# Nurse practitioners per 100 physicians, by state Counts of five health professions in the United States from the CMS National Plan and Provider Enumeration System (NPPES), September 2026 monthly full replacement file, with Census 2024 state populations as denominators. Every person counted here is an individual **enumerated and not deactivated** in NPPES. An NPI is not a licence, never expires, and is surrendered only by an affirmative deactivation, so these counts are larger than any count of people at work. Do not read them as such. Headline: Mississippi has 86.1 nurse practitioners per 100 physicians, Hawaii has 25.1. No state has more nurse practitioners than physicians. ## Files ### providers_by_state.csv One row per state, 50 states plus the District of Columbia, ranked by nurse practitioners per 100 physicians. Territories and armed forces postal codes follow in a flagged block (`row_type = territory`) and are in no ranking. A final row (`row_type = no_practice_state`) holds the 1,709 counted individuals whose record carries a practice-location value that is not a US state code. | Column | Meaning | |---|---| | `state` | Two-letter code as it appears in the practice-location field | | `state_name` | Full name | | `row_type` | `state`, `dc`, `territory` or `no_practice_state` | | `population_2024` | Census `POPESTIMATE2024`; empty where Census publishes none | | `population_source` | Which of those two applies | | `physicians` | Individuals enumerated and not deactivated, primary taxonomy 207* or 208* | | `nurse_practitioners` | Same, primary taxonomy 363L* | | `physician_assistants` | Same, primary taxonomy 363A* | | `dentists` | Same, primary taxonomy 1223* | | `pharmacists` | Same, primary taxonomy 1835* except 183700000X | | `_per_100k` | That count divided by `population_2024`, times 100,000 | | `nps_per_100_physicians` | Nurse practitioners divided by physicians, times 100 | | `pas_per_100_physicians` | Physician assistants divided by physicians, times 100 | | `rank_nps_per_100_physicians` | 1 to 51 over the states and DC, highest first, ordered on the unrounded ratio | ### enumeration_by_year.csv Enumeration year 2005 to 2026. The count is people who took their NPI in that year and who today are enumerated, not deactivated, and hold that primary taxonomy. | Column | Meaning | |---|---| | `enumeration_year` | Year from `Provider Enumeration Date` | | `physicians` to `pharmacists` | As above, by year of enumeration | | `students_390200000X` | Enumerated that year, primary taxonomy today is the NUCC student code | | `enumerated_that_year_holding_a_physician_primary_taxonomy_today` | Identical to `physicians` by construction; kept because the brief names it | | `enumerated_that_year_holding_a_physician_taxonomy_in_any_slot_today` | Physician code in any of the 15 slots, not only the primary | | `nps_per_physician` | Nurse practitioners divided by physicians; blank from 2022 on | | `ratio_confounded_by_the_student_taxonomy` | `Y` from 2022 on; see METHOD.md | | `partial_year` | `Y (to 2026-09-13)` on the 2026 row | ### enumeration_by_year_state.csv Enumeration year by state, physicians and nurse practitioners only, every year 2005 to 2026, states and DC plus a flagged territory block. `nps_per_100_physicians` is left blank where the physician cell is under 5 and on every row for enumeration year 2022 or later. Use it for the 2020-and-later cut, which spans those years and so is published as two counts and their difference, with no ratio. ### sex_by_year.csv Female share of newly enumerated by profession and year. The denominator is rows carrying `Provider Sex Code` M or F for that profession and year; the field is binary, with no third value, and rows with no value are outside the denominator. The first year each profession's new cohort is at least half female: dentists 2019 (51.2%), physicians 2021 (50.9%), pharmacists 2006 (51.0%). Physician assistants and nurse practitioners are majority female from the registry's first year, so neither has a crossing. This is a flow, the composition of each year's new enumerations. It is not comparable with a stock figure such as the ADA's 39.4% female among professionally active dentists. ### facts.json Machine-readable dump of every figure quoted in the report, including the funnel, the taxonomy audit, the national totals and all five ranked lists. ## Small cells Any cell under 5 is printed as `<5`, and a per-100,000 rate is left blank wherever its numerator is masked. A true zero is printed as `0`. ## Sources - CMS NPPES September 2026 monthly full replacement file, file date 2026-09-13 https://download.cms.gov/nppes/NPI_Files.html - NUCC Health Care Provider Taxonomy code set 26.1: https://www.nucc.org/ - Census state population estimates, `POPESTIMATE2024` https://www2.census.gov/programs-surveys/popest/datasets/2020-2024/state/totals/NST-EST2024-ALLDATA.csv ## Reproducing ``` python3 tools/report_np_ratio_facts.py ``` One streaming pass over the 9,798,758-row slim file. Method, limitations and the comparison against FSMB, AANP, NCCPA, ADA, BLS and AAMC are in METHOD.md. Licence: CC BY 4.0. Cite ProfessionLens. The underlying files are public United States government and NUCC data.