A couple years after I joined Kannact, we had made real progress…
Patient satisfaction had climbed to 96% and NPS reached 82.
Net churn stayed below 2%.
Coach efficiency improved more than 3×.

But even with those gains, the employer benefits space was changing quickly. Giants in the industry had already captured much of the market, and client contract renewals were dropping despite strong outcomes. It became clear that incremental improvements wouldn’t be enough. We needed to rethink our entire business strategy.

So we started looking beyond employers and mapping where real gaps existed… places where our experience in patient engagement, chronic condition management, population health, and virtual support could create meaningful clinical and financial impact.

We had already shown we could close HEDIS gaps inside Kannact’s employer-sponsored program. Among patients who stayed with us for over two years, 85% of hypertensive patients achieved blood pressure control, and patients with diabetes saw a 1.7-point reduction in HbA1c, surpassing national HEDIS benchmarks. We were also seeing 28% fewer emergency room visits among supported patients, driven by earlier intervention and steadier follow-up. Together, these outcomes translated into roughly $7k PMPY in reduced annual medical costs.

At the same time, value-based care was accelerating. Primary care clinics and health systems were taking on more risk and needed partners who could extend care beyond the clinic walls.

That combination sparked the idea for a virtual clinic.

If we could partner with providers to help their patients stay healthier between visits, it would directly improve their value-based care performance. Those improvements translate into shared savings and real revenue under value-based contracts. And that’s exactly what these organizations cared about.

Validating the opportunity

We began landing demos with large health systems and independent clinics, especially organizations that had tried to build remote patient monitoring (RPM) programs internally and found them too costly or difficult to scale, or those facing high total cost of care, readmissions, and capacity constraints.

I ran these demos alongside our CEO. I walked teams through the end-to-end experience and workflows, and he tied the model to financial outcomes and potential ROI. Even though the clinic was still largely conceptual, the response was strong. Leaders could see how our coaching model, device infrastructure, and demonstrated health outcomes could support their goals.

Because our model was fully remote and dedicated solely to this work, we could offer more touch points, more follow-up, and more day-to-day support than most health systems could realistically provide. That mattered. Leaders saw it as a scalable way to extend care without adding operational burden.

Those demos validated product-market fit well before the clinic technically existed. They gave us the confidence to evolve from an employer-focused coaching program into a true virtual clinic.

And that’s how Starlight was born.

The Gap We Needed to Close

Our program excelled at lifestyle support and long-term engagement, but it didn’t yet meet the clinical or operational standards required to function as a virtual clinic.

At the time, we were missing:

• Licensed providers to deliver Telehealth Evaluation & Management (E/M) visits
• Documentation, time tracking, and clinical oversight needed for billing E/M, RPM, and chronic care management (CCM) codes
• Structured insurance verification workflows
• A clean referral and intake process aligned with primary care and cardiology workflows
• Clinical documentation standards and a way to share structured notes back to providers
• Escalation pathways and medical decision making rules for high-risk situations
• Compliance workflows for telehealth, scope of practice, and multi-state care

Becoming a virtual clinic meant rethinking the entire model, end to end.

Creating Referral Pathways That Helped Us Become Part of Their Workflow

A clinic’s relationship with a patient starts the moment their provider makes a referral. That handoff needed to feel effortless for clinic staff and predictable for the referring team. So we built the referral experience first, because everything else depends on getting that moment right.

Providers could:

• Refer directly from their EHR, or use a secure landing page when their system didn’t support that
• Add notes or instructions about why they were referring the patient, including focus areas or context our team should pay special attention to
• Trigger an automatic text message to the patient with next steps on joining Starlight

Behind the scenes, we built:

• A predictable outreach sequence within hours of the referral
• A structured handoff that pulled in the clinical and contextual data we needed from the start
• Escalation rules for clinical concerns that needed to be shared with the referring provider
• Clear communication back to the referring provider, including regular reports on engagement, individual patient updates, and population-level trends

This made our clinic feel like an extension of their practice, not an external vendor. Instead of a fragmented referral, it created a trusted, predictable handoff that set the tone for everything that came after.

Creating a Transparent Insurance and Coverage Experience

Once a referral came in, the next step was helping patients understand their coverage. Cost came up in almost every first interaction, and for many people, this was the opportunity to build trust and the moment when they decided if they actually wanted to join.

To design this experience and teach it to my team, I spent weeks learning the entire process myself. I created a secure form to collect insurance details, tested the accuracy of cost estimates across multiple eligibility channels, and mapped out which service type categories our RPM and E/M codes fall under so we could return reliable out of pocket estimates.

I also called patients directly to explain deductibles, coinsurance, and cost expectations in plain language.

From that work, we built an intake model that delivered:

• A clear out-of-pocket estimate
• Transparent explanation of benefits
• Options for financial assistance
• A self pay pathway when needed
• Personalized follow up when coverage was unclear

Patients understood their coverage, and that transparency built trust quickly, contributing to referral conversion rates approaching 70%, even in complex clinical populations.

Capturing the Right Consents Before Care Began

Once coverage was clear, we needed to gather the consents that allowed us to deliver care safely and compliantly. This step had to be simple for patients and thorough enough to support clinical regulatory compliance.

I designed a consent flow that captured:

• Informed consent for Telehealth & RPM services
• Financial responsibility acknowledgement
• Notice of Privacy Practices
• Authorization for release of protected health information
• Caregiver involvement authorization

Placing this early in the intake ensured our clinical team had what they needed before the New Patient Visit. It also created clarity for patients so nothing felt unexpected later in their journey.

Designing the New Patient Visit

Next, we needed a medical visit that was warm, structured, and clinically sound. It was also a required step for establishing patients under the care of our Nurse Practitioners (NPs) so we could bill compliantly for RPM, CCM, and ongoing clinical support.

In the beginning, there were a lot of no shows. Although the visit was required for us, it was clear that patients didn’t yet see the value in it. We needed them to experience it not as a box-checking step, but as something truly helpful. A visit that allowed us to dig deeper into their health history and goals, create a personalized RPM plan, and make sure they had the right devices and testing frequency for what they wanted to work on.

Once we onboarded our NPs, I collaborated closely with them to define:

• Care team scripting to clearly communicate the purpose of the visit, its role in the patient’s care plan, and the importance of attending
• Pre-visit patient workflows covering administrative readiness and technical access to Zoom, including setup and light troubleshooting
• A smooth post-visit handoff to the coaching team with the right context and next steps

This visit became the foundation for clinical care. Once refined, no-shows dropped by >30%, and the visit became the foundation for ongoing clinical care rather than a box-checking step.

Creating Internal Systems That Functioned Like a Clinic

Once the patient experience was defined, the next step was building the operational backbone. A virtual clinic only works if the systems behind it are as rigorous as the care in front of it.

Building the engine for compliant billing

In a virtual care model, billing is where most “care-like” systems fall apart.

Billing the new patient visit (E/M code) is relatively straightforward. There’s a defined visit, clear documentation, and a direct code to submit.

RPM is more nuanced. The codes aren’t tied to a single encounter… they depend on what happens across an entire month: time spent reviewing data, time spent communicating with the patient, and the number of days a device is used.

To make this work compliantly at scale, we created:

• Clinical guidelines for prescribing the right devices based on payer requirements.
• Documentation requirements for billing, compliance, and clinical accuracy
• A way to track interactive communication time, chart review time, and daily device usage
• Monthly audits by a supervising or collaborating physician when required under NP licensing, payer rules, or internal quality standards

We then automated the billing layer itself:

• Real-time E/M claim submission for new visits
• A separate monthly RPM billing job that looks back over the prior 30 days to calculate eligible RPM codes based on total interaction time, chart review, and device-use days

Defining critical alert and escalation pathways

Remote care requires a plan for every scenario. I designed escalation models for:

• Critically high and low glucose readings
• Hypertensive urgencies
• Gaps in device readings
• New symptoms after hospitalization

Each pathway included how to reach the patient, what to ask, when to escalate, and how to loop in caregivers and providers.

Integrating with our referring providers’ EHRs and workflows

To support large health systems, our documentation needed to flow back into their world. I designed our integration strategy using:

• EpicCare Link for provider-facing chart access and clinical context
• Health Information Exchange networks to supplement longitudinal patient records
• Workflows for sharing visit notes and clinical alerts back to referring providers
• Forward-looking data models designed to map cleanly to FHIR as we scaled

This made our work visible and trustworthy inside the systems providers already used.

Building a Post Discharge Pathway for Higher Acuity Patients

As we partnered with larger systems, we began supporting patients during the fragile transition from hospital to home. The first 30 days mattered most.

A more hands-on patient experience designed for acute care

I helped shape the scripting and requirements for our post-discharge program and wellness checks, which focused on:

• Reviewing discharge summaries
• Clarifying medication changes and tapers
• Screening for red flag symptoms
• Confirming follow up appointments
• Reviewing wound care and home instructions
• Identifying caregiver support

Integrating a 24 hour nurse line

We partnered with a 24 hour nurse line so patients had immediate clinical support when something felt off, whether a wound looked concerning, pain escalated, a new symptom appeared, or medication side effects changed. It also gave them clear guidance on when to seek care in person, which helped avoid unnecessary ER visits.

Monitoring with the right intensity

Supporting patients after a hospital stay required a different rhythm than chronic care, so we adjusted our touch points to match the pace of healing.

During the first one to two weeks, we checked in daily or near daily to catch issues early and keep patients grounded while everything still felt new. We built short term monitoring plans that adapted to their symptoms and device trends, and we created rapid escalation rules so any concerning change was addressed quickly.

As patients stabilized, we tapered our outreach so the support stayed present without feeling overwhelming.

We adjusted monitoring intensity during the early weeks and tapered as patients stabilized. More than 65% of referred patients participated for at least 7 days (the most critical ones), and outcomes improved through early detection and steady, reassuring support.

The Clinic We Built

Building the Starlight Clinic wasn’t a single project. It was the work of threading strategy, clinical quality, operational design, and deeply human care into one coherent system. We created a model that health systems could trust, patients could understand, and teams could deliver every day without guesswork.

My role was to take an idea that lived in a slide deck and turn it into a functioning clinic: one with clear referral pathways, transparent onboarding, licensed clinical visits, reliable RPM operations, and transitional care strong enough to reduce readmissions. Along the way, I shaped the scripting, workflows, data models, escalation rules, and experience design that now anchor the program.

The result is a clinic that feels both modern and grounded in real care… a model that supports providers, strengthens patient safety, and blends technology with the kind of day to day support most health systems didn’t have the capacity to provide.

It is one of the things I’m most proud of creating.

The Impact

Today, the clinic we built launched programs and partnerships with Providence, MultiCare, MedStar, and CHRISTUS Health, health systems serving hundreds of thousands of patients across multiple markets.

Seeing our model work in real health system environments at that scale has been one of the clearest signals of product-market fit and a strong validation of what we built, with a clear path to expansion across additional states and partners.