 | ● |  |
|
|
|
| | | News and views you ought to know about:- If you’ve sensed healthcare AI has been going extra-gangbusters over the past year or so, your perceptions are impressively attuned. Spending on the technology in fiscal 2025 came close to tripling 2024’s total. The total outlay for 2025 was $1.4 billion. And healthcare organizations sped away from other economic sectors, adopting AI 2.2 times faster than the economy as a whole. The findings are from a report posted this week by the AI-focused venture capital firm Menlo Ventures in San Francisco. The authors state that the lion’s share of that $1.4B spend has gone not to pilot projects but to full-on rollouts, aka “production.” What’s more, providers account for 75% of the sum ($1B). The most frequently implemented categories are ambient clinical documentation ($600M) and coding and billing automation ($450M). The fastest growing are patient engagement and prior authorization. Other trends of note as reported and sub-headlined by Menlo:
- Providers accelerate, payers proceed with caution. Buying cycles for AI tools are 18% shorter for health systems and 22% shorter for outpatient providers, both of which are rushing to capture operational gains, Menlo finds. Buying cycles for payers are 20% slower as they remain largely in piloting and experimentation mode.
- Startups capture vast majority of AI spend, but incumbents are striking back. Startups currently dominate, capturing 85% of healthcare AI spending to date, Menlo notes. However, most customers report a slight preference for buying AI from their incumbent electronic health records (EHR) provider.
- Ambient scribe adoption faces retention risk. Menlo finds 67% of outpatient providers now using ambient scribes expect to switch vendors within three years.
- “Healthcare’s AI moment is here,” the authors comment. “Providers are seeing products that deliver ROI and witnessing peers adopt at scale, and buying cycles have compressed from 12 to 18 months to under six. The conditions are aligned for rapid acceleration.” However, they add, 80% of the market “remains untapped. The next waves will come from companies automating services budgets, building voice interfaces for patient engagement, solving prior authorization at scale and improving drug discovery.”
- The report is posted in full for free.
- Legal proscriptions that bar bots from guiding clinical care—whether to offer diagnoses or recommend treatments—infringe on patient autonomy. That’s the opinion of an analyst with the libertarian Cato Institute think tank. The viewpoint holder is Jeffrey Singer, MD, who happens to be a long-serving general surgeon along with a Cato senior fellow. “While such laws might shield existing providers from competition,” Singer maintains, “autonomous adults have the right to decide who and what they turn to for healthcare advice and treatment.” To be sure, Singer allows, lawmakers and policymakers should push for algorithmic transparency and patient safety. However, he avers, they also should “avoid enacting regulations that interfere with the natural development of the patient-bot relationship and unnecessarily obstruct AI innovation.” Get his fully fleshed-out argument here.
- UnitedHealth Group isn’t giving up on AI in claims processing. Remember, the company is being sued for using AI to deny payments for care rendered to Medicare Advantage patients. The suit was filed in November 2023 and is still in the maw of a federal court in Minnesota. Two years later, Bloomberg finds United piloting a system called Optum Real, which “aims to distill health plans’ complex rules around what is covered into information that doctors and billing staff can use in real time to tell whether a claim is likely to be paid.” The system has been in place at a 12-hospital system based in Minneapolis since March. Optum Real has “reduced claims denials meaningfully across more than 5,000 visits” in the health system’s outpatient cardiology and radiology departments, according to the health system’s CIO. “We avoid that disappointment, that frustration with the clinicians and the patient of having that denial experience,” the CIO tells a Bloomberg healthcare reporter. “It’s less paperwork on both sides.”
- Small wonder someone came up with a large language model that can draft customized appeal letters in minutes. As you might infer from the work appeal is doing there, the AI is aimed at patients who’ve had their health insurance claims denied. It’s also free. The brainchild of a frustrated family member who grappled with the denial experience on behalf of his cancer-patient wife, the platform is called Counterforce Health. Founder Neal Shah points out that 99% of patients or family caregivers do not appeal denied claims. “Of the tiny fraction who do, 40% win,” he points out for CNET. “That tells you most people are intimidated, don't know their rights or just assume a denial is final. And it also shows how frivolous so many of these denials are.” More on “AI fighting AI” in healthcare here.
- Organizations on the buying side of healthcare mergers and acquisitions are unwise to give AI diligence short shrift. After all, the odds of legal exposure are considerable given the real risks of compliance and class action concerns. Such ugliness could get especially nasty if it implicates AI-equipped software that has any interaction at all with protected health information. The heads-up is from attorneys at Los Angeles-based SheppardMullin. “AI is an evolving legal and operational risk area in healthcare transactions,” the authors write. “Conducting effective due diligence review of AI in a proposed transaction calls for a buyer’s and its counsel’s detailed understanding of the technology itself, as well as potential risks and liabilities surrounding its use. This rapidly developing area of law will continue shaping the regulatory landscape of the healthcare field, but with the right preparation, the diligence process will minimize a buyer’s exposure and best position it for post-closing success.” Of course they would say that. They’re with a law firm seeking healthcare clients. That doesn’t mean they’re not right. Hear them out here.
- Also worthwhile:
- Noteworthy research news:
- From AIin.Healthcare’s sibling news outlets:
|
| | | |
| |
| | | Speechmatics focuses on the notes. Doctors focus on the patients. With 93% overall accuracy, 17% fewer word errors vs the next best, and a 96% redcall for medical terms, Speechmatics’ transcription is helping doctors put their focus back on patients. Real-time speaker diarization helps distinguish between doctor and patient even in cross-talk and with background noise, while drug names, ICD-10-CM and procedure codes land with timestamps. The result is usable, ambient note-taking that doctors can trust. Read more.
- Don’t Let AI Touch Patient Data
Shadow AI is sneaking into hospitals and clinics faster than IT Teams can track. Download the free Security Checklist from Fellow to uncover where AI notetakers may be recording or storing sensitive conversations. Use it to flag compliance risks early and keep your organization on the safe side of privacy laws. Before Shadow AI spreads inside your team, download the checklist to reduce your risk here.
Nabla launches Nabla Connect: The fastest way for any EHR to deliver world-class ambient AI
Nabla Connect is a new plug-and-play module that lets EHR vendors embed Nabla’s trusted AI assistant directly into their platform, making integration fast, compliant, and effortless.
In just a few days, EHRs can: - Integrate ambient AI without heavy engineering
- Enhance clinician experience with seamless, accurate note generation
- Stay future-proof through continuous updates
Learn more about Nabla Connect
|
| | | |
|
| |
|
|
