Growth has flattened — MSOs

wRVUs per FTE physician are up. Total medical revenue per FTE physician is not.

PRESENTATION — WHAT THE OPERATOR SEES

Production is climbing and the schedule reads full, but total medical revenue per FTE physician has not moved in two years and the operating margin per physician is worse than it was. Everyone in the group is working harder than they were in 2023. The board wants to know why that has not shown up in the number. It is the presentation we see most often in multi-specialty groups and MSOs, and it has five separable causes.

DIFFERENTIAL — LIKELY CAUSES, MOST LIKELY FIRST

01
Referral capture — demand the group already generated is not converting into completed encounters

A referral is a physician-generated, named-specialist, in-network demand event. If it does not become a documented completed encounter, the wRVU is never produced and the revenue never exists — no additional demand generation changes that arithmetic.

Of 103,737 referral scheduling attempts, only 36,072 (34.8%) resulted in documented completed appointments. 18,531 (17.9%) were cancelled, 4,117 (4.0%) were no-shows, and 40,377 (38.9%) had no documented status at all. This was inside a single integrated health system, where capture should be easiest.

Patel MP et al., "Closing the Referral Loop: an Analysis of Primary Care Referrals to Specialists in a Large Health System." J Gen Intern Med. 2018;33(5):715-721. One academic health system, 20 high-volume specialties, July 2015 – June 2016. Not a national rate.

NOT THIS IF — Not this if referral orders placed are falling at the same rate as completed encounters. Capture failure shows up as a widening gap between orders placed and encounters completed, not as a parallel decline in both.
02
Yield per unit of production is falling, so flat dollars against rising wRVUs is the market, not the group

Growth measured in dollars can be flat while growth measured in production is real. Each wRVU is simply worth less than it was, so more production buys the same revenue.

wRVUs per FTE physician are up 9% since 2023 while net patient revenue per provider wRVU has decreased 1% over the same period; physician compensation is up 6%. Median net patient revenue per wRVU was $76.63 in Q2 2025, against $76.83 in Q2 2024 and $82.61 in Q2 2023.

Kaufman Hall / Vizient Physician Flash Report, Q4 2025 and Q2 2025. Sample size is not disclosed in this series — quote the quarter, never "Kaufman Hall says."

NOT THIS IF — Not this if net revenue per wRVU inside the group is holding or rising by service line. Then the flat total is a volume or mix problem, not a yield problem.
03
Service-line mix has shifted underneath a fixed capacity allocation

Demand moves between service lines faster than provider capacity and space are reallocated. A group can hold total encounters flat while the encounters it holds are worth progressively less.

Behavioral health visits rose 11.4% year over year while primary care visits fell 5.6%. Ambulatory surgery centers accounted for 50.8% of ASC-eligible surgeries in 2024. Median total medical revenue per FTE physician moved differently by category in 2024: primary care $760,383 (+3.80%), nonsurgical specialties $697,712 (+0.01%), surgical specialties $687,652 (−2.74%).

Trilliant Health, 2025 Trends Shaping the Health Economy (underlying claims dataset size not disclosed); MGMA DataDive Financials & Operations, 2024 data.

NOT THIS IF — Not this if encounters and wRVUs per service line are moving in the same direction and proportion as the prior year. Mix shift is visible only when the lines are separated.
04
Access and intake capacity has been squeezed out from under the providers

Support capacity per unit of production has fallen. The provider can produce; the desk cannot book, confirm, and re-book at the rate the provider can absorb, so capacity leaks as unbooked and unconfirmed slots.

Support staff per 10,000 provider wRVUs fell 13% in two years, to 2.99 (Q2 2025). Medical group leaders' four ranked patient-access priorities for 2026 are no-shows 27%, online scheduling 24%, phone access 22%, wait times 21% — every one of them an intake problem, none of them a demand-generation problem.

Kaufman Hall Physician Flash Report, Q2 2025; MGMA Stat poll, December 9, 2025 (236 applicable responses).

NOT THIS IF — Not this if third-next-available for new-patient slots is short and phone abandonment is low. Then the desk is keeping up and the constraint is upstream.
05
A genuine demand ceiling in the service area

The group has taken the available share of the addressable population and further growth requires a new site, a new service line, or a new payer contract rather than better conversion.

Ranked last deliberately. More than a third of patients are referred to a specialist each year and specialist visits constitute more than half of outpatient visits — referral is a high-frequency event, which is what makes a 34.8% documented completion rate financially enormous and makes a true demand ceiling the rarest of these five.

Mehrotra A, Forrest CB, Lin CY. "Dropping the Baton: Specialty Referrals in the United States." Milbank Q. 2011;89(1):39-68 (abstract-level).

NOT THIS IF — Not this until documented referral completion has been measured and is high, and new-patient wait times are short. Almost nobody who believes they have hit a demand ceiling has measured either.

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.

01 · Referral capture — demand the group already generated is not converting into completed encounters

Take one closed month of referral orders placed as the denominator — orders, not inquiries. Follow each to one of four states: documented completed encounter, cancelled, no-show, no documented status. Report all four shares in the same sentence as the denominator.

CONFIRMS IF

Documented completed encounters are a minority of orders placed, and the no-documented-status bucket is large enough that the group cannot say whether the patient was seen. Patel et al. found 34.8% and 38.9% respectively inside one integrated system.

EXCLUDES IF

Completion is high and the no-documented-status bucket is small. Then the referral loop closes and the plateau is elsewhere.

02 · Yield per unit of production is falling, so flat dollars against rising wRVUs is the market, not the group

Net patient revenue per wRVU, by service line, quarter over quarter for eight quarters. Use wRVU, not total RVU — total RVU includes practice-expense and malpractice components and will not answer this question.

CONFIRMS IF

wRVUs per FTE physician rise while net revenue per wRVU falls or stays flat. That is the market pattern Kaufman Hall reports and it means effort is not the lever.

EXCLUDES IF

Net revenue per wRVU is holding or rising. Then flat dollars means flat production or an adverse mix, both of which the next two tests separate.

03 · Service-line mix has shifted underneath a fixed capacity allocation

Encounters, wRVUs, and net revenue per encounter by service line, year over year, with provider FTE and room capacity alongside each line.

CONFIRMS IF

Total encounters are flat but the distribution has moved toward lower-yield lines, or capacity is still allocated to a line whose encounter volume is declining.

EXCLUDES IF

Line-level shares are stable within a point or two. Mix is not the story.

04 · Access and intake capacity has been squeezed out from under the providers

Third-next-available appointment for new patients by service line; call abandonment rate and average speed of answer at the scheduling desk; share of encounters booked without staff intervention.

CONFIRMS IF

New-patient third-next-available is long in the same service lines whose providers report open capacity — the classic signature of a booking constraint, not a demand constraint. Note that 71% of medical groups report fewer than one in four patients using digital tools to schedule (MGMA Stat, July 2025 — response count not disclosed for that poll; treat as directional).

EXCLUDES IF

Short waits, low abandonment, and providers who report full panels. Then intake is not the binding constraint.

05 · A genuine demand ceiling in the service area

Only after the four tests above. This is demand mapping by market and service line: attributed or served population against provider supply in the service area, using HRSA Area Health Resources Files for provider density.

CONFIRMS IF

Capture is already high, waits are already short, and served share of the addressable population is high. Then growth requires a site, a line, or a contract.

EXCLUDES IF

Any of the four prior tests came back positive. Address those first — they are cheaper and they are measurable this quarter.

WHAT RESOLVES EACH

What resolves this, and how you will know it resolved

Referral capture — demand the group already generated is not converting into completed encounters	Rx 03 · patient conversion →	Instrument the referral loop so orders placed, appointments scheduled, and encounters completed are one reconciled series with a stated denominator. What does not work: adding acquisition spend against an unmeasured capture rate — it raises the numerator of a fraction whose denominator is broken.
Yield per unit of production is falling, so flat dollars against rising wRVUs is the market, not the group	Rx 04 · marketing attribution →	Net revenue per wRVU and per encounter by service line, tied to the income statement. This is a measurement and mix decision, not a marketing one — we would say so rather than sell against it.
Service-line mix has shifted underneath a fixed capacity allocation	Rx 01 · patient acquisition →	Demand mapping by market and by service line, so capacity follows demand rather than history. Sequenced after intake instrumentation, never before it.
Access and intake capacity has been squeezed out from under the providers	Rx 03 · patient conversion →	Online booking and routing, same-day capacity matching, and call-centre instrumentation. The group's own leaders already rank these first — no-shows, online scheduling, phone access, wait times were the top four access priorities for 2026.
A genuine demand ceiling in the service area	Rx 01 · patient acquisition →	Only when capture is documented and access is short. If the four prior causes are unresolved, more demand is the expensive way to find out they were unresolved.

WHAT "RESOLVED" LOOKS LIKE — Total medical revenue per FTE physician, held or grown against a declining market — with documented referral completion measured as the leading indicator

MEDIAN

$760,383 primary care · $697,712 nonsurgical specialties · $687,652 surgical specialties (MGMA DataDive Financials & Operations, 2024 data)

TOP DECILE

MGMA does not publish a decile for medical groups. It publishes quartiles by wRVU productivity: 4th-quartile multispecialty groups reported total medical revenue per FTE physician 94% greater than 1st quartile, operating expenses 54% greater, and profit per FTE physician 154% greater at $374,577 (MGMA DataDive Pro Cost and Revenue, 2016 data — use the structure, not the dollar levels).

TARGET

A distribution position, not a promise: move from below the MGMA median toward third- and fourth-quartile productivity economics. Context for what "flat" is worth — surgical median revenue per FTE physician fell 2.74% in 2024, and MGMA records two straight years of declining revenue per FTE physician among physician-owned groups. Holding flat against that is a real result and should be reported as one.

MGMA DataDive Financials & Operations (2024 data); MGMA DataDive Pro Cost and Revenue (2016); MGMA Stat, "Does your margin have breathing room?"

HOW THIS DIFFERS BY SCALE

How this differs by scale

Single site	One site, one referral desk. The referral loop can be closed by hand in a spreadsheet in a week, and usually should be before anyone buys software. The plateau here is almost always access: one phone line, one scheduler, and a third-next-available nobody has ever measured.
Group	Three to fifteen sites. Referral capture becomes an inter-site problem — a referral from one site to a specialist at another crosses a handoff that nothing owns. The plateau shows up as total flat while individual sites diverge, which is why the multi-site-variance differential usually needs reading alongside this one.
Platform	Fifty-plus providers, often mixed ownership across acquired groups. Total medical revenue per FTE physician stops being one number — it is a weighted average of several economics, and the weighted average moves when the mix of acquired sites changes, independent of anything operational. Measure same-store per-FTE separately from platform per-FTE or the plateau cannot be diagnosed at all.

OTHER PRESENTATIONS — MULTI-SPECIALTY MSO

A differential narrows the field. It does not replace the examination — that is what the six weeks are for. Every figure above is an industry reference range, not a client's numbers; those stay sealed. Sources are set out at /sources.

APPLY — 6 / QUARTER →