Inside our April 2026 panel with EthAum Venture Partners — and the metrics our Founder & CEO Chrissa McFarlane says payers should be measuring instead of “engagement.”
In April 2026, Patientory teamed up with EthAum Venture Partners — a Singapore-based deep-tech fund — to host a candid, cross-border panel on a question every payer and health system is wrestling with right now: which AI health data KPIs actually move the needle?
Moderated by Pankaj Gupta (EthAum Venture Partners), the panel brought together five leaders working on the problem from different angles: Basak Altan (design strategist focused on AI in healthcare), Dr. Navneet Kathuria (Commissioner New York State Insurance Fund), David Dobbs (former Chief Data Officer at a Blue Cross Blue Shield plan and ex-UCSF Director of Data & Analytics), Manoj (GM & SVP of Product at HealthEquity), and our own Founder & CEO, Chrissa McFarlane.
The full conversation is on YouTube — embed below — but here’s what stood out, and why it matters for anyone shaping payer strategy, population health, or healthcare AI in 2026.
▶ Watch the full panel: https://www.youtube.com/watch?v=wuuLA8YroKQ
The premise: in US healthcare, AI is no longer a “nice to have”
Pankaj opened the panel with a framing the rest of the discussion built on: payers and health systems are sitting on oceans of claims, clinical, social-determinants, and now patient-generated data. Because of that, AI isn’t optional anymore — it’s the engine that turns data into measurable value.
But not all KPIs are created equal. Plenty of dashboards still measure activity that looks like progress without driving any. The panel set out to pressure-test which metrics actually translate to better outcomes, lower cost, and more trust between members, providers, and payers.
When her turn came, Chrissa cut straight to the three KPIs Patientory is operationalizing with payer partners today.
KPI #1 — Engagement-to-adherence conversion rate
Most digital health programs report engagement: app opens, coaching sessions logged, wearables paired, modules completed. But Chrissa’s point was sharp:
Engagement alone doesn’t drive outcomes. Behavior change does.
The metric that matters is the conversion from engagement to adherence — how effectively member interactions (app usage, coaching, wearable data) translate into real behavior: medication adherence, daily activity, care-plan compliance. This is the KPI that connects digital touchpoints directly to cost reduction and clinical improvement, which has become a top priority for payers operating in value-based arrangements.
If a population is highly “engaged” but adherence isn’t moving, the program isn’t working — and the engagement number is hiding that.
KPI #2 — Real-time risk detection & intervention latency
The second KPI is about speed: how quickly can a payer identify a rising-risk member from longitudinal data signals, and how quickly can they intervene?
This isn’t a vanity metric. The largest cost drivers for payers — avoidable ER visits, readmissions, disease progression — are downstream of latency. Every hour between a signal (a missed refill, a vitals anomaly, an early ED encounter) and an actual intervention is opportunity lost.
Reducing intervention latency is where AI earns its keep. It’s also where the data layer underneath everything has to be in place: stale claims data on a 30–90 day lag can’t power the same kind of intervention that near-real-time device or engagement data can.
KPI #3 — Longitudinal data completeness per member
The third KPI is the one most easily overlooked: the percentage of members with continuous, multi-source data across claims, clinical records, wearables, and behavioral signals.
Models are only as effective as the data they ingest. Clean data in, clean data out.
The differentiator in 2026 isn’t more data — it’s connected, continuous data across the care journey. Coverage of claims is table stakes; what makes AI predictions actually useful is when the model can see medication-adherence patterns, real-world activity, recent labs, and care-plan interactions side by side, over time.
Without that completeness, predictions are educated guesses dressed up as confidence scores.
The thread tying all three KPIs together
A theme ran through the entire panel — Basak Altan called for a shift from “AI-ready data” to decision-ready data, David Dobbs walked the group through his TACT framework (Trustworthiness, Accuracy, Completeness, Timeliness), and Dr. Navneet reframed data quality as the way clinicians “minimize the uncertainty gap” before they make a decision.
Chrissa’s framing tied it all together:
“AI readiness isn’t just about more data. It’s about connected, timely, and behaviorally rich data that reflects what’s actually happened with a member from day to day.”
Three KPIs, one underlying truth: AI’s value to payers and health systems is gated by whether the data layer can deliver connected, timely, contextualized signal at the moment a decision needs to be made.
Why this matters now: interoperability is finally at the forefront
Chrissa pointed to one structural shift that makes 2026 different from prior years: interoperability is no longer aspirational. CMS’s push toward FHIR-based data sharing, combined with the maturity of decentralized infrastructure, means payers and providers can now realistically expect the kind of multi-source completeness AI requires.
That changes the math. Claims data delayed 30 to 90 days behaves very differently from near-real-time device or engagement data — and that gap is where payers are leaving the most money and outcomes on the table.
Where Patientory fits in
Patientory is built for exactly this moment. We are a global health data infrastructure platform that aggregates health information into a personalized health data wallet members can manage and control — designed to prevent chronic disease at scale.
Our decentralized infrastructure is what makes the multi-source completeness story practical: securely aggregating EMR data, wearable signals, lab results, and behavioral data into a single, member-controlled view that AI models — and the care teams behind them — can actually act on.
A few markers of momentum worth sharing:
- Notable investments from Kaiser Permanente and Blue Cross Blue Shield
- Recent expansion into the Asia market, beginning with South Korea
- Continued board service on CareEquality, the government-adjacent body advising on patient access to data
- Active participation in the CMS health tech ecosystem in the United States
Watch the full panel
The 60-minute conversation goes deeper into data quality frameworks, the consumer-experience angle that Manoj raised, the clinician-trust perspective from Dr. Navneet, and how each panelist measures ROI on AI investment in real organizations.
If you work in payer strategy, population health, healthcare AI, or product at the intersection of all three — it’s worth your time.
▶ Watch on YouTube: https://www.youtube.com/watch?v=wuuLA8YroKQ
Keep the conversation going
If anything from the panel sparked a question, a partnership idea, or a “we should talk” moment — we’d love to hear from you. Reach out to the Patientory team at support@patientory.com or learn more at patientory.com.
A genuine thank-you to EthAum Venture Partners for co-hosting, to our co-panelists for a sharp, candid exchange, and to everyone who joined us live.