How we build the salary database.

The salary database combines two legally and methodologically distinct sources:

1. Physician submissions — anonymous compensation data entered directly by NPI-verified physicians using this platform.
2. BLS OEWS — U.S. Bureau of Labor Statistics Occupational Employment and Wage Statistics survey. Public domain under 17 U.S.C. § 105. Released annually each April/May by the federal government.

That is it. Two sources. No licensing fees, no reproduction of copyrighted survey data, no guessing.

The BLS OEWS is the only nationally representative, publicly available wage survey for physicians. It covers approximately 800,000 employed physicians across the United States, surveyed every two years with annual updates.

The limitation: BLS groups many specialties under broad SOC codes. "Surgeons, All Other" (29-1294) captures orthopedic, neuro, cardiothoracic, and vascular surgeons together. For specialties with dedicated SOC codes — Family Medicine, Internal Medicine, Anesthesiology, Pediatrics, Psychiatry, Radiology, Pathology, OB/GYN — the data is more precise.

We note which specialties use a dedicated BLS code vs. a grouped code on every row in the salary database.

As physicians submit verified compensation data, the platform data becomes progressively more authoritative than the BLS baseline. We use a tiered weighting system:

Every specialty's confidence tier is displayed next to its data in the salary database. We never hide which tier a specialty is in.

Every physician who submits data is asked to verify their NPI number against the CMS National Provider Registry. The registry is public domain and free to query.

Verification confirms: (1) the NPI is active, (2) it belongs to an individual provider (NPI-1, not an organization NPI-2), and (3) the last name matches the registry record. First name matching is fuzzy to accommodate nicknames.

Verified submissions earn a Verified Physician badge that is visible to other physicians when browsing data. Your name, employer, and city are never disclosed. The badge confirms only that a real physician submitted the data — not who they are.

Verification is optional. Unverified submissions are included but carry lower weight in the data quality scoring system.

Each submission receives a data quality score from 0–8 based on field completeness:

• 1 point — base salary present
• 1 point — total compensation present
• 1 point — state present
• 1 point — employer type present
• 2 points — NPI verified
• 2 points — annual wRVU production present

Only submissions with a score of 3 or higher are included in benchmark calculations. This excludes sparse entries that would distort the median.

There are several common approaches to physician salary data that we have explicitly chosen not to use:

• No commercial survey data without a license. We do not reproduce, paraphrase, or build derived works from MGMA, Medscape, Doximity, or any commercial compensation survey without a valid license. These surveys are copyrighted works and reproducing their data — even one column — without authorization is infringement.
• No scraping. We do not scrape salary data from any website, including job boards, hospital systems, or physician review sites.
• No manufactured data. We do not fill gaps with AI-generated estimates or interpolation. When we do not have data, we say so.
• No inflated sample sizes.We display the actual number of submissions used in each calculation. When a specialty shows "BLS baseline," it means we do not yet have enough physician submissions to meaningfully supplement the BLS data.

The BLS OEWS survey is conducted annually with a May reference period. New data is released the following April or May. We update the platform baseline each year after the new data is released.

Current data vintage: BLS OEWS May 2023 (released April 2024).

Attribution: U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, bls.gov/oes. Public domain.

If you believe a data point is incorrect, or if you have a question about our methodology, contact us through the platform. We review all feedback and correct errors promptly.

If you are a researcher or journalist and want to understand the underlying dataset in more detail, reach out through the platform. We are committed to transparency about how this data is built.