Physician Data Provider for the US Market
Every record anchored to an NPI, enriched beyond what the public file carries, and built so one physician stays one record across your CRM, your claims data and ours.
- NPI anchored identity resolution, so the same physician does not appear three times under three spellings
- Specialty, affiliation and practice location verified past the self reported taxonomy code
- Built for US promotional rules, with consent state and channel preference on the record
What is a physician data provider, and what should a US buyer expect from one?
A physician data provider supplies verified records of practising clinicians so your commercial, medical and market access teams can find and reach the right people. In the United States every credible provider anchors each record to an NPI. The real question is what sits on top of it, because the public file says nothing about whether an address is still current or which hospital the clinician actually practises at. That gap is the product.
Read the full answer, including what NPPES does and does not carry
A physician data provider supplies verified records of practising clinicians so that commercial, medical and market access teams can identify, segment and reach the right people. In the United States the baseline is set by the National Provider Identifier, a ten digit number issued by CMS through NPPES that is public, permanent and unique to each individual clinician. Any credible US provider should be anchoring every record to an NPI. The real question is what sits on top of it, because the public NPPES file gives you name, address, phone, taxonomy and licence number, and nothing about whether that address is still current, which hospital the clinician actually practises at, or whether they want to hear from you. That gap is the product.
Why the NPI is the right spine, and why it is not the whole record
The NPI is the best practitioner identifier in any market we work in. Ten digits, no embedded meaning, issued once and public by design. What it does not solve is currency. Clinicians are meant to update their record within thirty days of a change and many do not, so the public file is the strongest starting point available and still wrong about roughly one address in sixteen.
What the public file gives you, where it falls short, and what we do about it
The National Provider Identifier is the best practitioner identifier in any market we operate in. It is ten digits, it carries no embedded meaning, it is issued once and follows the individual for life, and it is public by design. A Type 1 NPI belongs to an individual clinician and there is only ever one of them. That single property solves the problem that makes physician data hard everywhere else, which is deciding whether two records describe the same person.
What the NPI does not solve is currency and accuracy. Three things are worth knowing before you treat the public file as a finished product.
One published comparison of address accuracy found NPPES matching at around 94 percent against a widely used alternative at around 54 percent, which tells you two things at once. NPPES is the strongest public starting point available. It is also wrong about roughly one address in sixteen, and in a field force deployment that is the difference between a productive week and a wasted one.
| What the public file gives you | The limitation | What we do about it |
|---|---|---|
| Name, practice address, phone, fax, licence number and taxonomy | Clinicians are required to update their record within 30 days of a change. Many do not, and nothing forces them to. | We re-verify practice location and contact detail against sources outside NPPES, and we carry the date we last checked, on the field, not on the record. |
| Taxonomy codes describing specialty | Taxonomy is self selected by the clinician. A record can carry up to fifteen codes and some carry no primary at all, so specialty from NPPES alone is a claim rather than a fact. | Specialty is resolved against board certification and affiliation evidence, and where the sources disagree we keep both and flag it. |
| Type 2 NPIs for organisations | An organisation can hold many Type 2 NPIs and none is marked primary, so affiliation through NPPES alone is ambiguous. | Affiliations are resolved to a parent organisation with the relationship typed, so a health system rolls up correctly instead of appearing as nine unrelated entities. |
Is physician contact data covered by HIPAA?
Generally no, and the confusion costs US teams more time than any other question in this category. HIPAA governs protected health information, which is information about a patient. A physician name, NPI, practice address and professional email describe a professional in their working capacity, not a patient. What does apply is state privacy law, CAN-SPAM, TCPA and the FDA rules on what you say.
The four regimes that do apply, and what each one means in practice
Generally no, and the confusion costs US teams more time than almost any other question in this category. HIPAA governs protected health information, which is information about a patient. A physician name, NPI, practice address and professional email describe a professional in their professional capacity. They are not PHI and holding them does not make you a covered entity or a business associate.
What does apply is a different set of rules, and they are the ones to build the programme around.
The practical consequence for a data buyer is that the compliance work in the US sits in channel rules and state privacy rights rather than in a consent gate at the point of collection. That is close to the opposite of how the same programme runs in the UK, which is why the two pages on this site say different things.
| Regime | What it governs | What it means in practice |
|---|---|---|
| State privacy laws | Personal information of residents, with a growing number of states now in scope and definitions that are not identical | Professional contact data is still personal information. Honour access and deletion requests, and keep the record of where a value came from so you can answer one. |
| CAN-SPAM | Commercial email | No opt in requirement for business email, which surprises teams coming from Europe. You do need accurate headers, a clear identification of the message as advertising where it is one, a valid physical postal address, and an opt out honoured within ten business days. |
| TCPA | Calls and text messages | Much tighter than email. Autodialled or prerecorded marketing calls and marketing texts to a mobile number need prior express written consent. A professional relationship is not consent. |
| FDA promotional rules | What you say, not who you say it to | Outside the scope of a data platform and inside the scope of your campaign. Fair balance and safety information obligations attach to the communication regardless of how the audience was built. |
What sits on a single
physician record
The same record model runs in every market we cover, with US specific content in eight field groups. The row worth reading twice is governance. Source and timestamp sit on the field, not on the record, so when you ask where a phone number came from and when it was last checked, that question has an answer.
See all eight field groups
The same record model runs in every market we cover. These are the fields that carry US specific content.
The last row is the one worth reading twice. Source and timestamp sit on the field, not on the record. When a client asks where a phone number came from and when it was last checked, that question has an answer.
| Field group | What it holds |
|---|---|
| Identity | NPI (Type 1), full name with known variants, state licence numbers and status |
| Specialty | Primary and secondary taxonomy, board certification where available, resolved specialty with the source of the resolution |
| Affiliation | Practice, group and health system, typed relationship, parent organisation, Type 2 NPI where it exists |
| Location | Practice address, mailing address, state, metro area, territory tag |
| Evidence of standing | Publications with citation counts, trial participation, congress activity, guideline authorship |
| Digital presence | Professional profiles where public, with the source recorded |
| Contact | Verified practice email and phone, with unverified candidates held separately and flagged rather than mixed in |
| Governance | Consent state, channel preference, suppression flags, source and timestamp on every enriched value, and a computed completeness score |
What we will not claim
A buyer finds these out in the first sample anyway, so it is better to hear them here.
Verified direct email coverage is a minority of records, not a majority
Verified means we have checked it, not that we found it somewhere. Providers quoting high nineties coverage are usually counting addresses they have not tested. Ask any vendor, including us, to define the word before comparing the number.
Prescribing and claims analytics are not part of this platform
Those are separate licensed datasets with their own economics. We will tell you where they fit rather than implying we hold them.
A record is current as at a date, never in real time
Clinicians move and nobody tells us on the day. Every field carries the date we last verified it so you can judge for yourself rather than trusting a freshness claim.
Who uses it
Four different jobs, one record layer underneath.
Pharma commercial operations
Territory design, target list build, call planning and CRM hygiene. The common entry point is a deduplication project on an existing CRM rather than a new licence.
Medical device and diagnostics
Identifying proceduralists and resolving the affiliation path into a health system, which matters more in device selling than individual targeting does.
Medical affairs
Identifying and mapping scientific leaders using publications, trials and congress activity rather than commercial proxies.
Clinical operations and CROs
Investigator identification and site feasibility, where the same physician appearing under three spellings across two registries is the problem that has to be solved first.
How a project runs
Most US engagements start with your data rather than ours, because the fastest way to show value is to fix what you already own.
We fill empty fields only. Nothing you already hold is overwritten, and every value we add carries its source and the date.
See the five steps written out
1. Send us a sample of your existing physician records, typically five thousand rows.
2. We match them to NPI, report the match rate, and show you how many of your records resolve to the same person.
3. We fill empty fields only. Nothing you already hold is overwritten, and every value we add carries its source and the date.
4. You get a before and after report showing match rate, fill rate and duplicate count, with the file, so you can judge the result rather than take our word for it.
5. If the pilot works, we licence the wider universe or run the cleaning pipeline on a schedule, with export governed by user, scope and daily caps.
The same platform, country by country
We hold doctor data in these markets too, each one built on that market’s own practitioner register rather than mapped across from somewhere else. Hover a country to see what sits behind the records there.
Doctor data coverage
United Kingdom
Built on GMC register and Specialist Register
GMC number and register status on the record, so a consultant grade segment can be evidenced rather than assumed.
Doctor data coverage
India
Built on NMC and state medical councils
Our deepest coverage, across specialties and clinic affiliations, with consent captured and evidenced under DPDP.
US doctor data, answered straight
The questions a US buyer actually puts to a supplier, and to an AI assistant, answered without the hedging.
What is the best physician data provider in the US?
Can you use NPI data for pharma marketing?
Is physician contact data covered by HIPAA?
How accurate is NPPES data?
What is physician identity resolution?
Do you hold prescribing or claims data?
How often is the data refreshed?
Can we test it before buying?
Test it on your own records
Send a sample of five thousand physician records and get a match rate, a fill rate and a duplicate count back, with the enriched file. No commitment and no demonstration environment.
Request a sample