Florida's license file and the federal NPI registry describe different administrative records. A dental hygiene license concerns professional authorization; an NPI identifies a provider in health care transactions. Neither record alone establishes that someone wants a new job. Comparing the files requires matching their definitions as well as their dates.
What is an NPI used for?
The Centers for Medicare and Medicaid Services describes the National Provider Identifier as a standard identifier for health care providers in covered administrative and financial transactions. It is associated with billing and other transactions, rather than serving as a count of everyone holding a state professional license. Source: CMS NPI standard.
That purpose matters when selecting a source for workforce research. A question about Florida dental hygiene licenses starts with the state's licensing records. A question about provider identifiers starts with the federal registry. Calling either file a complete picture of the profession would require evidence about its coverage that a file total alone cannot supply.
CMS also cautions that issuing an NPI does not validate a provider's licensure or credentials. A match in the registry therefore cannot replace a current license check. Source: CMS NPI Registry.
Why can the records disagree without either source being wrong?
Each source collects information for its own administrative purpose. A researcher then applies filters, such as profession or location, to produce a comparison. Differences may arise from the source definitions, the filters, or the age of the records. The comparison needs to document those choices before assigning an explanation to the result.
A Florida license is not evidence that the person's federal record uses a Florida practice address. Nor is the absence of a match in a particular extract proof that the person has no NPI. The search could miss the relevant record, or the records could carry different information.
Our files do not establish a person-by-person explanation for the gap between the saved populations. We therefore do not claim that most hygienists lack an NPI or that a particular billing arrangement explains every missing match. Those would require evidence beyond an aggregate comparison.
Why leave the original comparison ratio out?
A ratio makes a source comparison look precise even when the source dates and filtering rules are incomplete. The Florida DOH snapshot behind this article is August 3, 2026. The saved federal comparison in our source data does not establish a dated, reproducible population on the same terms.
We do not present that historical comparison as a current census. It also compares email-bearing state records with a federal provider population, which are different measures. A ratio between them would not describe additional people who will receive or respond to recruiting messages.
The useful finding is about source selection: a license file may answer a licensing question that a provider-identifier search cannot answer. The scale of that difference requires a documented comparison. Readers should be able to see which records qualified on each side before treating a ratio as evidence.
What can the Florida file establish about email?
Our public dataset keeps licensed people and licensed people with an email field in separate bands. The distinction is visible in Tampa-St. Petersburg-Clearwater: the Florida DOH file of August 3, 2026 places licensed people above 2,500 and people with an email field in the 1,001–2,500 band. In Orlando-Kissimmee-Sanford, both fall in the 1,001–2,500 band.
These ranges describe fields in a dated snapshot. An email-bearing record does not establish delivery, readership, or interest in a particular vacancy. Matching bands do not establish that the underlying populations are identical, either. Public bands cannot support an exact missing-email count.
We describe an address as the email in the license record. We do not label it personal or permanent without evidence about that address. A professional may use different contact details for different administrative purposes. Our aggregate tables cannot determine which inbox the person prefers for recruiting.
Does a matched phone number complete the contact record?
A number associated with a provider record needs its own context. It may refer to a practice rather than a private line. A matching name does not show who answers the number or whether the professional wants recruiting calls there.
For a hiring conversation, ask an interested professional which number they want used and when they are available. The request should identify the particular opening. Permission for that conversation should not be inferred from the existence of a number in another source.
Combining an email field with a telephone field does not produce knowledge of a person's preferences. It produces a combined record whose fields still need their original context. A researcher who drops that context can make the result look more certain than either source warrants.
Why does matching require continuing review?
Names and addresses can change, while records from separate sources may update on different schedules. A matching process needs to preserve where a field came from and when the source was captured. Otherwise, a later reader cannot distinguish new information from an older value carried forward.
Ambiguous matches need to remain ambiguous. Similar names or nearby addresses do not justify merging records into a confident identity claim. A mistaken combination could attach the wrong contact details to a person or send a licensing question toward an unrelated record.
Even a well-supported match has a limited scope. It can support the conclusion that records refer to the same person. It does not establish that the person is available, dissatisfied with their employer, or suited to a particular opening. Those questions require a conversation and a separate evaluation of the role's requirements.
What should a reader expect from a published comparison?
Expect a description of the population on each side, the date of each extract, and how the author treated uncertain matches. Counts of records, people, and contact fields need separate labels. A comparison that shifts between those units can make a large difference appear meaningful without explaining what changed.
For geographic licensing data, expect ranges. Our public data uses a floor of 11 and publishes bands above that threshold, as specified in the salary-data files. The Florida DOH snapshot is August 3, 2026. We do not publish individual names or a contact directory with these articles.
The license-count guide shows how to read those bands. For a question about an individual license, consult the current state record. For a question about a provider identifier, consult the federal registry. For a question about interest in a job, the person has to answer.