In states with corporate practice of medicine restrictions, a non-physician cannot own a medical practice directly. The standard workaround is two entities: a physician-owned professional corporation that owns the practice and all clinical decisions, and an entrepreneur-owned MSO that owns the brand, equipment, and non-clinical operations and charges the practice a management fee under a services agreement.
Two kinds of founders end up on this page. The non-clinician entrepreneur who wants to build a TRT, GLP-1, or concierge clinic and just discovered they may not be allowed to own one. And the physician or NP who keeps hearing "you need an MSO" from podcasts and consultants and isn't sure whether that's true or just structure for structure's sake. Both questions have clean answers once you see what the rule actually is and treat it like any regulation: a risk exposure you structure around, not a mystery.
What the corporate practice of medicine doctrine actually is
The corporate practice of medicine (CPOM) doctrine is a state-level rule (statute, case law, board policy, or some mix) holding that only licensed physicians, or entities owned by them, may own a medical practice or employ physicians to practice medicine. The premise is that a corporation answering to shareholders shouldn't be able to pressure a physician's clinical judgment. Premise aside, in a meaningful set of states it is live law, enforced by medical boards, attorneys general, and, most often, opposing counsel in a business dispute who notices your structure is defective.
Critically, CPOM exists on a spectrum, not a switch:
- Strong-CPOM states. States like California, Texas, and New York are commonly characterized as having robust doctrines with active enforcement. Non-physician ownership of a medical practice there is generally treated as off the table.
- Middle-ground states. The doctrine exists on the books but enforcement is thin, or statutory carve-outs (certain entity types, employment by hospitals or licensed facilities) soften it.
- Weak or no-CPOM states. Some states have no meaningful prohibition, and a non-physician owning a clinic entity raises no doctrine-specific issue at all.
Do not take any blog's state list as legal fact, including the one above. Enforcement postures shift, AG opinions get issued, and the details (what counts as "practicing medicine," which entity forms are allowed) vary enormously. The state classification is the first question you pay healthcare counsel to answer, and it's cheap relative to the cost of guessing wrong.
The standard answer: the friendly-PC + MSO model
When the state does restrict ownership, the industry-standard structure is two entities with a contract between them:
- The professional corporation (PC or PLLC), owned by a licensed physician, holds the medical practice. It employs or contracts the clinicians, owns the patient relationships and medical records, and makes every clinical decision.
- The management services organization (MSO), owned by the entrepreneur and any investors, owns everything non-clinical: the brand, the equipment, the lease, the non-clinical staff, the marketing engine, the billing infrastructure.
- The Management Services Agreement (MSA) connects them: the MSO provides its services to the PC for a fee, and the contract spells out exactly where the MSO's authority ends.
The physician-owned entity is often called a "friendly PC": friendly because the physician owner is aligned with the MSO through contracts, not because the MSO owns the practice (it can't, that's the whole point). The alignment is structural: a well-drafted MSA plus transfer restrictions, not a handshake.
PC/PLLC vs MSO: who owns what
| PC / PLLC (physician-owned) | MSO (entrepreneur/investor-owned) | |
|---|---|---|
| The medical practice itself | Owns it: patient relationships, clinical revenue | Never owns it |
| Clinicians | Hires, fires, supervises, credentials | May recruit & run payroll mechanics, never controls |
| Clinical protocols & treatment decisions | Exclusive authority | No authority |
| Medical records | Custodian (records belong to the practice) | Provides the systems, doesn't own the records |
| Brand, trademarks, website | Licenses them from the MSO | Owns them |
| Equipment & real estate leases | Uses under the MSA | Owns / holds the leases |
| Non-clinical staff | Typically none (usually employed by the MSO) | Employs front desk, marketing, admin |
| Marketing & advertising | Approves anything making clinical claims | Runs it |
| Billing & collections infrastructure | Fees are billed in its name | Operates the machinery |
How the money flows
Patients pay the PC. It's the medical practice, so clinical revenue lands there. The PC then pays the MSO a management fee under the MSA. Whatever remains in the PC pays clinician compensation, including the physician-owner's. The MSO's profit is the fee minus its costs; that profit is what the entrepreneur and investors actually own a claim on. How patient revenue arrives in the first place (memberships, per-visit charges, or insurance claims) is its own decision; the cash-pay vs insurance guide covers it.
Fee structures run a spectrum, and state law decides how much of it is available to you:
- Flat fee. A fixed monthly amount. Cleanest and most defensible; least aligned with growth.
- Cost-plus. The MSO's actual costs plus a defined margin. Defensible and scales with the operation.
- Percentage of revenue or collections. Most aligned with growth, but a number of states restrict percentage-based fees under fee-splitting prohibitions (rules against splitting professional fees with non-licensees). Where it's permitted, it's common; where it isn't, forcing it is a classic way these structures get attacked.
Whatever the formula, the fee generally needs to reflect fair market value for services actually rendered. A fee engineered to sweep essentially every dollar of profit out of the PC looks like disguised ownership, exactly the substance-over-form argument a board or a plaintiff's lawyer reaches for. Benchmark the fee, document what the MSO actually does, and let real profit sit in the PC. If the clinic ever bills federal healthcare programs, fee design also has to clear anti-kickback analysis.
The dividing line: what the MSO can and cannot control
Nearly every CPOM problem in a built structure traces to the same mistake: the MSO drifting across the clinical line. The classic division:
- PC only: hiring, firing, and supervising clinicians; clinical protocols and standing orders; diagnosis and treatment decisions; whether to accept or discharge a patient on clinical grounds; custody of medical records.
- MSO side: marketing and patient acquisition; scheduling infrastructure; billing and collections mechanics; real estate and equipment; IT, software, and vendor management; HR for non-clinical staff.
- Gray zone: structure it deliberately. Service pricing, adding or dropping service lines, clinician compensation design. These are business decisions with clinical fingerprints; the durable pattern is MSO proposes, PC decides, and the paper trail shows it.
The test regulators and courts apply is functional, not cosmetic: who actually controls the practice of medicine? A perfect org chart doesn't save a structure where the MSO's owner is, in substance, directing clinical operations over the physician's head.
NP-owned practices: full practice authority changes the picture
If the clinician in your plan is a nurse practitioner rather than a physician, the map redraws. In full-practice-authority (FPA) states, an NP can generally evaluate, diagnose, and prescribe without physician oversight, and in many of them own the practice entity outright. An NP founder in an FPA state may need nothing more exotic than their own PLLC.
In reduced- and restricted-practice states, the NP needs a collaborating or supervising physician, and some states also restrict who may own the professional entity, occasionally requiring physician ownership or co-ownership even when the NP delivers all the care. There, an NP-led clinic can need the same friendly-PC architecture a non-clinician would, with the NP's company playing the MSO role.
Two separate questions, both state-specific: can the NP practice independently, and can the NP own the entity? They don't always have the same answer, so verify both with counsel.
Continuity planning: the friendly PC's key-person problem
The structure's soft spot is that the entire clinical side sits in an entity owned by one person. If that physician dies, becomes disabled, loses their license, or simply walks, the MSO owns a brand, a lease, and a billing system attached to a practice it doesn't control and can't quickly replace.
The standard mitigation is a stock-transfer restriction agreement (or succession agreement) signed at formation: the physician agrees that on defined trigger events (death, disability, license action, termination of the MSA, departure), their equity transfers to a designated successor licensed physician, typically at a nominal or pre-agreed price. Pair it with a bench: know who your successor physician would be before you need one.
These agreements are themselves scrutinized in strong-CPOM states. A transfer mechanism that gives the MSO unilateral power to swap the physician at will starts to look like the MSO owns the practice in substance. Counsel's job is to make succession reliable without making it look like control.
When you don't need an MSO
Plenty of founders reading this don't need any of it. If you are a physician opening a practice in your own state with your own money and no non-physician investors, a single PC or PLLC is the whole structure. An NP founder in an FPA state that permits NP entity ownership is often in the same position. Adding an MSO to that picture buys you two tax returns, intercompany agreements, transfer-pricing discipline, and no benefit.
The MSO earns its complexity when at least one of these is true: a non-clinician owns or invests in the business; you're planning multi-state expansion (one MSO servicing a friendly PC per state is the scalable pattern); you want the management business to be separately sellable; or you want brand and equipment asset-protected away from clinical liability. If none of those apply today, don't build the structure on spec. You can layer an MSO onto an existing practice later, when a named reason shows up.
Records and systems: keep the software on the right side of the line
However the entities are drawn, one operational rule follows: medical records belong to the practice entity, and your systems should reflect the same separation your lawyers drew. The MSO's marketing team has legitimate business reasons to see appointment volume and revenue; it has no business inside charts. An EHR that hands every "admin" the whole database quietly erases the line your entire structure depends on, and an auditor or opposing counsel will notice.
Operator's note: Moonshot Clinic is built by an operator who runs this exact structure: an MSO alongside a clinician-owned practice. The platform is built to be multi-entity friendly, with role-based access that separates clinical records from business operations, and immutable audit trails showing who accessed what, when. The business side gets its numbers; the charts stay with the practice. If you're mapping out the clinical model too, start with the companion guide on how to start a GLP-1 clinic or see how concierge practices run on Moonshot.
This guide is educational content, not legal advice. Corporate practice of medicine doctrine is state-specific, enforcement varies widely and shifts over time, and the entity, fee, and succession details that make these structures work are fact-dependent. State characterizations here reflect how states are commonly described in the industry, not legal conclusions. Engage experienced healthcare counsel in the state where you'll operate before forming entities, signing a management services agreement, or taking investment.