Corporate practice of medicine by state for telehealth operators

Search for a list of corporate practice of medicine states and the results disagree with each other. One law firm's fifty-state survey puts the doctrine in force in a little over half the states, another counts fewer, and the trade press rounds to whatever number fits the headline. The disagreement is not carelessness. The doctrine is a spectrum rather than a switch. It lives in a mix of statutes, medical board rules, attorney general opinions and court decisions that are in some cases a century old, and there is no official register that says which states have it. An operator planning a national online clinic does not need the count. The operator needs to know what the rule does, which kind of state each target market is, and why one legal structure satisfies all of them.

What the rule stops, and what it does not

The corporate practice of medicine doctrine says that a business entity not owned by licensed physicians may not practice medicine, and in most versions may not employ physicians to practice it on the entity's behalf. The stated purpose is to keep a layperson's commercial interest out of clinical decisions. California's version is unusually blunt. Business and Professions Code section 2400 states that "corporations and other artificial legal entities shall have no professional rights, privileges, or powers," with a narrow exception for licensed charitable institutions and clinics that employ physicians on salary and do not charge patients for the professional service. A general stock corporation in California cannot hold a medical licence, cannot bill for a physician's work as its own, and cannot tell the physician what to prescribe.

The doctrine does not stop a non-physician from owning a business that sells services to a medical practice. It does not stop that business from owning the brand, the software, the marketing budget or the patient-facing storefront, and it does not stop the business from being profitable. What it stops is lay ownership of the clinical entity and lay control of clinical judgment: who is hired as a clinician, what the protocols say, whether a given prescription is written, what the chart records. The doctrine draws a line between the business of a clinic and the practice inside it. Everything on the business side is open to a non-physician owner.

That distinction is easy to state and easy to lose in practice. A founder who sets the prescribing threshold for a weight-loss program, or who overrides a clinician's refusal to prescribe, has crossed the line even if the entity chart is perfect. Boards look at who actually decides.

Three kinds of states

Sort the states by how hard the line is enforced and three tiers appear. In the strong tier, a statute or board rule states the prohibition outright, the board or attorney general has enforced it, and lay ownership of a practice is a licensing violation for the physician and often an unlicensed-practice offence for the entity. California sits here on the strength of section 2400 and the Medical Board's long enforcement record. Texas and New York are usually grouped with California in the law-firm surveys, each reaching a similar result through its own statutes and board rules.

In the limited tier, the doctrine exists, usually through an old court decision or a general professional-corporation statute, but enforcement is rare or the state has carved out broad exceptions for hospitals, nonprofits, licensed facilities, or any entity that leaves clinical judgment alone. The structure still matters in these states, but the risk concentrates in fee arrangements and control terms rather than in bare ownership.

In the third tier the state has no doctrine, or has abolished it, and a non-physician can own the entity that employs the physicians. Some of these states still restrict fee splitting, still require a professional entity for certain licence types, and still police who directs clinical decisions. So "no doctrine" describes the ownership rule and not the whole picture.

Which tier a state belongs to is a judgment call at the edges, which is why the published counts disagree. Two states with nearly identical statutes can sit in different tiers because one board enforces and the other does not. Legislatures also move, and bills that tighten the doctrine in response to private-equity and telehealth ownership have been introduced in several states in the last two years, so any list more than a few months old is suspect. The per-state notes on the state index track the current position.

Why an online clinic hits this rule in most states at once

A brick-and-mortar practice has one state to worry about. An online clinic is judged by where the patient sits, not where the company is incorporated. A patient in California is treated under California law, by a California-licensed clinician, through an entity that satisfies California's doctrine, even when the operating company is a Delaware LLC run from Miami. Open to patients in twenty states and twenty doctrines apply from the first day of marketing.

Two consequences follow. First, the strictest state in the footprint sets the floor. There is no practical way to run a strong-tier structure for California patients and a loose one for everyone else. The clinical entity, the provider contracts and the management agreement are the same documents in every state, and a board that finds a percentage fee or a lay-controlled protocol in one state's file has found it for all of them. Second, reading the doctrine state by state before choosing a structure is the wrong order. The operator ends up designing to the average and then rebuilding for the outliers, and the outliers include the largest patient markets in the country.

The strictest state in the footprint sets the floor, and the floor is cheaper to build once than to rebuild the month a California campaign goes live.

The structure that works everywhere

The structure that satisfies every tier is two entities bound by contracts. A professional corporation, or a professional LLC where the state uses that form, is owned by a licensed physician and does everything clinical: it employs or contracts the providers, holds the protocols, owns the medical records and issues the prescriptions. A management services organization, the MSO, is owned by the founders and investors and does everything else: brand, software, marketing, billing, payroll for non-clinical staff, working capital. The MSO sells those services to the PC under a management services agreement and charges a fee for them.

"Friendly" describes the physician owner's relationship with the MSO. A stock-transfer restriction or succession agreement keeps the PC from walking away if the physician retires, dies or falls out with the business, and lets the MSO designate a replacement licensed owner. The physician keeps genuine authority over clinical matters; the MSO keeps continuity. The friendly PC glossary entry and the longer piece on the MSO and the friendly PC walk through the documents one by one.

The reason it works in every tier is that it is built for the hardest one. In a strong-doctrine state, the arrangement respects the letter of the rule because the practising entity is physician-owned and clinical decisions are reserved to it in writing. In a limited-doctrine state, the same documents answer the narrower questions about control and fees. In a state with no doctrine the structure is more than the law requires, and the extra cost is small, because the operator has to stand up a clinical entity, provider contracts and a records custodian regardless. Building the strong-tier version once is cheaper than building a loose version and rebuilding it later.

Where the structure fails is in substance rather than on paper. A board looks past the entity chart to who decides. If the MSO's founder sets prescribing thresholds, vetoes a clinician's refusal, or ties the physician owner's pay to prescriptions written, the PC is a shell and the strong-tier states treat it as one. The management agreement has to reserve clinical decisions to the PC, and the day-to-day operation has to match the agreement.

Fee splitting is a different rule

Operators often collapse two rules into one. The corporate practice doctrine is about ownership and control. Fee-splitting rules are about money: a physician may not share professional fees with someone who did not perform the service, and in many states a management fee set as a percentage of collections counts as sharing. The two rules overlap in effect but differ in reach. A state with no ownership doctrine can still prohibit percentage fees, and a strong-doctrine state may tolerate a percentage in some settings and not in others.

The practical answer is the same under both rules. Set the management fee as a flat amount, or a cost-plus amount, that a third party would recognize as fair market value for the services actually delivered, and keep a record of how the number was reached. Percentage-of-revenue fees are the defect law firms flag most often in inherited structures, and the hardest to fix later, because the fix changes the economics of the deal after the deal is done. The federal picture points the same way. In 2025 the HHS Office of Inspector General issued Advisory Opinion 25-03 approving a telehealth arrangement involving multiple MSOs, and the law-firm summaries of the opinion stress that the fees fit an anti-kickback safe harbor. Most cash-pay clinics never bill a federal program, so the federal statute is rarely the live issue, but it is the template that state boards and state anti-kickback laws borrow from.

The same logic reaches the physician side. A medical director paid a percentage of revenue, or a bonus per prescription, has a fee-splitting problem and a corporate practice problem at once, because the payment both shares fees and gives the payer a lever on clinical judgment. A flat retainer for defined governance work avoids both.

Which states to open first

With the structure settled, the order of state expansion becomes a marketing and licensing question rather than a legal one. Three inputs decide it. Where the patients are for the vertical, since weight-loss demand, hormone therapy demand and hair-loss demand do not follow the same map. Where the provider bench is already licensed, since a clinician licensed in eight states covers eight states on the first day and adding a state through the Interstate Medical Licensure Compact takes weeks rather than days. And where the state-specific prescribing rules suit the intake model the brand plans to run, which is a separate question from the corporate rule and gets its own treatment.

The tempting shortcut is to open in the loose states first and deal with California later. That works only if the structure was built to the strong tier from the start. If it was, California is a licence and a marketing budget away. If it was not, California means redrafting the management agreement, re-papering the provider contracts and possibly moving the medical records to a new custodian while patients are mid-treatment. That migration costs far more than the original drafting would have, and most of the cost lands on operations rather than on the legal budget.

This is the order in which Tessic Health sets a client up. The MSO and friendly-PC documents are drafted for the client's ownership before any state goes live, the client's clinical entity contracts providers licensed in all fifty states, and the client chooses the expansion order by demand and unit economics rather than by doctrine. The brand, the entity, the patients, the records and the data belong to the client, so the structure travels with the client if the relationship ends.

Questions operators ask

Does incorporating in a state with no doctrine avoid the rule elsewhere? No. The doctrine attaches to where the patient is treated, and the entity that practises on a California patient must satisfy California. The state of incorporation matters for tax and governance, not for this.

Can the founder be the physician who owns the PC? Yes, if the founder holds a licence in the state or states where the PC practises. Several states require the owner to hold that state's licence in particular, which is why a national footprint often ends up with more than one professional entity, each owned by an appropriately licensed physician and each managed by the same MSO. A physician founder who also owns the MSO should expect the fee and control terms to be examined more closely, not less, because the two hats make the separation harder to demonstrate.

Can a nurse practitioner own the clinical entity? In some states, depending on the state's practice-authority rules and its professional-entity statute. Where an NP can practise independently and the professional-entity statute allows NP ownership, an NP-owned PLLC is a workable clinical entity for non-physician services. Where physician collaboration is required, the entity needs a collaborating physician arrangement on top of the ownership question. The answer is state by state and the state index is the place to check it.

Is a medical director enough? No. A medical director is a clinical-governance role inside the PC and does not change who owns the PC. An MSO that hires a medical director and calls itself compliant has the roles confused; the doctrine is about ownership and control of the practising entity, and a governance contract does not create one.

What happens if the physician owner leaves? That is what the succession agreement is for. It restricts transfer of the PC's shares, names the conditions on which the MSO may designate a replacement owner, and sets the price, usually nominal, at which the shares move. An operator inheriting a structure should read this document first. It is the one most often missing.