NCQA Health Plan Report Cards

The quality-verdict layer of American managed care. NCQA Health Plan Report Cards is the National Committee for Quality Assurance's public directory of rated health care organizations, and every entry in it is exactly one thing: a report card for a single rated plan, carrying that organization's NCQA rating or accreditation standing, its product-line category and the accreditation seal graphic shown beside the verdict.

Three properties make the directory usable as a dataset rather than a lookup page. Every card renders the same four attributes, so the whole population normalizes onto one row shape. The deployed build pins the displayed vintage at measurement year 2024, so every card in a pull shares one comparable period. And the directory's own tagging splits the population across commercial, Medicaid and Medicare Advantage lines, which turns plan type into a filterable dimension instead of an inference.

Within Datadory's managed health care shelf this is the outlier by design: the only dedicated per-plan ratings record among the slice's ten primary datasets (13 counting neighbors borrowed in), scored 6/10 against a catalog-wide average of 7.81. Where CMS tables count enrollees and KFF counts indicators, this record grades insurers one by one. See the managed health care data hub.

Get a sample of this dataset cut to the lines and plan types you follow.

One row per rated organization. The card's documented shape, as captured during the August 2026 research pass:

# one NCQA report card - documented shape, August 2026 research pass

card_grain   : 1 card per rated health plan / organization
org_name     : the plan's name as published          # populated with your sample
rating       : NCQA rating or accreditation status   # displayed for measurement year 2024
plan_type    : commercial | Medicaid | Medicare Advantage
seal         : accreditation seal graphic rendered beside the verdict

# observed from the deployed application during research
app_title    : Health Plan Ratings & Provider Directory - NCQA Report Cards
displayed_vintage : 2024                             # pinned by the current build
lookup_scope : United States

Read the block as an anatomy chart, not an illustration. The four card fields above are the entire delivered schema - identity, verdict, category and seal - and the fact that all four repeat identically on every card is what makes a directory of thousands of individual lookups enumerable. Per-card values were not capturable during research because the directory renders its results client-side; rather than pad this page with approximated verdicts, the fields ship typed and defined, and real card rows land with your sample in exactly this shape.

Four documented fields define the card, split into two jobs. Two identify the subject: Organization name, the health plan or practice as published, and Plan type, the product-line category that separates commercial books from Medicaid and Medicare Advantage ones. Two carry the verdict itself: Rating / accreditation status, the NCQA tier displayed for measurement year 2024, and Seal image, the accreditation graphic referenced on every graded card.

A note on confidence, because we would rather tell you than imply otherwise: the definitions were captured from the directory's own presentation during the August 2026 research pass, while per-card values were not, so the examples in the dictionary below populate with your named cut rather than being guessed at here. Anything beyond the core four folds under additional fields on request rather than padding every delivery.

Three dimensions summarize the footprint.

• Geography: United States throughout, at national altitude - the directory grades plans operating across the country, with the commercial, Medicaid and Medicare Advantage lines tagged per organization so a state-focused cut is a filter on the delivered rows rather than a separate hunt.
• Temporal: the current measurement year, 2024, displayed uniformly by the deployed build, which makes any pull internally comparable; earlier vintages exist as separately published editions running back to the 2013-14 Rankings era, so multi-year panels are assembled edition-over-edition rather than mixed inside one.
• Granularity: one report card per rated health plan or organization - the finest entity the program grades. That is the opposite end of the range from the slice's enrollment totals and state-year indicator panels, and it is precisely why this record cannot be substituted by either.

On quality: 6/10 on Datadory's rubric against the 7.81 catalogue average, the deduction reflecting thin verified depth per card rather than any weakness in the verdicts themselves.

API, files, or your warehouse. Daily, weekly, or hourly.

Name the lines, the plan types and the rating tiers you care about when you request the sample - the whole graded population, or a narrow cut such as Medicaid-plan cards only, or commercial cards below a named tier. The sample ships first either way, and the ongoing feed lands shaped identically, so anything prototyped on the sample survives into production unchanged.

Cards arrive normalized to the dictionary above: organization names resolved to a single canonical form, plan types typed as a filterable dimension, and seal references carried as structured columns rather than stripped graphics. Cadence is a settings conversation, not a re-integration project.

Per-plan verdicts earn their keep in six specific jobs:

• Competitor quality benchmarking - track rival carriers' ratings and accreditation standing card over card, release over release, for positioning evidence no enrollment table supplies; see competitive intel product teams use cases.
• Citation-grade sourcing - NCQA's own grading of a named plan is the quotable quality benchmark in health stories and peer-reviewed work; see journalists academics use cases.
• Insurer moat analysis - sustained accreditation standing pairs with enrollment and margin data as a durable-advantage input; see investors quants use cases.
• Network quality screens - filter candidate plans by tier before a benefits design or vendor selection shortlist; see market researchers use cases.
• Feature engineering - rating tiers join onto enrollment and utilization panels as categorical quality features; see data scientists use cases.
• Ratings-aware products - power plan-comparison views and quality badges on top of a maintained feed; see developers builders use cases.

Competitive intelligence and product teams lead at 3/3 relevance: rivals' ratings and seals, tracked card over card, are positioning evidence that arrives before press releases do; see competitive intel product teams. Journalists, academics and students match them at 3/3 - a national accreditor's own verdict on a named plan holds up in print and peer review alike; see journalists and academics. Market researchers and consultants (2/3) drop tier filters straight into plan-benchmarking decks; see market researchers. Investors and quant researchers (2/3) read sustained accreditation as a moat signal beside enrollment and financials; see investors and quants. Developers and builders (2/3) stand up comparison and badge features on a maintained feed; see developers and builders. Data scientists and ML engineers (1/3) get a compact categorical feature, not volume - pair it with the operational ledgers before modeling.

This record owns the per-plan verdict position, and nothing else in the slice contests it. The Data.Medicaid.gov CMS Medicaid Data Portal carries the operational ledger - managed care enrollment by state, program, plan and region from 2016 forward - which tells you how many lives a plan covers, never how well it cares for them. The KFF State Health Policy Medicaid & CHIP Indicators hold roughly 209 pre-computed state indicators including the Managed Care Market Tracker: ratios and program rules, again not per-insurer grades. The Data.gov managed care sweeps discover tables across government; they do not judge insurers.

The trade is symmetrical. If you need NCQA's judgment of a specific plan, this is the only source of it here; if you need volume, longitudinal depth or money tables, take this beside the operational records rather than instead of them. The full board sits on the best managed health care datasets ranking.

Four things worth knowing before you scope a cut.

First, the graded population is not published as a count: neither the number of organizations rated in measurement year 2024 nor a per-line breakdown appears on reachable surfaces, so size the universe by running your scope rather than trusting a headline figure.

Second, definitions are captured, values are confirmed: the four field definitions come from the directory's own presentation, and per-card examples populate with your sample rather than being approximated on this page.

Third, vintages do not mix: a pull reflects the displayed measurement year, and historical comparisons mean assembling separately published editions - tell us the years and we assemble the panel.

Fourth, grain is the point: one card per plan is the finest resolution in the slice, which is exactly why it joins cleanly upward to managed care organization enrollment ledgers and Medicaid managed care program tables rather than competing with them.