{ "openapi": "3.0.1", "info": { "title": "CMS Medicare Data Scraper - data.cms.gov API", "description": "Pull any of the 395 Medicare and Medicaid datasets on data.cms.gov: 3.4M clinicians, provider enrollment and ownership, hospital and nursing home quality, Part B/Part D utilisation and drug spending. Server-side filters, no API key.", "version": "0.1", "x-build-id": "9urR5SsSeGmLkbWi5" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/parseforge~cms-medicare-scraper/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-parseforge-cms-medicare-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/parseforge~cms-medicare-scraper/runs": { "post": { "operationId": "runs-sync-parseforge-cms-medicare-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/parseforge~cms-medicare-scraper/run-sync": { "post": { "operationId": "run-sync-parseforge-cms-medicare-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } } }, "components": { "schemas": { "inputSchema": { "type": "object", "required": [ "dataset" ], "properties": { "dataset": { "title": "Dataset", "enum": [ "catalog", "doctors-clinicians", "clinician-affiliations", "clinician-utilization", "provider-enrollment", "order-and-referring", "opt-out-affidavits", "revoked-providers", "revalidation-due-date", "revalidation-reassignment", "revalidation-clinic-groups", "pending-physicians", "pending-non-physicians", "missing-digital-contact", "taxonomy-crosswalk", "hospital-enrollments", "hospital-owners", "hospital-change-of-ownership", "snf-enrollments", "snf-owners", "snf-change-of-ownership", "home-health-enrollments", "home-health-owners", "hospice-enrollments", "hospice-owners", "rural-health-clinic-enrollments", "rural-health-clinic-owners", "fqhc-enrollments", "fqhc-owners", "opioid-treatment-programs", "home-infusion-providers", "provider-of-services-labs", "provider-of-services-facilities", "hospitals", "hospital-patient-survey", "hospital-complications-deaths", "hospital-infections", "hospital-readmissions", "hospital-timely-effective-care", "hospital-outpatient-imaging", "hospital-value-based-purchasing", "hospital-spending-per-beneficiary", "hospital-maternal-health", "nursing-homes", "nursing-home-ownership", "nursing-home-deficiencies", "nursing-home-fire-safety", "nursing-home-penalties", "nursing-home-quality-measures", "nursing-home-claims-measures", "nursing-home-survey-summary", "home-health-agencies", "home-health-patient-survey", "hospices", "hospice-provider-data", "dialysis-facilities", "inpatient-rehab-facilities", "long-term-care-hospitals", "medical-equipment-suppliers", "ambulatory-surgical-centers", "inpatient-psychiatric-facilities", "physician-by-provider", "physician-by-provider-service", "physician-by-geography", "inpatient-by-provider", "inpatient-by-provider-service", "outpatient-by-provider-service", "dme-by-supplier", "dme-by-referring-provider", "physician-supplier-procedure-summary", "part-d-prescribers", "part-d-prescribers-by-drug", "part-d-by-geography-drug", "part-d-opioid-rates", "medicaid-opioid-rates", "part-d-spending-by-drug", "part-b-spending-by-drug", "medicaid-spending-by-drug", "part-b-discarded-drug-units", "medicare-monthly-enrollment", "geographic-variation", "medicare-advantage-geographic-variation", "market-saturation", "hospital-service-area", "telehealth-trends", "hospital-cost-report", "snf-cost-report", "home-health-cost-report", "aco-participants", "accountable-care-organizations", "aco-financial-quality-results", "quality-payment-program", "medicare-covid-hospitalization", "custom" ], "type": "string", "description": "Which data.cms.gov dataset to pull. 92 datasets are wired up with named filters; every other dataset on the portal is reachable through \"Custom\". \"Catalog\" returns the metadata of all 395 datasets instead of data rows. Each group bills its own per-row event, so you only pay for the one you asked for.", "default": "doctors-clinicians" }, "maxItems": { "title": "Max rows", "minimum": 1, "type": "integer", "description": "Stop after this many rows. The row limit is the spending ceiling: nothing beyond it is written or billed. Free-plan runs are capped at 25 rows.", "default": 100 }, "state": { "title": "State", "type": "string", "description": "Two-letter state code, e.g. \"CA\". Applied by CMS on its own server. Available on 70 of the 92 datasets — the ones whose rows carry a state." }, "city": { "title": "City", "type": "string", "description": "City or town, matched as a substring so \"SAN\" finds San Diego and San Jose. Available on 56 datasets." }, "zipCode": { "title": "ZIP code", "type": "string", "description": "Five-digit ZIP. About half the CMS files store ZIP+4, so the query goes out as a prefix match and rows whose ZIP does not actually start with yours are dropped before they are written or billed. Available on 56 datasets." }, "npi": { "title": "NPI", "type": "string", "description": "A single 10-digit National Provider Identifier, matched exactly. Available on 30 datasets." }, "ccn": { "title": "CCN / provider number", "type": "string", "description": "CMS Certification Number — the six-character facility id used across Care Compare, cost reports and the Provider of Services file. Available on 46 datasets." }, "providerName": { "title": "Provider or facility name", "type": "string", "description": "Substring of the organisation, facility or group name. Case-insensitive. Available on 60 datasets — the person-level files expose the surname as Clinician last name instead." }, "lastName": { "title": "Clinician last name", "type": "string", "description": "Substring of an individual clinician's surname, for the person-level files. Available on 22 datasets." }, "specialty": { "title": "Specialty / provider type", "type": "string", "description": "Substring of the specialty or provider-type column, e.g. \"Cardio\" or \"Nurse Practitioner\". Available on 17 datasets." }, "drugName": { "title": "Drug name", "type": "string", "description": "Brand or generic drug name, matched as a substring, on the Part B, Part D and Medicaid drug files. Available on 6 datasets." }, "hcpcsCode": { "title": "HCPCS code", "type": "string", "description": "A single HCPCS / CPT procedure code, matched exactly, on the service-level utilisation files. Available on 5 datasets." }, "geography": { "title": "Geography", "type": "string", "description": "For the geography-level files, whose rows are a state or county rather than a provider: the geography description, e.g. \"California\" or \"Los Angeles\". Available on 7 datasets." }, "year": { "title": "Year", "minimum": 1990, "maximum": 2100, "type": "integer", "description": "Data year. On the 7 datasets that carry a year column this filters the rows; on the rest of the main CMS catalog it switches to that year's published version of the file — the Part D and Physician files go back to 2013. Datasets in the Provider Data Catalog have no archive and reject this." }, "search": { "title": "Full-text search", "type": "string", "description": "Free-text search across every column of the dataset, run by CMS. Main CMS catalog only — the Provider Data Catalog has no keyword endpoint and the run stops with a message rather than ignoring this." }, "filters": { "title": "Custom filters", "type": "array", "description": "One expression per line, applied on top of the named filters above and ANDed together. Write them as \"Column operator value\" using = != > >= < <= contains in \"not in\" or between, e.g. \"Tot_Clms > 1000\", \"Prscrbr_State_Abrvtn in CA,NY,TX\", \"Brnd_Name contains OZEMPIC\". A column that does not exist stops the run and lists the ones that do.", "items": { "type": "string" } }, "columns": { "title": "Columns", "type": "array", "description": "Only return these columns. Leave empty for every column. Narrowing the projection makes wide files (up to 473 columns) far faster to pull.", "items": { "type": "string" } }, "sortBy": { "title": "Sort by", "type": "string", "description": "Column to order by, applied by CMS. Combine with Max rows to take the top N — e.g. sort Part D prescribers by Tot_Drug_Cst descending." }, "sortOrder": { "title": "Sort direction", "enum": [ "asc", "desc" ], "type": "string", "description": "Direction for Sort by.", "default": "asc" }, "catalogSource": { "title": "Catalogs to list", "enum": [ "both", "main", "pdc" ], "type": "string", "description": "Which of the two catalogs to list in Catalog mode. CMS Data is the 159-dataset open-data portal; the Provider Data Catalog is the 236-dataset Care Compare feed. Ignored unless Dataset is \"Catalog\".", "default": "both" }, "catalogSearch": { "title": "Catalog search", "type": "string", "description": "Keep only datasets whose title, description, keywords or theme contain this text." }, "catalogTheme": { "title": "Catalog theme", "type": "string", "description": "Keep only datasets in a theme or keyword, e.g. \"Nursing homes\", \"Hospitals\", \"Medicare\", \"Dialysis facilities\"." }, "catalogModifiedAfter": { "title": "Modified after", "type": "string", "description": "Keep only datasets CMS updated on or after this date (YYYY-MM-DD). Use it to poll the portal for refreshed files." }, "onlyQueryable": { "title": "Only datasets with a live API", "type": "boolean", "description": "Drop the 28 datasets CMS publishes as ZIP or XLSX only, which this Actor can list but not query row by row.", "default": false }, "includeFields": { "title": "Include the data dictionary", "type": "boolean", "description": "Also emit one row per column of the dataset, with its machine name, human label, SQL type and position. Billed as Data dictionary field. Off by default.", "default": false }, "includeResources": { "title": "Include the downloadable files", "type": "boolean", "description": "Also emit one row per file CMS attaches to the dataset — the current CSV, every dated archive, and the data-dictionary PDF — with its URL and size. Billed as Dataset file. Off by default.", "default": false }, "customDatasetId": { "title": "Custom dataset ID", "type": "string", "description": "Only used when Dataset is \"Custom\". Either a main-catalog UUID (the id in https://data.cms.gov/data-api/v1/dataset//data) or a Provider Data Catalog id such as xubh-q36u. A full dataset URL works too." }, "customCatalog": { "title": "Which catalog the ID belongs to", "enum": [ "auto", "main", "pdc" ], "type": "string", "description": "Leave on Auto: a 36-character UUID is looked up in the CMS Data catalog, anything else in the Provider Data Catalog.", "default": "auto" } } }, "runsResponseSchema": { "type": "object", "properties": { "data": { "type": "object", "properties": { "id": { "type": "string" }, "actId": { "type": "string" }, "userId": { "type": "string" }, "startedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "finishedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "status": { "type": "string", "example": "READY" }, "meta": { "type": "object", "properties": { "origin": { "type": "string", "example": "API" }, "userAgent": { "type": "string" } } }, "stats": { "type": "object", "properties": { "inputBodyLen": { "type": "integer", "example": 2000 }, "rebootCount": { "type": "integer", "example": 0 }, "restartCount": { "type": "integer", "example": 0 }, "resurrectCount": { "type": "integer", "example": 0 }, "computeUnits": { "type": "integer", "example": 0 } } }, "options": { "type": "object", "properties": { "build": { "type": "string", "example": "latest" }, "timeoutSecs": { "type": "integer", "example": 300 }, "memoryMbytes": { "type": "integer", "example": 1024 }, "diskMbytes": { "type": "integer", "example": 2048 } } }, "buildId": { "type": "string" }, "defaultKeyValueStoreId": { "type": "string" }, "defaultDatasetId": { "type": "string" }, "defaultRequestQueueId": { "type": "string" }, "buildNumber": { "type": "string", "example": "1.0.0" }, "containerUrl": { "type": "string" }, "usage": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "integer", "example": 1 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } }, "usageTotalUsd": { "type": "number", "example": 0.00005 }, "usageUsd": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "number", "example": 0.00005 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } } } } } } } } }