Clinician Turnover Cost Calculator

Two headcounts, two turnover rates, two replacement costs, kept apart the whole way through and only added together at the very end.

What it calculates

For each profession, clinicians replaced this year equals headcount multiplied by the annual turnover rate. Turnover cost equals that number multiplied by the replacement cost you enter. The modelled annual cost of clinician turnover is the nurse figure plus the physician figure. A cost per point of turnover is also shown for each profession, so you can see what one point of movement in either rate is worth, the same way NSI reports it for RNs in its own survey.

Why nursing and physician costs are never averaged

A staff RN and an employed physician cost an order of magnitude apart to replace, and blending them into one figure per clinician would hide that difference rather than model it. This tool prices them as two separate line items throughout, and the results panel keeps them on separate rows before summing them.

Where the nursing defaults come from

The default replacement cost per staff RN, USD 60,090, and the default national RN turnover rate, 17.6 percent, are both from the 2026 NSI National Health Care Retention and RN Staffing Report, published by NSI Nursing Solutions, Inc. in March 2026 from survey responses covering 527 hospitals and calendar year 2025 data. NSI is a nurse staffing and recruitment firm, not a government body, and this is its own published industry survey rather than a federal statistic, stated plainly here rather than dressed up as one. NSI's figure covers recruitment, orientation, temporary coverage and lost productivity for a bedside staff RN; it does not include travel or agency nurse costs.

Where the physician defaults come from

The default national physician turnover rate, 7.6 percent, is from Bond AM, Casalino LP, Tai-Seale M, et al., "Physician Turnover in the United States," published in the Annals of Internal Medicine in 2023. The study built a national estimate from the complete set of traditional Medicare billing records and found the annual rate rose from 5.3 percent in 2010 to 7.6 percent in 2018, the most recent year it measured. It is a peer-reviewed academic estimate, the strongest kind of source this tool cites, but it measures physicians changing where they practice, not physicians leaving medicine altogether, which is the definition a replacement-cost model needs.

The default replacement cost per physician, USD 500,000, is the low end of the range the American Medical Association publishes in "How much physician burnout is costing your organization" (2018): USD 500,000 to more than USD 1,000,000 per doctor, covering recruitment, sign-on bonus, lost billings and onboarding. The AMA's own article does not attach an independent formula or a primary study to that range, so it is cited here as the AMA's own published estimate, not as a peer-reviewed finding, and the field is left wide open for a reader with a specialty-specific figure to use it instead.

The AMA and Annals of Internal Medicine comparison figure

The results also show physician headcount multiplied by USD 7,600, a separate figure from Han S, Shanafelt TD, Sinsky CA, et al., "Estimating the Attributable Cost of Physician Burnout in the United States," Annals of Internal Medicine, 2019. That study estimated burnout-attributable cost, from turnover and reduced clinical hours combined, at USD 7,600 per employed physician per year, averaged across every physician whether or not they actually left that year. It is shown beside the modelled turnover cost, never added to it, because it measures something different: an organization-wide average blending two effects across an entire physician workforce, versus this tool's figure, which only prices the physicians who actually departed.

What it refuses to do

It does not estimate how much any intervention, ambient documentation, a scribe, a wellbeing program or anything else, would reduce either turnover rate. No published, credible study puts a number on that reduction for either profession, and inventing one would hand a business case a figure it could not defend under questioning. What this tool gives instead is the baseline any such case has to beat: change the turnover rate inputs yourself to see how much the total moves, and treat the size of that move, not a percentage this tool asserts, as the bar an intervention has to clear.

What it leaves out

Advanced practice providers, allied health staff, and non-clinical turnover are not modelled; NSI publishes separate figures for APPs that a reader managing that population should use instead. Locum and travel coverage costs incurred while a role sits open are not included in either replacement-cost default, and neither is the cost of a bad hire made under pressure to fill a vacancy quickly, both of which are real and both of which push the true number higher than what this tool shows.