Blog

August 7, 2026

Healthcare Quality Measurement: Methods, Metrics & Why It Matters

Key Takeaways

Most healthcare quality measurement frameworks — including HEDIS, CMS Star Ratings, and CAHPS — were built to evaluate health plans and hospitals
Patient experience surveys like CAHPS measure how care felt, not actual clinical quality, and because of that, ratings built on them steer employees toward bedside manner rather than tangible outcomes
Measuring individual providers requires all-payer claims data, sufficient patient volume, and risk adjustment, a bar most standard rating systems were never designed to meet

Healthcare quality measurement is one of the most established disciplines in American healthcare. The National Committee for Quality Assurance (NCQA) has spent three decades refining HEDIS, its standardized tool for comparing health plan performance. CMS attaches star ratings to every Medicare Advantage plan, and Leapfrog grades thousands of hospitals twice a year. And yet almost none of that answers the question employers (and employees) should care about most: which specific doctor should an employee see?

Most quality measurement systems were built to evaluate health plans and hospital systems, not individual physicians, and institutional data says very little about what actually happens in the exam room. Before relying on any of these ratings, it helps to know what each framework is and is not designed to do.

What Healthcare Quality Measurement Evaluates

Nearly every quality measurement framework rests on three foundational categories: structure, process, and outcome.

Measure type	What it evaluates	Examples
Structure	Whether the right resources and systems are in place	Nurse staffing ratios, electronic health records, ICU specialist coverage
Process	Whether care followed accepted clinical standards	Blood sugar testing rates for diabetic patients, timely antibiotics before surgery
Outcome	What happened to the patient	Complication rates, readmissions, recovery time

The frameworks built on this foundation, namely HEDIS, CMS Star Ratings, and CAHPS, were designed primarily to compare health plans and institutions. Physician-level HEDIS measures do exist, but they are narrower in scope, and report card sponsors have not widely adopted them. Whether any of these tools belongs in an employer decision depends on knowing exactly what each one does and does not measure.

Structure and process measures are the easiest to collect, since the data comes straight from administrative records. But they only show whether the conditions for good care are in place, not whether good care was delivered. A hospital can staff well and follow every protocol, yet still produce mediocre results. Outcomes are the most meaningful measure of quality, but they are also the hardest to attribute to a specific provider, as a patient's recovery depends on a number of factors, including their underlying health, their other doctors, and random variation.

Where Individual Provider Performance Fits In

Doctor performance varies widely, even between physicians in the same specialty at the same hospital. Two in-network surgeons can produce very different complication rates, recovery times, and total costs for the same procedure. That variation is the reason evaluating individual providers matters. Institutional ratings average it away, and the doctor an employee actually sees determines the outcome they get.

Measuring an individual physician is also a more demanding exercise than measuring a health plan. A single doctor sees a limited number of patients, so drawing statistically meaningful conclusions requires several forms of data, including:

Claims data from every payer the doctor bills rather than one payer's slice
Risk adjustment, so a surgeon who takes on complex cases is not penalized for treating sicker patients
Enough volume to separate genuine skill from statistical noise

The individual provider measurements that exist today, physician-level HEDIS measures and CMS's MIPS scores, are narrow in scope, weighted toward Medicare patients, and limited to a small subset of clinical areas. Garner's methodology goes significantly further, scoring individual physicians across specialties using all-payer claims data and risk adjustment at the individual level.

The Most Widely Used Healthcare Quality Measurement Frameworks

Four frameworks dominate the conversation. Each works well for its original purpose, but none of them was built to help an employee find the right a doctor.

Rating system	What it scores	Measure count	Population covered
HEDIS	Health plans	90+ measures	235M+ enrollees in reporting plans
CMS Star Ratings	Medicare Advantage and Part D plans	Up to 43 measures	Medicare enrollees only
CAHPS	Patient experience with plans, providers, and facilities	Varies by survey version	Patients across all payer types
Leapfrog Safety Grades	Hospitals	22 safety measures	About 3,000 general hospitals
Garner's provider-level approach	Individual physicians	550+ clinical metrics	320M+ patients across all payers

‍HEDIS (Healthcare Effectiveness Data and Information Set)

HEDIS is NCQA's tool for comparing health plan performance, with more than 90 measures spanning prevention, acute care, and chronic condition management. More than 235 million people are enrolled in plans that report HEDIS results.

For employer decisions, its limits show quickly. HEDIS scores a plan, not the physicians inside it. The metric set leans heavily on process measures, and no HEDIS score will tell an employee which of the twelve in-network cardiologists near them delivers the best outcomes.

CMS Star Ratings

CMS Star Ratings assign one to five stars to Medicare Advantage and Part D plans based on clinical quality, patient experience, and administrative performance. In the 2026 ratings, plans with drug coverage are scored on up to 43 measures.

The system exists to help Medicare beneficiaries compare plans during enrollment. It says nothing about the specific doctors and specialists inside a plan's network, and its data comes entirely from a population that looks very different from a commercially insured workforce.

CAHPS (Consumer Assessment of Healthcare Providers and Systems)

CAHPS is AHRQ's family of standardized surveys that ask patients to report on their experiences with health plans, providers, and facilities. It captures how care felt. For example, did the doctor communicate clearly? Was the appointment easy to get? Was the staff respectful?

AHRQ is careful to note that patient satisfaction and patient experience are not the same thing, and neither one measures the clinical quality of the care delivered. A physician can have a warm bedside manner and a high complication rate. When employees pick doctors on ratings that are really experience surveys, they are choosing on personality, not outcomes.

Leapfrog and Hospital Safety Grades

The Leapfrog Group assigns letter grades to general hospitals based on 22 evidence-based patient safety measures. In its fall 2025 round, Leapfrog graded nearly 3,000 hospitals on how well they protect patients from errors, injuries, and infections, problems that affect about one in four hospital inpatients.

The grades tell you whether a facility has the systems in place to prevent avoidable harm, which is worth knowing before scheduling a surgery. However, they are not a measure of the individual surgeon, and physician performance varies enormously even within an A-graded hospital.

Why Standard Quality Metrics Fall Short for Employer Decision-Making

Though these frameworks can serve regulators, health plans, and hospital buyers well, they are often the only quality data a benefits team ever encounters. When quality comes up in a carrier RFP or a vendor pitch, these are the measures on the table, and none of them was designed to answer the question that matters most. Which in-network provider will deliver the best outcome for my employee?

The same three limitations appear in all four systems. They score institutions rather than individuals. Their data skews toward Medicare populations, while employer-sponsored insurance covers 154 million people under 65 whose demographics and care needs look different. And their metric sets are narrow enough that even a high-scoring plan or hospital can house physicians whose outcomes range from excellent to poor. Without provider-level data, employees fall back on reputation.

What Meaningful Provider-Level Quality Measurement Looks Like

Provider-level measurement that can support a real benefits decision has four requirements: all-payer claims data, enough patient volume per provider to reach statistically valid conclusions, risk adjustment for patient complexity, and a metric set broad enough to capture the full care journey.

Instead of describing how a plan performs on average, measurement built this way identifies which specific cardiologist, orthopedic surgeon, or primary care physician in a given market is a top performer and which is not. Two spine surgeons can look identical on every plan-level rating while one of them operates far more often than clinical evidence supports.

Garner built its methodology to meet this bar, applying 550+ proprietary clinical metrics to a dataset of more than 320 million patients to score doctors individually, specialty by specialty. For employers, that data becomes a practical benefit. Employees search for care in an app that surfaces Top Providers in their area, and Garner helps cover their out-of-pocket costs when they see one. Acting on the data becomes the financially advantageous choice, with no network changes and no plan redesign required.

How Benefits Teams Can Use Quality Measurement to Make Better Decisions

Quality data should inform three decisions.

Plan design. Which networks and tiers a plan builds around should reflect measured performance, not reputation. With healthcare costs still climbing, designs that concentrate volume with high-performing providers lower spend without cutting benefits. Better doctors resolve problems with fewer complications and less unnecessary care.

Provider steerage. Data only matters if employees act on it. Steering demand toward the best-performing doctors takes both guidance, meaning a simple way to find those doctors, and a reason to follow it, usually a financial incentive that rewards the higher-quality choice.

Vendor evaluation. Every care navigation and provider analytics vendor claims to have quality data. Ask what data the ratings are built on, and whether it measures individual providers or institutions. A vendor re-ranking public plan-level scores cannot tell an employee which doctor to see, no matter how polished the interface.

Healthcare Quality Measurement Is Only as Useful as the Data Behind It

Quality measurement frameworks are not interchangeable. The gap between institutional ratings and individual provider scoring is wide enough to produce materially different decisions, and materially different outcomes, for a health plan and the people on it.

Closing that gap means asking harder questions about the measurement itself. Does it draw on all-payer claims? Is it risk-adjusted? Does it score the physician, or the institution the physician works in?

Garner does all three. It scores individual providers on all-payer claims and risk-adjusted outcomes, delivered as an overlay on your existing plan, and it exists to answer the question standard frameworks cannot. Which specific doctor should my employee see?

Standard quality ratings tell you how the health plan performs. Garner tells you which doctor to see. Book a demo to see how provider-level data changes the benefits decisions you can make.

FAQs

What is a quality-based provider tier?

A quality-based provider tier is a subset of in-network providers grouped by measured clinical performance, with plan incentives that make the higher-performing tier cheaper for employees to use. Unlike cost-based tiers, which rank providers on negotiated prices, quality-based tiers rank them on outcomes, complication rates, and appropriateness of care. The full network stays intact, and the best-performing doctors become the most affordable choice.

Do higher-cost doctors deliver better care?

Not reliably. A doctor's price reflects negotiated rates, market leverage, and system affiliation more than clinical performance. Some of the best-performing doctors in a network charge average rates, while some of the most expensive deliver average or worse outcomes. Pairing price data with risk-adjusted quality data gives a far more accurate picture of value than cost alone.

What does "risk-adjusted" mean in healthcare quality measurement?

Risk adjustment means statistically accounting for how sick or complex a provider's patients are before comparing outcomes. Without it, a surgeon who takes on difficult cases would look worse than one who avoids them, even if the first surgeon is more skilled. Adjusting for patient mix keeps score differences tied to care quality rather than patient populations, which makes it a baseline requirement for fair physician-level comparison.

Can employees use quality data to choose their own doctor?

Yes, when the data reaches them in a usable form. Public report cards rarely help, since most score plans or hospitals rather than physicians. Benefits built on provider-level data let employees look up a specialty and location, see which nearby doctors rank as top performers, and act on that guidance, ideally with an incentive that lowers out-of-pocket costs.

What is the difference between a hospital rating and a doctor rating?

A hospital rating scores the facility, covering safety systems, infection control, staffing, and aggregate outcomes. A doctor rating scores an individual clinician. The two can diverge sharply, since a highly rated hospital can employ physicians whose outcomes span excellent to poor, and a strong physician can practice in a mediocre hospital. For choosing a specific provider, facility-level grades are a starting point at best.