Voice AI vs. Patient Portals: Why 85% Engagement Is Possible (and 15% Is What You're Probably Getting)

Here is the version of this story I tell at conferences, because it's the version that makes people uncomfortable in exactly the right way.

We built a mental health app for bipolar patients. I'm a clinical psychologist. I cared deeply about it. Thoughtful UX, evidence-based content, everything we knew. Fifteen percent engaged with it weekly. I told myself that was the benchmark. It wasn't. It was just the ceiling of what text-based digital engagement can do for someone managing a serious mental health condition.

Then we called them. AI voice. Their name. How are you doing. Eighty-five percent weekly engagement. The same patients. Different medium.

That number is why HANA exists. And it's also why I keep saying that patient portals and voice AI are not competing technologies. They're doing completely different jobs.

What is the actual patient portal engagement rate in 2024?

According to ONC's 2024 HINTS data, 65% of patients nationwide accessed their online medical records or patient portal at least once in 2024. That's up from 25% in 2014, which is real progress.

But access is not engagement. Among patients encouraged by their provider to use the portal, 87% logged in at least once. Among those not encouraged, 57%. Only 34% of patients are what ONC classifies as frequent users, logging in six or more times per year. For active, ongoing care management, portal-based engagement barely exists for most patient populations.

Why do patient portals fail at ongoing engagement?

Portals are passive. They wait for patients to come to them. That model works fine for lab results and prescription refills. It doesn't work for chronic disease management, post-discharge follow-up, or medication adherence, because those require active patient behavior maintained over time.

Active behavior requires motivation, digital literacy, remembered login credentials, and a working reason to open an app that mostly just holds test results. Most patients, especially older patients and those managing complex chronic conditions, don't maintain that behavior between visits. The portal has everything they need. They don't go back.

Voice AI inverts this. The call comes to them. It asks a specific question. It's done in under three minutes. No login. No app. No remembered password. The cognitive barrier is "answer the phone," which humans have been doing for a hundred years.

What does 85% weekly patient engagement actually mean in practice?

At HANA, 85% weekly engagement means 85% of enrolled patients have a meaningful voice interaction with their care program in a given week, consistently, across populations spanning five countries and three languages. The industry baseline for digital patient engagement in chronic care programs sits at 15 to 20%. That gap is documented in our research and is consistent across clinical settings.

What it means operationally: your care team has current data on 85% of your active patients at any given time. Not chart notes from six months ago. Not portal messages that went unanswered. Actual structured clinical check-in data from this week. That changes what's possible in between-visit management.

Which patient populations does voice AI engage better than portals?

Older patients, patients with limited smartphone literacy, patients managing multiple chronic conditions, and post-discharge patients are all categories where voice consistently outperforms portal-based engagement.

ONC data confirms that while 65% of patients nationally accessed a portal in 2024, the number drops significantly for populations without strong provider encouragement. Only 57% of patients not actively encouraged by their provider ever logged in. For a busy clinic, that encouragement burden is real. Voice AI doesn't require that encouragement because it initiates contact.

Is there a scenario where patient portals are the better tool?

Yes. Portals are the right tool for information delivery that patients actively want and seek out: test results, clinical notes, appointment summaries, prescription management. Patients with high portal motivation and digital comfort, often younger, more educated, managing a specific acute concern, use portals effectively and frequently.

The mistake is treating portals as a chronic disease engagement tool when the engagement data consistently shows they're not. Portals and voice AI solve different problems. Portals bring the health record to the patient. Voice AI brings the care team to the patient. Both have a place. HANA's use cases show where voice fits alongside existing portal infrastructure, not instead of it.

Key Takeaways

The 85% versus 15-20% engagement gap isn't a technology marketing claim. It's what happens when you stop requiring patients to take action and start delivering care to them. Patient portals are excellent tools for patient-initiated interactions. They're poor tools for clinician-initiated chronic care management at scale. Voice AI is built for the second job: the proactive, regular, documented touchpoint between visits that keeps patients engaged with their care plan and keeps care teams informed between appointments. If your current engagement numbers look like the industry baseline, the tool isn't the problem. The direction of contact is. Book a discovery call if you want to see what this looks like for your specific patient population.

FAQ

Why is voice AI engagement so much higher than app-based patient engagement? Voice removes every friction barrier that kills engagement: no login, no app to remember, no digital literacy requirement, no motivation to initiate contact. The care reaches the patient instead of waiting for the patient to reach the care. For chronic disease populations, where sustained behavioral change is the goal, that inversion is the entire mechanism.

Do patients actually accept and respond to AI voice calls from their care provider? Yes, at significant scale. HANA has processed over a million patient interactions across multiple countries and languages with zero critical adverse events, documented in our case studies. Patient acceptance rates are highest when the AI is introduced by the clinical team, identifies itself clearly, and asks specific clinically relevant questions. Generic robocall patterns are what patients reject. Structured clinical follow-up is what they accept.

Can voice AI and patient portals work together in the same care management program? Absolutely, and this is the optimal model. Voice AI handles the proactive outreach, symptom monitoring, and engagement layer. The portal handles patient-initiated information access, messaging, and records review. They address different patient behaviors and different moments in the care journey. HANA integrates with existing EHR and portal infrastructure rather than replacing it.