DX / Differentials / MULTI-SPECIALTY MSO

D-01 — PATIENT ACQUISITION COST · MULTI-SPECIALTY MSO

Cost per new patient is climbing — MSOs

Cost per encounter is rising — and the marketing line is the one number nobody publishes

DIFFERENTIAL — LIKELY CAUSES, MOST LIKELY FIRST

1. 01

   Operating cost per FTE physician is compounding far faster than revenue per FTE physician — and it is not marketing

   The per-encounter cost line rises because the fixed cost base per physician rises. Almost all of the increase sits in support staff cost and general operating cost, not in demand generation.

   Median total operating cost per FTE physician rose from $620,098 to $801,938 — a 29.3% increase — over five years, while total medical revenue per FTE physician rose 16.6% and US CPI rose 5.2% over the same period. Median total support staff cost rose 19.3%; median total general operating cost rose 32.7%. Over the longer 2011–2024 window, CPI rose 39.5% while operating cost per FTE physician rose 71.6% at physician-owned multispecialty groups and 83.3% at hospital- or IDS-owned groups.

   MGMA DataDive Cost and Revenue, via MGMA, "Data Mine: The administrative burden of operating a medical group" (David N. Gans); MGMA Stat, "Does your margin have breathing room?" (2011–2024 series).

   NOT THIS IF — Not this if operating cost per FTE physician is flat or tracking CPI. Then the per-encounter cost rise is a denominator problem — falling encounters against a fixed base — and belongs to the plateaued-growth differential.

2. 02

   The subsidy per physician FTE is the real cost line, and it is structural

   Groups are not usually losing money per encounter at the margin. They are carrying a fixed investment per physician that no encounter volume at current yield can retire.

   Median investment (subsidy) per physician FTE was $315,358 in Q4 2025 and $317,409 in Q2 2025, against $299,937 in Q2 2024 and $294,112 in Q2 2023. Median total direct expense per provider FTE reached $659,025 in Q2 2025, up 4% from $631,108. At system-affiliated medical groups, median loss per physician exceeds $249,000 — median total revenue per physician $719,901 against median total expenses per physician $1,036,238. Median total revenue per physician is up 9.1% versus pre-pandemic; median total expense per physician is up 26.5% over the same period.

   Kaufman Hall / Vizient Physician Flash Report, Q4 2025 and Q2 2025 (quarterly series — never quote one quarter as an annual figure); AMGA 2023 Medical Group Operations and Finance Survey, 5,700+ clinics.

   NOT THIS IF — Not this at a physician-owned independent group with no system subsidy line. The construct only exists where a parent is funding the practice.

3. 03

   Provider mix and compensation-to-production drift

   Cost per unit of production moves with who produces the unit. APP compensation has risen far faster than physician compensation on a per-FTE-physician basis, and the APP share of the provider workforce is contested between the two sources that measure it.

   Median NPP compensation and benefits per FTE physician rose 52.1% over five years against 10.2% for physician compensation and benefits per FTE physician (MGMA). APP compensation rose 19.39% between 2020 and 2024; median surgical specialist total compensation reached nearly $585,000 in 2024, up 5.57%. On workforce share, AMGA reports APC utilization rising from 36.8% (2020) to 45.9% (2024) while Kaufman Hall reports APPs at 40.7% of the workforce in Q4 2025 and 39.0% of provider FTEs in Q2 2025 — both named, both dated, roughly six points apart, almost certainly a denominator difference. Record both. Do not average.

   MGMA DataDive Cost and Revenue; MGMA DataDive Provider Compensation, 2025 report (220,000+ physicians and APPs); AMGA 2024 Medical Group Operations and Finance Survey (7,500+ clinics, 31,000+ providers); Kaufman Hall / Vizient Physician Flash Report Q4 2025 and Q2 2025.

   NOT THIS IF — Not this if the provider-mix ratio has been stable for three years. Comp-to-production drift needs a changing mix or a changing plan to produce a changing cost per encounter.

4. 04

   Revenue-cycle loss — the cost is per collected dollar, not per encounter

   An encounter that is delivered and denied costs everything it cost to deliver and yields nothing. That reads as rising cost per encounter when it is a collection failure.

   Controllable denials represent 17.1% of total denials. "Controllable" is AMGA's term and it is one of very few revenue-cycle benchmarks published anywhere for medical groups.

   AMGA 2023 Medical Group Operations and Finance Survey.

   NOT THIS IF — Not this if the net collection rate and days in A/R are stable and the controllable share of denials is low. Use the HFMA MAP Keys definitions — the values matter less than everyone using the same definition.

5. 05

   Actual marketing cost per encounter — unmeasured, and unbenchmarkable from public data

   The cost that the phrase "rising acquisition cost" usually means is the one nobody can compare. It can be built first-party; it cannot be sourced.

   No figure. Marketing spend as a share of revenue for medical groups sits behind MGMA membership and was not obtained. MGMA's public cost taxonomy carries a general operating cost line with no marketing breakout, and no free source publishes net revenue per encounter or encounters per FTE physician either — NAMCS, which used to supply ambulatory visit volume, collected no visit data from 2022 to 2024 after its redesign. Cost per encounter therefore cannot be honestly derived for medical groups from public data, and any vendor quoting an industry cost per encounter for medical groups is quoting something they cannot have.

   Stated as a gap: MGMA DataDive Financials & Operations (paywalled); CDC/NCHS NAMCS redesign. This differential names the absence rather than filling it.

   NOT THIS IF — This one is never excluded — it is the standing condition. The group can and should build the figure internally; it simply has nothing external to compare it against yet.

HOW TO TELL THEM APART

How to tell these apart in your own numbers

Each of these is a measurement you can run yourself, without us.

WHAT RESOLVES EACH

What resolves this, and how you will know it resolved

Operating cost per FTE physician is compounding far faster than revenue per FTE physician — and it is not marketing	Rx 04 · marketing attribution →	Cost per FTE physician and per wRVU tied to the income statement, monthly. This is a cost-structure problem and a marketing engagement does not fix it — we would say so. Named here because the reporting layer is what makes the other four causes separable.
The subsidy per physician FTE is the real cost line, and it is structural	Rx 04 · marketing attribution →	CFO- and board-ready monthly on a stated definition, quarter over quarter. The commonest failure is a subsidy number that changes definition between quarters and is then read as a trend.
Provider mix and compensation-to-production drift	Rx 05 · preparing for sale →	Comp-to-production is a compliance topic before it is an economics topic. It belongs with the diligence-grade documentation work, not with the growth work. Where a valuation firm holds the MGMA and SullivanCotter subscriptions, they are the right party — we would point there rather than substitute for it.
Revenue-cycle loss — the cost is per collected dollar, not per encounter	Rx 03 · patient conversion →	Front-desk feedback loop where eligibility, authorization, and registration errors originate — a large share of controllable denials are created at intake, not in billing.
Actual marketing cost per encounter — unmeasured, and unbenchmarkable from public data	Rx 04 · marketing attribution →	Build the first-party figure with the arithmetic shown and the definition fixed, so the group's own trend becomes the benchmark. What does not work: buying an industry cost-per-encounter comparison. It does not exist, and a vendor who produces one has invented it.

HOW THIS DIFFERS BY SCALE

How this differs by scale

Single site	One site. Cost per FTE physician is legible directly off the P&L and the biggest single lever is usually the denial and eligibility loop at the front desk. There is no subsidy line to analyse — the owner is the subsidy.
Group	Three to fifteen sites. Cost per FTE physician diverges by site well before anyone notices, because shared services are allocated rather than measured. Allocation method changes look like cost changes; fix the allocation before diagnosing the cost.
Platform	Fifty-plus providers, mixed ownership. Hospital- and IDS-owned units carry a structurally different cost curve from physician-owned units — MGMA measured operating cost growth of 15.8% (2021) and 19.9% (2022) at hospital/IDS-owned groups against 7.3% and 7.6% at physician-owned. A platform blending both will show a cost trend that belongs to neither. Report them separately, and report cost per new encounter by service line inside each rather than one blended figure across both.

OTHER PRESENTATIONS — MULTI-SPECIALTY MSO

• wRVUs per FTE physician are up. Total medical revenue per FTE physician is not.
• Referrals are being placed. Nobody can say whether the patient was seen.
• Same staffing ratio. Sevenfold different economics.
• Growth has to read as referral capture, not as acquisitions