An MRI of the knee: $227 in Estes Park, $12,405 in Lone Tree

An MRI of the knee is about as standardized as hospital care gets. A technician slides you into the same kind of scanner, runs the same sequence, and a radiologist reads the same images, whether the machine sits in a small mountain hospital or a large suburban one. The procedure code is identical everywhere — CPT 73721, a knee MRI without contrast — which is exactly what makes it a clean test of how Colorado hospitals price the same thing.

Run that test and the numbers are hard to believe. At UCHealth's small hospital in Estes Park, the median negotiated rate for that knee MRI is $227, by the hospital's own published file. At HCA HealthONE's Sky Ridge Medical Center in Lone Tree — about a hundred miles down the mountain — the median negotiated rate for the same scan is $12,405. That is 55× as much for the identical procedure, inside a single state, both numbers drawn from the files the law requires each hospital to publish.

Neither figure is the typical one. Across every Colorado hospital that reports a rate for this scan, the statewide median negotiated rate is $876 — sitting between the two extremes and closer to the floor than the ceiling. Estes Park is well below that middle; Lone Tree is more than fourteen times above it. The point of this article is not which hospital is "right" — it is that a number this consequential swings this wildly, and that a few minutes with the published data tells you where any given hospital falls.

What drives hospital-to-hospital spread

Before going further it helps to keep three numbers straight, because hospitals publish all three and they are easy to confuse. The gross charge is the hospital's sticker price — the full list amount almost nobody actually pays. The cash price is what the hospital will take from someone paying out of pocket. The negotiated rate is what a hospital and an insurer have agreed the insurer's plans will pay. The figures in this article are median negotiated rates — the middle of what insurers' plans have actually agreed to, not the sticker gross charge and not the cash price.

Hold the procedure constant, then, and the spread is about the hospitals, not the scan. Some of it may be the local market: a hospital in a community with one dominant system and little competition tends to carry higher negotiated rates than one in a crowded metro. A larger share likely traces to the pricing posture of the system that owns the hospital — the same posted strategy applied across all of its facilities. We ranked Colorado's hospital systems by exactly this measure in a separate look at the systems behind the price files, and the pattern there is the same one visible here: the high end of this knee-MRI range belongs to the system whose published rates run highest across the board, and the low end belongs to a small mountain hospital whose own published rate runs far below its system's statewide average. A system running high across its published rates does not necessarily mean its patients pay more — what any one person pays depends on their insurer, their deductible, and their plan's structure, which is why the ranking piece carries the same caveat.

The data shows where each hospital's published rate falls and which system it belongs to; it does not show why any one hospital chose its number, and we do not guess. What the files do say is narrow: the highest published knee-MRI rate in the state sits at a hospital in the system that posts the highest rates generally, and the lowest sits at a small hospital whose published rate runs far below its own system's typical level. Structure is the most plausible explanation for that range — the two extremes track the systems behind them, just as the systems-wide pattern above would predict — but the alignment is what the files show, not proof that the structure caused the gap. Motive, and cause, are not in the files.

The carrier layer

There is a second layer underneath the hospital-to-hospital gap, and it is the one that actually lands on your bill: which insurer you carry. A single hospital does not negotiate one knee-MRI rate — it negotiates a different rate with each insurer, so even at one building the published numbers can range widely depending on whose card you hand over. We took apart that within-one-hospital spread for a routine lab test in a look at what insurers really pay for a blood test, where the cheapest and priciest insurer rates at the same hospital differed by orders of magnitude. The same dynamic applies to imaging: the state median and the hospital median are useful orientation, but your plan's number is the one that matters.

That number is published too, and you can read it. On the knee-MRI procedure page you can pick your insurance carrier from the dropdown, and the table of Colorado hospitals re-ranks itself by what that carrier's plans actually pay for this scan, showing your carrier's published rate at each hospital instead of an overall median. It is the difference between knowing the scan runs anywhere from a couple hundred to twelve thousand dollars in the state and knowing what it runs at the hospitals near you, on the insurance you actually have.

How we computed this

Each hospital figure is that hospital's median negotiated rate for the knee MRI across all of its reported insurer rates, from its own published file. The statewide number is the median across every Colorado hospital that reports a rate for the scan. We summarize a hospital's rate only when it is backed by at least two reported lines, so a single stray figure cannot stand in for a hospital's price — both extremes here clear that floor comfortably. (The statewide median happens to match the Rest of Colorado figure shown in the chart above, though the two are computed differently — statewide across all hospitals versus that one region's hospitals alone.)

Where a paragraph states a multiple — "55× as much" — that ratio is computed from the unrounded hospital medians and then rounded, which is why it sits beside rounded dollar figures and will not exactly equal those rounded numbers divided by each other. Government payers — Medicare, Medicaid, Tricare, VA — are excluded from these medians as statutory fee-schedule floors. One honest caveat travels with that: a small number of Medicare-style program lines are filed under commercial-looking insurer labels and can survive into the commercial view, which we name rather than bury. Full definitions are on the methodology page.

What you can do

A statewide median is a starting point, not a quote. The useful next step is to find your own plan's number at the hospitals you would actually use. Open the knee-MRI procedure page, pick your carrier, and read the published rate at each Colorado hospital — then browse the procedure across Colorado's metros to see how the gap narrows or widens near you. If picking a carrier and reading the re-ranked table is unfamiliar, here is how to find your own number in about three minutes. A published rate is not a confirmation that your plan is in-network at that hospital, so verify network status with your insurer or the hospital before your visit.