The de-paneling year: the A/R tail, the attrition curve, the runway

The shape of the year: one trough, three curves

Chart the year of the exit and three curves appear on it. The accounts-receivable tail declines: old claims paying out on payer clocks you no longer control. The retained caseload declines too, as patients decide one by one whether to stay. And the self-pay rebuild rises from near zero, at whatever pace new patients sign on.

The first months flatter you. Checks keep arriving for panel-era dates of service, the schedule still holds patients who have not yet had to choose, and the bank balance can read close to normal while the engine under it has already been switched off.

Then the curves cross. Claim payments thin out on the payers' own schedules. Patients decide at two moments, the notice letter and the plan year's turn, and the self-pay line, whether memberships or per-visit fees, starts near zero either way. The stretch where declining tail plus shrinking caseload plus small rebuild add up to the year's minimum is the trough.

The A/R tail: what old claims still pay, and until when

Every date of service before your termination date is still billable, against deadlines that are mostly published and one that is not. For Medicare fee-for-service the outer bound is regulation: a claim must be filed no later than 1 calendar year after the date of service, with narrow exceptions for contractor error, retroactive entitlement, and certain recoveries 1Ref 1Centers for Medicare & Medicaid Services (codified regulation) (2010).§ 424.44 Time limits for filing claims..The outer regulatory bound on the accounts-receivable tail: a Medicare fee-for-service claim must be filed no later than 1 calendar year after the date of service, with narrow named exceptions.. Commercial contracts set their own filing deadlines, usually shorter, in the agreement you signed.

How fast the money comes back is a state-law question. California is one concrete example: a Knox-Keene-licensed health care service plan must reimburse a complete claim as soon as practicable and no later than 30 calendar days after receiving it 2Ref 2California State Legislature (2024).Health and Safety Code § 1371.One state's concrete prompt-pay example: a California Knox-Keene-licensed health care service plan must reimburse a complete claim as soon as practicable and no later than 30 calendar days after receipt., and the parallel statute for insurer-issued products runs the same 30-day clock, pay or contest in writing inside the window, with late payment drawing interest at 15 percent a year 3Ref 3California State Legislature (2026).Insurance Code § 10123.13.California's parallel clock for insurer-issued products: 30 calendar days to pay or contest in writing an uncontested complete claim, with a 15 percent per year interest penalty on late payment.. Your state's rule may differ, and may not exist in this form at all. Where a statutory clock does exist, a biller-of-one who knows it can cite it when chasing a slow payer, so look up your own state's prompt-pay provision before assuming a payer's timeline is discretionary.

One deadline is not published anywhere: how long a terminated payer keeps accepting claims for pre-termination dates of service. That run-out sits in your own contract's termination provisions and varies agreement by agreement. Pull the termination exhibit before the notice goes out, read the run-out language, and calendar what it says. If the contract is silent, ask in writing while you are still in network, and bill early enough that the answer stops mattering.

Work the aging report by payer, oldest date of service first, until everything before the term date has gone out the door; every unbilled visit ages toward somebody's filing deadline while you are busy building the new practice.

The tax year will not match the story you tell about the transition. A practice on the cash method reports income when the payment arrives, and a solo practice qualifies for that method comfortably: the eligibility test is a gross-receipts ceiling far above any one clinician's collections 4Ref 4Internal Revenue Service (2022).Publication 538 (01/2022), Accounting Periods and Methods.Cash-method mechanics for the trough year: income is reported when received, and eligibility for the cash method runs on a gross-receipts ceiling far above any solo practice's collections.. So the legacy checks that land after January are next year's taxable income from this year's decision, arriving in a year that may otherwise show unusually low receipts. Flag the pattern for whoever prepares your return before the year closes.

The attrition curve nobody publishes

No regulator, journal, or survey publishes a retention curve for practices that leave insurance panels, so treat any confident percentage as marketing. What can be said is structural: a patient can only be reimbursed for following you out of network if their plan design allows it, and many designs do not. Retention is a variable you solve for under more than one assumption, never a rate you look up.

Plan design decides who even has the option. In 2025, 46 percent of covered workers with employer coverage were enrolled in a PPO 5Ref 5KFF (Kaiser Family Foundation) (2025).2025 Employer Health Benefits Survey.PPOs were the most common employer-sponsored plan type in 2025, with 46 percent of covered workers enrolled., the most common employer-sponsored plan type; POS plans also carry out-of-network benefits. A patient whose plan has no out-of-network benefit at all can follow you only by paying the entire fee out of pocket. Sort your caseload by plan type before modeling anything; the ceiling is different for every practice, and yours is sitting in your eligibility records.

The first decision point is your notice letter, when every patient confronts the change at once. The second is the plan year's turn, when deductibles reset and a patient who had met theirs faces the full out-of-network math again. Check when each patient's plan year turns; it is not always January. A conversion that survives the letter can still lapse at renewal, so model the attrition in two waves and budget for the second.

Each patient you move off the panel is a clinical relationship you are ending, and the standing duty applies: notify far enough in advance that the patient can find another physician, and facilitate the transfer of care when appropriate 6Ref 6American Medical Association, Code of Medical Ethics (2016).Opinion 1.1.5, Terminating a Patient-Physician Relationship.The standing duty when ending a patient relationship: notify far enough in advance for the patient to secure another physician, and facilitate transfer of care when appropriate.. Naming in-network alternatives in the letter costs a paragraph and answers the question every non-converting patient has.

Medicare is a separate decision with a middle position

Dropping the commercial panels does not by itself change your Medicare status, and Medicare offers a middle position between fully in and fully out. A physician can go nonparticipating: still enrolled and still billing Medicare, with a regulatory ceiling on what a patient can be charged. The full opt-out with private contracts is a separate election, worth working through on its own before anyone signs anything.

The limiting charge caps what a nonparticipating physician or supplier may bill a beneficiary at 115 percent of the applicable Medicare fee schedule amount 7Ref 7Centers for Medicare & Medicaid Services (codified regulation) (1992).§ 414.48 Limits on actual charges of nonparticipating suppliers..The nonparticipating middle position's price ceiling: the limiting charge caps what a nonparticipating physician or supplier may bill a beneficiary at 115 percent of the applicable Medicare fee schedule amount., and the base is the amount applicable to a nonparticipating supplier, so check the arithmetic on your own codes before reading it as a flat 15 percent premium. That is a real difference from participating rates, and it is nothing like naming your own cash price. Run your Medicare volume under both positions before treating Medicare as just another panel on the list.

Nothing forces the exit to be total in one quarter, either. A practice can leave one payer at a time in a drop order read off its own 835s, shedding the worst contract first and letting each slice of the tail pay out while the caseload adjusts in pieces. Professions with carve-out panels run the same play on a smaller board; an optometrist leaving the vision plans while keeping medical panels is doing exactly this. The stepwise version spreads the same revenue loss across more months, so the trough is likely shallower and certainly longer; how much shallower depends on your drop order and on what tail each departure leaves behind. Whether that trade suits you is a question about your cash and your patience.

The rebuild: pricing what replaces the panels

The rebuild is the curve you control, and it is set by three choices: the model (memberships or per-visit fees), the price, and the pace at which new self-pay patients arrive. Write the replacement target first. Last year's collections, minus the billing apparatus that leaves with the payers, is the revenue the new model must reach before the move breaks even, and every pricing decision is downstream of that number.

Membership pricing is the DPC equation run against your own panel: the number of members you can serve well, times the monthly fee, must clear the target. Per-visit pricing runs on utilization instead: retained patients, times visits per year, times the fee. Neither equation is exotic. The danger lives in the inputs, because the retention assumption sits inside both. Run the equation at more than one retention rate and see how far apart the answers are: the gap between them is the size of the bet you are making, and where you price inside it is your call.

But out-of-network revenue is not always zero on the patient's side. A PPO patient who stays can often submit your receipt against out-of-network benefits, and a practice that hands every such patient a clean superbill, with codes, prices, and NPI, gives them the one document a reimbursement claim needs. Whether it lowers what staying actually costs them depends on their plan's out-of-network benefit and deductible, which live in that plan's own documents.

Sizing the runway

The runway is the cash that carries the practice across the trough, and the only defensible size comes out of your own ledger. The round numbers that circulate for how many months to hold are somebody's guess, so build the figure instead: monthly operating expense, times the months your model says the trough lasts, minus a discounted estimate of the tail still coming.

Four numbers, all already in your possession:

• Twelve months of 835s and collections, by payer. That is the revenue being given up, and the raw material for deciding whether the exit is total or stepwise.
• Monthly operating expense from the ledger, including your own draw at the level you can genuinely live on.
• The trough's length under at least two retention cases, spread wide enough to contain a bad outcome, since no published rate exists to borrow.
• The tail, estimated from the aging report against each payer's filing deadline, then discounted, because some of it will never pay.

Give the plan calendar room on both sides. The de-paneling year fits the same planning frame as a new practice's launch runway, and for the same reason: the expensive months come early and the revenue arrives later than the effort does. A clinician winding down toward retirement is running different math again; the two-year glide makes attrition the goal, and that reverses several of the numbers above.

And if the model says the numbers never recover, learn that before the notice letter goes out. Closing a practice is a different sequence with its own deadlines.