Independent practice

Eighteen percent

Independent practice is not dying of mystery. It is dying of a payment rate that has fallen a third against its own costs, an administrative load that scales with rules rather than patients, and a negotiating position of one.

82%of US physicians are employed by hospitals or corporate entities

33%decline in Medicare physician payment against practice-cost inflation since 2001

70.8%of physicians in acquired practices cited payer negotiating leverage as a reason

The numbers

As of January 2026, 82.0% of US physicians were employed by hospitals or corporate entities — 59.7% by hospitals and health systems, 22.3% by corporate owners. About 18% remain in physician-owned settings. Since 2018, roughly 253,000 additional physicians became employees and 85,000 additional practices were acquired. (Physicians Advocacy Institute / Avalere, 2026)

Corporate-owned practices — insurers, private equity, pharmacy chains — now outnumber hospital-owned ones, 33.2% against 30.6%. In rural areas, 80.2% of physicians are employed.

The AMA's benchmark survey tells the same story from the physician's side. Private practice fell from 60.1% of physicians in 2012 to 42.2% in 2024. Physicians holding an ownership stake fell from 53.2% to 35.4%. Solo practice went from 18.4% to 11.9%. Practices of ten or fewer physicians dropped from 61.4% of physicians to 47.4%. (AMA, 2024 Benchmark Survey)

The two datasets do not contradict each other, though they look like they might. The AMA surveys physicians about the practice they work in and counts physician-owned practices at 42.2%; the Physicians Advocacy Institute counts employment relationships and ownership of practice entities, which captures corporate and hospital acquisition that leaves the practice name on the door. Read together, the direction is unambiguous and the exact level depends on what you are counting.

Why they sell — asked directly

The AMA asked physicians whose practices were acquired between 2014 and 2024 for their reasons. The answers are unromantic:

70.8% — the need to better negotiate higher payment rates with payers
64.9% — improved access to costly resources
63.6% — better management of payers' regulatory and administrative requirements
55.1% — easier participation in risk-based payment models

Three of the four are about payers. Only one is about capital.

The payment arithmetic

The Medicare conversion factor was cut in 2021, 2022, 2023, 2024 and 2025 — five consecutive years — before rising in 2026. Adjusted for practice-cost inflation, the AMA calculates that physician payment has declined 33% between 2001 and 2026 (AMA, 2026), and 18% over the shorter 2015–2026 window (AMA chart). Medicare physician payment fell from 82% of private rates in 2011 to 68% in 2024.

2026 brought relief that is worth reading precisely: a one-year statutory increase of 2.50%, a MACRA baseline that now splits by APM participation, and a −2.5% "efficiency adjustment" applied to work RVUs for non-time-based services — with evaluation and management, care management, behavioural health, telehealth-list and maternity codes exempt. (CMS CY2026 PFS final rule)

The 2026 increase is a temporary statutory bump of roughly the same order as one year of practice-cost inflation. It does not recover the preceding decade, and it expires.

Meanwhile 90% of medical groups reported operating costs above the prior year, averaging about an 11% increase. (MGMA, 2025)

What technology honestly cannot fix

It is worth saying plainly, because vendors in our position rarely do.

Software does not change your fee schedule. The single largest reason practices sell is negotiating leverage against payers, and no EMR creates leverage. Only scale, or a market position, or a regulator does.

Software does not fix a payer mix. A practice whose economics depend on Medicare Advantage rates in a concentrated market has a contracting problem, not a technology problem.

Software does not replace a retiring physician. Succession is the quiet driver behind a great many sales, and no product addresses it.

What it can change

The third-ranked reason for selling — better management of payers' regulatory and administrative requirements, cited by 63.6% — is the one where the answer is partly buildable. So is the labour cost hidden inside it.

Consider what the administrative load actually consists of. Thirteen hours a week per physician on prior authorization. Around 5.8 hours of EHR time for every eight scheduled. Denials and appeals named the biggest revenue leak by 48% of groups. Forty percent of practices reporting they have hired or reassigned multiple full-time administrative staff per physician to manage payer rules, audits, appeals and reporting. (MGMA, 2026)

That is the layer where a small practice is structurally disadvantaged, because compliance cost scales with the number of rules, not the number of physicians. A twelve-hospital system amortises a prior-authorization team across a thousand doctors. A four-physician practice amortises it across four.

Consolidation is, in large part, a story about fixed administrative costs meeting a small denominator. Anything that converts fixed administrative labour into software is a direct argument against selling.

The adoption gap that worries us

Here is the uncomfortable finding. AI comfort among physicians at small group practices runs at 43%, against 65% at enterprise organisations — a 22-point gap. In the same survey, nine in ten small-practice physicians expressed fear about the difficulty of staying independent. (athenahealth / Harris Poll, 2026 — vendor-commissioned, and we would say the same of our own surveys.)

The barriers reported by practice administrators are cost, limited EHR integration, unclear ROI and accuracy concerns — in that order. Which means the tools most capable of narrowing the administrative gap are arriving fastest at the organisations that need them least.

If that persists, AI becomes another consolidating force rather than a countervailing one. That is a real risk and it is not obvious which way it resolves. What we can say is which side a product's design puts it on: whether it requires an implementation team, whether it prices per seat in a way that punishes small denominators, whether it assumes an analytics department exists to measure it.

Where we stand

ChartVoyant exists because a four-physician practice should not need a twelve-month implementation and a dedicated informaticist to get the documentation, coding and billing load down. The published implementation research is dated but the order of magnitude is telling: one careful study put EHR implementation at $46,659 per physician over sixteen months, with productivity still around 4% below baseline a year in — though the authors' own conclusion was that the revenue and productivity loss proved less burdensome than practices had feared. (AHRQ, 2011) Either way, it is a number a small practice has to plan around rather than absorb.

Eighteen percent is a hard figure to read optimistically. But the reasons practices give for selling are specific, and one of the biggest of them is administrative labour that software can genuinely absorb. That is worth building for.

Built for the practices that stayed

ChartVoyant is an EMR for independent ambulatory practice — voice-first documentation, built-in billing, and no enterprise implementation to survive first.

Try the live demo Talk to us

Sources

Physicians Advocacy Institute / Avalere Health. "Physician Employment Trends and Practice Acquisitions: 2018–2026," April 2026. physiciansadvocacyinstitute.org
AMA. "Physician Practice Characteristics in 2024" (Physician Practice Benchmark Survey). ama-assn.org
AMA. National Advocacy Update, June 2026 — Medicare payment against practice-cost inflation. ama-assn.org
AMA. History of Medicare conversion factors. ama-assn.org
AMA. Medicare updates compared to inflation, 2015–2026. ama-assn.org
CMS. Calendar Year 2026 Medicare Physician Fee Schedule final rule (CMS-1832-F) fact sheet. cms.gov
MGMA Stat. Medical practice operating costs, June 2025. mgma.com
MGMA. 2026 Regulatory Burden Report, April 2026. mgma.com
athenahealth / The Harris Poll. 2026 Physician Sentiment Survey. athenahealth.com
Fleming NS, et al. Cost of EHR implementation in primary care practices. AHRQ Publication 11-0081-4-EF, 2011. digital.ahrq.gov

Keep reading

Where AI in healthcare actually worksEighty-one percent of physicians now use AI professionally. The gap between what it does well and what it is sold as doing has never been wider. What an AI-native EMR looks like on a TuesdayNot the demo version. A walk through one ordinary visit, and the places where the design either earns its keep or quietly stops being used. The documentation burden, measuredEveryone agrees the EHR eats the day. Far fewer people have looked at what the logs actually say — or at what ambient AI has been measured to give back.