The Member Everyone Is Managing – and No One Really Knows

Transcript

Chapter 1

The Illusion of Management: Why Five Departments Miss the Real Story

Healthcare is complicated. Not because we don’t know what we want to accomplish. We want better outcomes. Better experiences. Lower costs. Smarter operations. The complication is everything that has to happen behind the scenes to make those things possible.

This is Healthcare Power Ups, a podcast about how healthcare actually works. In each episode, we’re going to take one part of the healthcare system, pull it apart, and figure out what really happens underneath. Not the PowerPoint version. The real version. The people. The processes. The technology. The economics. And most importantly, why it matters.

Today, we’re talking about care management. Now imagine something. A member has recently been discharged from the hospital. She has a cardiac condition. She also has diabetes. She’s taking several medications. She has a history of depression. She lives alone. And after a recent health event, she’s having difficulty getting around her house.

Now here’s the question: Who is managing her? The utilization management team knows about the hospital stay. The care management team knows about some of her chronic conditions. The quality team knows she has open care gaps. Her behavioral health history may live somewhere else. Her physicians know pieces of her clinical story. The pharmacy has another piece. And the member herself knows something almost none of those systems fully understand: What it’s actually like to be her.

Everyone may be managing something. But is anyone managing the member? That question was central to a discussion with four of Sagility’s clinical leaders: Krithika Srivats, SVP of Clinical Practice; Lori Campbell, VP of Care Management and Quality; Susan Ellis, Senior Director of Clinical; and Lisa Darrow, Clinical Manager.

Because care management sounds deceptively simple. Find people who need help. Connect them with a care manager. Develop a plan. Help them follow it. Improve their health. Except healthcare rarely works that neatly.

Let’s start with how we decide who needs help. For years, healthcare organizations have relied heavily on diagnoses and utilization to identify members for care management. A member has diabetes. Put them into this category. Congestive heart failure? Another category. Multiple hospital admissions? Higher risk. Add medications, emergency department visits and other indicators, and eventually the member gets assigned a risk score. It’s logical. It’s measurable. And increasingly, it’s incomplete.

Chapter 2

From Open Loops to AI Orchestrated Care

Because a diagnosis tells you what someone has. It doesn’t necessarily tell you what they need. Two people can have exactly the same diagnosis and require completely different interventions. One person with diabetes may understand the condition, take medications consistently, see their physician regularly, have stable housing, reliable transportation and a strong support system. Another person with the same diagnosis may struggle to afford medication, have difficulty getting to appointments, live alone and be managing depression at the same time. Same diagnosis. Very different risk. Very different needs. And that’s where traditional rules based care management starts running into trouble.

Healthcare has gotten very good at identifying conditions. The harder challenge is understanding people. As Krithika described it, the missing piece is simple: The member. What is this person already doing to manage their health? What barriers are they facing? What happened before they joined this health plan? What is happening outside the medical system? What resources do they have? What don’t they have? And perhaps most importantly: What is most likely to change their outcome?

Some of that information lives in claims. Some lives in clinical records. Some lives with providers. Some lives in pharmacy data. Some may be found in assessments or conversations. And some of the most important information may not exist in a neatly coded field anywhere. That’s a problem because traditional healthcare systems were largely designed around transactions. A claim. An authorization. An admission. A prescription. A quality measure. Care management is trying to understand something very different. A journey. And journeys don’t fit neatly into transactions.

Consider what happens inside a typical health plan. Utilization management may be evaluating whether services are appropriate. Quality teams may be tracking gaps in care. Behavioral health teams may be working on another set of needs. Care managers are trying to coordinate interventions. Member engagement teams are communicating with the same population. Each function has a legitimate job. Each may have its own technology. Its own workflow. Its own data. Its own measures of success. And that’s exactly where the problem begins. Because the member doesn’t experience healthcare as five departments. The member experiences one life.

Imagine that member we met at the beginning. She’s discharged from the hospital after a stroke. One system knows about the admission. Another knows about her medications. Another contains information about her previous care. Another knows she has unresolved quality gaps. Somewhere else is information suggesting she lives alone. And perhaps another system knows that she has struggled with depression. If those systems don’t connect, each team can perform its individual job correctly and still produce the wrong overall outcome.

The care manager may focus on preventing readmission. The quality team may focus on closing a care gap. Another team may focus on medication adherence. But nobody realizes the most immediate problem: She lives alone in a two story home and currently can’t safely navigate the stairs. That changes the question. Instead of asking: What program does this member qualify for? A better question is: What needs to happen next for this person?

Maybe she needs a short period of inpatient rehabilitation. Then physical therapy. Then home health. Maybe her progress needs to be monitored and the plan adjusted. Maybe transportation needs to be addressed. Maybe medication support matters. Maybe behavioral health needs to be part of the plan. The right answer isn’t one program. It’s an orchestrated sequence of interventions built around the person. And that is a fundamentally different way to think about care management.

It also helps explain why simply adding more care managers won’t solve the problem. Clinical capacity is scarce. Nurses and care managers spend significant portions of their time gathering information, reviewing records and documenting what happened. Documentation is necessary. In healthcare, if something isn’t documented, from an operational and regulatory perspective, it may as well not have happened. But every minute a clinician spends assembling information is a minute they’re not spending applying clinical judgment or engaging with a member.

This is one of the clearest opportunities for AI. Not replacing the clinician. Preparing the clinician. Imagine opening a member’s case and, instead of searching through multiple records, seeing a clinically useful picture of that person’s journey. Recent admissions. Diagnoses. Medications. Care gaps. Utilization. Relevant social circumstances. Previous interventions. Current care plan. Potential risks. And the information isn’t simply dumped onto the screen. It’s synthesized.

Now imagine the system helping prepare documentation after an interaction. Or answering a clinician’s question about an unfamiliar medication. Or surfacing considerations when multiple conditions overlap. Or gathering information conversationally from a member before the care manager joins. That’s more than administrative automation. It’s a way of extending scarce clinical expertise.

And Krithika raised an important problem here. Healthcare often talks about wanting clinicians who can see the whole person. But consider what clinicians are actually being asked to know. Cardiology. Musculoskeletal conditions. Dermatology. Behavioral health. Pharmacy. Social needs. Utilization patterns. Benefits. Provider networks. Community resources. The goal isn’t for every nurse to become an expert in every specialty. That’s impossible. The goal is for clinicians to have enough intelligence around them that they can recognize what matters, ask the right questions and navigate the member toward the appropriate expertise.

Traditionally, that breadth comes from experience. Sometimes decades of it. AI has the potential to change that equation. It can help close the information gap between what an individual clinician knows and what the member’s situation requires them to understand. That doesn’t eliminate expertise. It makes expertise more scalable.

But there’s a catch. Giving a clinician better information isn’t enough. Because care management has another longstanding problem: We are very good at creating care plans. A care plan says the member needs to see a specialist. Take a medication. Schedule physical therapy. Complete an assessment. Follow up with a physician. Change a behavior. Wonderful.

Susan made a distinction that sounds obvious until considering how much of healthcare still doesn’t work this way: A care plan has to drive action. It isn’t enough to tell a member: “Remember to schedule this appointment.” Did they schedule it? Did they go? It isn’t enough to make a referral. Did the member connect with the service? It isn’t enough to recommend medication support. Did the member actually get the medication? It isn’t enough to perform outreach. Did anything change because of it?

That’s the difference between an open loop and a closed loop care model. Open loop care management records what we intended to happen. Closed loop care management knows whether it happened. And then it knows what happened next. That creates a much higher standard for the technology and operating model behind care management. The system has to identify an opportunity. Trigger the appropriate intervention. Reach the member through the right channel. Capture what happened. Connect with providers and other parts of the care team when necessary. Confirm whether the action occurred. Measure the result. And if it didn’t work? Adjust.

That last part matters. Because real people don’t follow workflows. Their lives change. Their conditions change. Their motivation changes. Their support systems change. The intervention that made sense three weeks ago may no longer make sense today. So a truly intelligent care management system cannot simply execute a predetermined care plan. It has to continuously learn what is happening and determine what should happen next.

That’s where the idea of personalization starts becoming much more interesting. Healthcare uses that word constantly. Personalized care. Personalized engagement. Personalized medicine. But true personalization isn’t sending someone’s first name in a text message. It is understanding that the same intervention will not work for every person with the same condition. And it is understanding how someone wants to engage.

One member may answer a phone call from a nurse. Another won’t answer an unknown number under any circumstances. Another may happily interact digitally at ten o’clock at night. Some activities may be completed through self service. Others may benefit from conversational AI. Some require a clinician. The question becomes: What is the lowest friction, clinically appropriate way to move this particular person toward the next best action? That’s a very different operating model from assigning everyone to the same outreach queue.

And then we arrive at the question healthcare executives inevitably ask: Did any of this work? Did we reduce avoidable emergency department visits? Did we prevent a readmission? Did the member get healthier? Did we close the care gap? Did utilization change? Did total cost of care improve? Did the member have a better experience? Those questions expose another weakness in traditional care management. Healthcare doesn’t lack data. It often lacks the ability to turn data into action.

Krithika and Lori both emphasized the importance of reporting, but not reporting for reporting’s sake. A dashboard nobody uses is just decoration. The real question is whether the right information is embedded into how the operation is managed. Do frontline teams see it? Do managers use it? Do providers have access to what they need? Are the metrics connected to the actual interventions being performed? Can the organization trace an action to an outcome?

Because there’s an enormous difference between saying: “We contacted 10,000 members.” And saying: “We identified these members, performed these interventions, confirmed these actions occurred, and saw these outcomes change.” The first measures activity. The second begins measuring value. And that distinction may become increasingly important as care itself changes.

For decades, care management has largely been organized as a distinct function inside health plans. But Lori and Krithika see those boundaries becoming less meaningful. Utilization management knows something. Quality knows something. Behavioral health knows something. Providers know something. The member knows something. Care management shouldn’t have to reconstruct the entire story every time someone needs help.

The future looks much more like a connected clinical ecosystem. Lower risk needs may increasingly be addressed at the point of interaction through digital engagement, providers, remote monitoring, contact centers and self service. Higher acuity members may receive much more coordinated support across clinicians, providers and other services. And care itself may increasingly extend beyond traditional healthcare settings. Into the home. Into digital channels. Into community resources. Into specialized partner programs.

That’s important because no single organization can, or should, provide every intervention a member might need. A member may require physical therapy. Behavioral health support. Home based services. Remote monitoring. Dementia support. Medication assistance. Transportation. Specialty clinical expertise. The emerging model isn’t necessarily about building all of those capabilities inside one organization. It’s about orchestrating an ecosystem around the member.

One entity may remain accountable for understanding the member’s journey and the outcome. But the intervention itself can come from whichever part of the ecosystem is best equipped to deliver it. The member goes out for an intervention. Information comes back. The member’s profile gets smarter. The care plan changes. The next action follows. That starts to look very different from traditional care management.

In fact, maybe “care management” eventually becomes too narrow a term for what we’re describing. Because we’re not really talking about managing a program. The model is really about continuously answering three questions: What does this person need? What should happen next? And did it work? Data helps answer the first. Intelligence helps answer the second. And closed loop operations answer the third.

Put those together, and you get something much closer to a learning clinical ecosystem. One that doesn’t simply identify risk. It understands context. It doesn’t simply create care plans. It drives actions. It doesn’t simply perform outreach. It measures what happened. And it doesn’t ask clinicians to carry the entire complexity of healthcare in their heads. It gives them intelligence that allows them to focus on the part technology can’t replicate: Clinical judgment. Human connection. Trust.

Because maybe the biggest problem with care management was never that healthcare didn’t have enough programs. Maybe it was that we organized the programs around the healthcare system instead of organizing the healthcare system around the person. The member with diabetes doesn’t wake up thinking about diabetes care management. The member recovering from a stroke doesn’t think in terms of utilization management, quality management and behavioral health. They wake up thinking: Can I get downstairs? Can I get my medication? How am I getting to my appointment? Why do I feel like this? Who do I call? What am I supposed to do next?

The healthcare system has built an extraordinary number of functions to answer those questions. The next challenge is making them behave like one system. Because when everyone is managing a piece of the member, the real risk is that nobody sees the whole person. And perhaps the future of care management begins when we finally do.

And that is today’s Healthcare Power Up. One piece of healthcare operations. Pulled apart. Put back together. And hopefully, a little easier to understand than it was twenty minutes ago. Until next time, keep asking the question that tends to make healthcare a lot more interesting: What actually happens next?