Between appointments, very little is visible.
Small changes begin to take shape. A missed medication. A shorter walk. A skipped meal. A gradual withdrawal from daily routines. These moments rarely trigger alerts or surface in time. More often, they build quietly until the consequences become harder to avoid.
For older adults aging at home, this period between visits is where health trajectories are shaped. Yet for most health plans, it remains largely out of reach.
This is not just an engagement gap. It reflects a deeper limitation in how care is structured today.
The gap that defines outcomes
Across Medicare Advantage, Medicaid managed care, and dual-eligible populations, the same dynamic shows up again and again. The members who carry the highest risk are often the hardest to reach through traditional engagement channels.
Portals go unused, apps are ignored, and outreach depends on the member opting in. Many simply do not.
At the same time, care management teams are working within real constraints. There are not enough hours or staff to maintain consistent, meaningful contact with a large and growing aging population between visits.
The result is a blind spot at exactly the point where intervention would matter most.
Changes in behavior, adherence, and wellbeing tend to happen gradually and outside of clinical settings. By the time they appear in claims data, assessments, or escalations, the opportunity to intervene early has often passed.
For health plans, this gap directly affects quality measures, member experience, and utilization, all driven by what happens when no one is present.
What changes when engagement lives in the home
A different model is beginning to take shape, one where engagement is continuous rather than episodic.
AI companion technology shifts the dynamic entirely. Instead of waiting for the member to engage, it becomes part of the home environment and part of daily life. It is not another app to open or a call to schedule, but an ongoing presence that interacts naturally throughout the day.
Unlike passive monitoring tools, this model is built on active, ongoing interaction.
Within this model, engagement becomes less transactional and more relational, driving meaningful behavioral change that helps keep seniors stable in the home. Over time, that consistency is what enables real influence on daily habits and outcomes.
An AI companion like ElliQ engages older adults through natural conversation, generating an average of 45+ interactions per day. That level of proactive, consistent engagement is difficult to replicate through traditional channels and is key to reinforcing behaviors that support independence.
This frequency is not just a metric. It is what allows support to happen in context. Medication routines can be reinforced at the right moment, daily activity can be encouraged naturally, and subtle changes in behavior can be noticed as they begin to emerge.
It also creates something that has historically been missing. Visibility.
With continuous interaction comes a continuous stream of real-world engagement data. For care teams and health plans, this provides a clearer understanding of what is happening between visits, without adding operational burden or requiring additional staffing.
Engagement is no longer something that is inferred after the fact. It becomes something that can be observed and acted on in real time.
What the evidence is beginning to show
As this model has been deployed at scale, the results are beginning to move beyond theory.
In a three-year study conducted by the New York State Office for the Aging, participants using ElliQ reported a 96% improvement in overall health and wellness and a 95% improvement in quality of life. Reported loneliness, a factor closely tied to hospitalization risk, cognitive decline, and chronic condition progression, decreased by 93 to 94%. More than 80% reported feeling more connected to the outside world.
At the same time, large-scale public sector deployments, including the Washington State Department of Social and Health Services Medicaid long-term services and supports program, are bringing this approach to high-need populations living at home.
For health plans, the implications extend beyond engagement alone.
Consistent, high-frequency interaction supports medication adherence, increases completion of health assessments, and reinforces behaviors tied to preventive care. Early signals of change in mood, activity, or routine can be identified sooner, allowing for earlier intervention.
While long-term cost impact continues to be studied, the direction is becoming clearer. When engagement becomes continuous, the likelihood of avoidable escalation decreases.
A different way to think about population health
The organizations seeing the strongest results are not treating this as a standalone solution or a limited pilot. They are integrating AI companion technology into a broader population health strategy, extending care into the home, where members spend most of their time and where outcomes are ultimately shaped.
For plans serving Medicare Advantage, Medicaid, dual-eligible, and rural populations, this represents a shift in how engagement is defined. Instead of relying on periodic touchpoints or member-initiated interactions, engagement becomes continuous and embedded in daily life. It begins to function as an always-on layer that supports members between visits while giving care teams a clearer, more actionable understanding of what is happening in real time.
The question is no longer whether older adults can be consistently engaged. Increasingly, the evidence suggests they can. The more relevant question is whether existing models are designed to reach them in the moments that matter, and whether plans are equipped to turn that ongoing visibility into meaningful action.