We Read UnitedHealthcare's Own Rate Files. It Pays Less Than Medicare for an Office Visit.

Since July 1, 2022, the federal Transparency in Coverage rule has required every non-grandfathered commercial health plan in the United States to publish what it has negotiated with every in-network provider. Not a range. Not an average. The actual negotiated rate, by billing code, refreshed every month, posted publicly, free to download, no login. Practically nobody in medical billing has ever opened one. There is a good reason for that. A single UnitedHealthcare national network file runs about 15 gigabytes compressed and expands past 400 gigabytes. Cigna states plainly on its own compliance page that its files may reach one terabyte each. These are not spreadsheets. Excel will not open them, and neither will your practice management system. So the rule created a strange situation. The information asymmetry that defines payer contracting, where the payer knows what it pays everyone and you know only what it pays you, was legally abolished four years ago. In practice it survived intact, protected by file size. We decided to actually read them. This article reports what we found in UnitedHealthcare's files, how we got the numbers, and what a practice should do with them. The full state by state breakdown lives at what payers actually pay, and the extraction method is documented at our rates methodology.

The figures on this page come from UnitedHealthcare's Choice Plus national network file for the August 2026 reporting period, cross-checked against the Navigate national network for the same month. Before that we ran the same extraction on four UnitedHealthcare networks for the July period: Choice Plus, Options PPO, Navigate, and Choice HMO, 49.1 gigabytes in total. We verify every file byte for byte against the size UnitedHealthcare's own index reports and run a full integrity check before parsing a single record. From each file we extract every negotiated rate for a set of high volume CPT codes, filtered to professional billing class, keeping only flat dollar negotiated amounts. We then resolve each provider group to a state through the national NPPES registry, which holds 9.67 million provider records, and aggregate by code and state. Two decisions shaped the numbers, and both matter enough to state plainly. First, we report office rates only. UnitedHealthcare publishes different rates for the same code depending on where the service happens, and the gap is not small. A 99213 performed in an office pays about 1.45 times what the same code pays in a facility setting. Blending them produces a number that describes nobody. A practice billing from its own exam room gets the office rate, so that is what we publish. Second, we publish one network rather than an average of several. Averaging medians across networks is not a real statistic, and no practice is contracted to an average. We publish Choice Plus, UnitedHealthcare's flagship national commercial network, and use the others as verification. That verification is what we would point a skeptic toward, and it runs two ways. Across networks, the July extraction had Options PPO agreeing with Choice Plus on 306 of 306 comparable cells within 15 percent, Navigate on 98 percent, and Choice HMO on 98 percent. Across time, we re-extracted the Navigate network a month apart: the median ratio between July and August was 1.0000, every one of 250 state and code figures moved less than 5 percent, and 62 percent were identical to three decimal places. Separate downloads parsed a month apart do not agree that precisely by accident. It also tells you something useful in its own right, which is that negotiated rates move slowly enough that a figure here does not go stale in weeks.

Here is what UnitedHealthcare's Choice Plus network pays in an office setting, as the median across 51 states, against the Medicare non facility amount for the same code. Every figure below rests on 29,000 or more individual negotiated rates. The pattern is the finding. On established patient visits, the two codes that carry the bulk of volume in most practices, UnitedHealthcare pays less than Medicare. On new patient visits it pays essentially the same as Medicare. On the procedure in our sample it pays a premium. Read the last column carefully. A 99213 pays below Medicare in 35 of the 51 states we measured. This is not a handful of bad markets pulling down an average. It is the normal case.

Look at where the below Medicare rates fall and the structure stops looking accidental. A typical established primary care practice bills 99213 and 99214 for the large majority of its encounters. New patient visits are a small share of volume by definition, because a patient is new exactly once. Procedures depend on specialty and many practices bill few of them. So a payer that pays 101% on new patients and 111% on injections, while paying 88% on the code you bill most days of the week, can describe its contract as competitive on a code by code basis and still pay you less than Medicare across your actual case mix. Nothing about that is illegal or even unusual. It is ordinary contract construction, and it works because most practices never assemble the weighted picture. This is also why the standard advice to compare your contract against a Medicare percentage is close to useless without volume weighting. A contract quoted at 105% of Medicare tells you nothing until you know which codes carry the 105% and which carry the 88%. Run your own numbers before your next renegotiation. Pull your top ten codes by annual volume from your practice management system, multiply each by the published rate on our rate pages, and compare that against the same volumes at your Medicare locality rate. That single calculation reframes most contract conversations, and it takes an afternoon.

The state variation in these files is larger than most practice owners expect, and it is larger than Medicare's own geographic adjustment. On the same UnitedHealthcare network, a 99213 pays $71.41 in South Carolina and $194.38 in Alaska. That is not a typo and it is not an outlier we failed to clean. Alaska genuinely sits at the top of the commercial rate distribution for reasons that have to do with provider scarcity and cost of delivery. Alabama sits near the bottom on nearly every code we measured. Medicare adjusts for geography too, through the geographic practice cost index, but its spread is far narrower. Commercial negotiation produces variation several times wider than the cost differences Medicare recognizes. Two practices doing identical work, with identical credentials, on the same payer's network, can be paid twice as differently depending on which side of a state line they sit. For a practice this has a practical consequence. If you are told your rates are set by a regional fee schedule and are non negotiable, the published files let you check that claim against what the same payer pays in neighboring states. Sometimes the regional story holds. Often it does not. Every state figure we hold is published at our rate pages, including the 25th and 75th percentile so you can see the spread inside a state, not only the midpoint.

We would rather state the limits ourselves than have someone else find them. These are negotiated rates, not payments. What lands in your bank account also depends on eligibility, medical necessity determinations, bundling edits, patient responsibility, and whether the claim survives adjudication at all. A published rate is a ceiling with conditions attached. The figures describe one payer's networks. UnitedHealthcare is the largest commercial insurer in the country, which is why we started there, but Aetna and Cigna publish the same files under the same rule and their numbers will differ. We are working through those next. Rates are specific to a network and a reporting period. These come from the August 2026 files. Payers renegotiate continuously and republish monthly, so a figure that is right today can move. And a median is a midpoint, not your rate. Half the negotiated rates in a state sit below the number we publish and half sit above. Where you fall inside that spread depends on your bargaining position, your group size, and what you agreed to, sometimes years ago. That is precisely the thing the published percentiles let you check for the first time. What these files do settle is the question of whether you are being told the truth about the market. If a payer representative tells you the rate offered is the standard rate for your area, that claim is now checkable against the payer's own published data. It was not checkable before.

Three concrete steps, in the order we would do them. First, find your position. Pull your top five codes by volume, look up the published range for your state at our rate pages, and mark where your contracted rate falls inside it. If you land below the 25th percentile, you have a documented case rather than a feeling. Second, weight it. Multiply each code by your annual volume at both your rate and the published median. The dollar gap across a year is the number that gets a payer's attention, and it is almost always larger than practice owners expect, because the discount concentrates on high volume codes. Third, ask in writing. Contract negotiation moves when a specific, evidenced request lands in front of someone with authority. A request that cites the payer's own published rates for your state, for the codes you actually bill, is a different conversation from asking for a general increase. If that sounds like work you do not have time for, that is fair, and it is a large part of what we do. We handle payer contracting and rate benchmarking for practices alongside billing, and the analysis above is the same one we run for clients. Send us your top codes and your state and we will tell you where you sit, whether or not you ever become a client. One last thing worth saying. We published this knowing UnitedHealthcare will not enjoy it. The data is theirs, published under their own legal obligation, and we have shown our method in full at our methodology page so anyone can check the work. If we have it wrong, we want to know, and we will correct it in public.