The programs that will struggle most through 2026 are not the ones that bought the wrong machine in 2021 or 2022. The machines were mostly fine. The programs that will struggle are the ones that signed the wrong deal around the machine, against an assumption set that no longer exists.

Here is how those deals happened. Capital was approved. The vendor walked the committee through the proforma. Volume assumptions were favorable. Reimbursement assumptions were favorable. The program signed. The machine was probably the right machine, and the clinical case for it was probably solid. Neither of those is where the problem lives.

The vintage matters for a mechanical reason. A deal signed in 2021 or 2022 was the last one priced entirely against the old environment: the old delivery codes, the old rate structure, the 2022 price list. Then the lifecycle ran. Twelve to eighteen months from signature to a commissioned machine, and a year of warranty after that. Which means 2026 is the first or second year the program has paid the service line out of pocket. The escalator started compounding in the same year the revenue assumptions underneath it died. That is not bad luck. That is the arithmetic of every deal signed before the 2026 environment was visible.

The problem is the structure underneath the machine. I spent twenty years building the vendor side of these proformas, and the assumptions in them are chosen to get to yes. Five strands hold a capital deal together, and 2026 has already cut through several of them.

The pricing baseline. The negotiation set your price against the 2022 list and the discount that list was built to absorb. That baseline never gets revisited, and every downstream number in the deal keys off it.

The proforma volume mix. The revenue projection was built on delivery codes that no longer exist. The January overhaul retired the old IMRT and 3D delivery codes into complexity tiers, and payers are already working the top tier down. A proforma that penciled treatment revenue in 2022 code language stopped being a forecast in January. It is a historical document now.

The reimbursement line. If the proforma carried a rate assumption at all, it assumed stability. What actually happened is a 2.5% conversion factor increase that applies to 2026 only and expires December 31, with 2027 statutory updates too small to cover the hole unless Congress acts again. And the 2027 proposed rules are arriving now. The hospital outpatient rule dropped July 2, the fee schedule rule is due within weeks, and the finals land by November 1. No deal signed in 2022 modeled any of this. That is the revenue base your fixed payments sit on anyway.

The service escalator. The operating cost trajectory was set with an escalator indexed to a number that now compounds against a smaller revenue base every year. This strand is where the recoverable money usually is. One service agreement I reviewed this spring ran $4.4 million a year with roughly $500,000 a year of room in it, and the escalator was doing most of that damage on its own.

The bundle and the supervision model. The software bundle locked in vendor relationships the program would negotiate differently if it were buying the same package today. The supervision assumptions under the staffing plan were written before the rule makers started revisiting coverage requirements.

None of that is the equipment. All of it is the work that should have happened before the signature, run against the math that operates the program now.

So here is the move, and it fits in one afternoon. Pull the original proforma out of the capital file. Put it next to the last twelve months of actuals. Mark every assumption that missed by more than ten percent: volume by code tier, payer mix, rate, service cost, software spend. That marked-up page is the agenda. Every line that missed is either a renegotiation item with the vendor or a correction that belongs in the FY27 budget before the cycle opens, not during it.

If the architecture still holds, the decision still works, and you have the document that proves it. If it does not, you want to know while the 2027 numbers are still proposals, not after the final rules confirm them in November.

The 2027 rules are not going to rescue programs running on 2022 deal structures. The work between now and November is making sure the deals signed in 2026 do not become the same problem in 2029.

The capital bundle, run against 2026 math.

The Pitch: "One number. One vendor. One service contract. The machine, the planning system, the OIS, and the AI layer come bundled. You get a discount on the bundle you do not get procuring the pieces separately."

The Reality: The bundle discount assumed the reimbursement environment the deal was signed in. The proforma underneath it assumed top-tier delivery utilization that has since landed lower and codes that no longer exist. The escalator compounds against a smaller revenue base every term. The AI line that was not broken out at signature is still not broken out, and it now carries a premium inside the SaaS renewal that nobody on the program side has independently priced. One number is easy to approve. Seven prices invite questions. That is the design.

With Heather Turner, RT(T), PMP

The Mirage of the Staffing Shortage: Why your staffing crisis might actually be a scheduling crisis.

Many oncology administrators across the country are focused on the same line item: high travel therapist costs and unplanned overtime. When a single traveler contract can easily drain $15,000 a month from your operating budget, and incremental afternoon overtime pushes around $73 an hour per RTT, the financial bleeding feels impossible to stop. The immediate reaction is always to hire more core staff. But before you open three new requisitions in a tight labor market, you need to look at your daily volume curve.

When I worked the treatment floor, I knew exactly when the day would fall apart. It wasn't because we had too many patients; it was because we had too many patients at 10:00 AM.

Many programs staff for their peak hours rather than their average volume. They build a team large enough to handle the mid-morning rush, which leaves therapists underutilized at 7:30 AM and 4:30 PM. You don't always have a staffing shortage. Often, you have a scheduling bottleneck.

Before you sign off on another expensive recruitment sign-on bonus, look at your scheduling templates through an operational lens:

Flatten the Curve: Work with your scheduling team to lock down your templates. Do not book three complex stereotactic cases back-to-back simply because the slots were open. Space them out. Be aware of what your therapists can handle.

Stagger the Shifts: The traditional 8-to-4:30 model can create artificial constraints. Transitioning even two therapists to a 7-to-3:30 or a 9-to-5:30 shift flattens the labor cost curve without requiring you to add staff.

When you align your staff availability to your actual treatment density, the "shortage" often disappears. Protect your bottom line by protecting your scheduling template first.

If an older capital deal is running in the program and the original proforma has never been put against a year of actuals, the Medsolve 30-minute program review is the conversation.