Reducing the cost of care comes down to something deceptively simple: making every interaction count, whether that means getting a member to the right specialist, catching a care gap before it turns into an ER visit, or giving a clinician the full picture instead of half of one. Care coordination has always been one of healthcare’s strongest tools for improving outcomes, but what’s changing is how much further that tool can now reach.
Members don’t experience healthcare the way plans organize it
Health plans are built around products, benefit tiers, and departments. Members don’t think in any of those terms. To them, healthcare is one continuous experience, and they expect it to feel that way regardless of who’s paying for what.
That disconnect shows up most clearly in how plans think about member engagement with new tools and who they assume will use them. The safest assumption, that high-risk, high-need members will resist new technology, keeps turning out to be wrong. Home-based care programs built around exactly that population are seeing some of the strongest engagement, not the weakest. A few plans are now pairing wearables with in-home coordination for these members, giving care teams a live read on someone’s condition between visits instead of waiting for the next appointment.
Channel matters as much as content. How plans reach members matters just as much as what they offer them. The right intervention delivered through the wrong channel can still miss the member entirely. Effective care coordination means understanding not only what a member needs, but how and when they are most likely to engage.
Personalizing the channel alongside the message can make even simple outreach more effective.
Context matters
Clinical data can tell you what is wrong, but it does not always explain why someone is struggling to address it. It could be a lack of transportation, difficulty accessing food or housing, or simply a work schedule that makes it hard to get to an appointment. Plans often don’t know about these challenges unless a case manager takes the time to ask.
When plans screen for social needs and connect people to services alongside clinical care, they get a clearer view of what may be keeping members from getting better. That visibility can reduce avoidable spend and help members access the care they need but may not have been receiving, especially when clinicians can see social and environmental barriers alongside the clinical record.
The same visibility matters when managing spend. The cost of the same service can vary significantly depending on where and by whom it is delivered. The goal was never more data. It’s the right piece of it, reaching the right person at the time they need it.
AI works best with Human judgment
Every serious conversation about AI in care coordination eventually lands in the same place: it is an accelerator, not a substitute. It can summarize calls, consolidate a scattered record, flag a care gap before it’s missed, and reduce the time staff spends on paperwork so they can focus on people. That shift is already measurable at the payer level. According to HealthEdge’s 2026 Healthcare Payer Survey, 94% of payers are either live with AI or actively adopting it.
There’s a hard line, though. AI can support a recommendation to approve, but it doesn’t get to deny one. Every recommendation must be explainable and traceable to its source, with a person accountable for decisions carrying clinical, financial, or regulatory weight. What’s changed is not who decides.
It’s that the reasoning behind the decision is now visible.
AI is only part of the Equation.
Putting AI into care coordination is one thing. Making it work within the realities of a health plan is another. Data may sit across disconnected systems, workflows can span multiple teams, and new capabilities need the right governance and skills around them. Gartner’s 2026 AI Maturity research for US healthcare payers points to similar constraints, identifying gaps in data, governance, and skills that can limit AI readiness and scale.
Those realities also look different from one plan to another. The operating environment of a national plan with a large self-funded population can be very different from that of a regional, fully insured plan. That shapes where AI can create value, how it fits into existing workflows, and how quickly it can scale.
Training is just as important. AI cannot remain the domain of an innovation team or a small group of specialists. The people doing the work need to understand how to use it, where it can support their decisions and where human judgment still matters. And plans need clear baseline measures before deployment so they can determine whether AI is improving efficiency, care, and member outcomes.
Move deliberately, not slowly
Healthcare caution is typically warranted. But there comes a point when it becomes counterproductive, and the risk of inertia begins to outweigh the risk of action. At the scale health plans operate today, relying on manual processes alone is increasingly difficult to sustain.
Moving faster does not mean abandoning governance. It means creating room to test, learn, and adapt while maintaining the clinical oversight that healthcare demands. The health plans that get this right will be those that use technology to make coordination easier for the people delivering care and more connected for the members receiving it.
Care coordination done right can become a strategic asset, improve outcomes, and lower the cost per member per month.
