ksKearny StreetManagement
Back to Insights
Strategy8 min readMay 14, 2026

HEDIS starts in the network: designing for quality measures

Kearny Street Management

Most plans treat HEDIS as a year-end scramble. The measurement window closes, the abstraction vendors descend on charts, care-gap outreach campaigns fire off, and everyone holds their breath waiting for the audited rates. When the numbers come back short, the instinct is to double down on the same downstream fixes next year: more outreach, more chart chases, more incentives.

The uncomfortable truth is that many of your HEDIS results were decided long before any of that, at the moment you built your network. The providers you contract, where they sit, how engaged they are, and whether their systems can send you clean data determine how many gaps you can close and how much of that closure you can actually prove. If you want better HEDIS scores, the highest-leverage work happens at the design stage. Here is how to build a network with quality measures in mind from day one.

What HEDIS is and why the network sets the ceiling

HEDIS, the Healthcare Effectiveness Data and Information Set, is the standardized set of quality measures maintained by the National Committee for Quality Assurance (NCQA). It spans dozens of measures across domains like preventive screening, chronic condition management, behavioral health, and medication adherence. For Medicare Advantage plans, HEDIS results feed directly into CMS Star Ratings, and Star Ratings drive both enrollment and the quality bonus payments that make or break a plan's economics.

The precision of HEDIS is not optional. Every organization that reports HEDIS must pass an independent HEDIS Compliance Audit that reviews its information-system capabilities and reporting processes before results can be certified. A rate you cannot substantiate through auditable data is a rate you cannot claim.

This is why the network sets the ceiling. A HEDIS measure is only satisfied when the right service is delivered to the right member and captured in data the plan can retrieve and defend. Every part of that chain, the delivery and the capture, runs through your contracted providers. A network that is too thin, poorly located, disengaged, or unable to transmit clean data caps your achievable score no matter how hard you work the back end.

Which measure categories are network-sensitive

Not every HEDIS measure depends equally on network composition, and knowing which ones do lets you focus your build. The most network-sensitive categories share a trait: they require a specific service, from a specific type of provider, delivered inside a measurement window.

  • Preventive screenings such as breast cancer, colorectal cancer, and eye exams for members with diabetes. These require the member to physically reach a specific provider type, so access and geography drive the achievable rate directly.
  • Chronic condition management such as blood pressure control, diabetes care including HbA1c testing and retinal exams, and kidney health evaluation. These depend on primary care panel capacity, empanelment, and the availability of the relevant specialists.
  • Behavioral health engagement and follow-up after hospitalization or an emergency visit for a mental health or substance use condition. These are chronically undersupplied and now more heavily weighted, and the follow-up measures are extremely sensitive to whether behavioral health capacity even exists in-network.
  • Medication adherence and management, which depend on the prescribing patterns of your contracted providers and their willingness to engage in medication reconciliation and review.
  • Transitions and follow-up after acute events, which require coordination among hospitals, primary care, and specialists who are actually connected to one another.

Gap closure is a function of network composition and engagement

A care gap is a member who is eligible for a HEDIS-measured service and has not received it inside the window. Closing that gap requires a provider who can deliver the service and a member who can reach that provider. When the network is well composed, gap closure is a matter of outreach and scheduling. When it is not, gap closure hits a wall you cannot spend your way past.

Consider the diabetic retinal exam. If your network has too few ophthalmologists and optometrists in the counties where your diabetic members actually live, no volume of reminder calls will close those gaps, because there is nowhere to send members. The same logic governs colonoscopy access, mammography, and behavioral health follow-up. Composition sets the ceiling; outreach can only fill the room beneath it.

Engagement is the second half. A contracted provider who never returns your calls, ignores gap lists, and will not participate in quality initiatives is, for HEDIS purposes, a partial dead end. The measures that improve fastest are the ones owned by engaged primary care groups who see the gap reports, act on them, and are aligned through value-based arrangements that reward closing gaps. Network design is not just who signs a contract; it is who will actually work the panel with you.

The data path is as important as the care path

Here is the trap plans fall into repeatedly. The care was delivered, the member got the screening, the gap is genuinely closed, and the plan still fails the measure, because the service was never captured in retrievable data. Under a HEDIS compliance audit, undocumented care does not exist.

There are two ways HEDIS captures a service: administrative data, meaning claims and encounters, and supplemental data such as electronic clinical feeds, registries, and validated chart abstraction. Administrative capture is cheaper, faster, and more defensible. That makes a provider's ability to submit clean, coded, timely claims and encounters a quality asset, not just a billing convenience. A provider whose encounter data is late, miscoded, or missing forces you into expensive chart chases to prove care that already happened.

This has a direct design implication. When you evaluate a provider or group for the network, evaluate their data capability alongside their clinical one. Can they submit encounters cleanly? Can they support an electronic clinical data feed for supplemental capture? A network of high-quality providers with poor data hygiene will still underperform on HEDIS, because you cannot prove the quality you are delivering.

Designing for quality from day one

Building for HEDIS from the start does not mean a separate initiative bolted onto network development. It means folding a quality lens into decisions you are already making about who to contract and where.

Practically, that means overlaying your target membership's likely care needs onto the network map: where the diabetics, the members due for cancer screening, and the members needing behavioral health actually live, and whether the network puts the right provider types within reach. It means prioritizing primary care groups with the capacity and appetite to manage panels, not just accept them. It means treating behavioral health access as a first-class requirement rather than an afterthought, given how heavily those measures now weigh. And it means writing data expectations and, where possible, quality-linked incentives into contracts from the outset, so engagement is structural rather than something you beg for each measurement year.

How we approach it

We build networks with the quality scorecard in view from the first county analysis. When we assess whether a provider or group belongs in your network, we look past the fee schedule to the questions that decide HEDIS: does this provider sit where your measured members live, will they engage with gap closure, and can their systems return the clean encounter and clinical data that make closure provable?

That produces a network where quality is a property of the composition itself, not a rescue mission every fourth quarter. Outreach and gap-closure programs still matter, but they work with the grain of the network instead of straining against its limits. The plans that win on Star Ratings are not the ones that abstract the most charts. They are the ones whose networks were designed, from day one, to make quality both deliverable and provable.


Ready to start?

Two weeks. A build plan worth running.

Fixed fee, no commitment past the diagnostic. You walk out with a plan — whether you run it with us or not.

Schedule the diagnostic