SAM.gov excluded providers in NDH
Active SAM.gov exclusion records (HHS LEIE + OPM FEHBP debarment + DOJ + others, aggregated) whose NPI also appears in the federal NDH bulk export. Closes the third of four federal database checks named in 42 CFR § 455.436.
Headline
4,517 distinct NPIs on SAM.gov's active exclusion list appear as NDH practitioners — 3,765 still flagged `active=true` in the 2026-05-08 bulk export. SAM aggregates HHS LEIE + OPM FEHBP debarment + other federal-agency actions. The OPM slice is net-new federal-screening signal beyond what LEIE alone surfaces.
4.5K / 7.1K = 63.95%
unit: count
What this means
Regulators
42 CFR § 455.436 names SAM as one of four federal databases for monthly Medicaid screening. Matches between active SAM exclusions and NDH directly measure federal-level alignment; persistent matches indicate cadence drift between excluding-agency action and NDH publication.
Payer data teams
The OPM-debarred slice is the operationally interesting one — those providers are barred from FEHBP but may still be in commercial network listings if your data feed treats SAM as out-of-scope. Treat OPM exclusion as an independent signal from LEIE, not a duplicate.
Provider data teams
If your NPI is matched here, sam.gov/search/?index=ex is the authoritative lookup. SAM exclusions can come from agencies other than HHS — read the excluding_agency and exclusion_type fields before assuming an OIG-LEIE issue.
Researchers
NPI population in SAM is the structural ceiling — only ~4% of SAM rows carry a real NPI, because SAM is a multi-domain feed (sanctions, contractor debarment, foreign-asset blocks). Healthcare-relevant matches are concentrated in the HHS and OPM agency slices.
Null hypothesis
Zero NPIs on SAM.gov's active exclusion list also appear in the federal NDH bulk export. Federal directory and federal exclusion list are in agreement.
Denominator
Active SAM rows (record_status = "Active") with a populated, real-format NPI. Approximately 7,063 of the 167,262 SAM rows in the V2_26120 extract; the remaining 96% are non-healthcare exclusions (OFAC sanctions, EPA contractor debarments, etc.) and are out of scope for AINPI's NPI-keyed match.
Data source
SAM.gov Public Extract V2 (sam.gov/data-services/Exclusions/Public V2) loaded via `analysis/ingest_sam_exclusions.py`, joined to NDH `practitioner._npi`. The HHS slice overlaps substantially with H24 LEIE; the OPM slice (FEHBP debarment under 5 USC 8902a) is net-new federal-screening signal not visible from LEIE alone.
Notes
SAM source: sam.gov/data-services/Exclusions/Public V2 (V2_26120 extract). Total SAM rows: 167,262; with real-format NPI: 7,063 (4.2%); active + with NPI: 7,063. The 96% of SAM rows without a populated NPI are non-healthcare exclusions (OFAC sanctions, EPA contractor debarment, etc.) and are out of scope for this NPI-keyed match. The HHS-slice match overlaps substantially with the H24 LEIE finding by design — the same exclusion appears in both feeds. Each match is a data-quality flag, not a fraud determination — investigation, hearing rights, and reinstatement claims belong to the excluding agency. sam.gov/search/?index=ex is the authoritative source for any individual verification.