Is your compensation model rewarding the wrong doctor?

Every practice eventually faces the same question: How should we divide what we earn? Fee-for-service collections, ancillary income, care-management fees, quality bonuses and shared savings all have to be allocated somehow, and the temptation is to find one clean number that settles the argument. Revenue per patient is the number practices reach for most often. It is easy to pull from the practice management system, easy to display and easy to defend as objective.

It is also the wrong instrument for the job. Revenue per patient is a serviceable operational statistic but a poor allocation basis, because most of what it measures has little to do with the physician it is being used to judge. A practice that builds its revenue split, profit split or savings distribution on it will move money in directions the partnership would not choose if it understood what the number contained.

What the number actually contains

Collections per patient is the product of three things: the work performed, how that work was documented and coded, and what the payer behind the patient agreed to pay. Only the first is clinical. The other two are administrative and contractual, and in most practices, they account for the bulk of the variation between patient panels.

So, when a practice divides money by revenue per patient, it believes it is rewarding productivity. What it is actually rewarding is a blend of contract quality, panel composition and billing-system attribution logic — factors that vary enormously across a partnership and that almost no individual physician controls.

Payer mix comes first

The largest driver of collections per encounter is who pays the bill.

Identical work, identically documented and delivered with identical skill, is reimbursed at materially different rates depending on the contract behind the patient. Commercial plans, traditional Medicare, Medicare Advantage under delegated or capitated arrangements, Medicaid, and self-pay can differ by a factor of 2 or more for the same service. Regional plan variation and contract carve-outs add further noise. So does the share of revenue that arrives as per-member-per-month care-management fees rather than claims. Add it all up, and revenue per patient becomes largely a report on the practice's contract portfolio.

Panels are assembled over years through practice marketing, referral relationships, employer and hospital arrangements, network participation and the ordinary impact of geography and timing. A split that pays more for higher collections per patient is paying for how contracts happened to distribute across panels — a practice-level outcome — as though it were individual effort.

Panel composition is not performance

The second driver is who is in the panel.

A panel weighted toward complex chronic disease produces higher collections per encounter: higher-level visits, more frequent visits, more ancillaries and more billable care management. A panel weighted toward healthy adults and preventive care produces less. Neither reflects diligence. Panels differ because of age distribution, community demographics, referral patterns and how patients were assigned when they entered the practice.

More importantly, higher collections in a complex panel arrive attached to higher costs — more staff time, more refills and prior authorizations and more coordination between encounters. Longer visits mean fewer patients per day and higher revenue per patient, while overhead stays fixed for every provider. The practice collects more per patient and spends more per patient, and the metric reports only one side of that. A panel with lower collections per encounter may consume proportionately less support and contribute more margin per hour of capacity used. Used as an allocation basis, revenue per patient credits illness burden as if it were efficiency.

Attribution artifacts

Much of what shows up in the figure reflects how the billing system assigns credit rather than who performed the work.

Ancillary revenue provides the clearest example: Revenue from in-office procedures, injections, vaccines and point-of-care testing is credited to the ordering physician, even though the practice supplies the staff, equipment and infrastructure to perform those services — and bears the associated costs. Because billing attribution does not necessarily distinguish between the clinician’s role in ordering a service and the staff, equipment, overhead and operational resources required to deliver it, assigning all related revenue to the ordering clinician can overstate individual contribution and understate the practice-level investment required to provide those services.

Shared savings is the worst case

Allocating shared savings by revenue per patient is the least defensible application, because savings are generated by a different mechanism than revenue.

Savings arise from reduced total cost of care: avoided admissions, fewer emergency visits, lower post-acute utilization, tighter referral patterns, better medication management and timely follow-up after discharge. These savings come from the full range of care activities, and most of that work never surfaces as revenue per patient — the phone calls, portal messages and care-management outreach that prevent the next admission generate no billable encounter at all.

Attribution compounds it. Savings, quality bonuses and risk-adjustment payments arrive months or years after the work and are assigned by methodologies few partners can reconstruct. Attribution follows plurality of primary care services, benchmarks reflect historical regional spending, and risk scores reflect documented diagnoses rather than clinical labor. When those dollars land in one column because of attribution logic and are then divided by a revenue-per-patient formula, the practice has stacked two unrelated accounting conventions and called the result performance. Savings should be distributed on the drivers of savings: panel size and member months, risk-adjusted attribution, quality performance and participation in the care-management work that produces the result.

The cost side is missing

Revenue per patient says nothing about what each physician costs the practice.

Support intensity varies: medical assistant time per encounter, nurse triage, documentation or scribe support, scheduling accommodation, front-desk handling, dedicated staff, specialized supplies, physician-specific equipment and space. Those costs are real, they differ measurably and none of them appear in a collections figure.

Fixed overhead is the larger blind spot. Practice expenses barely move with modest differences in hours or volume. Rent, the electronic medical record contract, core staffing, billing infrastructure, malpractice coverage, utilities, accounting, compliance and management costs remain in place. Capacity is built for a full schedule; when a schedule contracts, the capacity does not disappear, and the practice absorbs it. A physician who sees more patients per day at a lower average revenue per visit is often contributing more net profit than one with fewer, higher-revenue visits — because fixed overhead is spread across more encounters. The per-visit figure can point in exactly the opposite direction from the practice's own economics.

Growth and behavior

Panel growth is the practice's future revenue and a meaningful share of its enterprise value. New-patient acquisition, retention and the absorption of unassigned or transferred patients are genuine contributions. A formula built on average revenue per encounter values them at zero, and sometimes as a penalty, because new patients are expensive to onboard and slow to become profitable.

Any allocation methodology also shapes behavior. It can push clinicians toward whatever the metric rewards, at the expense of essential work the metric doesn't measure. Practices rarely intend those messages but they surface anyway, in scheduling friction with quiet disputes over who takes what.

Allocate profit, not revenue

Revenue per patient belongs on an operational dashboard, not as the primary basis for compensation or profit allocation. It can be a useful diagnostic measure highlighting documentation opportunities, contract performance issues, changes in patient mix, or gaps in service capture. But as an allocation methodology, it can confuse differences in circumstances, contracts and measurement rules with individual performance.

The distinction to hold onto is that profit is not revenue. Profit is what remains after the practice pays for the infrastructure that made the revenue possible — space, staff, technology, billing, insurance, compliance, management — and almost none of that cost structure varies with which payer happened to be behind which patient. A partnership dividing profit is dividing a residual, and a residual can be allocated fairly only if both sides of the equation are visible. Revenue per patient shows one side and hides the larger one.

A practice should stop looking for a single number that settles the split and build an allocation that mirrors how it actually earns money:

Contribution = collected revenue − directly attributable expenses − allocated share of fixed overhead

Measure collected revenue by physician. Subtract the costs directly attributable to each — dedicated staff, specialized supplies, physician-specific equipment, space and genuinely variable costs that rise with volume. Allocate fixed overhead on a stated, consistent principle: equally among full partners, by clinical full-time equivalent, by square footage or a blended method. Then divide what remains.

Keep the streams distinct, since fee-for-service production, ancillary income, care-management fees and shared savings are generated by different mechanisms. And where panels differ materially in payer mix, comparisons should be normalized using measures such as work relative value units, visit volume or risk-adjusted panel size, so that differences in reimbursement are not mistaken for differences in current physician effort or performance.

None of this prevents a practice from rewarding leadership, governance, mentorship, administrative work or a negotiated schedule. Those are legitimate decisions, and they can be funded openly from a defined line item where every partner can see them. The failure mode is embedding such judgments inside a metric that looks objective but is not.

A partnership can survive disagreement about how to divide profit. It has a harder time surviving the discovery that the formula everyone signed was measuring something other than what it claimed.

Robert Resnik, M.D., MBA, is a board-certified internal medicine physician practicing in Cary, North Carolina. He earned his medical degree from Eastern Virginia Medical School and completed his residency at East Carolina University. He also holds an MBA from Duke University.