Health systems hold a large and growing number of arrangements with physicians: employment agreements, professional services agreements, medical directorships, call coverage, space and equipment leases, recruitment agreements and various forms of incentive arrangement. Each is negotiated with attention to the applicable compliance frameworks, and the terms are usually defensible. The difficulty is what happens over the following three years.
Physician contracting is unusual in that the same counterparty frequently holds several distinct agreements with the organization, entered at different times by different departments. A hospital may have an employment agreement with a physician, a directorship at one facility, call coverage at another and a lease for office space, with no single record that shows all four together.
This article looks at the administrative failures that generate most of the exposure, and at what a contract system needs to hold to prevent them. The audit-facing view is covered in stark-anti-kickback-hipaa-clm-audit-checklist.html.
The administrative failures that create exposure
These are ordinary contract management failures. They matter more here because the applicable frameworks are technical and a lapse in documentation is itself the problem.
- Expiry without renewal, the agreement term ends, the service continues and payments continue, leaving a period with no signed writing covering the arrangement
- Undocumented compensation change, a rate is adjusted by amendment or by practice without contemporaneous valuation support for the new amount
- Unevidenced duties, medical direction hours, committee participation or administrative deliverables described in the agreement with no records showing they occurred
- Fragmented inventory, arrangements held in departmental files so the aggregate position with a physician cannot be assembled on request
- Stale valuation support, a fair market value opinion obtained at inception and never refreshed across multiple renewals as market compensation moved
Expiry is worth separating from the rest because it is the most common finding and the least defensible. There is no argument to be made about a period during which no signed agreement existed. Prevention is entirely mechanical: a tracked expiry date, an owner, a lead time sufficient for renewal approval, and an escalation that reaches someone with authority before the date passes rather than after. Renewal lead times are the specific control, and they are usually set too short. A thirty-day reminder does not leave room to refresh valuation support, confirm the continuing need and route the approval, so the renewal is signed after the date has passed or not at all.
Reviewing a single physician arrangement in isolation can produce a defensible conclusion that does not survive aggregation. Total compensation across an employment agreement, a directorship and a call coverage arrangement may look different from each component. Any review process should begin by assembling every arrangement with the physician and any entity they control, which requires a contract inventory keyed to the individual rather than to the department that signed.
What the contract record should hold
Beyond the executed document, five elements make an arrangement defensible without a reconstruction exercise.
- Valuation support linked to the term. The fair market value analysis relied on, dated, tied to the compensation it supports and refreshed whenever the compensation changes.
- Commercial reasonableness rationale. A contemporaneous record of why the arrangement was needed, what business purpose it serves and why the scope is appropriate.
- Performance evidence. Time logs, meeting records or deliverables for duties the agreement requires, stored against the arrangement rather than in a departmental folder.
- Aggregate view by physician. Every arrangement with the individual and related entities, with total compensation visible across all of them.
- Amendment and approval history. Who approved each change, when, and against what valuation support, so the decision chain can be reconstructed years later.
A five-step roadmap to controlled physician contracting
Step 1, Build the complete inventory
Locate every arrangement with a physician or physician-controlled entity across employment, services, directorships, leases and recruitment. Reconcile against payroll and accounts payable to find arrangements that exist as payments without a contract record on file.
Step 2, Key the inventory to the individual
Restructure the records so all arrangements with one physician resolve to a single view, including related entities. Department-keyed filing is the reason aggregate questions cannot be answered, and it is a data model problem rather than a policy problem.
Step 3, Attach the supporting file to the agreement
Link valuation support, commercial reasonableness rationale and approval records to the contract record itself. A supporting file held separately by the compliance team will drift from the agreement population within a year.
Step 4, Instrument expiry and review dates
Set renewal lead times long enough to complete valuation refresh and approval before expiry. Escalate to a named owner with authority, and report on arrangements approaching expiry weekly rather than monthly.
Step 5, Run a periodic aggregate review
On a defined cycle, review total compensation and arrangement count by physician rather than by agreement. This surfaces the accumulation that individual reviews cannot see, and it is the review most likely to find something worth acting on.
See physician arrangements in one aggregated view
Walk through inventory by individual, linked valuation support and expiry tracking with escalation.
Why this is a system problem rather than a diligence problem
Compliance teams in health systems are generally well informed about the applicable frameworks and careful in reviewing individual arrangements at the point of approval. The failures described here occur after approval, in the ordinary administration of agreements that were correctly constituted when signed. That distinction matters because it determines what fixes the problem. More training on the requirements does not prevent an agreement from expiring unnoticed, and a more rigorous approval process does not create the aggregate view that individual review lacks. What prevents these failures is contract infrastructure: a complete inventory keyed to the physician, supporting documentation stored with the agreement rather than beside it, expiry and review dates instrumented with owners and escalation, and a change history that records what was approved and on what basis. This is the same infrastructure any organization needs for high-consequence agreements, and the healthcare specificity lies in what is tracked rather than in how. It is also worth noting that the effort is front-loaded. Building the inventory and keying it to individuals is the hard part, and it is usually harder than expected because arrangements are distributed across departments that each believe theirs is the complete set. Once built, maintaining it is routine, and the aggregate review that was previously impractical becomes a scheduled report.
Failure modes and controls
The table pairs each administrative failure with the control that prevents it, which is generally a system capability rather than a policy statement.
| Failure | Control | Evidence produced |
|---|---|---|
| Agreement expires, payments continue | Tracked expiry with escalation | Continuous signed coverage |
| Compensation changed without support | Amendment blocked pending valuation | Dated valuation per compensation level |
| Duties not evidenced | Deliverable tracking on the record | Time logs and meeting records |
| Arrangements not aggregated | Inventory keyed to the individual | Total compensation by physician |
| Valuation support goes stale | Refresh trigger on renewal | Current analysis at each term |
The compliance frameworks governing physician arrangements are demanding but stable, and most health systems understand them well. What generates findings is administration: agreements that lapse, changes made without refreshed support, duties that are not evidenced and an inventory that cannot be aggregated. Those are contract system capabilities, not knowledge gaps. Building the inventory keyed to the individual is the difficult first step, and almost everything else follows from having done it. The second step, attaching the supporting file to the agreement rather than holding it separately, costs very little and prevents the drift that otherwise reappears within a year.
For related guidance, see The health system guide to physician agreement controls, Physician and provider agreement tracking for health systems and GPO and supply contract price compliance tracking.
