by Colin Woodward

  • Carl M. found these two highly influential books revealed how America became the global hegemon that it is today

• Padel

  • A double-recommendation!! coming from me and Adam W. Adam says it’s the sport of the future. I concur. It’s faster and more fun than Pickleball, has a very easy learning curve compared to Tennis, and is a great way to meet new people. If you’re ever in Miami, let me know and we will get a game in on the floating padel island 🎾

• Brouwerij 't IJ in Amsterdam

  • Beer connoisseur Jake T. says Brouwerij ‘t IJ (i have no idea how to pronounce this tbh) is a 300+ year old brewery in the heart of Amsterdam built next to a beautiful Dutch windmill that they used to use for grain milling. The beer and vibes are outrageously good. Absolutely must-stop if you’re in Amsterdam

• Queenstown, NZ

  • Samantha N. says this city is incredibly deserving of its title as “the adventure capital of the world.” There are so many different & unique ways to experience the awe-inspiring landscape of New Zealand… one of the only places where she “completely unironically cried tears of wonderment.”

• Smoking your own meats

  • Jonathan K. argues that the healthiness vs. flavor trade off is the best with smoking. No oil, butter or sugars required, and the result is the juiciest/best texture of any method to cook meats

• Visit the Dalmatian Coast

  • According to AJ L., the Dalmation Coast is an underrated travel destination rich with history, culture, great food/wine, and an abundance of islands and beaches

• The Nimble Kitchen Cookbook

  • Meagan W. says what sets this book apart is that it provides you with recipes that use things you might typically throw away! Carrot tops, apple skins, etc. Meagan is always trying to find ways to create less waste and this cookbook definitely helps you get creative with it!

• Daily Outside Time

  • Julia K. says she finally put some common advice into practice this year and found a lot of value in it! In her words, “Spending time outside, whether on a hike or on a coffee walk through the city, helps connect me to something bigger than myself and recenter.”

• Morning Pump Up Playlists

  • Reformed marching band enthusiast Christina M. thinks you absolutely need to have a morning playlist to get fired up for your day. Her mix of “a little bit of everything” provides a little source of motivation to get out of bed and get the day going

  Subscribe now

Thanks for reading. See you… next year 🤪!!

One thing that I often recommend is to flesh out a list of 3-5 “favorites” in three categories: businesses, executives, and investors. I think this is a good exercise for two reasons.

First, it helps you figure out if you have enough baseline curiosity about the people and levers that drive businesses to find a VC career engaging. Early-stage investing mostly sucks. The feedback loops are super slow and mostly negative. You can develop sharp perspectives but typically lack sufficient control to action those insights. You’re constantly trying to find non-consensus alpha when it seems like everyone around you is getting rewarded for indexing whatever prevailing trend is taking shape at the time. To power through the slog, you have to become a pure simp for capitalism, so that at least you might be able to relish the journey. This exercise is a good barometer of this specific type of curiosity.

Second, the exercise forces you to think in narratives. All investing is grounded in storytelling, and this is doubly true for early-stage investing which is data sparse to begin with. When you build a list of people and groups you really admire, it forces you to think about telling the story of why they are so interesting, which is a fundamental muscle you want to develop because it informs almost everything - sourcing companies, pitching them to your team, pitching them to potential partners/customers, recruiting talent for them, selling them.

So in essence, it’s about figuring out if you find the stories of company building interesting and if you want to tell those stories every day. With that context, the Mt. Rushmores of companies/executives/investors that you pick don’t have to be the most “successful,” just ones that you are fascinated by. And you don’t really have to be fascinated by every aspect of their existence - even 1-2 things that you really find fascinating about them is enough.

I get that it feels weird to fawn over the accomplishments of people that in a macro sense are sort of your peers. But I figure that most very successful people are students of their game - why shouldn’t you be a student of yours?

This idea has come up a few times for me in the last month, so I decided to put mine down on paper, mostly for my own benefit. Hopefully you find it interesting too. For my little Mt. Rushmore of each category, I tried to pick 3 “conventional”/well-known names and then 1 less obvious name to mix it up because I’m just so fun and random like that. I started drafting this out and it got super long, so I’m going to break each category into a separate post, starting with four companies I’m fascinated by.

My picks are as follows (drum roll please):

Very Fascinating Businesses: Costco, Epic Systems, Renaissance Technologies, Major Food Group

A little more on all of these follows.

I’d love to hear who you think merits Mt. Rushmore consideration if the concept resonates with you. Anyways, thanks for reading ✨✨.

Subscribe now

I. Businesses I Am Fascinated By

Costco

Fascinating because: Costco is a great example of how single-mindedness of purpose drives operational clarity, which in turn has enabled the brand to bleed into the identity of its customers.

Costco is my favorite company. There are so many aspects of the operation that spark joy on which a ton of ink has been spilled - market leading comp for employees, promoting from within, flexing its leverage on suppliers and passing on the savings to customers, the curated SKUs, the absurdly accommodating return policy, the treasure hunt experience, the $5 chicken!

The unifying thread across these operating principles is a fanatical focus on delivering customer value. This single-mindedness does two important things.

One, it clarifies every business decision. A sacred North Star provides a precise scale by which to weigh costs and benefits. In Costco’s case, the North Star is simply what will make the membership more valuable.

Second, this single-mindedness expressed over and over across so many facets of the business engrains this message into the minds of customers. And it is done so effectively that I think Costco achieves something very unique in business - it transcends into the customer’s identity layer. People don’t just shop at Costco, they are Costco Shoppers. Being a Costco member conveys something fundamental about people’s values - that they care about quality and value, ideas which have universal appeal, including and especially for Costco’s relatively affluent customer base. Very few brands achieve this.

Perhaps this is why the wholesale-core Kirkland Drip trend emerged - we are dying to show people that we’re not like regular consumers, we’re cool value-conscious consumers 😎.

Epic Systems

Fascinating because: Epic is an incredible example of building a dominant company “brick-by-brick.”

It blows me away how under-the-radar Epic is outside of healthcare circles. It is probably one of the most successful technology startups of all time. Its dominance is so prolific that Bill Gurley gave an entire talk complaining that Epic’s success surely must be the nefarious result of regulatory capture.

Epic, like so many successful startups, is wholly shaped in the image of its founder Judy Faulkner:

Faulkner, a computer programmer who founded Epic in a basement in 1979, has bucked convention from day one, starting with the company dress code: “When there are visitors, you must wear clothes.”

While competitors like Cerner (now Oracle Health), Allscripts (now Veradigm) and Athenahealth, grew by taking venture capital money, executing mergers and acquisitions, going public and then private again, Faulkner has eschewed any outside interference. This mentality is enshrined in the first two of Epic’s Ten Commandments, which hang on the wall in a campus bathroom:

1. Do not go public.

2. Do not be acquired.

…

Faulkner has retained all of Epic’s voting shares, which, upon her death, control will be shifted to a trust governed by four of her family members – her husband Gordon, 80, and three children – and five long-time senior manager employees at Epic. In line with Epic’s Ten Commandments, she says the trust rules prohibit an IPO, sale or acquisition, and stipulate that the next CEO “has always got to be a long-term Epic employee and a software developer.” A separate oversight board of up to three of Epic’s long-time healthcare customers will serve as trust protectors and “sue anyone who doesn't vote according to the rules.”

Source: Forbes

In a market characterized by mass consolidation via M&A, Epic has never made an acquisition. Not one. The graphic below comparing the origin/history of Allscripts/Veradigm vs. Epic highlights this insane dynamic:

This in-house, full control mentality is central to what Faulkner says sets Epic apart from its competitors: Since Epic has never acquired another company, it hasn’t had to build bridges to interface from one type of software to another.

Source: Forbes

Mapping out Epic’s fully-integrated ecosystem further emphasizes the point - as illustrated by Brendan Keeler:

Epic's monolithic architecture is truly a wonder to behold, allowing for tight integration between different modules and ensuring data consistency across the entire system. Notice in the diagram how all applications and modules are built on that same underlying database (Chronicles), using a shared data model:

…Epic's dominance in healthcare IT is no accident. Moreover, it’s not the result of nefarious covert operations and market manipulation. It's the result of a carefully crafted strategy that leverages integrated systems, network effects, a weak and diminishing pool of competition, and a deep understanding of healthcare's unique needs. From its monolithic architecture to its community-wide interoperability through Care Everywhere, Epic has built a system that's more than the sum of its parts.

Source: “An Epic Saga: The Origin Story” by Brendan Keeler / HealthAPI Guy

In short, Epic’s success is not regulatory capture - it is a masterclass in developing durable moats and network effects through thoughtful strategy, focus, and execution. The dedication required to ignore the noise and keep building brick-by-brick is truly remarkable.

Renaissance Technologies

Fascinating because: Renaissance Technologies (“RenTech”) is a fascinating case study on the power of compounding small advantages over time and incentive design for human capital.

RenTech merits fascination based on performance alone - the sheer consistency of market-beating returns from its flagship Medallion fund is mind-blowing:

I have RenTech in my fascinating company bucket rather than the fascinating investor bucket because it is the system/organization that I find so uniquely compelling vs. the raw investment approach.

First, if we take founder Robert Mercer at his word, the Company’s outperformance illustrates a really simple insight that applies broadly to basically all of life: a slight advantage compounded over a sufficiently large number of iterations yields significant returns. That’s it, that’s the whole model.

I find this concept existentially soothing. You don’t have to win every time or dominate every single situation (you couldn’t if you tried). Rather, the key to success in so many disciplines from professional investing to professional poker1 is to build a slight edge and then just give yourself enough reps/iterations/time to have the odds compound in your favor, outlast competitors, and deliver big outcomes.

The second thing that is super interesting about RenTech is how they have architected a system of financial incentives to keep new generations of talent hyper-motivated to continue delivering this performance. Ben and David from the Acquired podcast make the case really well in their episode on RenTech:

David: … I think the third puzzle piece of what makes RenTech so unique and defensible is Medallion’s structure itself, that it is an LP-GP fund with 5% management fee and 44% carry.

…

David: It’s all themselves, it’s all insiders. Why do they charge themselves 44% carry and 5% management fees? I think Jim [Simons] talks about this that, oh, I pay the fees just like everybody else.

… What is happening here? Okay, here’s my hypothesis. This is not about having crazy performance fees. This is not about having the highest carry in the industry. This is a value transfer mechanism within the firm, from the tenure base to the current people who are working on Medallion in any given year.

Here’s how I think it works. When people come into RenTech, they obviously have way less wealth than the people who’ve been there for a long time, both from the direct returns that you’re getting every year from working there and just your investment percentage of the Medallion fund… If you work there, your 401(k) is the Medallion fund…

Given that, though, how do you avoid the incentive for a group of talented younger folks to split off and go start their own Medallion fund?

Ben: Especially when they all have access to the whole code base. The whole thing is meant to function like a university math department, where everyone’s constantly knowledge-sharing because we’re going to create better peer-reviewed research when we all share all the knowledge all the time. You would think that’s a super risky thing to give everyone all the keys.

David: … Let’s round that up and just add them and say 49% [5% mgmt + 44% carry] of the economic returns in any given year go to the current team, and 51% of the economic returns go to the tenure base.

What is the equivalent here? I think it’s like academic tenure kind of thing. The longer tenure you are at the firm, the more your balance shifts to the LP side of things…

Ben: Essentially David, the real magic is they’ve got one fund, it’s evergreen, and when you start at the firm, you’re only getting paid the carry amount, but over time you become a meaningful investor in the firm and you shift to that 51%. You’re the LP, and then over time you eventually graduate out entirely and you’re only an LP.

Source: Acquired

Incentives design is so critical to investment structuring and scaling an organization. As the great Mungerism goes - show me the incentive and I’ll show you the outcome. RenTech is a great example of utilizing incentives design to protect its moat.

Major Food Group

Fascinating because: Major Food Group (“MFG”) illustrates the power of extreme attention to detail to cultivate hospitality, a concept that is fundamental to all commerce.

The success of MFG is particularly interesting to me because it has been achieved in a conventionally terrible business - restaurants. The MFG empire now spans the globe with dozens of well-executed concepts with dedicated followings. They’ve started membership clubs where eyewatering fees yield members the perk of getting preferred access to other MFG restaurants so they can give MFG even more money. They’re even building a branded residences here in Miami. Soon you can literally live at Carbone (personally I’ve been yearning for my nights to be more Dionysian).

Why are they so successful? How are they getting people to gladly fork over $50 for a little spicy pasta?

When you ask people why they enjoyed dining at Carbone or another MFG restaurant, people often talk about “the vibe” just being awesome. I think that’s important. Studying MFG’s approach, “the vibe” is actually the culmination of an extreme, systematic attention to detail iterated across the entire customer experience. In this interview from 2021, Mario Carbone talks about his approach to opening The Grill in midtown NYC which emphasizes this hyperfocus (starts at 30:45):

This is such an important idea I come back to a lot. I think basically every single business is a hospitality business. Even the technology and healthcare services business I invest in at work are fundamentally about hospitality. I’ve seen over and over again that customers don’t really buy products and features, they buy trust in a counterparty that makes them feel understood and is energized to solve their problem. Delivering that feeling, “the vibe,” to customers across every touchpoint seems to be what sets great sales teams apart, especially in the early stages of a Company’s commercial life. Thus, I’m super fascinated by groups like MFG that have systematized the vibe.

You’d best believe in the feelings economy folks, you’re in it 💫🪐✨.

If you spend even 30 seconds learning about the present state of the biopharma research industry, you will undoubtedly hear: “80% of clinical trials fail to meet enrollment targets on time.”

This frequently parroted statistic offers an unfair characterization of the clinical research industry.

Rather, the fact that research stakeholders - including pharma and biotech companies (research “sponsors”), contract research organizations (CROs) and research sites - do successfully enroll 1 out of 5 trials in light of the multiple layers of complexity that they have to deal with is a miracle.

It’s like MLB batting averages - numerically low success rates can still be impressive when the underlying thing is hard to do. And research stakeholders are doing the equivalent of stepping up to the plate blindfolded with their arms tied together and a pool noodle for a bat. Oh, and they’re on fire.

Why are modern trials so complex? For starters, we could say that clinical research just is inherently complex. Consider a few of the deeply intertwined activities that take place across the life of a trial:

• Pharma and biotech sponsors must design trial protocols that are specific enough to support the clinical goals of the trial, but flexible enough to accommodate recruitment targets and proactively avoid protocol deviations in deployment

• Trial sponsors need to decide which sites to partner with and where/how to deploy the trial

• Patients need to be identified, matched to trials, screened and enrolled

• Participants need to adhere to the trial activities delineated in the protocol, often in spite of hurdles like transportation and costs

All of these activities are intertwined - e.g. poor site selection could lead to delays in meeting enrollment targets, poor protocol design could lead to expensive and time-consuming amendments down the line.

On top of the inherent hurdles to running a clinical trial, I see three prevailing layers of complexity at work:

• Clinical complexity

• Operational complexity

• Market complexity

What alarms me is that across these three areas, signs point to clinical trial complexity increasing in the coming decade.

At the same time, increasing biopharma research complexity should serve as a tailwind for emerging clinical development technology companies. Specifically, companies that deliver technology in a way that actually reduces site/sponsor burden have promising outlooks. As an investor in the space, I’m optimistic about what the next decade of clinical research infrastructure “modernization” could yield, to say nothing of the transformative advancements in human progress that better infrastructure could unlock.

More on the drivers of this complexity wave and potential advancements towards modernized clinical research infrastructure follows below. As always, thank you for reading and drop me a line if this is interesting to you (📧 vpradhan@sopriscapital.com).

Clinical complexity

I think of clinical complexity as the challenges associated with designing a scientifically meaningful trial.

Data suggests that trial designs are becoming more complex. For example, IQVIA tracks a multi-variate complexity “index” metric which serves as a proxy for industry-wide trial complexity. This index suggests that trial complexity has steadily climbed over the past decade.

One key driver of protocol complexity is the number of endpoints being measured in the trial. For example, a study from Tufts University’s Center for the Study of Drug Development (CSDD) reports that the average number of endpoints measured in a Phase 3 trial has increased 37% since 2015.

Why are trials measuring more stuff? Some reasons include:

• Increased focus on capturing patient experience and quality of life measures as relevant trial endpoints

• Increased focus on precision medicine, which requires more endpoints as sponsors seek to stratify participants based on biomarkers, genetics and other factors

• Increased use of composite endpoints, where multiple primary endpoints are combined to measure outcomes

The focus on these drivers is likely to intensify.

For example, patient experience and quality of life measures will be critical as pharma and biotech companies navigate increased price scrutiny, as positive data in these areas can greatly strengthen the positions of biopharma companies in price negotiations. As sponsors seek to make more progress in precision medicine, particularly in areas like oncology, the need for granular patient stratification will increase.

Another driver of clinical complexity is eligibility criteria for a trial. Here, we see additional evidence for increasing complexity. For example, industry-wide there is a major focus on driving greater diversity in trial participant populations. This is great, because poor representation in clinical research has historically led to major issues and consequences for patients.

But increasing representation in research appears to be much easier said than done - whereas discussion around greater diversity in trials has been prevalent over the past decade, data suggests that Black and Hispanic trial participation rates have actually declined over that period.

While challenges around representation extend far beyond protocol design, recent FDA guidance suggests that eligibility criteria (even if ultimately broadened to include more participants in line with the guidance) will require greater scrutiny.

Across multiple vectors, the clinical complexity involved in designing a trial protocol is trending up.

Operational Complexity

As illustrated in the graphic below, clinical trials face a multitude of logistical challenges.

An emerging theme in clinical trial operations is the idea of making trials more “patient-centric” so that it is easier for patients to participate in trials.

“Traditional” site-based trials present several logistical hurdles to participation for eligible patients. For example, according to Antidote, an estimated 70% of potential clinical trial participants live more than two hours away from a study center. This is a huge problem - if it’s hard for patients to get to trial activities, recruitment and trial adherence declines.

In recent years, fully decentralized or virtual trials (DCTs) emerged as a potential answer to the patient-access challenges of site-based trials. The idea was that if trial activities could be coordinated outside of the confines of a research site, then the geographic reach and success rate of trials should increase.

Unfortunately, as many folks in the industry will have seen firsthand, the promise of fully decentralized trials did not really pan out, in no small part due to the logistical issues of delivering a successful trial. For example, a recent survey by Health Advances found that clinical research executives “tend to be disappointed by DCTs’ ability to increase patient retention because the reduction of in-person visits can make it more challenging to identify and course-correct non-compliant patients.”

But despite the areas in which fully-decentralized trials fell short of expectations, executives seem to acknowledge that the future of trials will be “hybrid” and will entail more decentralized activities.

Decentralizing the activities of a trial creates new challenges. For example, questions surround how to collect and manage data from patients outside of the clinic, how to create touchpoints with clinical providers when necessary, and how to manage distribution of trial supplies. In effect, a focus on patient-centric trial design has created a last-mile delivery problem for research stakeholders, who have been left to figure out how to cobble together various assets (e.g. wearable devices, mobile clinics, in-home visits) to assemble a viable trial.

In addition, decentralized trial activities could create new data management problems for trial stakeholders.

“A remote or virtual trial introduces other solutions that are now sources for data, and all of that has to be integrated and managed… Somewhere along that chain of support in executing a protocol will be people who are charged with ensuring the quality and integrity of the data that’s gathered… It’s pretty remarkable how many different sources are now contributing data that have to be curated, compiled and then turned into an analysis data set.”

Source: Centerwatch

Great advancements have taken place over the past several years in developing clinical-grade digital measurements that can be used in trials. This is exciting, because it opens up the frequency and types of data that could be collected from patients, making research outcomes more robust. But deploying those measurements and effectively managing the data is yet another area where research complexity is likely to increase.

All told, operational complexity looks to be trending up too.

Market Complexity

At a high level, the clinical research market seems like a straightforward two-sided market.

Like many two-sided markets, greater fragmentation means more complexity. And clinical research sites are highly fragmented.

So if you’re a pharma company trying to connect into a specific site (or vice-versa), the landscape looks more like this.

Oh and by the way, you have little to no visibility into the patients at the sites, making it hard to decide which sites to engage. Most sponsors have to figure out which sites to engage with non-clinical performance proxies at best and pure guesswork at worst.

Fragmented landscapes often trend toward consolidation - the resulting scaled operators theoretically benefit from efficiencies and economies of scale. Sure enough, on both the supply side (sites) and demand side (trial sponsors), forms of aggregation have certainly emerged - e.g. site networks and CROs, respectively. In theory, this should reduce complexity because it reduces the disparate touchpoints between the two sides of the market.

But this is only the case if supply (different sites) and demand (different trials) can be effectively integrated. In practice it looks like the aggregators have simply ingested the complexity themselves, with efforts to actually reduce complexity through integration falling short.

For example, private equity backed site-network platform Cenexel reportedly tested the market for a potential exit earlier this year, seeking a >$1bn valuation. Instead, according to Axios, the platform met resistance from prospective buyers due to insufficient integration:

“…prospective buyers are concerned about CenExel's business not being integrated enough… One source says CenExel hasn't invested enough into its technology backbone, with the business having "seen lots of EBITDA adjustments."

Source: Axios

Why is this complexity-reduction-via-integration approach so difficult in practice?

Take an example of a site network trying to understand the patient census spread across its different sites. An emerging trend in the industry is the recognition that patient electronic health record (EHR) data could have valuable insights for research stakeholders. Theoretically, a site network should be able to access the EHR data of its patients to determine how many patients they have that are eligible for a given trial protocol. But individual sites in a site network often have disparate systems - e.g. sites utilizing different EHRs that do not connect to one another easily.

This makes it nearly impossible to have the kind of real-time, single source of truth that would streamline the market complexity of clinical research. This means that consolidated groups can actually be more complex to run than if you ran a standalone site with a single set of systems.

This matters in particular because M&A consolidation of research sites has exploded in the last decade.

In many ways, clinical research sites are ripe for the private equity playbook:

• Acquire sub-scale assets with predictable/proven financial profiles at attractive multiples

• Drive financial improvements through the asset base via enhanced operations and economies of scale

• Achieve multiple arbitrage through enhanced scale at exit

The highly fragmented landscape dominated by standalone research sites suggests that there is still massive runway for continued consolidation.

But if these roll-up platforms cannot adequately integrate assets, we can expect the prevailing market complexity to persist, if not compound. On the other hand, this suggests that PE-backed platforms that do leverage technology to integrate their networks will be positioned to maximize value at exit, and should prioritize this as part of their value-creation strategies.

Complexity as a Tailwind for Technology Adoption

This dynamic of rapidly increasing complexity in clinical research should set up the playing field in favor of emerging clinical development technology companies that promise to help research stakeholders simplify this chaos.

And yet, in speaking to sites and sponsors, a paradoxical view emerges in which there is simultaneously:

• Whitespace for technology adoption (tons of “pen-and-paper” work taking place)

• Technology vendor fatigue burdening research stakeholders

Here’s a great perspective shared by Jeff Kingsley, Chief Development Officer of Centricity Research:

As protocols continue to grow more complex, scaling requires greater specialization and increasingly sophisticated tech stacks. Jeff explains, “We still have lots of (tech) players in the space, which makes life harder for the site. More usernames and passwords are just the tip of the iceberg. The bigger issue is that when you have to be 'adept' at using numerous platforms, the reality is that quality and efficiency suffer because you can’t be proficient in 14 different systems.” It seems that a major challenge is the lack of good guidance.

Lack of guidance spills into every task, creating non-value added work, without regard for the reality that sites run more than one study at a time. Jeff illustrates, “I get messages like, ‘I need you to sign on to EDC and sign off on blah, blah, blah patients,’ without specifying the trial or EDC platform. They assume you’re only handling their trial. With 600 concurrent trials, this is a huge time-waster. I have to sift through emails to figure out what they’re referring to, which adds unnecessary waste to the process.”

Source: ProofPilot* Blog

A look at the startup landscape across clinical development chain supports this view - while dozens of startup companies have emerged in recent years to solve pain points for research stakeholders, the current market is populated with a host of disparate point solutions, occupying adjacent spots on the value chain.

The view I have come to about opportunities in the clinical development technology space is that it is actually less about what you do and way more about how you do it. Meaning, companies need to actually reduce the complexity burden for their customers to generate strong, sticky traction. And this might mean doing a lot of “things that don’t scale.”

For example, while research stakeholders might theoretically be able to generate value from technology on a self-serve basis via great product design, channel checks with customers suggest that they often have a lot of competing priorities, which makes it hard to fully realize value from tech on their own. My sense is that technology companies that embed a robust services layer into their offering to ensure customers actually realize the value from the product will be really well positioned to win share in the market. This is a case where startups may need to trade gross margin for adoption, speed-to-value and retention.

What might successful startups look like at scale?

In the near to medium term, I think there are 9-figure value opportunities to build the “glue” for supply and demand aggregators to reduce complexity via better network integration.

Over the long term, I think there could be a 10-figure or even an 11-figure value opportunity to build some kind of tech-enabled middleware platform connecting the two sides of this marketplace. Maybe this kind of platform even disintermediates the incumbent supply and demand aggregators over time.

Whether these platform opportunities emerge as the result of consolidation of existing startups or from the emergence of a purpose-built, end-to-end platform, or a combination of both, reducing complexity for biopharma research stakeholders is a very promising market opportunity, and one that I am closely watching as an early-stage investor.

Ultimately, science is hard. In the absence of some world-changing discovery catalyst, every marginal scientific breakthrough becomes inherently harder to achieve (the oft-cited “Eroom’s law” of drug discovery). But entrepreneurs who can simplify the multi-variate complexity that plagues research today for pharma companies, research sites and most importantly - patients, will find themselves rewarded while accelerating this critically important engine of innovation and human progress.

* Denotes a Sopris Capital Portfolio Company

Disclaimer: This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

Next week, Apple is officially launching its new Vision Pro mixed reality (MR) headset. While it is not the first MR device to hit the market (e.g. Microsoft’s HoloLens has been out for years), for many people, Apple’s device might be their first foray into “extended-reality” (XR).

We will have to wait and see if the Vision Pro is Apple’s next great consumer hit. In the meantime, there are a few emerging use cases for MR in healthcare - specifically, various flavors of medical education and clinical training - that I think are worth paying attention to.

What is Mixed Reality?

MR is different than virtual reality (VR). The main difference is that with MR, devices project interactive visual elements into the real-world environment whereas VR provides full immersion into a fully virtual environment.

As an aside, I think most companies in this space will likely utilize the nomenclature of “XR” as it becomes a catch-all term. Also, X is a way cooler letter than M (see also: X-Games, Vin Diesel’s XXX, the short lived but electric XFL, X gon give it to ya).

The primary advantages of MR vs. VR are:

• Compute - for both MR and VR, all of the compute required to generate the experience takes place within the device. While device compute will almost certainly increase with time, there is always going to be some limit to device compute power, which will present tradeoffs. With respect to rendering environments, MR has a strong advantage over VR in that with MR, you don’t need to allocate precious compute resources on rendering the background environment. With VR, you do need to spend compute resources on the background to create the fully virtual field of view. The impact of this is that objects rendered on a MR device can be way more detailed than they would be on a VR device with equal compute, which is important for some of the use cases explored below.

• User Experience - a common experience for people putting on a VR headset for the first time is cybersickness - i.e. nausea and disorientation. MR typically induces less cybersickness for users. There are a couple of interesting reasons for this. The sheer act of overlaying content onto the real world reduces disorientation for the user because it is easier for the user to ground themselves in the physical world. MR typically incorporates a wider field of view, which also reduces user disorientation. Further, MR headsets typically utilize eye tracking and adjust the projected images accordingly, which reduces nausea-inducing visual conflict. Solving for less user nausea might unlock greater adoption, given how often cybersickness has been cited as a barrier to widespread usage of VR.

MR x Healthcare

A couple of years ago, I wrote that I thought that the most intuitive applications of “metaverse” technologies in healthcare are in the enablement of “immersive learning experiences”:

Metaverse use cases in healthcare would likely take advantage of the ability to:

• Create deep immersion via richly rendered 3D spaces

• Synchronously convene with both real peers and programmed entities in those spaces

• Interact with co-inhabitants via programmable inputs/outputs

• Simulate activities/events/situations that are rare or dangerous in real life

• Transform one-to-one experiences to one-to-many experiences

The confluence of the capabilities above unlocks a core power of the metaverse which I would call immersive learning experiences. The prospective healthcare use cases … enhance our ability to acquire skills/knowledge from difficult to access sources (due to geographic or supply constraints) in a memorable, impactful way.

The areas where I am seeing this idea of MR-enabled immersive learning experiences play out most prominently so far are in:

• Anatomical Education

• Clinical Simulations

• Surgical Planning and Operation

• Behavioral Healthcare

I’ll cover the first three in greater detail below; I think the utilization of MR for behavioral healthcare is part of a broader discussion around mental health digital therapeutics and I have not studied this as closely. As always, feel free to let me know your thoughts or in a comment or an email!

Abra Cadavera 🪄

Medical education already relies on multiple forms of immersive learning experiences.

For example, in post-secondary and medical school education, students take anatomy courses. A common standard teaching instrument for anatomy education in medical schools is the cadaver lab.

Work in the cadaver lab involves dissecting and labeling body parts on a body donated to science. This has actually been a standard of education for centuries. And yet, cadaveric dissection has multiple drawbacks:

• It’s hard to see specific organs and systems. Embalmed cadavers have compressed organ presentation and the grey-ish color of a cadaver obscures distinct body parts that would typically be more visually distinct in a living person. Further, it’s impossible to see certain systems, like the circulatory or nervous systems, in a cadaver. The presentation of any particular body may be so distinct from the anatomical standard that it can convolute the student’s understanding of how the body systems could look in another person.

• It’s expensive. Cadavers can cost $2-3k per “unit” and require significant storage infrastructure. Particularly in urban settings, it can be very expensive to build and maintain adequate storage capacity for cadavers. Some medical schools report difficulty in sourcing cadavers at all due to a variety of supply chain issues.

• It's time consuming. Cadaveric dissection takes a long time which limits the amount of content medical schools can pack into curricula. Further, students have limited time to spend examining cadavers (you can’t just bring the cadaver home with you lol).

Most of the proponents of cadaveric dissection that I have spoken to defend the practice primarily based on the idea that it is a rite of passage. The idea is that working with a body that somebody donated to science can lend a sense of gravity to the work students are doing at medical school. I think that’s a fair claim, but it suggests that the true value of traditional cadaveric dissection is limited. Thus, it’s no surprise that academic institutions are increasingly looking to adopt digital anatomy training tools.

For example, one digital tool to teach anatomy that has achieved decent traction in the last several years is the Anatomage table. Anatomage tables are basically ginormous iPads which enable students to view a full body scale digital rendering.

The advantages of the Anatomage table are primarily related to the digital nature of the form factor. These devices enable students to view life-size renderings based on a deep library of anatomical assets. This is a step up from traditional cadavers.

However, one of the major pitfalls of the 2D form factor is the inability to sufficiently depict depth. This is where 3D visualizations are a major level up and where mixed reality offers a leap forward in capabilities. Also, like a traditional cadaver, you can’t take this giant iPad home with you.

With MR platforms like HoloAnatomy by AlensiaXR1, students can literally poke into the body which enables a deeper understanding of how body systems work together. The technology is mobile and enables a far more convenient experience for students seeking to master anatomical content.

Early adopters of MR technology in anatomy education indicate that students absorb the content much faster, which enables institutions to enhance the curriculum by incorporating additional programming.

Just like the simulations

Another widespread component of medical school curricula are patient simulations with “standardized patients”. This is where medical schools bring in actors to portray patients with particular conditions (the famous Seinfeld scene comes to mind).

I’m rehashing this point from my last article because I think it bears repeating that one of the most exciting potential use cases of MR is in advanced clinical simulations. For example, the University of Illinois uses GigXR in their medical school education program to train students in “real-world” scenarios.

As I also discussed in my last post, these simulations are only going to get more advanced as advancements in LLM’s enable life-like conversations with simulated entities. For example, check out the capabilities from Convai (the group that partnered with NVIDIA in that demo I linked in the last post).

The possibilities blow my mind. Dropping these emerging AI capabilities into patient simulations could significantly increase the value prop of MR simulations, further enhancing their value as tools for clinician training.

MR in the OR

Another setting showing promise for mixed reality technology is surgical operating rooms.

Mixed reality enables a patient’s medical imaging to be visualized in 3-D to assist in surgical planning. Companies like Augmedics, OssoVR, ImmersiveTouch, Medivis and MediView offer platforms that could transform the standard of surgical care, improving pre-operation planning/training and remote collaboration to reduce the risk of procedural errors. In the future, these technologies might assist providers during the actual procedure navigation.

These kind of tools are going to become increasingly important as surgeries, medical devices, and pharmaceutical products become increasingly complex. The WSJ published a great piece highlighting this trend last year:

A major obstacle looms for the drugs of the future. Not enough doctors know how to administer them. 

For just one rare neurodevelopmental disorder, known as Angelman syndrome, clinical trials are testing four cutting-edge therapies. Twenty more research programs are under way and could yield treatments ready to move into human testing in the next several years, according to Allyson Berent-Weisse, chief scientific officer of the Foundation for Angelman Syndrome Therapeutics, or FAST, and the mother of a child with the disorder.

These promising therapies include antisense oligonucleotides (or ASOs) which affect how proteins inside cells are made; Crispr gene editors that can cut, rewrite, or replace faulty genes; and other novel approaches.

But there’s a hitch. Some of these drugs, as well as therapies in development for other conditions, are administered through lumbar punctures, which aren’t a routine part of doctors’ daily practices. They may involve infusions of a gene packaged inside the shell of a virus, which can lead to potential immune complications and can require close monitoring of patients. Some drugs need to be stored in special conditions, which can take complicated planning and coordination that few doctors have experienced.

“The bottleneck used to be the science and finding drugs,” says Dr. Berent-Weisse. “The new bottleneck is having enough people trained to give them.”

Source: WSJ

Groups like Gilead and Roche have developed training centers to teach physicians and surgeons how to execute some of these complex procedures, training that should certainly yield greater adoption of their products and better care for patients. Integrating MR into these efforts could help accelerate learning curves on these breakthrough therapies.

the medicine of toMoRrow

An overarching question I have about adoption of MR in medical education is around urgency. What is the “why now” that drives increased usage of this nascent technology by customers? After all, VR has been around for a while and certainly has yet to see widespread uptake.

My non-scientific prediction is that many people will head to the Apple store at the mall at some point this year and try on a headset; if they recognize that MR is a big step forward in terms of usability, it might get them to start thinking about how they can incorporate this form factor into their professional work.

But considerations around what the future of medicine will look like might be a sufficient catalyst as well. Consider that a pre-med student entering university this fall is actually going to be practicing medicine as a physician like 10-15 years from now. Thus, educators have to think about investments on decades-long time horizons. And if the prevailing view becomes that MR is going to be a key technology in the delivery of care in future decades, then there will be a greater urgency to get this technology into the hands of aspiring clinicians today.

Moreover, the clinical talent shortages that plague healthcare (which I cover a lot here) will likely also drive greater adoption of MR technology for medical education. This is because as the system contends with more acute physician shortages, care delivery will naturally be forced to decentralized. As care delivery decentralizes, medical education and training will need to decentralize too.

I speculated a bit in my piece on the rapid growth of NPs and PAs within the clinical labor force that “on the job” clinical training is going to become more and more important, particularly as physician shortages inevitably increase the clinical scope of groups like NPs and PAs. Leveraging technology like clinical simulations can help augment the experience set of practicing healthcare workers of all types in support of this expanding scope.

Just this week, the Bloomberg Philanthropies announced the launch of a $250mm investment into expanding medical vocational training for students as early as high school as part of an effort to combat talent shortages.

As the healthcare industry seeks to attract prospective talent, technology will play an important role in engaging, training and uplifting that talent.

The search for a catalyst, of course, is entirely moot if this new slate of MR devices turns out to be so good that they replace computers as the primary way we engage with the digital world (and maybe even the physical world). In that case, you are all cordially invited to my birthday party this summer in the metaverse as we take yet another leap towards becoming those hover chair people from Wall-E.

Share

Subscribe now

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

Juniorization referred to a perceived recomposition of the Wall Street workforce as younger, less experienced staff replaced older, more experienced staff.

When I first heard the term, I simply shrugged it off into the recesses of my mind and returned to devouring the remains of the septuagenarian managing director whose job I stole with my robust powerpoint logo arranging skill set.

But recently, I have been thinking about how “juniorization” is an apt description of a broad trend taking place across healthcare.

As the healthcare industry contends with a widespread shortage of physicians, so-called “mid-level” providers, like nurse practitioners (NPs) and physician assistants (PAs) are playing an increasingly important role in the provision of care. For people in healthcare deserts, like rural regions which span 97% of the US by land, mid-level providers serve as an absolute lifeline by representing what is often the ONLY clinical care readily available.

I am super interested in this trend, having seen the impact that NPs and PAs are having across various clinical settings. Many emerging care delivery models that we evaluate at Sopris Capital rely on these clinicians to provide services cost-effectively. We’ve invested in multiple companies where mid-level providers are critical to operations, like TapestryHealth (senior care) and Mindful Care (mental health urgent care).

Continued below is some discussion on why this trend is taking place, and where I see some potential challenges/opportunities for new companies to help support a “juniorized” workforce.

There’s an APP for that

The reason this trend is taking place is simple: healthcare as an industry is being asked to do more, with less. Demand for healthcare services continues to grow as the US population ages. While the physician workforce appears to have entered an era of chronic supply shortages, the labor pool of mid-level advance practice providers (APPs) is growing rapidly. Naturally, more volume will have to shift to where the labor actually exists in order to meet demand.

The physician workforce is quite old. For example, according to a 2021 survey, ~47% of physicians across all specialties are aged 55+, with some specialties exhibiting alarmingly high ratios1.

Meanwhile, there are not enough new doctors entering the workforce to replace aging docs as they retire. Several factors contribute to this, including long and expensive education requirements, limited residency slots, barriers to in-sourcing immigrant talent, over-indexing on specializations, and historical moratoriums on new medical schools.

Against this backdrop, APPs provide tremendous support to healthcare’s labor infrastructure. APPs are trained and scoped to deliver significant care at significantly lower cost than physicians.

A lot of physician groups will get upset and say, “Well, you know that an APP is not a physician,” and I will never say that they are. They don’t have that level of training, but the data shows they can do 85% of what a physician can do. The other 15% should be done by a physician and by elevating the complexity of what the physician does in the course of their daily work by offloading a lot of that other 85% to the APP, then we have everybody working to the top of their scope and we still increase access.

Source: Health Leadern interview with Allison Dimsdale, DNP, NP-C, AACC, FAANP

Unlike the physician labor pool, the labor pool of APPs is growing rapidly.

The combination of streamlined education requirements and an attractive ROI for prospective PAs or NPs all but guarantees that the growth of the APP labor pool will persist.

We project that these trends will continue through 2030. The number of full-time-equivalent physicians is expected to continue growing by slightly more than 1% annually, as increased retirement rates are offset by increased entry, whereas the numbers of NPs and PAs will grow by 6.8% and 4.3% annually, respectively. Roughly two thirds (67.3%) of practitioners added between 2016 and 2030 will therefore be NPs or PAs, and the combined number of NPs and PAs per 100 physicians will nearly double again to 53.9 by 2030. These shifts will probably be even more pronounced in primary care, where physician supply has been growing more slowly than in other fields and NPs tend to be more concentrated.

Source: Growing Ranks of Advanced Practice Clinicians

Bridging the GAPPs

This ongoing recomposition of the clinical workforce poses multiple challenges and opportunities. I think that the most significant pain points are in: recruiting & retention, training & development, and compliance & billing.

Recruiting and Retention

With clinical workforces acutely strained during the pandemic, technology-enabled healthcare staffing platforms & marketplaces garnered significant interest and funding over the past few years - e.g. IntelyCare, ShiftMed, ShiftKey, Nomad Health.

However, based on a quick scan of online NP/PA forums/blogs and talking to my NP/PA friends, the most common resources to find full-time APP roles are still traditional recruiters and industry-agnostic job boards like Indeed and LinkedIn. This implies that there might be room for dedicated recruiting resources for APPs, particularly tools that enable APPs to manage their own careers.

Training and Development

Given that APPs by definition enter the workforce with less experience, training should be a key focus area for organizations seeking to maximize their APP talent.

Robust training offers both a differentiator in recruiting talent and a tool to ensure that recruited providers are able to deliver high quality care. For example, Mindful Care has developed a thorough monthslong onboarding curriculum which helps new-grad psychiatric NPs ramp up in their roles within the organization.

Many forms of continuing education tools could be relevant in elevating APP talent. One emerging area that I am particularly interested in is next-generation clinical training simulations. In speaking to practicing physicians, they express consternation around the fact that APPs have substantially less clinical exposure than physicians. A lot of “good medicine,” they argue, is applying hours of experience to understand the right way to dialogue with a patient: asking questions and interpreting the patient’s feedback appropriately. This is an area where simulation technology could come into play.

Simulations offer clinicians the ability to receive feedback and guidance at an unprecedented scale. For example, medical schools and universities are already utilizing mixed-reality hardware powered by clinical simulation software like HoloPatient by GigXR to put students into real-world scenarios. The technology represents a substantial upgrade to the current standard at med schools of hiring paid actors to represent standardized patients (sorry Kramer).

These scenarios can be infinitely customizable, repeatable and scalable, advantages to even real-life patient encounters. And these simulations, which are still in the earliest innings of development, will likely become even more powerful over time as they begin to incorporate next-generation AI capabilities.

I would not be surprised if the adoption of simulation tools accelerates as organizations seek to develop their non-physician provider teams.

Credentialing and Billing

Scaling an APP based workforce introduces complexity around licensing, credentialing, compliance and billing.

States have differing regulations around the autonomy with which APPs can practice, with varying degrees of physician supervision required. Beyond state regulations, individual payors can have their own requirements around physician supervision. Navigating this complexity while seeking to maximize revenue opportunities (e.g. leveraging “incident-to” billing) creates a significant administrative burden.

Relieving the administrative and compliance burden for organizations deploying APP talent is an interesting opportunity space for new companies. For example, Zivian Health offers a platform to streamline APP/MD collaboration while helping organizations scale their supervising MD capacity. These kinds of resources should enable organizations to devote their time and attention to core operations, instead of managing bureaucratic bloat.

Insatiable APPetite

In closing, the growing role of NPs and PAs is inevitable as supply constraints force greater decentralization of healthcare. This decentralization will have impacts well beyond mid-level providers as the roles of groups like care coaches, care coordinators, and non-clinical caregivers become elevated as well.

Together, these groups will form a critical connective tissue that is foundational to US healthcare. Companies that can build solutions for the growing pains of these groups are bound to find fertile ground for growth.

Subscribe now

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

There seem to be two emerging views on where the venture capital asset class is heading.

On one hand, there’s a view that the worst is over - that we are “mostly through this downturn” as Union Square Ventures’ Fred Wilson commented last month, with the caveat that it could still take a few quarters to see this play out in venture capital activity.

There’s some data to support this. For example, Carta finds that total capital raised in Q2 increased QoQ. SVB (lol) suggests that we have found a floor with 2023 healthcare fundraising on pace to match 2021.

The other view is that we are positioned for prolonged pain in early stage capital markets - a multi-year “slow bursting” as Bill Gurley phrased it in a recent interview.

There are also data points supporting this view. While capital deployment may be increasing, a significant portion of recent financings are part of a “tidal wave” of down rounds. Startup layoffs, “unicorn” fire sales, and significant turnover at investment firms are widespread. In healthcare in particular, many “darling” startups of the past decade are imploding in front of our eyes (e.g. Pear Therapeutics, Bright Health, Babylon Health).

The lack of consensus is exacerbated by the opacity of private markets. Whereas public markets capitulate transparently, the dynamics of private markets have worked to conceal the reality of the private asset bubble bursting (e.g. non-disclosed insider rounds which use intricate structures to keep headline valuations flat).

Ultimately, the bearish view feels more convincing, particularly for healthcare startups. Healthcare continues to offer attractive long-term secular tailwinds and a litany of opportunities for technological disruption, but prevailing macro conditions will likely exert an outsized impact on early stage healthcare companies, which I explore in greater detail below. Investors that continue to price deals “to perfection” and fail to account for what is, in many ways, unprecedented macroeconomic uncertainty, are going to destroy a lot of value.

If you’d like to chat about this topic (or idk really anything 🥺), please don’t hesitate to reach out. Thanks for reading!

Motion in the Ocean

There seems to be a tendency for operators and venture capital allocators to look at the present market conditions as just another part of the market cycle, one that should eventually correct to the good old times. This psychology is reflected in the highly structured bridge rounds that are emerging across early stage tech, where companies are hoping to buy time until things “come back”.

But this perspective fails to capture just how fundamentally economic conditions have changed. The shift isn't simply a linear compression of risk appetites; instead, a new baseline framework for inflation and interest rates establishes an entirely different set of investment parameters compared to the previous decade. One of Howard Marks’s recent memos neatly summarizes this:

It’s hard to understate how meaningful this shift is; it’s literally uncharted territory for most professional investors.

So we’re in a new paradigm where indexing off of valuation inputs from the preceding decade does not make sense anymore. Compounding this challenge is that the near-to-medium term macro outlook seems to skew to the downside.

For all of the recent optimism around waning inflation, a “soft landing” is far from certain. Wage gains (hello UPS) and increased production costs related to geopolitical decoupling are sticky and could apply lasting inflationary pressure, long after the transient deflationary benefits of falling commodity prices and demand destruction are realized. As BlackRock explains:

We’ve moved from a regime where demand shocks dominated the macro environment to one where supply constraints dominate. These constraints … are set to stoke ongoing inflationary pressures.

This has two implications.

First, central banks will likely have to systematically hold policy tight to lean against such price pressures. That’s in stark contrast to the past 30 years when broad disinflation prompted central banks to keep monetary policy easy. And in the event that rate hikes caused financial cracks – such as the collapse of a few U.S. regional banks earlier this year – or faltering economic activity, central banks would have been quick to come to the rescue with rate cuts to help stimulate activity. See the chart. That’s no longer the case. And it is a sea change from the low interest rate environment that prevailed before the pandemic. We see policy rates staying higher for longer in major economies.

Source: BlackRock 2023 Midyear Outlook

Outside of inflation, there are other looming market risks that could impact investor risk appetite, including aging populations (which are more sensitive to inflation, thus limiting central bank maneuverability), consumer spending pressures (e.g. student loan payments resuming), and declining corporate capex/opex investments as we approach a big scary corporate “debt wall.”

These macro risks pose two broad implications for early stage companies.

First, the marginal buyer of assets is evaporating.

Growth stage funds are struggling to meet fundraising targets. Even the pLaYiNg DiFfeReNt GaMeS investors are (finally) severely marking down investments made at the top of the market. This suggests that sponsor-backed buyouts and growth rounds will be harder to come by. “Strategic” investors will also be under greater scrutiny. This means fewer M&A exits, fewer “flyer” venture capital investments, and fewer science experiments off the balance sheet in non-core categories. This is particularly impactful on healthcare where the majority of lucrative exits happen via strategic M&A. The shrinking buyer set will apply further pressure to valuation multiples, a particular concern for those venture capital firms which have built their entire models on passing off the hot potato to the next guy/gal.

The second major implication is that achieving earnings growth will be harder and harder.

Sales cycles across industries are elongating. This has a severe impact in healthcare where the sales cycles are already glacial. Indeed, all types of customers in healthcare - payors, providers, life science companies, etc. - are becoming more judicious about vendor spend. This means more small pilots instead of large first contracts. More ROI validation before customers are willing to expand. Reduced ability to count on YoY seat/license/membership expansions at growing customers as an embedded revenue growth lever.

The slow sales and implementation cycles that plague healthcare create a dynamic where compared to other industries, it just takes longer to understand if healthtech startups have achieved compelling product-market-fit. All of these macro headwinds work to make this problem worse.

On balance, this means that the filter on companies across stages will get tighter and tighter, and valuation multiples will likely remain compressed for all but the companies that have robust indications of product-market fit, attractive returns on capital, and de-risked plans on how to deploy new capital to fund growth.

For investors, this emphasizes the need for price discipline, something which venture capital firms are notoriously bad at.

Public market value investors will be familiar with the concept of investments being “priced to perfection” vs. having an embedded margin of safety. This means that even great companies can be bad investments if the entry price is too high. This also means that difficult situations can yield attractive returns, if the price is right.

If you had bought Enron equity in those days between the collapse of the Dynegy merger and the bankruptcy… you would have had over a 10X return if you [were] willing to hang on for 7 years.

Source: Acquired Podcast - “Enron: The Complete History and Strategy”

Unfortunately, due to the perverse incentives at play for most venture capital strategies, investors have been implicitly encouraged to eschew any semblance of price discipline. That’s because the conventional VC model passes the risk buck almost entirely onto the entrepreneur. I’m not sure how durable that model is going to be going forward.

Surviving the chop

Early stage healthcare companies in this environment should look to modulate fundraising and burn rate based on business model conviction.

Do you have tremendous pull from the market that you can can’t satisfy with existing infrastructure and resources? Great, then make like a beyblade and let it rip. Not there? I’d suggest calibrating very precisely around what the Company wants to achieve with the capital to avoid getting caught “over your skis” - i.e. running out of runway with no ability to tap existing investors for additional cash and not enough to show to garner outside investment interest.

In the face of elongating sales cycles and growing commercial pressures, healthcare startups that can achieve any of the following will be very well positioned to attract capital at premium valuations:

• Unique distribution models that make getting to a contract an easy yes

  • Example*: TapestryHealth elevates the standard of care via its tech-enabled platform for SNF residents at little to no cost to the SNF owner, making it a “no-brainer” for the SNF

• Legitimate validation of hard-dollar ROI, ideally with referenceable customers

  • Example*: Azra demonstrated in a referenceable case study with HCA the ability to identify more oncology patients and reduce patient leakage via its AI technology which reviews pathology reports

• Rapid “speed-to-value” for customers to reach the stickiness stage faster

  • Example*: Razormetrics has emphasized rapid implementation and action yielding the ability to demonstrate ROI on drug spend for health plans and self-insured employers in <120 days (hey, that’s fast for healthcare!)

* These are all Sopris Capital portfolio companies since those are the companies I’m most familiar with

In the absence of these things, companies may find other ways to “de-risk” revenue growth. I always lqtm when I think of this tongue-in-cheek “hierarchy of disclosure.”

It’s meant to be derogatory to overpromising startups with little substance to speak of, but it’s also… kind of what early stage healthcare companies need to do? Helping investors gain confidence in customer demand is essential, and there might be creative ways to do that outside of signed contracts (e.g. validation via subject matter experts, recorded demos with folks who would be customers). Investors will (and should) haircut these efforts, but it can help build conviction on the margins.

Good times ahead…

Over the long term, there will continue to be opportunities for new companies to build enterprise value by solving acute pain points throughout the healthcare ecosystem. After all, private investments in healthcare have historically outperformed other sectors across multiple market cycles.

We can’t forget that despite representing ~20% of US GDP, healthcare has historically represented only 6% of technology spending, implying substantial underinvestment and runway for increased tech adoption (I mean, have you ever been to a doctor’s office lol).

The problem is that many investors over the preceding decade have translated the large TAM and overarching long-term tailwinds in healthcare to mean that “healthtech goes brrr,” leading to significantly overcapitalized, overvalued early stage companies forced to grow into unrealistic expectations. And often, this has come at the expense of patient care.

Failure to account for macro headwinds in investment underwriting leads to poor outcomes for all stakeholders. Investors who can exert price discipline are going to be better positioned to achieve attractive risk-adjusted investment returns while delivering on the promise of improving care and outcomes.

Share Vick's Picks

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

Don’t laugh, but I have been obsessed with mortality lately. Not in an “early-2010s scene kid scrolling moody tumblr pages while listening to linkin park” kind of way; more in a “experiencing unbridled dread while watching ‘80 for Brady’ at the movie theater last week” kind of way.

On the bright side, this minor existential crisis (I’m fine!!) prompted me to explore some of the interesting developments taking shape in the realm of end-of-life care.

These developments are important because despite the gains achieved in longevity and end-of-life care quality over the past several decades, terminal illness can still elicit deep sorrow as illustrated by the tragic story of the Gilland family from a few weeks ago:

Jerry Gilland, a 77-year-old terminally ill man, made a pact with his wife about three weeks ago, the authorities said: If his failing health did not improve, he wanted his 76-year-old wife, Ellen Gilland, to kill him.

His health continued to decline, and so shortly before 11:30 a.m. on Saturday, Ms. Gilland brought a gun inside AdventHealth Hospital in Daytona Beach, Fla., went to Mr. Gilland’s room on the 11th floor and fatally shot him in the head, the police said. She had planned to fatally shoot herself, too, but in the end, “she couldn’t go through with it,” said Chief Jakari E. Young of the Daytona Beach Police Department.

…

Chief Young said it appeared that the couple had planned for Mr. Gilland to shoot himself, but he lacked the strength. “So she had to carry it out for him,” he said.

Source: NYT

Overall, there are still substantial gaps to close in addressing physical, mental and spiritual needs of the terminally ill.

This post is a discussion of where end-of-life care is today, some interesting startups emerging to elevate the standard of care, and some not so far-out ideas that could shape end-of-life care over the next decade.

Subscribe now

End-of-life care: a continuum

End-of-life care typically refers to the final months of an individual’s life, but it is part of a broader continuum of care for “life-limiting” diseases that encompasses palliative care, hospice and bereavement.

Specifically, there are four categories of life-limiting diseases/conditions:

As a result of improvements in infant mortality and broad-based longevity gains achieved over the past century, death has shifted generally into an “old-age phenomenon.”

As a result, end-of-life care in the US is largely covered by CMS. Medicare spends >$20bn per annum on hospice care alone. While that number seems high, there is a meaningful argument that greater emphasis on end-of-life and hospice care reduces the overall cost burden on the healthcare system. 10% of an individual’s lifetime medical spend takes place in the final year of life, largely driven by curative interventions aimed at prolonging lifespan (while often reducing quality of life during that period). For example, 10% of Medicare recipients had surgery in the final week of life.

How is end-of-life care trending? Overarching trends seem positive.

The tools we have today to quantify the quality of an individual’s passing are kind of weird. One established metric is to look at the “site of death” as a leading indicator for the individual’s quality of life at the time of passing. Ideally, patients are able to pass away in the comfort of their own home. Surveys indicate 70%+ of people prefer to pass in-home, and in-home deaths are generally considered less taxing for family and caregivers. “Worse” outcomes include passing away in a nursing home or a hospital setting.

By this measure, quality of end-of-life outcomes has materially improved over the past decade, with deaths at home surpassing in-hospital deaths.

Further, while the increase in the number of individuals receiving hospice care is partially driven by an aging population overall, there’s evidence that utilization rates are also increasing (excl. pandemic). As the US population continues to age and utilization rates right-size post-pandemic, the number of people enrolling in hospice every year should continue to grow rapidly.

There are still real challenges in end-of-life care: poor communication between providers/patients/caregivers/family, an anemic staffing environment (in line with just about every corner of care delivery), limited deployment of technology/data/analytics, significant racial disparities in access to /utilization of care. COVID-19 had a major impact on the industry, changing workflows and exacerbating staffing issues.

Solving some of these issues can work to align end-of-life care more tightly with people’s key priorities - ensuring adherence to their end-of-life care preferences while relieving the burden on the ones they leave behind.

Elevating the standard of care

There are several companies forming to address some of these gaps. Here are some interesting areas where startups are innovating around end-of-life care today:

Advance Care Planning

Advance care planning (ACP) involves creating alignment around care goals and future care decisions should an individual’s health deteriorate due to a life-limiting illness. ACP is a billable service that can be rendered by a provider. As part of ACP, individuals can create Advance Directives which are forms of legal documentation of a person’s wishes about their medical treatment based on their values and preferences (e.g. living will, durable power of attorney for health care).

Improving ACP is a key lever to enhancing overall end-of-life care quality. >60% of people in declining health report feeling insufficient control over their medical care. This might be because most people still have never have a discussion about end-of-life care with a medical professional, despite really wanting to:

Effective ACP can help patients and their families feel more in control. Companies like VyncaCare, WiserCare, and Koda Health are increasing access to ACP by digitizing the process, proactively identifying patients who might benefit from ACP today, and connecting individuals to ACP facilitators.

Tech Enabled Hospice

There are multiple opportunities for technology to elevate hospice care. Similar to other areas of healthcare, better provider technology (e.g. more intuitive clinical software/EHRs, workflow automation) can reduce administrative burden enabling more time spent with patients. Better family/caregiver support technology can improve engagement and communication, reducing the burden on family members/caregivers supporting a loved one in hospice.

Guaranteed is developing a next-generation hospice care platform with the goal of doing exactly the above - improving the hospice experience for both providers and patients by leveraging technology. For example, the Company enables providers with a purpose-built EMR to improve key workflows, while providing patients and family members with a proprietary app for telemedicine visits and access to educational resources.

Bereavement Care

Bereavement care companies provide tools for family members and caregivers to better manage both the emotional challenges of losing a loved one as well as the literal logistical challenges of losing a loved one. For example, Empathy enables companies to offer an employee benefit to help navigate both administrative issues (e.g. funeral planning, closing subscriptions/financial accounts) and the emotional pain of loss. With the average family spending 12+ months resolving financial matters for a passed loved one, streamlining these activities is a legit value add.

Lantern connects individuals with a 1:1 guide to help navigate loss. Autumn seeks to make “loss more livable” with an end-of-life marketplace that connects bereaved people to loss-related service providers with an interface that feels intuitive to e-commerce natives.

With a spectrum of models ranging from high-touch to self-serve emerging in the market, it will be interesting to see what resonates most deeply with consumers.

Let’s get mystical

I had a few ideas which I have yet to see in the market but could be compelling commercial opportunities in the future.

Culturally Competent Hospice 🫶🏻🫶🏽🫶🏿

Non-white racial groups materially under-utilize hospice care.

It’s not surprising when you consider the myriad areas where there are deeply entrenched racial disparities across the US healthcare system, but the drop-off in the utilization of the Medicare-covered hospice benefit is particularly pronounced. Sherman Leung and Dan Arteaga wrote this detailed piece unpacking some of the underlying drivers of this phenomenon. For example:

Much of this is cultural and tied to spiritual beliefs as well for many minorities, many of whom value life-prolonging care, full code statuses, and hold beliefs of miracles, God’s will, and/or fatalism. “There’s a lot of African Americans who think saying ‘no’ to medical care is like saying ‘no’ to God. It’s like you are trying to take things in your own hands and not trusting God’s timing” (source) One of the biggest sticking points is around reimbursement policies that require patients to discontinue curative treatment in order to qualify for hospice care. Though some families have found ways to reframe hospice care to align with their values, this choice between treatment vs hospice is a large barrier for many patients who consider hospice care.

This being said, providers themselves are also to blame for end-of-life discrepancies in whose preferences they choose to listen to: “It appears that providers adhered to the EOL (end of life) preferences of white Americans more often than African Americans” (source 1, source 2)

Source: Margins of Medicine

Many health tech startups have emerged to close racial disparities in healthcare by providing “culturally competent” primary and specialty care (e.g. Hurdle, ReKlame Health, Spora Health). I think a similar model could emerge to drive greater hospice utilization and better end-of-life care outcomes for non-white patient populations. There might also be opportunities to coach clinicians across the continuum on having more productive, culturally sensitive discussions around end-of-life care with their patients.

Psychedelic Medicine 🍄

Psychedelic medicine is gaining steam as a potential tool to improve behavioral health outcomes. While most discussion and capital formation around the use of psychedelic medicine is focused on addressing treatment-resistant psychiatric conditions, the hospice and end-of-life setting may be a prime candidate for psychedelic treatments.

Recent research suggests that there might be significant psychosocial benefits to patients receiving even a single dose of psychedelic treatment as part of end-of-life care:

When administered under psychologically supportive, double-blind conditions, a single dose of psilocybin produced substantial and enduring decreases in depressed mood and anxiety along with increases in quality of life and decreases in death anxiety in patients with a life-threatening cancer diagnosis. Ratings by patients themselves, clinicians, and community observers suggested these effects endured at least 6 months.

Source: “Psilocybin produces substantial and sustained decreases in depression and anxiety in patients with life-threatening cancer: A randomized double-blind trial.” Journal of Psychopharmacology

Source: “Rapid and sustained symptom reduction following psilocybin treatment for anxiety and depression in patients with life-threatening cancer: a randomized controlled trial.” Journal of Psychopharmacology.

While there’s much more work to be done in validating these treatment pathways, end-of-life care might emerge as a setting that is amenable to experimentation given the underlying prognosis of patients.

Virtual Reality & the Metaverse 🥽

As I wrote last year, the Metaverse™ is relevant to healthcare because it offers unique capabilities around convening geographically dispersed people together into a synchronous, live experience within a rich, visual environment.

In end-of-life care, VR could be used to provide people with experiences that they are physically unable to partake in as their health declines. These experiences could offer meaningful benefits related to physical pain management and quality of life:

Virtual reality (VR) provides temporary escapes from pharmacologically resistant pain and allows for experiences and journeys patients may not access in any other way. Enabling wishes through virtual worlds may also offer additional benefits such as controlling psychological and physical symptoms.

Source: BMC Palliative Care

Consider this story from 2019 as an illustrative case study:

When hospice nurse Laurie McKay arrived at the emergency room, her patient -- a man in his early sixties with terminal cancer and now a fractured hip -- told her: "I knew you would be coming someday, but today my wife and I were supposed to be getting on a cruise ship."

McKay, who is chief nursing educator for Continuum Care Hospice based in the San Francisco Bay Area, didn't want the couple to miss out on that last cruise to Alaska together, so she turned to a tool that Continuum had started using with its patients -- virtual reality.

She made an appointment to visit the couple once they were back at home. Using Samsung Gear VR headsets and Google Earth VR, she'd mapped out all the ports where the cruise would have stopped, giving the couple the 360-degree views of ocean, waterfalls and ice caves they might have had in person. McKay also showed the man his childhood home in the present day, and the marina in California where the boat he'd been working on was docked.

"These were experiences he thought he would never be able to see completed," McKay said.

…

McKay -- the nurse who plotted out the cruise ports for the couple who missed their trip -- said the man's wife reported that he told everyone who came to visit him about the VR experience. And when he died, she even talked about it at his service.

"[He thought] he was going to end up simply being in a bed in his home waiting to die," McKay said. "Instead he found he was able to live and participate and find enjoyment each day that he was given."

Source: CNET

Future experiences could be self-guided, therapist-guided, or “multiplayer” with an individual’s loved ones. Such experiences could truly transform the way that we experience our final months of life, enabling a wealth of adventures and shared memories with the people we care about.

Embracing the Inevitable

While I personally have a loooong way to go in getting comfortable with mortality, I’m optimistic that end-of-life care will be transformed by technology over the coming years such that there will be a breadth of tools available for individuals coming to terms with loss. In the meantime, let’s get busy livin’.

You might also find interesting:

🧠 End Well - End Well is a nonprofit focused on exploring new frontiers in end-of-life care. Some great resources and presentations from thought leaders in the field

📚 Being Mortal by Atul Gawande - kind of a cop-out to include this book almost everyone knows, but it’s still an insightful read about end-of-life care through the lens of the health care system. If you haven’t read it yet, I’d give it a go

📺 Limitless with Chris Hemsworth (Ep 6 - Acceptance) - this series on Disney Plus is about Chris Hemsworth trying to find ways to increase longevity, but the final episode shifts gears to exploring how to come to terms with one’s own mortality. They do some really thought-provoking exercises which I think are worth checking out

🏈 80 for Brady - just kidding. please do not watch this

Share Vick's Picks

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

Staffing shortages have erupted into a five-alarm fire complicating operations for care delivery and services organizations in every corner of the healthcare ecosystem.

The business impact of staffing shortages is severe. Hospitals and health systems have experienced significant margin contraction due to outsized short-term contract labor spend (e.g. travel nurses). Care delivery companies from behavioral health to skilled nursing are struggling to meet demand because of persistent personnel turnover. Digital health companies that sell their offerings into health systems are experiencing protracted sales cycles because their customers’ singular near term priority is the resolution of their own staffing problems.

Staffing shortages are a pull-forward, not a blip

The optimistic view is that acute staffing shortages are a one-off COVID-driven supply shock. Over time as the pandemic subsides, labor supply should normalize back to pre-pandemic levels, easing the cost burden on operators. There’s some evidence to support this view; for example, leading hospital operator HCA reported on its most recent earnings call that peak contract labor utilization is squarely in the rear-view mirror, implying that the worst of labor cost inflation should be behind the industry.

However, there’s also a lot of reason to believe that labor cost inflation is not going anywhere.

Rather than characterizing labor shortages as a one-time shock, it’s more accurate to view this as a “pull-forward” of supply shortages that were going to impact the market in the future based on prevailing supply/demand trends. That is, just as it pulled forward demand for lots of things (e.g. telehealth, e-commerce, WFH), the pandemic simply pulled forward declines in clinical labor supply that would have been realized over several years.

Projections have long indicated that rapidly rising demand for clinical services would soon (i.e. in like 10-15 years) outstrip clinician supply. Consider the supply of registered nurses. Growth projections for the supply of registered nurses were already tepid pre-pandemic. The pandemic has effectively pulled forward several years of RN exodus. Thus, critical levels of supply/demand imbalance are materially closer than pre-pandemic projections.

It’s hard to see how this trend reverses over the near to medium term.

One reason for that is that supply shortages create a negative feedback loop. With fewer available staff, remaining staff has an even higher workload burden, driving more burnout and more turnover. Insufficient staffing is the most commonly reported reason for clinical staff quitting. Just this week, 15,000+ nurses in Minnesota staged a walk-out in protest of insufficient staffing levels across area health systems. A recent survey quantifies the sentiments on the picket line, suggesting that the risk of nurses dropping out of the workforce is at an all time high:

Nearly one-third (28 percent) of nurses indicated their desire to leave the profession had increased dramatically since the pandemic, while those who said their desire to stay had increased since the pandemic dropped from 24 percent last year to 4 percent this year.

Source: Florida Atlantic University & Cross Country Healthcare Joint Survey, Sept 2022

Further enhancing the pricing power of clinicians is the unprecedented flexibility that clinicians of all levels have today to choose how and where they want to work. For example, physicians and mid-level providers of all specialties can pick up part-time or full-time work via a plethora of telemedicine platforms in the market. This is particularly appealing to physicians who might be burned out in their current setting of care (which is like, >half of physicians!).

All of this means that healthcare services companies need to think very hard about the value proposition they offer to clinicians.

In startupland, it can be hard to see the impact of this problem when looking at a care delivery company that is onboarding a few clinicians per month; but as a company scales and starts to add 10-15+ clinicians per month, ensuring that the right infrastructure is in place is critical to recruiting & retaining valuable clinical talent and driving commercial momentum.

Your margin compression is my tailwind

Ultimately, there are several stakeholders that are well positioned to enhance their market power in the face of prolonged staffing shortages over the medium term. Some of these are:

CMOs & Clinical Leaders at Early Stage Companies

Elevating clinical leadership to the helm of the organization is critical for any healthcare startup that wants to survive the long haul. Sustainable business models need to be rooted in good medicine. And in order to achieve that, clinical leadership should be able to have direct input into the business model at all stages of development.

The absence of clinical input into areas like bizops and product yields situations where business outcomes completely skew the workflows of front-line clinicians in ways that can actually be harmful to patients. A perfect example of that dynamic run rampant is D2C telepsychiatry companies like Done and Cerebral effectively running online Adderall pill mills in support of hypergrowth aspirations. From D2C to DEA!!

Clinical leaders can help establish clinical protocols and assist with designing key business levers like provider utilization, onboarding and ongoing training. Strong clinical leaders can also help inspire confidence in prospective clinical hires, providing a competitive advantage with respect to recruiting. As a result, demonstrated clinical leaders will be coveted more than ever by high-growth healthcare startups.

Contract and Shift-Based Staffing Tech Platforms

If the monster growth rates posted by public healthcare staffing firms in the past several quarters is any indication, increasing complexity for clinician recruitment and retention will significantly expand the pie for firms in the healthcare staffing value chain.

A litany of digital marketplaces and enterprise software applications have emerged to support the gig-ification of healthcare labor. Companies like ShiftMed, CareRev, TrustedHealth and Incredible Health help healthcare workers match with on-demand or contract labor opportunities. On the flip side, a multitude of next-generation vendor management software platforms and tech-enabled managed service providers are seeking to digitize the management of contingent labor on the back-end for health systems. Although at some point it’s natural to expect consolidation with so many players in this fragmented space, there is so much room for continued digitization of staffing workflows that there will likely be sufficient whitespace to accommodate multiple winners in these categories.

Rentable Provider Networks (AWS for Healthcare)

If you can’t build it - rent it! As provider recruitment and retention becomes more complex and costly, it will become less compelling for care delivery companies to build their own networks in-house. This is compounded by the fact that in a rising interest rate environment, capital is more expensive and upfront investments into fixed costs (like a W-2 provider network) are harder to economically justify.

Instead, digital health companies can look to rent provider networks for a multitude of enterprise telemedicine use cases (e.g. tele-consults, data annotation) from companies like Medcase*, Wheel and SteadyMD. Companies that need clinical capacity but don’t have the bandwidth or in-house core competency to manage the tremendous complexities involved with physician recruitment, credentialing and retention can leverage these on-demand provider networks to support their business models, enabling them to narrow their focus on what makes their “beer taste better”.

Upskilled Non-Clinical Labor

One idea that could help relieve pressure on constrained labor supplies is upskilling existing parts of the non-clinical labor pool. For example, life coaches could be trained and certified (=/= licensed) to provide basic cognitive behavior therapy, greatly expanding access to therapy for patients. Non-clinical resources could be cost-effectively deployed in patient’s homes to perform data gathering activities or serve as a conduit to care (think a non-clinical person brings a tablet to a patient’s home so that they can tele-interface with a physician); Atlanta-based HealPros* leverages a similar model to solve major pain points for health plans. While there will always be a need for expert licensed physicians and mid-level providers to make clinical decisions, upskilling the labor pool can help patients get more cost-effective care while providing more people with a compelling career track in healthcare.

Physician Workflow Automation Tools

The biggest driver of physician burnout is administrative burden. It’s kind of funny that while physician time is more valuable than ever, physicians are spending inordinate amounts of time dealing with bureaucratic tasks.

This is a broad category with many interesting ideas/companies percolating. Some cool ideas out there now: AvoMD enables physicians to build automated workflows that can be deployed at the point-of-care. HealthNote is automating the patient intake process. Abridge’s AI platform transcribes and structures patient/physician interactions to streamline charting, notes and claims. Azra AI* screens unstructured patient data to flag and surface patients at risk of developing cancer.

There are so many ways that digital health companies can solve the major pain points of bureaucracy and capture huge market value. A caveat to this optimism is that given the glacial pace of healthcare innovation, these technologies will likely take a long time to achieve widespread adoption. Just consider how long it took to achieve widespread EHR adoption even with the help of regulatory mandates.

Still, marginal gains in this area can unlock significant value.

If you build it…

Healthcare’s labor problems are not going anywhere - companies that can build a durable advantage in provider/physician recruitment and retention, or help other healthcare stakeholders in this area, are well positioned for the coming storm ⛈️.

*denotes a Sopris Capital portfolio company. It’s called full disclosure people!

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.

why waste time dial lot number when few number do trick?

Earlier this month, SAMHSA announced that the National Suicide Prevention Lifeline will be overhauled into 988, “an easy-to-remember three-digit number for 24/7 crisis care” powered by a network of >200 call centers. The move simultaneously underscores the massive scale of the country’s mental health crisis while illuminating the gaps that exist in our emergency mental health infrastructure.

For example, skeptics point out that crisis call centers are already severely understaffed, leading to high (~20%) abandonment rates on calls into the Lifeline; the proposed investment is unlikely to provide sufficient resources to satisfy increased call volume expected from the 988 overhaul.

More importantly, even when we are able to successfully intervene with an individual experiencing a psychiatric emergency, we often have nowhere to direct them to receive care:

We do know that when someone calls and connects with a caring and trained human they often can benefit, but if they need more than a crisis counselor, the rest of our outpatient and inpatient mental health infrastructure is still woefully inadequate. This is not dissimilar to what happened with COVID-19. Sure, we could set up a memorable shortcut to reach an understaffed COVID hotline but what good will that do if the hospitals are full and all the ventilators are spoken for?

Source: Carlene MacMillan, MD and O. Scott Muir, M.D - The Frontier Psychiatrists

The inadequate supply of emergency psychiatric care is partially why individuals with high acuity mental health needs experience poor clinical outcomes and disproportionately outsized health care costs. And while surging capital formation in the behavioral health sector is democratizing access on the lower acuity end of the care continuum, it may actually be warping incentives for providers and exacerbating constrained access to high acuity care where financial realities (low reimbursement, unpredictable volume, clinical labor cost inflation) stifle private sector investment.

Instead, innovators in high acuity behavioral health care will likely need to build longitudinal care models (e.g. care teams that can address an individual’s holistic needs) in order to bend the cost curve and improve outcomes. Getting this right could drive tremendous value to stakeholders, representing a big opportunity for entrepreneurs and investors.

More on this below. If you’re working on something interesting in high acuity behavioral health care, I’d love to learn about it.

Where are all the psychiatric hospitals?

Psychiatric emergencies are defined as disturbances of thought, mood or behavior that require immediate intervention; typically the individual poses a threat to their own safety or the safety of others.

Patients with a psychiatric emergency must be stabilized via counseling or medication to the point that they are no longer a threat to themselves or others. That commonly requires a 24/7 inpatient treatment facility with dedicated psychiatric care resources.

There are three major models/settings for emergency psychiatric care where a person could be safely stabilized during a crisis - co-located treatment within an existing medical emergency department (ED), dedicated psych wing within a medical ED/hospital, or a standalone psych facility.

While standalone psychiatric facilities are ideal because they have dedicated focus and resources, the number of standalone sites has declined monotonically decade after decade. From 1970-2014, the total number of inpatient psychiatric beds decreased 64%. Meanwhile, ED utilization for mental health issues is skyrocketing. From 2006 to 2014, overall utilization of the ED increased 15%, but ED visits for suicidal ideation and intentional self-inflicted injury increased by 415%.

Why are there so few dedicated inpatient psych facilities? With the disclaimer that I’m not a historian, a crude summary of the 70+ year timeline of how we got here is: in the 1950s most individuals with severe mental illness (SMI) - and many people who did not have SMI (e.g. loosely defined “senile” geriatric population, orphans, wards of the state) - were institutionalized. This led to a massive population (>500k people) in mental institutions. These institutions were generally overcrowded/underfunded and poorly run. This was objectively a terrible standard of care. Around this time, advancements in pharmaceuticals made it possible to manage SMI conditions in less restrictive settings. The 1963 Community Mental Health Act was enacted with the goal of decentralizing psychiatric care by moving patients from state-run facilities into local/community care settings. Additional drivers of deinstitutionalization included the rise of professional psychiatry in outpatient settings, cultural criticism of institutions and the advent of Medicare/Medicaid (which did not cover patients' costs for living in state mental institutions). Unfortunately but unsurprisingly, the government fumbled the transition to community based care via a classic combo of underfunding/mismanagement and <40% of the originally contemplated capacity for community-based psychiatric care was ever created.

As a result, we’re left with the status quo where the standard of care for a person experiencing a psychiatric crisis is pretty much 1) showing up to the ED and being boarded for literally days and weeks until a psychiatrist is available to provide care or 2) going to jail because the only “resource” to deal with the individual is law enforcement (in 44 states, a jail or prison holds more mentally ill individuals than the largest remaining state psych hospital).

Big dollars

Like many areas of healthcare where cost burden exhibits a pareto-like distribution, individuals with high acuity mental health needs have disproportionately high cost of care.

Individuals with schizophrenia or severe bipolar disorder represent <3% of US adults, but these conditions are the primary or secondary diagnosis for almost 10% of annual ED visits and represent ~$30bn of hospitalization spend alone. This doesn’t include the impact borne by overburdened ED’s, the judicial system, and social services.

One reason driving the exorbitant cost of hospital care for individuals with SMIs is the recurring cadence with which these individuals are showing up to the ED. 23% of individuals with schizophrenia discharged from the ED are readmitted within 30 days. In NYC, 30% of all patients who made five or more ED visits from 2007 to 2010 had schizophrenia or bipolar disorder.

This matters a lot to health insurers. And not just Medicaid (contrary to prevailing assumptions); while Medicaid membership exhibits slightly higher prevalence of of SMIs, most individuals with SMIs are actually covered by private insurance.

Successfully managing these crisis flashpoints of care, including post-discharge aftercare, could drive tremendous cost savings while also improving clinical outcomes for patients.

Solutions?

It’s unlikely that the industry will manifest significant capital formation behind new crisis care centers. This is because on a fee-for-service basis, the point-of-care economics of managing psychiatric emergencies are not compelling. >50% of the nation’s 1,700+ inpatient psychiatric facilities had negative net operating margins in 2016. A study of children’s hospitals found significantly lower financial margins for mental health and suicide attempt hospitalizations (~11% of all hospitalizations at these sites) than for other medical hospitalizations.

Source: Journal of Hospital Medicine, Dec 2020

In the absence of de-novo psychiatric bed capacity, there are still opportunities for private capital investment to drive meaningful value creation at these crisis care flashpoints. Here are some ideas I’ve seen in the market:

• Urgent Mental Health Care - replicating the urgent care model in primary care (i.e. walk in or same-day/next-day appointments) for psychiatric care needs could help de-escalate situations before they rise to crisis level. Examples: Mindful Care (Sopris Portfolio Company) provides next-day access to psychiatric health care via its network of brick-and-mortar locations and telehealth.

• Steering Care Away from the ED - many individuals don’t need to be transported to the ED to receive care; the ET3 model for EMS opens the door to delivering care (via telemedicine) to the patient where they are. Something ET3 for 988 with warm handoffs to on-demand psychiatrists could be an interesting model, particularly in regions with highly concentrated payor mix. Examples: companies like PriorityOnDemand are positioned to deliver on this.

• Coordination of Existing Supply - often, stakeholders don’t know what bed capacity in the community is open to take in an individual in a crisis. Enabling interoperability between disparate systems and data sources could work to compress the timeline to successful referrals. Examples: OpenBeds (now part of BambooHealth) provides real-time psychiatric bed inventory data to stakeholders to help direct patients to care venues with available capacity

• ED Integrated Telepsychiatry - deploying telepsychiatry can help fill the gap for hospital emergency departments that don’t have access to a full-time psychiatrist, reducing wait times for patients. Examples: QLER Telepsychiatry embeds its psychiatrists on-demand for partner hospitals to resolve this access problem

• Longitudinal Wraparound Care Coordination - this is probably the biggest idea/opportunity I can think of. Given that individuals with SMIs are prone to super-utilization of emergency services, creating a care team that wraps around the patient with a holistic approach (e.g. addressing social determinants of health) could have a huge impact to outcomes and costs, particularly if that relationship can be sustained over multiple months or years. Imagine a situation where an individual who otherwise would be in-and-out of the ED can call on their trusted care advocate/coach as symptoms escalate. Care teams can also play a big role in post-discharge aftercare planning to avoid recurring rehospitalizations. Examples: Companies like Amae Health and Firsthand appear to be building a platform like this (I think)

• High Acuity Substance Use Disorder (SUD) Care - the SUD treatment space is in dire need of greater supply of clinically validated, high acuity care, particularly in the earliest phases of an individual’s journey through treatment. Examples: Gallus Detox (Sopris Portfolio Company) utilizes a data-driven approach to medical detox to drive significantly better outcomes and adherence to aftercare.

The usual obstacles to adoption of mental health solutions will be even more pronounced for innovators in high acuity care. For example, non-existent outcomes measurements + data will be a hurdle to demonstrating value. Developing solid outcomes measures will help innovators communicate value back to stakeholders (i.e. payors). Increasing cost of clinical talent will be particularly challenging as the pool of clinicians who can/want to treat SMIs is smaller than the general pool of mental health clinicians, which is already really small. Models that can leverage lower cost resources (e.g. non-clinical care advocates or coaches) should find more sustainable economics.

All told, there is a massive opportunity to create financial and societal returns by elevating the standard of psychiatric crisis care; as we look to the next leg of behavioral health care innovation, this is an area to watch.

Disclaimer:

This content is being made available for educational purposes only and should not be used for any other purpose. The information contained herein does not constitute and should not be construed as investment advice, an offering of advisory services, or an offer to sell or solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning economic trends and performance is based on or derived from information provided by independent third-party sources. The author believes that the sources from which such information has been obtained are reliable; however, the author cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based.