Supported document type
Detect tampered and deepfake Health Insurance Card
A health insurance card is checked in seconds at a reception desk, usually from a photo on a phone, and that is the gap forgeries exploit. The member and group IDs are edited so treatment is billed against a policy the patient is not enrolled on, or the coverage dates are moved forward so a lapsed plan reads as current. A card also has physical properties, printed stock and a laminate and sometimes a photo panel, that a screen capture cannot reproduce.
API slug: health_insurance_card
Last updated:
What a genuine health insurance card contains
A health insurance card is proof of enrolment at the point of care. It shows the subscriber's name, the member identification number, the group number, the plan name and tier, the effective date and any termination date, and a dependent suffix or list where the policy covers a family. The payer's name and logo sit alongside the network name, the copay amounts for an office visit, a specialist, and emergency care, and the pharmacy routing codes. Member and provider service numbers are printed on the reverse, with a photo panel on some schemes.
Almost none of those values are free-form. The member ID follows the payer's own structure, often with a check character, and the pharmacy routing code identifies a specific benefit manager that the payer actually contracts with. Copay figures belong to the named plan tier, so raising the tier without raising the figures produces a card describing two different plans. A dependent suffix maps to one person on the policy, and the effective date has to precede the date care is given for the card to mean anything at all.
How health insurance card forgeries are made
Member and group ID edits
The identification numbers are changed so care is billed against someone else's active policy, which is the highest-frequency fraud on this document. An edited ID commonly breaks the payer's own format or its check character, and the altered digits sit at a different weight from the rest of the field.
Coverage date extension
The effective or termination date on a lapsed card is moved forward so coverage appears current. Only a few characters change, and because most desks check that a date is present rather than what it implies, the edit is cheap to attempt and frequently succeeds until the claim is rejected.
Plan tier upgrading
The plan name is raised to a tier with broader benefits, usually without adjusting the copay figures beneath it. The result is a card whose tier and its own benefit table describe different products, which is a contradiction the card cannot resolve on its own.
Screen capture presented as a physical card
An image of a card is edited and shown on a phone rather than produced in the hand, which removes every physical cue at once. Screen capture leaves its own evidence: subpixel structure, display moire, and an absence of the card stock and laminate reflections a photographed card carries.
Photo panel replacement
On schemes that carry a photograph, the image is swapped so a card belonging to one member is presented by another. Photo zone analysis compares the panel against the card around it for resolution, noise, compression, and edge continuity where the replacement was seated.
Cards built from a payer template
A card is assembled from scratch using a payer's branding and layout, so no single element has been altered. What gives it away is uniformity: real cards are printed once and then photographed, and a composited card lacks the consistent lighting, focus falloff, and surface behaviour of an object in front of a lens.
What TamperCheck checks on a health insurance card
TamperCheck runs 200+ forensic checks across three layers and returns a risk score from 0 to 100 with plain-English findings tied to specific regions of the document. These three carry the most weight on this document class.
Portrait substitution and face-swap analysis
The document is isolated from whatever it was photographed on, then the portrait region is assessed against the card surface around it. A pasted or generated face rarely matches the substrate it was placed onto.
Letterhead, logo, and issuer plausibility
Issuer branding is checked for resolution and compression consistency against the body text it sits on. A logo lifted from a website carries the signature of its source, not of the document.
Date and validity consistency
Issue, expiry, and period dates are checked against one another and against the file's own creation metadata, which catches backdating and validity extension.
And many more checks
The three above are the layers that carry the most weight on a health insurance card. Every upload runs the full suite of 200+ checks regardless of document class, spanning file structure and metadata, pixel-level forensics, font and text rendering, optical and print characteristics, provenance signals, AI-generation signatures, and many more checks.
Who verifies a health insurance card, and why
Providers and benefits administrators validating coverage at enrolment. In each case the document is being used to unlock money, access, or a legal status, which is exactly what makes it worth forging.
How to verify a health insurance card in 4 steps
Check the fields against each other
Read member ID, coverage dates, and plan tier together rather than one at a time. Forgers typically change one value and leave the rest of the document describing the original.
Inspect the file metadata
Open the document properties and look at the producer, creation date, and modification date. A document produced by a consumer PDF editor, or created long after the date printed on its face, is worth a closer look.
Ask for a second document
Request a corroborating document from the same issuer or an adjacent period. Forgery effort concentrates on one file, so inconsistencies surface as soon as there are two to compare.
Run a forensic check
Manual review catches obvious edits but not field-level pixel manipulation or synthetic generation. TamperCheck runs 200+ forensic checks on a health insurance card and returns a risk score from 0 to 100 within seconds, at $0.50 per document.
Frequently asked questions
How do providers verify a health insurance card?
Providers run an eligibility check with the payer using the member and group numbers, which confirms enrolment and coverage dates from the payer's own records rather than from the card. The card itself is checked for format, for agreement between plan tier and copay figures, and where it arrives as an image, for whether it was photographed as a card or captured from a screen.
Can a health insurance card be faked?
Yes, and it is easier than most insurance documents because the card is small, the review is brief, and it is increasingly presented as a phone image rather than as physical stock. The common versions are an edited member ID, so care is billed to another person's policy, and a moved coverage date on a plan that has already terminated.
What does an edited member ID look like forensically?
It looks like a small patch of the card with its own history: the changed characters carry different compression, sharpness, and often a marginally different baseline from the digits beside them, even when the substituted font is a close match. The finding comes back tied to that region, naming the member ID as the field that was altered.
Can you tell if an insurance card photo came from a screen?
Yes. A photograph of a screen carries subpixel structure and interference patterns from the display's own pixel grid, and it lacks the specular highlights and slight surface curvature a laminated card shows under a light. These are measurable properties of the image rather than judgements about how the card looks.
TamperCheck analyses health insurance card uploads with a hybrid forensic and AI pipeline tuned for this document class. Upload endpoints accept PDF and common image formats; class is inferred automatically. See the API & MCP documentation for authentication, async jobs, and webhooks.