Problems · 8 min read
No published data says what you will lose. Your own books say what you can afford to lose, and that is the number that decides it.
There is no credible published figure for how many patients a dental practice loses when it drops a PPO plan. That is the honest answer, and it is worth saying before anything else, because almost every article on this question opens with one.
What does exist, and what almost nobody computes, is the number on the other side of the trade. Your books already hold the write-off you are absorbing on that carrier, the share of production those patients represent, and therefore the exact amount of attrition the decision can survive. That number is knowable to the dollar. The one everyone quotes is not.
Owners are not asking this because a consultant put it in their head. They are asking because the arithmetic of the last five years has gotten worse in a specific, documented way.
The American Dental Association's Health Policy Institute tracks a reimbursement rate index across all payer types. In its Q2 2026 State of the U.S. Dental Economy update, surveying 552 dentists in private practice, the picture is this:
| Reimbursement, all payer types, since Jan 2021 | +19% |
| Dental equipment and supply prices, same period | +23% |
| Hourly earnings of dental office staff, same period | +23% |
| Overall inflation, same period | +27% |
Reimbursement has grown slower than the cost of running the practice and slower than the dollar itself. In the same survey, low reimbursement and insurance pressure is the single most cited reason dentists give for feeling skeptical about the sector, named by 34.7% of respondents, ahead of patient affordability, rising costs and the growth of corporate dentistry.
So the impulse is well founded. The question is whether dropping a plan is the right response to it, and that is a different question with a different answer for every practice.
Search this topic and you will meet the same figure repeatedly: assume you will lose about 30% of that plan's patients. It appears in consultant guidance, in vendor content, and in at least one accounting firm's advice to dentists.
It is asserted every time and sourced none of them. There is no study behind it, no sample, no methodology and no publication. We looked for one and could not find it, which does not prove it is wrong, only that nobody has shown their work.
The closest thing to evidence is a first-person account, and it points the other way. Writing in Dental Economics in July 2024, Alexander Matheson, DMD, described dropping seven PPO plans while keeping one that paid within 15% of his full fee. He wrote that he had anticipated losing up to 30% of his PPO patients and that the actual loss "came in well under that figure." That is one practice in one market, and it is an anecdote rather than data, which is exactly how it should be read. It is still more evidence than the 30% has.
Treat the figure as what it is: a planning assumption someone once made that got repeated until it sounded like a finding. It is not a reason to stay and it is not a reason to leave.
Four things, all of them specific to your practice, and all of them sitting in data you already have.
The real write-off rate on that carrier. Not the average across all your plans, and not the number in a fee schedule comparison. The actual contractual adjustment those patients generated over the trailing 12 months, divided by what you billed them at your full fee. If your practice management system reports adjustments by carrier and your books carry insurance write-offs as their own line rather than buried in a netted revenue figure, this is a report, not a project.
What share of net production those patients represent. This is the exposure. A plan that is 6% of your production and a plan that is 34% of it are not the same decision, and a practice that has never broken production out by payer mix cannot tell which one it is looking at.
Whether patient count and production concentration match. They usually do not. A carrier can be 25% of your patients and 12% of your production, or the reverse. The version that matters for this decision is production, because that is what you would be putting at risk, and headcount will overstate or understate it every time.
What you actually collect, against what you produce. In network, the contractual adjustment is taken up front and the rest is fairly reliable. Out of network, the patient owes you the whole fee and gets reimbursed by the carrier afterward, if at all. That moves collection risk from the insurer to the patient, and it shows up in your collection ratio and your accounts receivable aging rather than in your fee schedule. If you want a structured way to see where your practice sits on these before you model anything, our free Dental Practice Benchmark Scorecard walks the same ratios.
Once you have the write-off rate, the question stops being "how many patients will I lose" and becomes "how many can I lose and still be even." That second question has an arithmetic answer.
The logic is simple. In network, every dollar you produce for those patients gets cut by the contractual adjustment before you collect it. Out of network, the dollars you keep are whole, but you only get them from the patients who stay. Break-even is the retention rate where those two totals meet.
At a 35% write-off, roughly two thirds of that carrier's patients have to stay for collections to hold. At a 25% write-off, closer to four fifths. At a 45% write-off, a little over half. The deeper the discount you are absorbing, the more attrition the decision can survive, which is why the same choice is obviously right for one practice and obviously wrong for another with the same patient count.
It is also worth noticing that a 35% write-off puts break-even attrition in the low thirties. That is very close to the 30% everyone repeats, which may be how a rule of thumb got mistaken for a measurement.
Here is the part the genre gets backwards, and it matters in your favor.
At break-even you are collecting the same money from fewer patients. That is not a wash, it is a better hour. You delivered less dentistry for the same collections, so the clinical supply and lab cost attached to the work you no longer did comes off with it, and chair time opens up. On contribution you finish ahead of where you started.
So the honest risk is not what happens at break-even. It is what happens below it, and what you do with the capacity above it.
Below break-even the arithmetic turns hard quickly. Rent does not fall, your team does not shrink and the equipment note does not care how many people sat in the chair. Almost everything in total overhead is fixed against a decision of this size, so a shortfall in collections lands almost entirely on profit rather than being cushioned by lower costs.
Above break-even, the freed chair time is worth exactly what you put in it. Filled with work at your full fee, it is the best outcome this decision offers. Left open, you are still fine on the money, but the decision has quietly changed from a growth move into a decision to work less for the same collections. That is a legitimate thing to want. It is only a problem if you did not know that is what you chose.
These are patterns rather than rules, and your own numbers are what actually decide it.
The math tends to favor practices carrying a deep discount on a plan that is a small share of production, practices already turning patients away, practices in markets where a meaningful share of employers buy plans with real out-of-network benefits, and practices whose new patients arrive by referral rather than by network directory.
It tends to go badly for practices where the plan is a large share of production, for practices with open chair time they have not been able to fill, for practices whose patients hold plans with no out-of-network benefit at all, and for practices that have never separated production from collections closely enough to know which one moved. The ADA's Q2 2026 survey found 26% of dentists reporting they were not busy enough and could have treated more patients. For a practice in that position, deliberately shedding demand is a hard case to make.
One thing is not a pattern and applies to everyone. Read your contract before you read anything else. Notice periods, termination windows and any obligation to complete treatment already in progress are contractual, they vary by carrier, and they set the calendar the rest of this runs on.
Going out of network changes the shape of your financial statements, not just the size of the numbers, and it is worth knowing which comparisons break.
Contractual write-offs on that carrier stop. Gross production and net production converge for those patients, which means your net production line steps up even where collections do not, and every ratio measured against net production shifts with it. Your production and collections now tell a different story than they did, and comparing this year's overhead percentage to last year's is comparing two different denominators.
Patient receivables grow and age differently, because the patient now carries the balance the insurer used to. Bad debt exposure rises where contractual adjustment used to sit, and those are two different things in your books with two different meanings. None of that is a reason to avoid the decision. It is a reason to fix the reporting before you make it, so that six months later you can actually tell whether it worked.
P.S. Reciprocity Accounting breaks your production and write-offs out by carrier every month, so a question like this one gets answered from your own trailing 12 months instead of from somebody's rule of thumb. See how we can help your practice.
Nobody has published a defensible answer, and any article that gives you one without naming a source is repeating an assumption. The figure you will see most often is 30%, and it has no study behind it. What you can compute is the opposite side of the question, which is how much attrition your write-off rate lets you absorb before collections fall. That number comes from your books and it is specific to you.
There is no threshold that settles it, because the write-off is only half the input. A 45% adjustment on a plan that is 5% of your production is a smaller decision than a 30% adjustment on a plan that is a third of it. Look at the two together, then at whether you have the chair time to refill. A deep discount on a small plan is the easiest version of this decision, and it is also the one that moves the least money.
For the patients who stay, yes, because the contractual adjustment that used to reduce gross production to net production disappears. That makes the year-over-year comparison misleading in your favor, and it is worth flagging in your own reporting. Collections are the honest measure of whether the decision worked, and profit is the honest measure of whether it was worth doing.
Practices that do this deliberately usually do not. Working one plan at a time gives you a real observation of your own attrition rate before you commit the rest of the book to it, and after the first one you are no longer estimating. The practitioner account cited above ran over roughly two years and kept the one plan that paid close to full fee, which is a common shape.
Not quite. Dropping one plan is a contract decision about one carrier. Going fee-for-service is the end state where you hold no network contracts at all, and it is usually reached in steps rather than in one move. The financial modeling is the same in either case, just repeated for each plan you are considering.