Why Gale

Why are the best getting paid up to 40% less?

A network that promises to fill a caseload prices every clinician at the average and keeps the difference. Independent, cash-first practice keeps the rate — and the patient. The case follows, every claim cited.

every number below is numbered and sourced to real literature · the demonstration data is synthetic

New to Gale

Everything in one place.

Start practice →

Already on Gale

Log in to your practice.

Log in →

See a sample patient chart — the EHR, no signup →

The spread

A network that pays everyone the average pays the best clinician the least

~37%
Kept by the platform, one documented session
Alma billed Cigna $151.74, paid the therapist $95
84%
Never told how the fee was split
survey of 667 therapists
50%
Earn the same or less than solo practice
same survey
$139.75
Self-pay per session, vs $99.75 insured
across 113.6M sessions

The aggregators sell a filled caseload. The price is a rate sheet with a row for a license, not for being good. In one documented case Alma billed Cigna $151.74 for a session and paid the therapist $95 — a ~37% margin the clinician never sees 1. The split is rarely disclosed: 84% of 667 surveyed therapists were never told how fees were divided, and half earn no more than they would alone 2. The rate is set by license and billing code, not experience — and it moves without consent. From July 2026 Aetna pays a 60-minute session no more than a 45-minute one for Alma providers, and stops varying pay by credential at all 3.

Meanwhile the same hour is worth far more in cash than through insurance: $139.75 self-pay against $99.75 insured, across 113.6 million sessions 4 — a gap of 25 to 35% 5. A uniform rate is a fair deal for the average clinician. For anyone above it, it is a subsidy paid to the network.

On the insurance rail, 46 weeks$53,130
On Gale, cash rate minus the flat 3.5% all-in fee$58,595
What the rail costs you, a year$5,465

$116 kept on a $120 session — a flat 3.5% all-in, card processing included. $11 more per session than the insurance rail.

The rate card · pre-commercial demonstration

Every paid transaction — cash or insurance, any patient
flat 3.5% all-in
New York & Texas
$1.00 + costs at cost
Software
$0

One rate, one line: card processing is included in the 3.5%, a claim that never pays costs $0, and the fee is itemized on every payout row. In New York and Texas, where state law does not allow percentage-based platform fees, Gale's compensation is a fixed $1.00 per paid transaction with third-party costs passed through at cost — why the fee differs by state.

A planning sandbox, not a forecast — demonstration figures, not observed performance. The insurance-rail default is anchored to reported per-session reimbursement (SimplePractice: self-pay $139.75 vs insurance $99.75; Heard: ~$105; Medicare 90837 ceiling $167 — see the numbered sources); the Gale fees reconcile to the engine (`fees.py`). It assumes a patient who pays the cash rate and models no overhead or no-shows. Nothing you set here leaves your browser.

Figure 1. The cost of the network rail. Per-session yield on the insurance rail against a self-set cash rate, for the same weekly sessions. The rail default is anchored to reported reimbursement — SimplePractice's $99.75 insured average 4, Heard's ~$105 5, and the $167 Medicare 90837 ceiling 6. A planning sandbox, not a forecast, and a pre-commercial demonstration; it assumes a patient who pays the cash rate. The figure carries Gale's cash-pay rate card, reconciled to the billing engine.

Comparing Gale with other offerings →

The front door

Cash patients can't find you.

46%
Start at Google
42% read reviews first
32%
Have asked an AI chatbot about health
double the year before
.66 / .22
Mentions vs links, for AI-answer visibility
across ~75,000 brands

Patients no longer arrive by referral. Nearly half start at an insurer's directory, half at Google, and 42% read reviews before they choose 7. A third have already put a health question to an AI chatbot — double the share of a year earlier 8. On both surfaces the individual practice loses to the directory: it cannot out-rank Psychology Today or Zocdoc, and AI answers favor whoever is mentioned across the web, not whoever bought the most links — brand mentions correlate .66 with AI-answer visibility against .22 for backlinks 9.

Gale is built to be that front door: a library of nearly four thousand cited health articles and a public profile for every US clinician, on one domain — so a cash patient searching a condition or a city meets a real practice, not only an aggregator. It is non-exclusive by design: a clinician keeps Alma, keeps Psychology Today, and adds the page that is theirs.

Where you stand

Improve your public profile — get a free report to see where you stand.

See your ranking →

The first expression

Your instant, autonomous practice

Gale is adaptive software that automates billing and RCM, fosters a renewable patient journey, and returns lost income (up to 2x) to the single provider practice.

← inside the systemdrag, click a stage, or watch it playa practice of one →
stage 1/8 Inside the systemday zero
Documentation, in-visit and after hours34%
Inbox and portal triage22%
Prior authorization16%
Billing and coding follow-up14%
Credentialing and compliance paperwork14%

share of the non-clinical week — illustrative composite

An employed clinician sees patients for a fraction of the week. The rest goes to documentation, prior authorization, billing follow-up, and credentialing paperwork — work that feeds the middle layers, not the visit. The cited numbers sit in the prose beside this figure; the bars are the shape of a week.

Burden removed
  • nothing lifted yet — this is the week as it stands
You now hold
  • nothing yet — start dragging
Figure 2. The provider journey, scrubbed. Eight stages from inside the system to a practice of one. It plays on its own — drag the slider, click a stage, or use the arrow keys to take over. Stage values are an illustrative sandbox, not a forecast; the cited economics live in the surrounding text.

Five practices, told properly

What this looks like, one clinician at a time

The proof is particular: five research-backed stories, each in the register of the study that started this — numbered citations, honest limits, one clinician at the center. The argument that follows is the average; these are the instances.

The argument

For every hour an American clinician spends with a patient, nearly two more go to the screen and the desk 10, and the day ends with another stretch of “pajama time” in the chart 11. Every visit pays a billing toll 14; every practice pays a five-figure annual tax just to talk to insurers 15; a third of national health spending is administration 16. Under that weight, physicians sold their independence — private practice fell below half 18, and by January 2026 82% of US physicians worked for hospitals or corporate owners 19 — which bought the system higher prices, not better care 22 23. None of this burden is medicine. All of it is software, process, and arbitrage. Gale's thesis: make the software free, make the back office automatic, charge one flat 3.5% all-in on the payment when the clinician is actually paid — and the independent practice becomes the best job in healthcare, and the best deal in it.

The burden

The clinician's day belongs to the machine

2:1
Desk hours per patient hour
49.2% of the day on EHR and desk work vs 27.0% face time
5.9h
Of an 11.4h day in the EHR
incl. ~1.4h after hours — 'pajama time'
40
Prior authorizations per week
13 hours of physician + staff time
26%
PA caused a serious adverse event
physician-reported, 2025 survey

Direct observation across four specialties found physicians spending 27% of the office day in direct clinical face time and 49% on the EHR and desk work — for every hour with a patient, nearly two more at the screen 10. Event logs of 142 family physicians put it at 5.9 hours of an 11.4-hour day inside the EHR, a quarter of it inbox management, with about 1.4 hours spilling past clinic hours into the evening 11. Prior authorization alone runs 40 requests per physician per week and 13 hours of combined staff time; 95% of physicians say it delays necessary care, and 26% have watched it cause a serious adverse event 12. This is not a temperament problem. The clerical burden itself drives burnout — physicians using computerized order entry carry 29% higher adjusted odds of it 13.

The toll

Every encounter pays the middle layers first

The price of the paperwork is measurable. Billing and insurance-related activity costs $20.49 and 13 minutes per primary-care visit — 14.5% of professional revenue, rising to a quarter for emergency visits 14. Interacting with health plans costs a practice about $68,000 per physician per year 15. Zoom all the way out and administration consumed 34.2% of US health spending — $812 billion in 2017, against 17% in Canada 16. Automation has clawed back $258 billion of avoided administrative cost, and the industry's own index still counts $21 billion a year sitting in transactions nobody has automated 17. The toll is real money, collected from the smallest practices least able to pay it.

The exit that wasn't

So clinicians sold the one thing that was theirs

42.2%
Still in private practice, 2024
down from 60.1% in 2012
82.0%
Employed by systems or corporates
as of Jan 1, 2026 — up 85% since 2018
−33%
Medicare pay in real terms
2001–2025, adjusted for practice-cost inflation
$17k
Per physician per year
just to keep the EHR running (2010-era baseline)

The squeeze worked. Private practice fell from 60.1% of physicians in 2012 to 42.2% in 2024, and the AMA names the drivers plainly: inadequate payment, costly resources, administrative burden 18. By January 2026, 82% of American physicians were employees of hospitals, health systems, private-equity firms, or insurers 19. Behind that migration sat two decades of Medicare payment falling 33% in real terms 20 while the software a practice could not legally do without cost on the order of $32,000 per physician to install and $17,000 a year to keep 21. Employment was never the preference; it was the only door left open.

The arbitrage

The system charges more for the clinician's work than the clinician ever could

What did consolidation buy? Hospitals in monopoly markets price 12% higher than hospitals facing four or more rivals, and nearby mergers raise prices more than 6% 22 — while acquisitions showed modestly worse patient experience and no mortality or readmission gains 23. The arbitrage is codified: Medicare itself pays 105% more for the same mid-level office visit in a hospital outpatient department than in an independent office, and 141% more for the first hour of chemotherapy infusion 24. Private employers pay hospitals 254% of Medicare for the same services at the same facilities 25. The same hands, the same work — priced by who owns the building. That spread is not value. It is the rent the system collects on the clinician's license.

And the punchline has been in the literature for a decade: the smallest practices — one and two physicians — produce 33% fewer preventable hospital admissions than practices ten times their size, and physician-owned beats hospital-owned 26. The independent clinician was never the problem. The economics around them were.

What Gale does about it

Free software. An automatic back office. One honest fee.

Gale is the practice OS for a clinician who works for no one: licensing and credentialing run as a tracked pipeline, never auto-attested; a practice page built to be found; telehealth where the chart lives; a scribe that drafts the note in the visit, where the clinician stays the final editor of every note; billing that verifies eligibility before the visit and sends a clean claim after it; follow-ups triaged before they reach the inbox. The software costs nothing. Gale earns one flat 3.5% all-in per paid transaction — cash or insurance, card processing included, only on transactions that actually pay — its rate card is shown above as a pre-commercial demonstration. It never holds the clinician's payer contracts or takes a spread on the session. No subscription, no per-seat rent, no commission on new patients. Add up what a practice stops paying in the practice cost calculator, or see how the billing models compare.

Gale is built for the therapist-led independent practice today — the prescriber rail (e-prescribing, EPCS-ready) is a named upcoming cohort, not yet open. If that's your practice, the honest answer is not yet — the rest of the OS is not gated behind it.

Honesty rails, inherited from the research lineage and binding here: the demonstration data on every Gale surface is synthetic; economic figures shown in product are a planning sandbox, not a forecast; a claim that cannot be cited is removed rather than approximated. As licensed clinicians join, each Gale article a clinician reviews will carry their own name and credential — a real, verifiable reviewer, never a borrowed byline.

References

Every number above, sourced

  1. 1.ClearHealthCosts (Gold J et al.) (2025). Therapists have misgivings on the platforms: Alma, Headway etc. and the business of therapy. ClearHealthCosts. linkPer a therapist's Cigna Explanation of Benefits reported by ClearHealthCosts, Alma billed $151.74 for a CPT 90837 psychotherapy session and paid the therapist $95 (a $56.74, roughly 37% platform margin); on a CPT 90834 session Alma received $125 and paid $79.Health-journalism report (not peer-reviewed); the dollar figures come from one therapist's own Cigna Explanation of Benefits, not an audited aggregate.
  2. 2.Psychotherapy Action Network (PsiAN), reported by ClearHealthCosts (2025). Practice-management platform survey of 667 therapists. ClearHealthCosts / Psychotherapy Action Network. linkA Psychotherapy Action Network survey of 667 therapists found 84% were not informed about fee-splitting arrangements before joining a practice-management platform, and 50% report earning the same or less than in independent practice.Advocacy-group survey reported via health journalism (not peer-reviewed); figures attributed to PsiAN's survey of 667 therapists.
  3. 3.Navigating the Insurance Maze (2026). More clinician pay cuts? Aetna flattens Alma providers. Navigating the Insurance Maze. linkEffective July 15, 2026, Aetna will pay CPT 90837 at the same rate as the shorter 90834 for Alma-contracted providers and stop varying rates by clinician degree or credential type.Trade blog (not peer-reviewed); the specific rate change was corroborated by an APA advocacy letter to Aetna dated June 2026.
  4. 4.SimplePractice (2025). State of Private Practice 2025 Report. SimplePractice. linkAcross 113.6 million sessions on SimplePractice in 2025, self-pay averaged $139.75 per session versus $99.75 for insurance reimbursement, a roughly 29% gap.Vendor industry report (not peer-reviewed); platform-wide analysis of 113.6M sessions across 245,000+ clinicians.
  5. 5.Heard (2026). The Heard 2026 Financial State of Private Practice Report. Heard. linkIndividual therapy sessions commonly run $130 to $185 private-pay, while insurance reimbursement ran 25 to 35% below private-pay rates (for example, a $150 cash session reimbursed near $105).Vendor survey report (not peer-reviewed); approximately 2,000 therapists surveyed.
  6. 6.Centers for Medicare & Medicaid Services (Physician Fee Schedule), via MedFeeSchedule (2026). CPT 90837 Medicare reimbursement (non-facility), 2025 to 2026. CMS Physician Fee Schedule (aggregated by MedFeeSchedule). linkMedicare's non-facility payment for CPT 90837 (60-minute psychotherapy) was $154.29 in 2025 and $167.00 in 2026.Figures derive from the CMS Physician Fee Schedule; CMS publishes them only via its dynamic look-up tool, so a third-party aggregator is cited and cross-checked against a second aggregator.
  7. 7.Harmony Healthcare IT (2024). Healthcare moving statistics: finding a new doctor. Harmony Healthcare IT. linkWhen finding a new doctor, 46% of surveyed Americans use their insurer's directory, 46% use Google or a search engine, and 42% read online reviews.Consumer survey (not peer-reviewed); n=1,030 US adults via Prolific, April 2024.
  8. 8.Rock Health (Zweig M et al.) (2025). The tortoise and the hare of care: health AI insights from Rock Health's 2025 Consumer Adoption Survey. Rock Health. link32% of US adults reported having turned to an AI chatbot for health information in 2025, double the 16% share a year earlier.Industry consumer survey (not peer-reviewed); n=8,000 US Census-matched adults, fielded December 2025.
  9. 9.Ahrefs (2026). An analysis of AI Overview brand-visibility factors (75K brands). Ahrefs. linkAcross roughly 75,000 brands, presence in Google's AI Overviews correlated 0.66 with third-party brand mentions versus 0.22 with backlinks, mentions outweighing links roughly three to one.Vendor correlational study (not peer-reviewed); Spearman correlations across approximately 75,000 brands.
  10. 10.Sinsky C, Colligan L, Li L, Prgomet M, Reynolds S, Goeders L, Westbrook J, Tutty M, Blike G (2016). Allocation of Physician Time in Ambulatory Practice: A Time and Motion Study in 4 Specialties. Annals of Internal Medicine, 165(11): 753-760. doi:10.7326/M16-0961Direct observation of 57 physicians (430 hours, 4 specialties): during the office day physicians spent 27.0% of time on direct clinical face time and 49.2% on EHR and desk work — for every hour of direct clinical face time, nearly 2 additional hours go to EHR and desk work; the 21 physicians keeping after-hours diaries reported 1-2 hours of after-hours work each night, mostly EHR tasks.
  11. 11.Arndt BG, Beasley JW, Watkinson MD, Temte JL, Tuan WJ, Sinsky CA, Gilchrist VJ (2017). Tethered to the EHR: Primary Care Physician Workload Assessment Using EHR Event Log Data and Time-Motion Observations. Annals of Family Medicine, 15(5): 419-426. doi:10.1370/afm.2121EHR event logs of 142 family physicians over 3 years: clinicians spent 355 minutes (5.9 hours) of an 11.4-hour workday in the EHR — 269 minutes during clinic hours plus ~86 minutes (~1.4 hours) after hours ('pajama time'); clerical/administrative tasks (documentation, order entry, billing and coding, system security) accounted for 44.2% of total EHR time (157 min/day), inbox management another 23.7% (85 min/day).
  12. 12.American Medical Association (2026). 2025 AMA Prior Authorization Physician Survey. American Medical Association. linkPractices complete an average of 40 prior authorizations per physician per week, consuming 13 hours of physician/staff time weekly; 40% of physicians have staff who work exclusively on PA; 95% report PA delays access to necessary care; 79% report patients abandoning treatment due to PA; 26% report PA led to a serious adverse event; 80% say PA increases physician burnout; only 33% believe the June 2025 insurer reform pledge will make a meaningful difference.Industry survey (web-based, fielded December 2025, N=1,000 physicians: 400 primary care/600 specialists, Medscape panel; released May 13, 2026), not peer-reviewed.
  13. 13.Shanafelt TD, Dyrbye LN, Sinsky C, Hasan O, Satele D, Sloan J, West CP (2016). Relationship Between Clerical Burden and Characteristics of the Electronic Environment With Physician Burnout and Professional Satisfaction. Mayo Clinic Proceedings, 91(7): 836-848. doi:10.1016/j.mayocp.2016.05.007National survey of 6,375 physicians in active practice: 84.5% used EHRs and 82.5% used CPOE; physicians using these systems were less satisfied with time spent on clerical tasks, and CPOE use carried higher burnout risk (adjusted OR 1.29; 95% CI 1.12-1.48; P<.001) — the clerical/electronic burden itself drives burnout.
  14. 14.Tseng P, Kaplan RS, Richman BD, Shah MA, Schulman KA (2018). Administrative Costs Associated With Physician Billing and Insurance-Related Activities at an Academic Health Care System. JAMA, 319(7): 691-697. doi:10.1001/jama.2017.19148Time-driven activity-based costing at an academic system with a certified EHR: billing and insurance-related (BIR) activities cost $20.49 and 13 minutes per primary care visit (14.5% of professional revenue), $61.54 per ED visit (25.2%), $124.26 per general inpatient stay, $170.40 per ambulatory surgical procedure, and $215.10 per inpatient surgical procedure (3.1%); range 3.1%-25.2% of professional revenue.
  15. 15.Casalino LP, Nicholson S, Gans DN, Hammons T, Morra D, Karrison T, Levinson W (2009). What Does It Cost Physician Practices To Interact With Health Insurance Plans?. Health Affairs, 28(4): w533-w543. doi:10.1377/hlthaff.28.4.w533National survey of ~900 physicians and practice administrators: practices spent an average of $68,274 per physician per year interacting with health plans (physicians themselves ~142 hours/year), totaling $23-31 billion nationally — the $31 billion figure equals 6.9% of all US spending on physician and clinical services.
  16. 16.Himmelstein DU, Campbell T, Woolhandler S (2020). Health Care Administrative Costs in the United States and Canada, 2017. Annals of Internal Medicine, 172(2): 134-142. doi:10.7326/M19-2818US insurers and providers spent $812 billion on administration in 2017 — $2,497 per capita, 34.2% of national health expenditures — vs $551 per capita (17.0%) in Canada; US physicians' insurance-related costs were $465 per capita vs $87 in Canada; cutting US administrative costs to Canadian levels would have saved ~$600 billion.
  17. 17.CAQH (rebranded DataSpring) (2026). 2025 CAQH Index. CAQH/DataSpring. linkAutomating administrative transactions helped US healthcare avoid $258 billion in administrative spending (a 17% increase in cost avoidance year over year), with a remaining $21 billion annual savings opportunity from fully automating the manual and partially manual transactions that remain.Industry benchmarking report (released Feb 19, 2026; data from 600+ provider organizations and health plans covering 63% of insured lives), not peer-reviewed. CAQH rebranded as DataSpring; the 2024 Index (prior edition) reported $222B avoided and a $20B opportunity, with 70 minutes saved per visit under fully electronic workflows.
  18. 18.Kane CK (American Medical Association) (2025). Policy Research Perspectives — Physician Practice Characteristics in 2024: Private Practices Account for Less Than Half of Physicians in Most Specialties. American Medical Association. linkIn 2024 only 42.2% of physicians were in private practice (wholly physician-owned), down 18 percentage points from 60.1% in 2012 (~80,000 fewer physicians); only 35.4% held an ownership stake in their practice vs 53.2% in 2012 and ~76% in the early 1980s; the AMA names inadequate payment rates, costly resources, and burdensome regulatory/administrative requirements as the drivers.AMA survey report (Physician Practice Benchmark Survey, 2024 wave), not peer-reviewed.
  19. 19.Physicians Advocacy Institute / Avalere Health (2026). PAI-Avalere Health Report on Physician Employment Trends and Practice Acquisitions: 2018-2026. Physicians Advocacy Institute (analysis by Avalere Health). linkAs of January 1, 2026, 82.0% of US physicians (550,494) are employed by hospitals/health systems (59.7%) or other corporate entities such as private equity firms and insurers (22.3%); 63.9% of physician practices are owned by hospitals (30.6%) or corporate entities (33.2%); employed physicians up 85% (+253,000) since 2018.Industry report (PAI-commissioned Avalere analysis), not peer-reviewed; newest edition, released April 2026. The widely cited 77.6% figure is the prior (April 2024) edition, as of Jan 1, 2024.
  20. 20.American Medical Association (2025). Medicare physician pay has plummeted since 2001. Find out why.. American Medical Association. linkAdjusted for inflation in practice costs (Medicare Economic Index), Medicare physician payment effectively declined 33% from 2001 to 2025; physicians are the only Medicare providers without an automatic annual inflationary update, with current-law updates of just 0.25% (0.75% in APMs).AMA advocacy analysis (page dated April 21, 2025) based on the Medicare Economic Index, the official measure of practice-cost inflation; not peer-reviewed. MedPAC has likewise repeatedly acknowledged the payment-update gap.
  21. 21.Fleming NS, Culler SD, McCorkle R, Becker ER, Ballard DJ (2011). The Financial And Nonfinancial Costs Of Implementing Electronic Health Records In Primary Care Practices. Health Affairs, 30(3): 481-489. doi:10.1377/hlthaff.2010.0768EHR implementation cost an estimated $162,000 for an average five-physician practice (~$32,400 per physician) plus $85,500 in first-year maintenance (~$17,100 per physician); ongoing maintenance ran ~$1,425 per physician per month ($1,225 of it software licensing, hosting, and vendor support) — roughly $17,000 per physician per year.Cost levels are 2010-era (26 north Texas primary care practices); cite as an order-of-magnitude baseline for what practice software has cost independent physicians.
  22. 22.Cooper Z, Craig SV, Gaynor M, Van Reenen J (2019). The Price Ain't Right? Hospital Prices and Health Spending on the Privately Insured. The Quarterly Journal of Economics, 134(1): 51-107. doi:10.1093/qje/qjy020Using claims covering 28% of employer-sponsored-insurance enrollees: prices at monopoly hospitals are 12% higher than in markets with four or more rivals, and in 366 mergers (2007-2011) prices rose over 6% when merging hospitals were geographically close (≤5 miles) — market power, not value, drives commercial hospital prices.
  23. 23.Beaulieu ND, Dafny LS, Landon BE, Dalton JB, Kuye I, McWilliams JM (2020). Changes in Quality of Care after Hospital Mergers and Acquisitions. New England Journal of Medicine, 382(1): 51-59. doi:10.1056/NEJMsa1901383Across ~250 acquired hospitals (2009-2013 transactions), acquisition was associated with modestly worse patient experience and no significant change in 30-day readmission or mortality rates — consolidation raised prices (per Cooper et al.) without measurable quality gains.
  24. 24.Medicare Payment Advisory Commission (MedPAC) (2022). Aligning Fee-for-Service Payment Rates Across Ambulatory Settings (Chapter 6, June 2022 Report to the Congress: Medicare and the Health Care Delivery System). MedPAC. linkIn 2022 Medicare paid 105% more for a midlevel office visit in an on-campus hospital outpatient department than in a freestanding physician office, and 141% more for the first hour of chemotherapy infusion; of 169 service APCs, 57 (most frequently delivered in physician offices) could safely be paid at physician-fee-schedule rates — alignment would have cut 2019 Medicare program spending by $6.6 billion and beneficiary cost sharing by $1.7 billion.
  25. 25.Whaley CM, et al. (RAND Corporation) (2024). Prices Paid to Hospitals by Private Health Plans: Findings from Round 5.1 of an Employer-Led Transparency Initiative. RAND Corporation Research Report RRA1144-2-v2; also RAND Health Quarterly 2025;12(2):5. linkIn 2022, employers and private insurers paid on average 254% of what Medicare would have paid for the same hospital services at the same facilities — 254% for inpatient facility services, 279% for hospital outpatient facility services, and 184% for associated professional services — across >4,000 hospitals in 50 states; several states exceeded 300% of Medicare.Most recent published round as of June 2026 (Round 6 expected September 2027). Figures verified via the peer-indexed RAND Health Quarterly mirror (PMC11916091) and RAND's May 13, 2024 press release; rand.org blocks automated fetches.
  26. 26.Casalino LP, Pesko MF, Ryan AM, Mendelsohn JL, Copeland KR, Ramsay PP, Sun X, Rittenhouse DR, Shortell SM (2014). Small Primary Care Physician Practices Have Low Rates Of Preventable Hospital Admissions. Health Affairs, 33(9): 1680-1688. doi:10.1377/hlthaff.2014.0434Among 1,045 primary care practices with ≤19 physicians linked to Medicare data: practices with 1-2 physicians had 33% fewer preventable (ambulatory care-sensitive) hospital admissions than practices with 10-19 physicians; practices with 3-9 physicians had 27% fewer; physician-owned practices had fewer preventable admissions than hospital-owned practices.

26 sources, numbered by first appearance. Every entry verified 2026-06-11 against PubMed / PMC / publisher pages (214 in the full bibliography).