Article
Health benefit costs are set to rise 6.5% in 2026 — the steepest increase in fifteen years — pushing average cost per employee above $18,500 (Mercer, National Survey of Employer-Sponsored Health Plans, 2025). Employers are being asked to absorb that increase, shift it to employees, or find a different way to manage it. We think there's a fourth option: change who's managing the care.
Eighteen months ago, we started with a small, direct-enrollment cohort in Chicago to establish the model — proving out the patient experience, the data and analytics, the financial outcomes, and building the tech stack needed to operationalize it before scaling. Here's what we found.
Why we started small
Medicine rewards patience, not speed. Before we asked any employer to trust us with their population's health and their healthcare spend, we needed to know the model actually worked — for patients first, and for the numbers second.
That meant building the infrastructure before the enrollment: a hybrid virtual-and-in-person care model with a mobile clinic, an analytics layer that tracks every encounter at the same level of detail as a claim, and a way to feed health plan data back into the care delivery tool itself — so we could see cost containment opportunities as they emerged, not months later in a claims report. The cohort was the proving ground for that system.
The results
Across our first ~50 enrolled patients:
- $1,889 average savings per patient per year, based on captured claims — care delivered through Partner Medical and diverted from higher-cost settings, measured against published local hospital pricing at the CPT-code level.
- 100% patient retention.
- 5-star average patient rating.
One patient, describing a difficult medication reaction managed through the practice, put it this way:
"I've never had a better experience with any medical team as I've had with Partner... Daily check-ins, suggestions, a non-judgmental attitude, and problem solving — I was able to get the help and support I needed. You can see that this is a team who cares about their patients. You're a person to them, not just another 'hi and bye' patient."
Every number above is calculated the same way for every patient, and the methodology — visit categorization, pricing source, diversion logic — is shared openly with our partners. This is meant to be a transparent process, not a black box.
This isn't a new bet — it's a proven approach, assembled correctly
It's fair to ask whether one cohort is enough to trust. But Partner Medical isn't asking the market to believe in something untested. We built this model by combining three approaches to care delivery that each already have an independent, well-documented track record of reducing cost:
- Direct primary care. A Society of Actuaries / Milliman actuarial study found employer-sponsored DPC associated with a 12.6% reduction in total claims costs and a 40.5% reduction in ER costs (SOA/Milliman, 2020).
- Onsite and near-site clinics. A claims-based analysis of more than 148,000 individuals found onsite and near-site care reduced total cost of care by an average of 25%, driven by a 39% increase in preventive visits and fewer ER visits and hospital admissions (Premise Health, 2023).
- Telemedicine. A peer-reviewed University of Pennsylvania study found telemedicine episodes cost roughly 25% less than comparable in-person primary care, ER, or urgent care visits — $380 versus $493 per episode (Perelman School of Medicine, 2023).
Each of these models works — to a point, and each has a real gap:
- Direct primary care gives patients access and a relationship, but it isn't built or aligned for cost containment itself. Without a mechanism steering specialty and ER referrals, leakage and suboptimal referrals are common.
- Onsite clinics solve access and continuity, but the economics only work at real scale — an employer needs several thousand on-site lives to justify the build, and even then, they're now running a clinic operation alongside their actual business.
- Near-site clinics avoid that build cost but reintroduce the other problems: leakage, no real continuity or relationship, opaque data, and the same referral and benefit-plan-design gaps that drive people toward high-cost sites of care.
- Telemedicine solves convenience but not the underlying need. Most primary care still requires in-person exams, and without continuity, virtual-only models see high rates of referral to more expensive downstream care.
We built the end-to-end employer tool that closes those gaps: on-site access, virtual access, and a single concierge doctor relationship, unified into one care delivery model — with the referral discipline, continuity, and cost containment none of these approaches deliver alone.
On top of that, we add two things none of these models offer on their own: transparent, encounter-level analytics shared openly with our partners, and a tool that feeds the employer's actual benefit plan design into the clinical workflow — so the provider isn't just treating the patient, but actively directing care toward the lowest-cost, clinically appropriate setting in real time.
Our cohort's results — $1,889 in average savings per patient — are the first proof point of that combined model in practice. They sit squarely inside the range each individual approach has already demonstrated independently. That's the point: we didn't invent a new theory of cost containment. We built the delivery system that finally puts the proven pieces together.
What's next
The cohort phase is complete. We're now partnered with self-insured employers, delivering this same model — and the same level of savings — at scale: concierge-level care delivery, transparent encounter-level analytics, and full risk-sharing with a guaranteed return on investment built into the contract. Not a point solution added to a benefits stack, but a clinical partner in cost containment, accountable for the same outcomes employers are.
As our partnerships grow, so will the data. We'll be publishing more outcomes as our impact expands across new employer populations.
We're building a custom ROI analysis for groups of 50+ employees, at no cost or commitment, in partnership with your benefits team.
Sources: Mercer, National Survey of Employer-Sponsored Health Plans (2025); Society of Actuaries / Milliman (2020); Premise Health, claims-based analysis of onsite/near-site care (2023); Perelman School of Medicine, University of Pennsylvania, telemedicine cost study (2023).
