Data.gov — Managed Care Full-Text Search

A two-word query with the whole government behind it. This record is the standing 'managed care' sweep over the federated US government catalog — an index holding 552,271 datasets at the August 2026 review — and it resolves to hundreds of matched records drawn almost entirely from two publishers: the Centers for Medicare & Medicaid Services federally and California's Department of Health Care Services at state level, with NIH-indexed literature citations riding the same ranked list. See federated catalog search.

Every match arrives dressed in DCAT-US metadata: publisher name and government level, theme keywords, issued and modified dates, per-file media types, popularity signals, a harvest timestamp and a direct-download flag — see DCAT metadata record. That wrapper is what makes the sweep usable as a discovery layer rather than a bookmark folder.

Within Datadory's catalog this is one of the 10 primary datasets (13 counting three borrowed from adjacent industries) in the Managed Health Care slice, out of 1,744 cataloged overall, and it scores 9/10 against a catalog-wide average of 7.81 — see the managed health care data hub.

Get a sample of this dataset cut to your agencies, states and subject lines before anything else.

Five verified rows from the research pass, shown in the delivered frame:

# row 1 - the state-side money table
title        : Medi-Cal Managed Care Capitation Rates - Geographic Managed Care (GMC)
publisher    : California Department of Health Care Services
org type     : State Government            # has_download: true

# row 2 - the beneficiary-facing federal table
title        : Managed Care Information for Medicaid and CHIP Beneficiaries by Year
publisher    : Centers for Medicare & Medicaid Services
org type     : Federal Government          # has_download: true

# row 3 - the program umbrella record
title        : Medicaid Managed Care
publisher    : Centers for Medicare & Medicaid Services
org type     : Federal Government          # has_download: true

# row 4 - the recurring national series
title        : Managed Care Enrollment Summary
publisher    : Centers for Medicare & Medicaid Services
org type     : Federal Government          # has_download: true

# row 5 - the penetration series
title        : Share of Medicaid Enrollees in Managed Care
publisher    : Centers for Medicare & Medicaid Services
org type     : Federal Government          # has_download: true

Read the list as a map of the aperture. Four of five rows are federal program tables — enrollment counts, beneficiary information, penetration shares — and the fifth is the one the narrower Medicaid Managed Care query never surfaces: Medi-Cal capitation rates broken out by delivery model, the price side of the ledger. All five carry the download flag, so every row lands as a usable table rather than a description of one.

Get a sample of this dataset and real rows arrive in exactly this frame.

Twelve documented fields, each defined down to type during the research pass — the table below is the full dictionary. Three do the analytical heavy lifting. publisher turns a keyword search into a publisher census: it is the column that separates CMS-authored national series from California DHCS state files without reading a single abstract. organization.organization_type makes the federal-versus-state-versus-local split a filterable value rather than an inference — the practical fix for a ranking that deliberately mixes governments. And dcat.distribution[].mediaType carries each distribution file's MIME type, so a CSV-only pipeline and an XLSX-bound analyst triage the same record differently and both walk away correct.

The quieter columns earn their keep too. modified holds the publisher-reported change date, while last_harvested_date stamps when the catalog last re-checked the record — the pair together tells fresh data from stale metadata. _score exposes the relevance ranking itself, which is how you audit why a literature citation outranked an enrollment table. slug gives a stable identifier in a catalog where titles recur year after year under near-identical names, and has_download answers the only binary question most pipelines ask. Fields beyond the twelve fold under additional fields on request rather than padding every delivery.

• Geography: United States across every government level — federal publishers led by CMS, HHS and NIH, states led by California DHCS, plus local government contributors. Individual underlying datasets run national or single-state; the sweep covers both in one ranked list.
• Temporal: enrollment and program tables from 2016 forward through 2024–2026 submissions, with each record carrying its own publisher-reported modified date and a last_harvested_date stamp observed current into July 2026. Recurring annual series sit beside one-off program-feature matrices in the same results.
• Granularity: catalog-level records pointing at datasets of any grain underneath — national summary tables, state-program-plan-year capitation rows, beneficiary-month counts. Plan analysis against the specific datasets a search surfaces, not against the index layer itself.

Against the wider Datadory catalog — average quality score 7.81 across all 1,744 datasets — this record scores 9/10, with all twelve fields documented to type. The score reflects what a discovery layer does well: complete dictionaries, provenance on every row and a result set that reaches past the federal core into state money tables.

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

Name the agencies, the states and the subject terms when you request the sample — the full managed care cross-section, or a narrow cut such as CMS national enrollment tables for one analysis frame plus the Medi-Cal capitation files beside them. The sample ships first either way, and the ongoing feed lands shaped identically, so joins prototyped on the sample survive into production unchanged.

Records arrive normalized to the field dictionary above: publisher names resolved, government levels typed, media-type listings expanded per distribution file, and harvest stamps preserved as structured columns rather than flattened strings. Cadence changes are a settings conversation, not a re-integration project — see developers builders use cases.

A catalog-wide sweep earns its keep in five specific jobs:

• Market studies with both sides of the ledger — CMS enrollment tables supply the people, Medi-Cal capitation-rate files by plan model supply the price, and one query returns both; see market researchers use cases.
• Citation-grade sourcing — every statistic traces back to its harvesting agency through DCAT metadata, which is where fact-checks stop arguing; see journalists academics use cases.
• Pipeline scoping — enumerate which agencies publish machine-readable managed care tables, in which formats, before a line of ingestion code gets written; see data scientists use cases.
• Payer-side rate environment tracking — MCO enrollment and spending tables feed utilization and rate assumptions in payer analysis; see investors quants use cases.
• Contract mapping by inference — plan-model breakdowns indicate which carriers hold managed care contracts state by state; see competitive intel product teams use cases.

The common thread: this record tells you what managed care data exists and who owns it, which is the step every one of those jobs runs first.

Market researchers and consultants get the widest aperture: CMS national series and California's capitation-rate matrix by delivery model — Geographic Managed Care, Two-Plan, Regional/Rural Expansion, COHS — reachable through one standing query instead of portal-by-portal hunting. Journalists, academics and students get named agencies and typed government levels behind every figure, citable provenance for stories and theses built on Medicaid managed care statistics.

Developers and builders get format triage up front — media types on every distribution file, so pipeline decisions happen before download rather than after. Investors and quant researchers get the utilization and rate-environment inputs payer models run on. Data scientists and ML engineers get a discovery half that hands off cleanly: find the tables here, then join their contents on geography and time. Persona playbooks sit on the data scientists and journalists academics pages.

Four things worth knowing upfront.

First, this is an index, not a warehouse. Records describe datasets living on their own agencies' terms, so structure and completeness vary by publisher; the sweep guarantees you can find a managed care dataset, not that any two will agree on shape.

Second, the ranking mixes genres by design. Two generic words pull CMS program tables, state money files and NIH-indexed literature citations into one list — the organization.organization_type field is the filter that separates them when precision matters.

Third, titles recur. Several CMS series republish year after year under near-identical names, so panel builders should dedupe on slug and modified rather than on title strings.

Fourth, totals are measured, not published: there is no standing count for the query, so treat headline figures as photographs taken during the August 2026 review. Name your scope when requesting the sample and the extract arrives cut to exactly that frame — see managed care organization for the entity the whole slice orbits.

Provenance note — compiled and maintained by Data.gov as the federated index of American government data, with each record owned by its publishing agency. One operator runs the index; CMS, California DHCS, NIH and their peers own what the records describe, which is why the DCAT-US schema stays constant while the content beneath it varies. Source background sits on the Data.gov source profile.

Methodology note — the five sample rows above, the twelve-field dictionary and the publisher mix cited throughout were verified during the August 2026 research pass. Publisher-level figures read as catalog registration scale, not as row counts inside any dataset.

Completeness note — Datadory scores this record 9/10 with all twelve fields documented to type. Fields outside the verified core ship with your sample on request rather than being approximated here.

Where to go next — the Data.gov Medicaid Managed Care Search narrows this sweep to its CMS core plus New York's outcome series, and the head-to-head lives on the full-text versus Medicaid search comparison. The Data.gov Health Insurance Search widens to the coverage-and-marketplace neighborhood, while the KFF Medicaid & CHIP Indicators add the analytical state-year layer beside it. Start from the best managed health care datasets ranking, or browse every record in the slice on the managed health care data hub.