Medicare knee arthroscopy fell by about half over the decade

Roughly 99,000 patients in traditional Medicare — the fee-for-service side, which bills Medicare directly — had a knee arthroscopy in 2015, an operation done through a small camera incision. In 2024, about 40,000 did. A drop that size normally means the denominator moved: fee-for-service has been shrinking for years as people choose private Medicare Advantage plans, so almost anything measured inside it looks like it is going away.

So take the denominator away. Per 1,000 fee-for-service patients, the rate went from about 3.0 in 2015 to about 1.4 in 2024 — a fall of roughly half, against the count's 59%. The decline is not people leaving traditional Medicare. It is the procedure being done less often.

What makes this one unusual is that the record around it is legible. Two widely covered trials — a 2002 comparison against sham surgery, a 2013 comparison against physical therapy — were published years before the volume moved, and the fall runs through the decade after. And this is where the reading has to stop. Coverage rules, referral habits, and practice patterns all shifted in the same window, and any of them could plausibly matter. The record measures the volume, not the reason.

What it does measure, it measures in a particular unit: a patient on a claim code, not an operation. Several arthroscopy codes are counted here and a patient counts once on each, so one session billing two of them shows up twice — which lifts the level in every year alike and leaves the trend where it was. On that unit, a common operation lost about half its rate in ten years, and about 40,000 traditional-Medicare patients still had one in 2024.