...

By Alexis Kayser

Healthcare Editor

0

Share

Newsweek is a Trust Project member

See more of our trusted coverage when you search.Prefer Newsweek on Googleto see more of our trusted coverage when you search.

This is a preview of the September 3 edition of Access Health—Tap here to get this newsletter delivered straight to your inbox on Thursday mornings.

Good afternoon. I spent the morning speaking with executives from some of the world's leading smart hospitals and health systems.

Read More on Health Care

One of them started by telling me his hospital isn't all that smart.

"Being smart, I believe, is admitting that you have many unsmart parts," Dr. Wonchul Cha, chief information officer at Samsung Medical Center, said during today's Newsweek webinar. Despite Samsung's reputation for digital maturity, Cha described his organization as still learning how to work alongside AI. The technology, meanwhile, isn't taking "baby steps," he said. "They're doing giant steps. And we are being dragged by this technology, actually."

I liked that admission because it cuts against how we often talk about health care innovation. "Smart" can sound like a destination, or a designation bestowed on organizations with the newest AI tools, robots and digital infrastructure. Cha's point was that the organizations furthest along may be the ones most willing to acknowledge how much they still have to figure out, including the downsides of technologies they've already deployed.

Data presented during the webinar reinforced that distinction. Statista's latest Smart Hospitals Digital Maturity Survey found that technologies like AI, information systems and automation are increasingly widespread among the more than 200 hospitals surveyed globally. Deep implementation is not. Teleconsultation, for example, is available in some form at 82 percent of surveyed hospitals, yet more than half use it for fewer than one in 20 consultations.

That gap between having technology and actually becoming a smart hospital became the throughline of my conversation with Cha and leaders from Sheba Medical Center, NYU Langone Health and Statista.

Professor Eyal Zimlichman, chief innovation and AI officer at Sheba and founder and director of ARC Innovation, said, "Pilots are relatively easy." The harder work begins when health systems try to connect new technology to legacy infrastructure, change clinical workflows and move from reaching one or two percent of patients to 20, 30 or 50 percent.

Even the best of the best are still grappling with the definition of digital maturity and agree that it's likely to change over time. What they could agree on today is that "smartness" isn't about how much technology a hospital can showcase, but how much it can make routine, measure, govern and improve.

Or, as NYU Langone's Chief Health Informatics Officer Dr. Paul Testa put it near the end of our conversation, a smart hospital "isn't a goal, it's a result."

If you missed today's webinar, you can view the full on-demand replay here.

Have a wonderful Labor Day weekend, and I'll see you next week!

In Other News

Major health care headlines from the week

• OpenAI launched a new Epic integration for ChatGPT for Healthcare, allowing organizations to bring authorized patient information into ChatGPT or embed ChatGPT directly into supported EHR workflows. Clinicians can use it to review patient histories, recent labs, medication changes and unresolved follow-ups, with responses pointing back to supporting chart information.

  • The company also introduced a Healthcare Public Data plugin connecting ChatGPT to nine official sources, including PubMed, ClinicalTrials.gov, CMS Coverage, RxNorm and DailyMed. The tool is designed to let clinical, research and operational teams compare information such as trial eligibility, medication data and coverage policies without searching each source separately.
  • OpenAI reported strong results from physician evaluations of the EHR-connected product. Across 4,363 ratings spanning 27 clinical use cases, physicians rated 99.1 percent of responses safe. In a separate evaluation using large U.S. health care datasets, more than 93 percent of responses for each of five tested data sources received a “good” or better accuracy rating.

• Indiana became the third state in the nation to end routine hospital diversion due to emergency department and hospital overcrowding, following Massachusetts and Rhode Island. Beginning Sept. 1, hospitals and EMS providers will restrict diversion to disasters and critical infrastructure failures.

  • State officials said diversion can create problems beyond the crowded hospital it is intended to relieve, including longer ambulance transport times, EMS shortages, delayed care, reduced continuity of care and potentially worse patient outcomes. Indiana health officials, hospitals, EMS organizations and other emergency-care groups jointly agreed to the change and will continue working on other approaches to ED crowding.

• The FDA has selected the first four companies for its TEMPO digital health device pilot: Dexcom, SonderMind, Limbic and Cadence Solutions. Their tools span diabetes and metabolic care, hypertension and behavioral health, including an AI voice agent that delivers cognitive behavioral therapy.

  • The pilot could offer a new regulatory path for emerging digital health tools. FDA plans to exercise enforcement discretion on some requirements, including premarket authorization, while manufacturers collect real-world performance data. The devices are intended for use through CMS’ ACCESS model for chronic disease care.
  • FDA ultimately plans to select up to 40 participants across four clinical areas, including cardiometabolic risks, cardio-kidney-metabolic disease, chronic pain and behavioral health. For the first four participants, effectiveness for the pilot uses has not yet been evaluated, the agency emphasized.

• Employer health benefit costs are projected to jump 8.2 percent per employee in 2027, the largest increase since 2003, according to preliminary results from Marsh’s latest survey of more than 1,800 employers. Without benefit or cost-reduction changes, employers expect costs to rise an average of 11 percent.

  • GLP-1s are a major cost driver. Marsh estimates rising use of GLP-1 weight-loss drugs will add a full percentage point to 2027 cost growth.
  • But they’re not alone. AI-enabled software that helps physicians document and submit claims, resulting in more and higher-level claims, and rising payments through the No Surprises Act dispute-resolution process could each add as much as another percentage point.
  • Employers are preparing to pass on some of the pressure while reconsidering how they buy care. Fifty-nine percent plan cost-cutting benefit changes, such as higher deductibles, while about two-thirds of large employers expect to increase workers' premium contributions. Employers are also showing greater interest in steering workers toward high-performing providers and, in some cases, contracting directly with hospitals and physicians.

Pulse Check

Executive perspectives on key industry issues

...

LeanTaaS announced this week that it acquired Aidin, a care-transition platform that coordinates more than 2 million referrals annually across a network of more than 21,000 post-acute providers. The deal expands LeanTaaS’ patient-flow business beyond the hospital, connecting its inpatient capacity-management technology with the process of finding and coordinating care for patients after discharge.

I spoke with Mohan Giridharadas, founder and CEO of LeanTaaS, about why hospitals can no longer build their way out of capacity constraints, where patient flow breaks down and why he believes an “EHR-first” strategy shouldn’t become an “EHR-only” strategy.

Editor's Note: Responses have been lightly edited for length and clarity.

Before we get into the acquisition, tell me about LeanTaaS and the problem you set out to solve.

I started LeanTaaS at the end of 2009, early 2010. In my prior life, I was at McKinsey for 18 years, where I led our lean manufacturing and lean service operations practices. The whole idea of starting LeanTaaS was: Imagine delivering operational excellence, but on a software-as-a-service platform instead of consultants with clipboards and spreadsheets.

We initially worked across industries, but in 2013, we applied the thinking of yield management and airline optimization to infusion centers. We figured out how to mathematically level-load chair occupancy. Wait times went down by half, capacity went up 15 percent and nurses who had routinely missed lunch because of the midday rush could actually take their breaks.

Once we realized we'd found lightning in a bottle, we pivoted the whole company exclusively to health care.

The common theme in everything we do is sophisticated data science to unlock capacity, because when you unlock capacity, flow gets better. The clinical brilliance of U.S. health care is astounding: robotic surgery, genomics, precision medicine, proton beam therapy. Meanwhile, operationally, surgeons are faxing orders and nurses are counting heads and beds on pieces of paper.

Our mission is to bring the operational sophistication of health care up to match its clinical brilliance.

So how does the Aidin acquisition fit into that mission?

Our inpatient optimization predicts and helps the flow of a patient from admission to discharge. But a very large number of patients, when they're discharged, aren't going home. They're going to a skilled nursing facility, long-term acute care, dialysis, hospital at home or some other post-acute service.

Today, that can mean the patient is stuck in the bed while the team calls different care providers and asks: Can you take this patient? Do you have capacity? Do you take this insurance? That cycle is manual and painful, and it takes a long time.

Aidin picks up at the wall of the hospital and runs a two-sided marketplace to make the downstream happen. By connecting what we do with what Aidin does, we can free up a bed faster and give the patient better choices for post-acute care.

Today, the process of finding a post-acute facility often starts when discharge is about to happen. We can predict a day or two in advance that someone is likely to need a skilled nursing facility and start teeing things up, including getting authorizations going.

Who owns that transition today? The hospital? The patient?

The hospital. There's a role called a case manager who serves as the quarterback when discharge requires this level of complexity.

The case manager works with patients to understand their needs. Some are clinical and some are operational: How close is it to my house? Can I visit my family member who's going to be admitted? What can I afford? What's the copay?

There are economic conditions, logistics conditions and clinical conditions, so the case manager acts as the quarterback, balancing the patient's needs and the hospital's requirements while understanding the available options.

How does combining LeanTaaS and Aidin change the case manager's role?

It won't necessarily change the role. It'll give them world-class tools by surfacing things more easily.

We closed the acquisition a week ago, so we haven't connected any software pipes yet. Today, Aidin and LeanTaaS operate as separate tools. But it's very clear what information needs to flow between them.

Initially, the users are also slightly different. A case manager tends to use Aidin more, while a unit manager or charge nurse tends to use LeanTaaS more. It's not mandatory that the user interface be identical and harmonized on day one, although over time our products will move toward a common look and feel.

What happens when the algorithm knows exactly where a patient should go, but there simply isn't a post-acute bed, home health worker or other resource available? Are there limits to optimization?

At the end of the day, we cannot manufacture capacity. When you run out of capacity, you run out of capacity. We don't want to pretend this is magic that can create a bed out of thin air.

What we can do is continuously expand the geographic footprint, the options and the kinds of services. Because Aidin is connected to 21,000 care providers and can dynamically look at a bigger range, we've got a better shot at finding an option.

The alternative isn't perfection. Without Aidin, if a case manager calls the five skilled nursing facilities in the neighborhood and all five are full, they're going to start searching their way outward, trying to find more locations without necessarily knowing what each facility does or whether it takes the patient's insurance. With Aidin's network, even that process can be better.

This acquisition seems to make an argument that hospitals can't solve their capacity problems entirely inside the hospital. Are health system leaders thinking broadly enough about post-acute care when they develop their capacity strategies?

Even if you stay within the health system, there's a lot of opportunity to improve because health care has so much operational complexity. Each department is stuck optimizing for itself and doesn't fully have a sense of what's happening upstream or downstream.

A bed gets blocked, which means there's no room in that inpatient unit. Somebody in the ED who needs to be admitted to that unit now gets blocked in the ED. Then the ED is blocked, and new patients coming in might leave without being seen.

It's hard to connect the dots all the way from that inpatient unit to the ED to the patient who left without being seen. Health system operations have plenty of challenges managing that complexity within their four walls. When you start going outside those four walls, it's a whole new degree of difficulty.

If the industry gets patient flow right, what changes beyond margins and length of stay? What does the patient actually experience differently?

For one, access changes. Patients getting stuck in a bed means fewer beds are available for incoming patients. Hospitals go on diversion and tell incoming transfers, "I'm really sorry, I can't take this patient. Send them somewhere else." So access improves.

Second, the patient experience improves. Patient satisfaction is driven by the clinical correctness of the treatment. Patient dissatisfaction is driven by waiting. When patients wait to get an appointment, to have a discharge happen or to execute that discharge, it leads to frustration.

There are also dramatic financial consequences. If the payer says it should be a three-day stay and three days becomes four or five, the hospital ends up absorbing that cost. There's also the opportunity cost: Had you gotten that patient out, somebody else could have taken that bed.

In the old days, hospitals dug themselves out of capacity challenges by building more hospitals and ORs, hiring more staff and deploying more equipment. Financial stresses have made that building boom more challenging.

If you can't build your way out of trouble and demand is still going up, you have to optimize your way out of it.

Both LeanTaaS and Aidin depend on information that ultimately lives in the electronic health record. How do your products fit into that environment?

We can fully coexist in an EHR environment because, at the end of the day, all the data comes from the EHR and users live in the EHR.

Our products can exist as a side panel within the EHR, read continuously from the EHR or have a direct link. We can deliver as seamless an experience as the health system wants. Sometimes they want it within the same tab or an adjacent tab. Sometimes they're happy to have it running somewhere else. We configure it based on what works best for the health system.

What advice do you have for health system technology leaders trying to connect these fragmented processes more seamlessly?

Health systems have to have a horses-for-courses approach. EHR-first is a very strong mindset, and it's a good mindset. But sometimes an EHR-first mindset slips into an EHR-only mindset, and that shuts off innovation because then innovation happens at the speed of the single EHR vendor.

Step-change transformations require entire ecosystems. Health system leaders who believe the EHR is all they need will find that it's a great system of record, but it cannot be the only system of intelligence.

Staffing sits in a different system. Devices sit in a different system. Ambient technology sits in a different system. Intelligence signals come from dozens of places outside the EHR.

Relying on the EHR to take you on its own all the way home is probably not going to get you there.

C-Suite Shuffles

Where health care leaders are coming and going

• Jenn Schwartz began her tenure as president and CEO of ChristianaCare on Tuesday.

  • Schwartz joined the Newark, Delaware-based health system in 2018, most recently serving as executive vice president and chief strategy officer. She is an attorney by training and previously worked as Lourdes Health System’s general counsel.
  • She succeeds Dr. Janice Nevin, who retired after 12 years in the role.

• Stephen Hunter was named senior vice president and chief innovation and transformation officer at Hospital for Special Surgery, based in New York City.

  • He has served as the organization’s chief digital strategy officer since March 2025, according to his LinkedIn profile, and previously held leadership roles in digital strategy and innovation at Highmark Health.

• The American Medical Association tapped Jennifer Goldsack to lead its Center for Digital Health and AI.

  • Goldsack is the founder and CEO of the Digital Medicine Society (DiME) and “a leading voice on the responsible adoption of digital health and AI in healthcare,” according to the AMA. She will be the first official leader of the Center, which launched last year to help physicians navigate the growing use of AI and other digital technologies in medicine.

This is a preview of the September 3 edition of Access Health—Tap here to get this newsletter delivered straight to your inbox on Thursday mornings.

View 0 Comments

Request Reprint & Licensing

Submit Correction

View Editorial & AI Guidelines