A 30-Year-Old Provision That Congress Keeps Blocking
In 1996, Congress wrote a provision into HIPAA that called for a universal patient identifier — a unique number assigned to every patient in the United States. The idea was straightforward: reduce medical errors, eliminate duplicate records, and make healthcare data exchange seamless. Then, almost immediately, Congress turned around and banned HHS from spending a single dollar to implement it.
That ban has been renewed every year in appropriations bills for nearly three decades. I've watched this cycle play out from inside healthcare organizations where patient matching failures cause real harm — and real HIPAA liability. The universal patient identifier remains one of the most consequential unfinished chapters in American healthcare policy.
If you work with PHI in any capacity — as a covered entity, a business associate, or an IT vendor — this debate affects you directly. Here's what you need to know about where things stand, what's at stake, and why patient identification errors keep showing up in breach reports.
What Is a Universal Patient Identifier?
A universal patient identifier would be a single, unique number assigned to every individual in the U.S. healthcare system. Think of it like a Social Security Number, but designed specifically for healthcare. It would follow a patient across every provider, payer, and health information exchange they interact with.
Section 1173(b) of the original HIPAA statute — 42 U.S.C. § 1320d-2 — explicitly directed HHS to adopt standards for a unique health identifier for individuals. The provision sits right alongside the identifiers HHS did implement for employers (the EIN) and providers (the NPI). The patient identifier is the missing piece.
Why Congress Has Blocked It Every Single Year
Privacy concerns killed the universal patient identifier before it ever had a chance. In the late 1990s, advocates warned that a single number linking all of a person's health records could become a surveillance tool. The fear was that a centralized identifier would make it easier for bad actors — or the government itself — to track individuals across the healthcare system.
In 1998, Congress inserted language into the Labor-HHS appropriations bill prohibiting HHS from using any funding to "promulgate or adopt any final standard" for a unique health identifier. That rider has appeared in every subsequent appropriations bill.
I've spoken at conferences where audience members physically recoil at the idea. The emotional reaction is real. But the technical reality in 2026 is that we already have far more invasive data collection happening through commercial health apps, insurance claims databases, and electronic health records — none of which are governed by a unified matching standard.
The 2023 Shift That Almost Changed Everything
In 2023, the House Appropriations Committee actually voted to remove the funding ban for the first time. The amendment, led by a bipartisan group, didn't survive the full appropriations process. But it signaled a real shift in how lawmakers view the issue. The American Medical Association, AHIMA, and CHIME have all publicly supported developing a universal patient identifier or an equivalent strategy.
As of 2026, the ban remains in place. But the conversation has changed from "should we do this" to "how do we do this safely."
The Patient Safety Crisis Hiding in Plain Sight
Here's what happens without a universal patient identifier: hospitals guess. They use probabilistic matching algorithms that compare names, dates of birth, addresses, and Social Security Numbers to link records. These algorithms are good. They are not good enough.
Studies from the Pew Charitable Trusts found that matching rates between healthcare organizations range from 50% to 80%. That means up to half of patient records may fail to match correctly when data moves between systems. The consequences are severe:
- Duplicate records — A patient has two charts at the same hospital, and a critical allergy is documented in the one the ER physician never sees.
- Overlays — Two different patients' records get merged into one. One patient receives treatment decisions based on someone else's lab results.
- Billing fraud exposure — Mismatched records create vulnerabilities that identity thieves exploit, leading to PHI breaches your organization must report under the HIPAA Breach Notification Rule.
I've consulted with a mid-size health system that discovered over 300,000 duplicate records in a single EHR migration. Three hundred thousand. Each one represented a potential patient safety event and a potential HIPAA exposure.
The $1.3 Million Problem No Algorithm Can Solve
Patient identification failures don't just create clinical risk — they create compliance risk. When records are mismatched, PHI ends up in the wrong chart, gets sent to the wrong patient portal, or gets disclosed to the wrong family member. Each of those events can trigger an OCR investigation.
Consider the 2019 OCR settlement with the University of Rochester Medical Center for $3 million. While that case centered on encryption failures, the root investigation was triggered by breach reports — the same kind of breach reports that patient misidentification generates. When PHI goes to the wrong person because your matching algorithm failed, you have a reportable breach under 45 CFR § 164.402.
OCR doesn't care that Congress blocked the universal patient identifier. Your organization is still responsible for safeguarding PHI. That means your workforce needs to understand how patient identification errors create HIPAA exposure. If your team hasn't been through recent training on PHI handling and breach scenarios, explore the HIPAA training catalog at HIPAACertify for courses that cover exactly these situations.
What Organizations Are Doing Instead
Since there's no universal patient identifier, the industry has built workarounds. Some are clever. Most are expensive. None are complete.
Referential Matching Databases
Companies maintain massive databases of consumer identity data — credit bureau records, address histories, phone numbers — and use them to triangulate patient identity. This works better than demographic matching alone, but it introduces new privacy concerns. You're now feeding patient data into commercial identity systems.
FHIR-Based Digital Identity
The HL7 FHIR standard supports patient matching through resource endpoints, and the Trusted Exchange Framework and Common Agreement (TEFCA) has pushed for better identity resolution. But TEFCA's Patient Matching requirements still rely on probabilistic methods without a unique identifier at the core.
Biometric Matching
Palm vein scanning, iris recognition, and fingerprint systems are deployed at some large health systems. They work well within a single organization. They don't solve cross-organizational matching because there's no shared biometric registry — which circles right back to the universal patient identifier debate.
How This Affects Your HIPAA Compliance Program Today
You don't need to wait for Congress. Patient identification is a HIPAA compliance issue right now, and your risk analysis should address it. Here's what I recommend to every covered entity I work with:
- Include patient matching accuracy in your HIPAA risk assessment. If you can't quantify your duplicate record rate, you have a gap in your analysis.
- Train your registration staff on identity verification. Front desk errors are the number one source of record overlays. Your workforce training should include specific scenarios involving misidentified patients and the HIPAA implications. The HIPAACertify training courses cover workforce obligations around PHI accuracy and breach prevention.
- Document your matching methodology. OCR expects covered entities to have reasonable safeguards. If your only safeguard is "we trust the EHR algorithm," that's not going to hold up in an investigation.
- Monitor your patient portal disclosures. Portal misdirections — where Patient A sees Patient B's records — are among the fastest-growing categories of reported breaches on the HHS Breach Portal.
Will 2026 Finally Be the Year?
I've been asked this question every year for a decade. The honest answer: probably not this year either. But the political dynamics have shifted. Healthcare interoperability mandates keep expanding. TEFCA is scaling. And the sheer volume of health data flowing between organizations makes the absence of a universal patient identifier more dangerous with each passing year.
The ONC's focus on information blocking and interoperability has created a paradox: the government demands that data flow freely, but it refuses to fund the one mechanism that would let it flow accurately. Something has to give.
What I tell my clients is this: don't build your compliance program around the assumption that a universal patient identifier will arrive and solve your problems. Build it around the messy reality that exists today. Train your people. Audit your matching rates. Document everything.
Because when OCR comes knocking after a misdirected disclosure, they won't ask whether Congress gave you the tools to prevent it. They'll ask what you did with the tools you already had.