KPSGILL policy proposal · model regulation
Corporate control of clinical judgement
The prohibition already exists. What is missing is a list of the levers it applies to — the target, the template, the compensation formula, the code set, the staffing, the record — and a duty to keep evidence of who pulled them.
The problem
California has a statute that bars a private-equity group or hedge fund from interfering with professional judgement, voids offending contract provisions, and names an enforcer CURRENT LAW. That is a genuine advance, and it is also where most such statutes stop: the prohibition is stated at the level of principle, and the conduct it prohibits happens through mechanisms that no one has enumerated.
A management services organisation does not send a memo instructing a physician to see more patients. It sets a productivity target, ties compensation to it, controls the schedule template, owns the coding configuration, and holds the payer contract. Each step is individually defensible. Together they determine how many minutes a patient gets. A statute that prohibits “interference” without naming those levers is enforced case by case, slowly, against whoever is unlucky.
Documentary baseline: the operative California statute with its named enforcer and its void-provision rule; the corporate-practice doctrine analyses in the policy library; and the enforcement matters where control over clinical decisions was the contested question. Each carries its own primary source in the record layer.
The recommendation
Implement the statutory prohibition as a regulation that names the mechanisms, requires disclosure of the ones that are legitimate, and makes the record of who set them discoverable. Not a new prohibition — an operational definition of the one that already exists.
§ 1. Enumerated levers. For the purposes of the statute, each of the following, whether exercised directly or through a management services agreement, is control over the practice of medicine: (a) setting or adjusting the number of patient encounters expected of a licensee per unit of time; (b) setting appointment length or schedule templates that determine encounter duration; (c) conditioning compensation on encounter volume, panel size, coding intensity or referral patterns; (d) configuring or restricting the diagnosis and procedure codes available to a licensee; (e) determining the clinical staffing available to a licensee; (f) controlling the content of, or access to, the medical record; (g) approving or denying the acquisition of clinical equipment or supplies; (h) determining which payers the practice contracts with, where that determination materially restricts the patients a licensee may treat.
§ 2. Permitted business functions. Nothing in § 1 prohibits ordinary administrative services — billing operations, human resources, facilities, information technology, capital provision — where they do not determine the clinical matters enumerated in § 1.
§ 3. Disclosure. A management services agreement covering a medical practice shall be disclosed to each licensee practising under it, and shall identify, for each function in § 1 that the agreement addresses, the party holding the decision. A licensee may not waive this section.
§ 4. Attribution of record. Where a target under § 1(a)–(c) is set, the practice shall maintain a record identifying the party that set it, the basis on which it was set, and the date. The record shall be produced to the department on request.
§ 5. Clinical override. A licensee’s determination that a patient requires more time, additional testing, or a different disposition than a target or template contemplates shall not be a basis for adverse action against the licensee. Documentation of the determination shall be sufficient.
§ 6. Void provisions. A provision of any agreement inconsistent with this article is void and unenforceable, and its inclusion is itself a violation.
§ 4 is the quiet engine. Most disputes about clinical independence fail on evidence: nobody can show who set the target. A retention duty converts an unprovable allegation into a document request.
Who bears what
Physician
§ 5 gives a defensible answer to the productivity conversation, and § 3 means she can read the terms that govern her own clinical latitude before she signs.
Patient
Encounter length stops being set by a party with no duty to the patient.
Investor
Capital provision and administrative services remain expressly permitted under § 2. What ends is control of clinical variables through administrative levers.
Competition
Physician-owned practices currently compete against groups that can extract margin by compressing encounters. § 1 removes that advantage rather than subsidising the alternative.
Equity
Volume-compression models concentrate where payer mix is weakest, so the patients with the fewest options are the ones whose visits get shortest.
Burden
§§ 3 and 4 impose real documentation duties on management companies, and almost none on clinicians.
The strongest arguments against
- § 1 is so broad that ordinary practice management becomes unlawful.
- § 5 lets a physician ignore any productivity expectation.
- Disclosure of management agreements exposes commercially sensitive terms.
- This will drive capital out of underserved markets that need it.
- Enumerating levers invites avoidance through unlisted mechanisms.
Answers
- § 2 exists for that reason and is drafted to be read alongside § 1. The line is whether a function determines a clinical variable, not whether a non-physician performs it.
- It does not. It protects a documented clinical determination in a specific case. A licensee who documents nothing has no protection under § 5.
- Only to the licensees practising under it, and only as to the functions in § 1. Nothing requires publication.
- A serious concern, and the honest response is that this is an empirical question that § 4 records would let a regulator answer — which is more than anyone can do today.
- Which is why § 1 opens with “each of the following” rather than “only the following”, and why the statutory prohibition remains the operative rule.
Metrics, sunset, open questions
Metrics. Share of management agreements disclosed under § 3; adverse actions in which § 5 was invoked and their outcomes; median scheduled encounter length by ownership type; enforcement actions in which a § 4 record was the evidence.
Sunset. No sunset on §§ 1, 2 and 6, which implement a statute. §§ 3 and 4 carry a five-year review against the enforcement record they generate.
Open questions. Should § 1(h) reach payer-contracting decisions at all, given that a practice must be free to decline unsustainable contracts? Does § 4 create a discovery risk for physician-owners who set their own targets? Should hospital employment be treated identically to management-company control, or does the medical-staff structure supply a substitute protection?
Related in the record: the corporate practice doctrine in California · who controls clinical judgment · productivity targets and independence · the enforcement layer, where the operative statute and its named enforcer are recorded.