Month: October 2025

Menlo Ventures: The State of AI in Healthcare

Posted on by Jason Barry

Of all the AI market overviews that have hit the wire recently, none have generated more buzz than Menlo Venture’s State of AI in Healthcare. One look at the report and it’s easy to see why.

First things first. Here’s a couple high-level callouts before we zoom in on the details:

  • Healthcare AI spending has already topped $1.4B in 2025 – 22% of healthcare orgs have now implemented domain-specific AI tools, a 7x increase over last year [Chart 1].
  • 85% of all healthcare AI spend is currently flowing into startups (faster cycles, clearer ROI), rather than incumbents (often layering AI on legacy platforms) [Chart 2].

Providers accelerate, payors deliberate. Providers dominate AI adoption in healthcare, especially health systems – supplying $1B of the $1.4B total spending.

  • Outpatient providers represent $280M, while payors surprisingly contribute just $50M.

The song remains the same. Menlo found that leading health systems are choosing AI based on themes we’ve covered plenty of times before. They prioritize:

  • Tech maturity – providers prioritize production-ready solutions that perform at scale. 
  • Risk level – tools that don’t directly interface with patients see less scrutiny.
  • Quick value – a 2025 favorite, rapid ROI and organizational confidence are essential. 

What solutions check all the boxes? Two categories account for the lion’s share of AI budgets, in large part because they quickly address acute operational pain points.

  • Ambient documentation ($600M), no surprise here. This puts it in perspective [Chart 3].
  • Coding and billing automation ($450M), hard to think of a quicker ROI.

Bonus chart. Here’s the closest we’ll ever get to official ambient AI market share [Chart 4].

IT is good, services are great. Total U.S. healthcare administration spending reaches $740B annually, yet IT spend represents <10% of that. The report has a top tier breakdown [Chart 5].

  • AI’s frontrunners found success carving into existing IT budgets, but the future victors could be the teams that convert services dollars into software dollars for the first time.
  • AI offers the ability to automate workflows that have always been “people-intensive” – prior auth, patient engagement, front-office RCM – and Menlo believes 80% of this market is still completely untapped.

The Takeaway

Healthcare’s AI moment is here, and most of its potential has hardly been touched.

Brook Lands $28M for AI-Driven Remote Care

Posted on by Jason Barry

Brook.ai just closed $28M of Series B funding to help extend care beyond clinic walls using its signature blend of remote care teams and AI. 

Chronic conditions require care continuity. Brook delivers that at scale by combining remote monitoring with in-house care teams and always-on support, plus an AI assistant that ties it all together.

  • The platform is purpose-built to create great patient experiences. All of the RPM devices work straight out of the box, with batteries and cellular connectivity already set up to remove as much friction as possible from the starting line.
  • The AI assistant answers any remaining questions and provides a direct line to Brook’s nurses. It also delivers recommendations and insights to keep patients engaged – all while offloading the administrative burden of coordinating care and reimbursement.

The results speak for themselves. Over half of Brook’s patients use the platform daily, and the consistent support has resulted in 82% patient retention with an “excellent” tier NPS of 66.

It turns out that getting patients to engage with their care is a great way to improve their health, and better patient outcomes lead to better provider outcomes. Brook’s partners have seen:

  • 90% reductions in congestive heart failure readmissions
  • 80% increases in controlled hypertension populations within six weeks
  • All-cause readmissions cut in half
  • $256k+ annual net value per 500 patients

The partnership model is the cherry on top. Brook allows providers to adopt remote care without building their own infrastructure, hiring staff, or managing complex billing processes.

  • It accomplishes that by shouldering the upfront CapEx investment, delivering the results it says it can, and only billing providers for the patients that it helps.
  • Providers get a prepackaged service that contributes margin from day one. Patients get care from the comfort of their home. Brook shares in the success.

Time to rinse and repeat. Brook has built up enough data and success stories to prove it can shepherd patients from the hospital to the home, and the Series B will help it scale that same care continuity to more patients across the country.

The Takeaway

Delivering continuous remote care, keeping patients engaged with their treatments, and making sure everyone is accurately reimbursed are all huge problems own their own. Brook pulled together a complete package to tackle all three, and it has a fresh $28M to take the model nationwide.

HLTH 2025 Recap and Major Announcements

Posted on by Jason Barry

Hard to believe that was already Digital Health Wire’s fourth trip to HLTH – a few more flights out to Vegas and we might be getting close to “legacy media” status.

The more the conversations change, the more they stay the same. Last year’s “Be Bold” theme and LLM mania gave way to tales from healthcare’s “Heroes and Legends” and an AI agent avalanche.

The industry is still facing most of the same challenges, but it was amazing to see how quickly new innovations are compounding in the solutions that it’s using to conquer them.

Here’s our (non-exhaustive) roundup of some of those solutions from the expo hall:

  • Arbital Health debuted its Merlin AI value-based care assistant built to make actuarial analysis transparent and actionable. Merlin AI interprets complex risk contract data, explains performance drivers, and instantly recommends next best steps.
  • Abridge is bringing CDS into the flow of clinical conversations through a partnership with Wolters Kluwer. The UpToDate integration adds to a white-hot October for Abridge, which already included heavy-hitter roll outs at UPMC and Northwell.
  • Brook.ai hauled in $28M of Series B funding to expand its personalized remote care platform to more patients and conditions. Providers can get up and running on the platform in 30 days with no CapEx required, and the results in the announcement speak for themselves. More to come on this one next week.
  • Cedar debuted an aptly named Cedar Cover enrollment tool to make sure patients aren’t caught in the rain when the dark cloud of Medicaid cuts gets here. Cedar Cover’s key capabilities include Medicaid enrollment, proactive renewal, and denials resolution.
  • Ellipsis Health is teaming up with NVIDIA to leverage its Parakeet ASR model and build out the AI infrastructure supporting Sage, an emotionally intelligent AI care manager that expands staffing capacity through proactive patient engagement.
  • GE HealthCare is collaborating with The Queen’s Health Systems and Duke Health to advance the development of its upcoming AI-driven hospital operations solution. The solution will leverage insights from both systems and the 500+ hospitals using Command Center to surface actions for improving care quality, patient flow, and resource utilization. 
  • hc1 unveiled hc1 IQ to deliver precision insights that transform lab data into life-saving action. hc1 IQ unifies lab, clinical, and supply chain data into a single AI-powered platform to produce enterprise-wide intelligence for improving outcomes.
  • IntelePeer launched a SmartAnalytics Starter Pack that gives providers the most approachable entry-point we’ve seen to the AI agent ecosystem. It includes one patient engagement automation from IntelePeer’s menu, along with real-time dashboards and call analysis to drive immediate operational improvements.
  • League introduced League Agent Teams to its AI-first consumer experience platform. The multi-agent system guides users through complex health journeys using a suite of AI agents fine-tuned for specific tasks and an orchestration layer to coordinate them.  
  • Lorikeet debuted a healthcare-specific extension for its Team of Agents (deja vu, definitely one of the biggest themes at HLTH), which coordinates actions across multiple agents to call vendors, text doctors, and take action to “actually solve customer issues.”
  • Nabla debuted Nabla Connect, a plug-and-play module that enables any EHR to seamlessly integrate ambient AI. Nabla already has one of the widest EHR footprints in the space, and CEO Alex LeBrun gave us the live walkthrough of how Nabla Connect extends that foundation to smaller EHRs looking to unlock the same capabilities.
  • Optum took the lid off its Optum Real claims system that delivers instant coverage validation. The multi-payor platform enables real-time data exchange between payers and providers, allowing any issues to be intercepted at the point of submission.
  • Penguin Ai is joining forces with UPMC Enterprises and leveraging its Ahavi data platform to validate new AI models in a secure testing environment. Ahavi will allow Penguin to refine its Small Language Models for real clinical use cases like record summarization and prior auth optimization.
  • Solera Health showcased its Precision Insights Suite for AI-driven cost containment and care navigation. The two core components include Precision Intercept, which identifies patients at risk of significantly increasing costs within a year, and Precision Navigate, which delivers personalized provider recommendations based on similar patients.
  • Suki launched a nursing-focused AI consortium with several leading health systems and AvaSure on the inaugural roster. The consortium will co-develop Suki for Nurses while integrating ambient AI capabilities into AvaSure’s virtual care platform.
  • Vital announced the launch of Vital Urgent Care, an AI-powered platform that provides patients in urgent care settings with real-time updates, wait times, and personalized guidance – without requiring a single download or additional staff bandwidth.
  • Wellsheet is rolling out across Ascension to give providers a unified view of previously overlooked data. The “GenAI front-end” allows care teams to access patient-specific EHR data on a single screen, accelerating diagnoses and clinical decisions.
  • Withings Health Solutions unveiled best practices for obesity care management programs built on a decade of partnerships across the segment. Those include: (1) continuous monitoring for real-time adjustments, (2) on-device communications, (3) data-driven outcome tracking, (4) addressing comorbidities, (5) measuring more than traditional BMI metrics. All great practices for maternal health as well!

Welcome to all the fresh faces scrolling through DHW for the first time, and shoutout to all the long-time readers we caught up with at the show – the OGs have officially been here longer than a lot of the exhibitors have existed!

We picked up as many announcements as we could carry, but if we missed anything exciting and you don’t see it below, hit reply and let us know what to circle back on next week.

2025 State of Digital Health Purchasing

Posted on by Jason Barry

What better way to kick off HLTH week than with a new survey from the Peterson Health Technology Institute showing that healthcare decision-makers are still hungry for new solutions?

PHTI’s 2025 State of Digital Health Purchasing Survey showed that health plans, employers, and health systems are all investing heavily in digital health, but spending patterns are starting to diverge.

  • The headlining stat: 61% of health plans and 44% of health systems plan on increasing digital health spending in the coming year, compared to just 14% of employers.

Employer spending has leveled off. Two-thirds of employers plan to hold spending steady as they maintain their current suite of solutions.

  • In contrast, health plans and health systems are ramping up to offer a wider variety of solutions (81%) and keep up with patient engagement (77%).
  • The top priorities across the board? Improve access, reduce costs, and strengthen user experiences, especially in areas like diabetes, mental health, and primary care.

Purchasers are more hawk-eyed than ever… especially when it comes to their contracts. Nearly half of purchasers are already using performance-based contracts, and the majority plan to use them in the next year.

  • 73% of contracts now have a duration under two years – up from 59% last year – leaving a short window for solutions to prove their value.
  • Short-term contracts, annual portfolio reviews, and a laser focus on engagement and ROI are the new normal. Building long-term relationships hinges on demonstrating both.

Vetting isn’t getting any easier. When comparing vendors, employers (66%) are more likely than health plans (23%) and health systems (42%) to cite cost as the deciding factor.

  • Payors and providers both prioritize a proven track record above cost, but the report points to other ways to gain an edge.
  • Although performance-based contracts are gaining traction, few purchasers are satisfied with their current models, and they’re looking for vendors that can offer either better outcome thresholds or attribution methods.

The Takeaway

Health plans, employers, and health systems all still have an appetite for new solutions, and PHTI just gave vendors a way to stack their decks with more data on each of their unique priorities.

2025 Trends Shaping the Health Economy

Posted on by Jason Barry

Trilliant Health just released its 2025 Trends Shaping the Health Economy Report, delivering a uniquely holistic perspective on the healthcare market through the lens of supply and demand.

Here’s the state of play. Health expenditures are growing faster than the rest of the economy, and they’re projected to represent 20.3% of GDP by 2033. 

  • The U.S. spends more and gets less than peer nations, which might have been tolerable when our federal debt was at $800B in 1980, but definitely isn’t now that it’s at $37T.

What levers can we pull? This year’s 115 page analysis looks into six macro trends driving the healthcare economy, and underlines each of them with concrete stories from real-world data.

  • 1) Affordability concerns are reshaping demand. Medical prices are up 54.5% since 2009, and they’re pressuring patients and employers to weigh their options – especially when rates for an inpatient procedure can vary as much as 7x within the same facility depending on the payor (p. 19).
  • 2) Stakeholders are slow to adapt to demographic trends. Mortality rates among adults aged 18-44 have been rising as the fertility rate falls, shrinking the share of Americans with employer-sponsored coverage (p. 22).
  • 3) Specialty care intervention is incentivized over primary care prevention. In 2024, behavioral health visits rose 11.4%, while primary care visits declined 5.6%, marking the first time behavioral health utilization surpassed primary care (p. 43).
  • 4) Fraud, waste, and abuse are pervasive. The share of high-complexity ED visits has risen sharply, increasing from 36.6% to 47.8% of visits between 2018 and 2024, underscoring the financial impact of upcoding (p. 57).
  • 5) Alternative therapies are accelerating. GLP-1 utilization increased 745% from 2018 to 2023, while bariatric surgery volumes were flat to declining, illustrating how high-margin procedures face growing competition from medications (p. 86).
  • 6) If the industry won’t deliver value, the government will. Federal programs have consistently failed to bend the cost curve (MSSP savings are less than 1% of Medicare spending), and there’s mounting political pressure for top-down structural reform (p. 89).

The Takeaway

The U.S. healthcare system is at a crossroads. As Trilliant put it so nicely, the choice for all health economy stakeholders is whether to implement “radical change from the inside” or “to be subjected to such change by external forces.”

Bain & Company: Top Healthcare IT Priorities

Posted on by Jason Barry

Payors and providers are fighting different operational battles, but they’re using the same two-letter weapon to come out on top: AI, you guessed it. 

A joint report from Bain & Company and KLAS found that 80% of payors and 70% of providers now have an AI strategy in place, up from just 60% last year.

  • Providers are up against structural workforce shortages and rising patient volumes, while payors are contending with higher medical loss ratios and more regulatory scrutiny.
  • Bain and KLAS’ survey of 228 U.S. healthcare execs suggests that all signs point to one solution, and that’s deploying tech to improve margins.

Where are payors investing? Care coordination (57%) and utilization management (55%) were the top IT investment priorities for the second straight year.

  • Payors place total cost of ownership, functionality, and scalability ahead of suite convenience, so best‑of‑breed is still the default buying motion.
  • Plans are leveraging AI for everything from member engagement (35%) and enrollment (26%) to risk adjustment (26%) and prior auth automation (20%).

Where are providers investing? Revenue. Cycle. Management.

  • Half of providers ranked RCM among their top IT priorities, placing it above clinical workflows (34%) and EHRs (32%).
  • RCM = ROI. Accurate documentation and coding results in cleaner claims and fewer denials, which directly translates to higher revenue and lower expenses.
  • It’s also a match made in heaven for AI automation, and RCM currently represents the four most common AI use cases: ambient documentation (62%), clinical documentation improvement (43%), coding (30%), and prior authorization (27%).

Here’s the kicker. Providers cite EHR integration and interoperability as their biggest pain points, so most of them prioritize their EHR vendors for new solutions.

  • Only 20% of providers are primarily best-of-breed buyers, and two-thirds of Epic customers would choose an Epic option that’s “good enough” over a better competing product.

The Takeaway

It’s getting pretty hard to not be bullish on AI. There’s still plenty of uncertainty, but both payors and providers now seem to agree that inaction is the riskiest action.

Rock Health Q3 Overview: Signals Out of Sync

Posted on by Jason Barry

Rock Health’s always-excellent digital health market overview painted an interesting picture for Q3, with venture funding continuing to climb despite several “signals out of sync.”

We’re steady on the surface. Digital health startups raised $3.5B across 107 deals in the third quarter, outpacing last year by a decent margin and bringing the year-to-date total to $9.9B across 351 rounds [Chart: Q3 Funding].

  • Deal volume continued to slow, but fewer raises yielded larger checks. Q3 saw 107 funding rounds, down from 120 in Q2 and 124 in Q1.
  • The average raise in 2025 now stands at $28.1M (up from $20.4M in 2024), and we’ve already seen 19 mega-rounds above $100M – surpassing last year’s total with a quarter left to go.

The middle is murky. Rock Health rolled up its sleeves and calculated widely variable trends in mid-market funding.

  • Series B deal flow has thinned, with just 30 raises through Q3, compared to an average of more than 60 annually over the past three years.
  • As fewer startups reach Series B, those that do are stretching the range of what a B round can be. Series B deal sizes so far in 2025 spanned $11M–$210M ($199M) – the widest spread since the boom of 2021.
  • Pair that with the persistent prevalence of unlabeled raises, and the thinning Series B pipeline suggests that startups are traveling increasingly winding roads to reach scale.

Activity is concentrating around workflows. The biggest theme of the Q3 report was that Clinical Workflow and Non-Clinical Workflow are now 2025’s two most-funded value propositions, capturing a combined 42% of the total funding [Chart: Value Propositions].

  • A $1.3B lead separates these value propositions from the rest of the pack, and workflow tools now appear to be in a league of their own.

Startups are heading horizontal. The report also highlighted a growing group of startups pushing into adjacent workflows, such as Abridge’s partnership with Highmark Health (expanding into prior auths) and Judi Health acquiring Amino (moving into patient navigation).

  • M&A volume is up 37% from last year, with 166 acquisitions through Q3 (already topping 2024’s 121 total), in large part due to these horizontal moves. 

The Takeaway

The numbers look steady, but the market is also steadily splitting in half. That means that the real story going forward won’t be whether digital health startups can attract investors (they can), but whether companies can demonstrate the impact needed to land on the right side of the divide. 

Assort Health Raises $76M for Experience Agents

Posted on by Jason Barry

A good patient experience starts with a good first impression, and Assort Health just closed $76M of Series B funding to prove its AI agents can deliver exactly that.

Accessing care is a painful process. The Assort OS platform delivers AI agents that can help manage patients’ needs, give staff more bandwidth for higher-level tasks, and reduce friction (AKA frustration) for everyone involved.

  • The agents go beyond scheduling, handling everything from care navigation and prescription renewals to physician referrals and lab tests. 

Goodbye hold music, hello agents. Providers have less staff to manage more patients, so they’re desperately looking for ways to offload the burden on the front office. Voice agents have become the go-to solution, and it’s tough to stand out.

Assort lists several core features that separate it from the pack:

  • Specialty-specific agents – PCPs and oncologists aren’t asking patients the same questions, and Assort adapts to the unique needs of each specialty.  
  • Seamless integration – Assort’s agents integrate directly into the EHR and practice management systems, allowing them to work within the unique clinical rules and workflows of each provider.
  • 90% resolution rate – The specialty-specific tuning allows Assort’s agents to maintain a resolution rate above 90%, while limiting errors like misdirected referrals.

Momentum builds momentum. There might be a ton of startups jumping into the voice AI arena, but not many are closing a Series B less than four months after their Series A. 

  • Assort’s investors will tell you that it’s “leading the re-platforming of patient engagement into the AI-native era.” At least they’re putting their money where their mouth is.
  • Assort now plans to use the $102M it’s raised over the last four months to get its agents into as many practices as possible before the competition gets their first.

The Takeaway

The race is on to transform the patient experience with the magic of AI, and Assort might just have enough agents (and VC dollars) to pull a rabbit out of its hat.

AI Learns the Natural History of Human Disease

Posted on by Jason Barry

Clinical decision-making relies on understanding patients’ past health to improve their future health, an impossible task without first understanding how diseases progress over time.

That’s where a new study in Nature suggests AI is ready to help.

It starts with generative pretrained transformers. Researchers built a GPT, dubbed Delphi-2M, to predict the “progression and competing nature of human diseases.” 

  • Delphi-2M was trained on 400k UK Biobank participants (which lean healthier than the average person), and then externally validated on 1.9M Danish patients.
  • The training was designed to predict a patient’s next diagnosis and the time to it, using only data readily available within the EHR: past medical history, age, sex, BMI, and alcohol/smoking status.

How’d it do? The results speak for themselves:

  • Delphi-2M was able to forecast the incidence of over 1,000 diseases with comparable accuracy to existing models that are fine-tuned to predict single diseases.
  • Death could be predicted with eerily impressive accuracy (AUC: 0.97), and the survival curves that it simulated lined up almost perfectly with national mortality statistics.
  • Comorbidities emerged naturally from the training, and Delphi-2M was able to understand the progression from type 2 diabetes to eye disease to nerve damage.
  • Delphi-2M’s ability to predict heart attack and stroke matched established scores like QRisk, and it even outperformed leading biomarker-based AI models.

Better forecasts inform better policies. If policymakers can consult the Oracle of Delphi to see how many people will develop a disease over the next decade, the authors conclude that they’ll also be able to implement better regulations to prepare. 

  • Not a bad theory, assuming models trained on historical data can make forecasts that hold up to evolving treatments and populations (and that politicians act in the best interest of the people:).

The Takeaway

AI is reaching the point where it can predict thousands of diseases as well as the best narrowly focused models, and that could have big implications for everything from early screening to policymaking.

Get the top digital health stories right in your inbox