Why Patient Portals Fail in Behavioral Health — and What Comes Next
The patient portal is one of the great disappointments of the last decade of health IT. Half of behavioral health providers surveyed by Cantata Health Solutions reported portal use below 10% at their organization. One in five agencies did not track portal usage at all. And 57% of consumers rarely used their portal consistently. At NatCon 2026, three speakers laid out why the patient portal has been failing behavioral health, and what has to come next.
Three things every behavioral health tech leader should know
- Traditional patient portals fail in behavioral health because they are transactional, fragmented, and provider-centric. Consumers show up, sign forms, and leave. Recovery, meanwhile, is happening in the days and weeks between sessions with no support tool in place.
- The next generation of engagement technology is not a portal. It is a consumer engagement hub. Mobile-first, multimodal, multilingual, self-directed, with a data feedback loop into the EHR.
- The design principles are set. The evidence is emerging. Recent studies of digital mental health apps show that between-session tools measurably improve engagement and treatment adherence.
Why Traditional Patient Portals Fail in Behavioral Health
Christy Winter, LMSW, is Chief Product Officer at Cantata Health Solutions, the company behind the Arize EHR platform. She is a social worker by training with a decade of provider-side experience before moving into product. Her opening framing at NatCon 2026 named the problem exactly.
"Recovery is happening in between those sessions. Portals aren't necessarily doing the job that we thought that they would. They've been more administrative in their intent. It's more about transactional. It's about sending documentation, obtaining signatures, checking on lab results."
— Christy Winter, LMSW, Chief Product Officer, Cantata Health Solutions
Winter identified four structural failures of the traditional behavioral health portal.
Transactional intent. Portals were built for the administrative functions of care (documents, signatures, appointment reminders, lab results), not for the therapeutic content of care.
Fragmentation. A patient with a behavioral health provider, a primary care provider, and a specialist might have three separate portals with three sets of credentials, none of which talk to each other.
Provider-centric design. Portals were designed for the organization to collect information from the consumer, not for a two-way exchange in which the consumer contributes their own data and controls what gets shared.
Accessibility barriers. Different browser requirements, mobile compatibility gaps, and login friction. Small technical failures aggregate into abandonment.
What Cantata Learned From Talking to 117 Providers and 212 Consumers
Cantata partnered with Matt Kudish, principal at the consultancy We Better Work and former CEO of NAMI-NYC Metro, and with OpenMinds to survey 117 senior leaders across community mental health centers, FQHCs, and CCBHCs, along with 212 behavioral health consumers across diverse treatment settings and recovery timelines. They then held focus groups to dig deeper.
Kudish walked through what they found.
"Nearly half of the providers we surveyed showed that portal use was below 10% at their organization. One in five didn't track any portal usage at all, and 57% of patients rarely use their portals consistently. Portals are everywhere, but actual meaningful engagement, not so much."
— Matt Kudish, Principal, We Better Work
On the consumer side, roughly 45% were logging in weekly or more often. But only 60% strongly agreed that the portal supports their recovery. Nearly one in three said it either did not, they were neutral about it, or they did not use it enough to have an opinion.
The most valuable data point in the survey may be the alignment between consumer requests and consumer challenges:
- 53% struggle with stress — 49% want mindfulness and coping tools
- 38% don't know if they're progressing — 47% want progress tracking
- 31% forget coping skills between sessions — 49% want reminders
"This alignment is exactly what we were looking for. When you engage consumer voice in the design of your product, you're building something that's addressing their needs, as opposed to guessing what their needs really are."
— Matt Kudish
The Five-Phase Consumer Journey
Instead of a monolithic portal, Cantata's design framework breaks the consumer's care journey into five phases, each with different digital touchpoints:
- Awareness and engagement. How the consumer discovers care and takes the first step.
- Intake and assessment. The initial contact through screening and treatment planning.
- Active treatment. Session-based work and between-session support.
- Monitoring and adjustment. Ongoing measurement of progress and course correction.
- Recovery and long-term support. Sustained engagement after acute treatment.
The tools that help someone manage a panic attack on a Saturday, Winter noted, are different from the tools that help them prepare for a Tuesday clinician session. A portal cannot be a single feature. It has to be a hub.
What a Consumer Engagement Hub Actually Contains
The Cantata team designed the app that supports Arize based on four guiding principles surfaced by both provider and consumer feedback.
Mobile-first. Behavioral health consumers are more likely to have a phone than a computer.
Multimodal. Text alone is not enough. Consumers want to journal by voice, by picture, by drawing, and yes, by typing, depending on mood.
Multilingual. Journaling and reminders need to work in the consumer's native language.
Consumer-controlled. Reminders and nudges are opt-in. Content sharing back to the provider is opt-in on a per-item basis.
Sarah Parks, a Senior Solutions Consultant at Cantata and a former Certified Alcohol and Drug Counselor, demoed the app during the panel. Highlights of what she showed:
- Mood check-ins at customizable frequency, with the ability to alert clinical staff if a score crosses a threshold.
- Journaling with text, voice, drawing, and photo modalities. Each entry has a per-entry privacy control (the consumer decides whether the clinician sees it).
- Coping skill practices, including breathing exercises and a bubble-pop micro-game for immediate anxiety relief.
- Assignments from the clinician (a PHQ-9, a homework exercise) that appear as scheduled tasks.
- Medication reminders with the consumer able to rename the alert (e.g., "take dog for a walk later" as code for "take medication") to preserve privacy.
- A 988 button that dials the Suicide and Crisis Lifeline directly if the consumer is in crisis.
- Light and dark modes, plus language toggles, so the interface matches the consumer's mood and comfort.
One design choice that came out of the focus groups was worth noting: the streak counter. Consumers told the team that traditional gamification streaks made them feel penalized for missing a day. Cantata's version instead counts total days engaged (e.g., "your 98th time doing this") without breaking a streak. A small change with meaningful psychological impact.
The Feedback Loop Back to the Clinician
The engagement hub is only half the value. The other half is the data feedback into the EHR.
Parks demoed the provider view. A clinician preparing for a session with a consumer named Sandra can pull up a "consumer console" that shows recent mood check-ins, journal entries the consumer opted to share, PHQ-9 and GAD-7 scores over time, medication adherence, and completed assignments. The clinician can also configure automated alerts. A rising PHQ-9 score triggers a text or email to the assigned staff, who can then route through the EHR to see the client's full record and follow up.
"Now when the client comes to their session, you're not spending the first five or 10 minutes saying, 'How was your week? Did you do this?' You now have that insight, and you can dig deeper. You can get personalized in your sessions and get the most out of that 30 or 45 minutes."
— Sarah Parks, Senior Solutions Consultant, Cantata Health Solutions
Just as important: the clinician can see what is not working. If the consumer is not journaling but is doing breathing exercises three times a day, that is a signal to lean into the intervention that is landing.
What the Emerging Evidence Shows
A January 2026 study in JMIR Formative Research examined between-session engagement on the Wysa mental health platform among 1,210 users who subscribed to text-based coaching sessions. The findings: users who engaged with adjunct app features (conversational agent, self-management tools, journaling) showed measurably higher adherence to their coaching sessions than users who did not.
A January 2026 scoping review in JMIR Mental Health synthesized 38 studies on mental health app integration in clinical practice. The two features practitioners cited most frequently as essential were engagement (79% of studies) and interoperability with the EHR (76% of studies). Both are directly aligned with the Cantata design principles.
The evidence base is early, but the direction is consistent. Between-session digital engagement improves adherence. Data flowing back to the clinician improves in-session efficiency. And the consumer's sense of agency, control over what gets shared, and personalized experience are what make the tool actually used.
Frequently Asked Questions From the Audience
Q: What about the open notes movement? Can consumers see the clinician's notes?
Kudish noted that this did not come up in the survey or the focus groups, though the 85% of consumers who reported comfort sharing wellness data did mention their preference for selectivity in what flows back to their provider. Bidirectional openness is not yet baked into the design, but consumer control over the outbound direction is.
Q: Consumers don't really own their data. What have you done about that?
Winter's answer was honest. The team has not solved the data-ownership problem, which is genuinely complex. What they have solved is the data-control problem. User preferences let consumers decide, entry by entry or globally, what flows to the provider and what stays private.
Q: Can this plug into an existing EHR that isn't Arize?
Not yet. The consumer engagement hub is currently integrated only with the Arize platform.
Q: How does the 988 button work? Does it record the crisis contact back to the EHR?
The button dials the 988 Suicide and Crisis Lifeline. The EHR-side documentation of the crisis contact is a feature Cantata is still building out.
Q: Do you have accessibility features for visual impairment, such as read-aloud?
Not yet. Winter acknowledged this as a real gap and noted the team is looking at browser-based accessibility components.
The Bottom Line
The patient portal was a first-generation answer to a problem that had not yet been correctly named. The correct name for the problem is between-session engagement, and it is going to be solved not by a document repository behind a login screen but by a consumer engagement hub that meets people where they actually are: on their phone, in the middle of a difficult Saturday, with a coping tool one tap away and a data feedback loop that makes their next session shorter, more focused, and more personal.