Data Methodology
How clinic records are built, what sources we use, what we deliberately leave out, and how we keep the data current.
Where clinic records come from
Every clinic profile starts with the CDC/NASS (National ART Surveillance System) dataset — the US government’s annual census of fertility clinics that perform assisted reproductive technology cycles. CDC/NASS gives us a verified starting point: clinic identity (name, address, CDC clinic ID), cycle counts, and success-rate data across defined age cohorts and treatment types.
CDC/NASS is the canonical seed. We do not start from web scrapes, patient review sites, or unverified lists. If a clinic is not in CDC/NASS, it can still appear in the directory — but only after manual review confirms it is an active fertility clinic in the United States.
How we enrich records
After the CDC/NASS seed is established, clinic-owned pages are the primary enrichment source. We extract:
- Services and treatment programs offered
- Pricing signals (where publicly stated by the clinic)
- Insurance participation (where publicly stated)
- Specialist names and credentials
- Office hours
- Clinic descriptions
All extracted values are paraphrased, not copied verbatim. Exact pricing, specific coverage claims, and accreditation status are never guessed — they are either sourced with evidence or left blank. For clinic chains or networks, we prefer location-specific pages over corporate pages for location-specific fields.
Aggregate review signals (overall rating and review count) are drawn from Google Places — aggregate numbers only. Individual review text and photos are not stored.
Source-of-truth hierarchy
| Field category | Primary source | Secondary source |
|---|---|---|
| Clinic identity & address | CDC/NASS | Clinic website |
| CDC clinic ID | CDC/NASS | — |
| Success-rate data | CDC/NASS | — |
| Services, programs, pricing, hours | Clinic website | — |
| Coordinates | Google Places | — |
| Aggregate rating | Google Places | — |
| State insurance mandate context | State government sources | ASRM, RESOLVE, KFF |
What we deliberately do NOT store
The following are excluded from durable storage by policy:
- Google review text (snippets, quotes, or full reviews)
- Clinic or patient photos
- SART directory content (any bulk reproduction requires a license)
- Raw Google Places API payloads beyond the fields listed above
- Website HTML caches committed to the repository
We exclude these categories either because we do not have a license to reproduce them, or because transient data introduces reliability and attribution risks that would degrade profile quality.
QA cycle
Before a clinic profile is published, it passes an internal QA review. QA checks:
- Required fields are present and sourced (name, address, state)
- Extracted values are consistent with the source page
- No guessed pricing or coverage claims
- CDC/NASS success-rate denominator definitions are preserved accurately
- The clinic’s operational status is current (not closed or relocated)
Profiles that have passed QA are marked internally. Profiles that have not yet completed QA may still appear in the directory for usability, but are displayed with a caveat noting the verification status.
Refresh cadence
CDC/NASS data is updated annually after the CDC publishes a new complete reporting year dataset. We always use the most recent complete dataset and do not backfill or blend reporting years for individual clinics.
Clinic website extractions are reviewed periodically, with priority given to corrections submitted by clinic staff or patients. If you have updated information, please submit a correction.
Clinic names, logos, and trademarks
Clinic names, logos, and trademarks displayed on this directory are the property of their respective owners and are used solely for identification purposes. Their display does not imply any endorsement of or affiliation with Explore Fertility. Clinic representatives who would like a logo removed or replaced may submit a request.
Corrections and source disputes
Corrections submitted by users, patients, or clinic staff are treated as intake artifacts until manually verified against a source. Accepted corrections are merged with a reviewer ID and date in the internal merge log. We do not accept unverified corrections directly into published data.