Rotator Cuff Tears Found In 73% Of Adults Over 40. Here’s Why That Might Not Matter

Nearly everyone over 40 has shoulder damage that would show up on an MRI scan. Yet, since it doesn’t hurt, most people have no idea.

A study scanned the shoulders of 602 randomly selected adults and found rotator cuff abnormalities in 99% of them, ranging from early wear and tear to partial fraying to complete tears. Crucially, these problems showed up just as often in people with zero shoulder pain as in those complaining of serious issues. After researchers crunched the numbers, even the worst tears had essentially no connection to whether someone’s shoulder actually hurt.

The findings, published in JAMA Internal Medicine, turn conventional medical wisdom on its head. For decades, doctors have looked at shoulder MRIs, spotted tears or signs of wear, and assumed they found the culprit behind a patient’s pain. Treatments followed: physical therapy, cortisone shots, even surgery. But if nearly everyone walking around has these same “abnormalities,” calling them abnormal stops making sense.

Led by orthopedic surgeon Thomas Ibounig at Helsinki University Hospital, the study recruited everyday Finns from a long-running national health survey: not people already seeking treatment for shoulder problems. Each person underwent a thorough shoulder exam, filled out questionnaires about pain and function, and underwent high-powered MRI scans of both shoulders. The median age was 58, and the group split evenly between men and women.

Rotator Cuff Damage Was Everywhere

When radiologists analyzed the scans, damage appeared almost universally. About 25% of participants showed early signs of tendon wear and tear. Another 62% had partial tears, akin to a rope that’s fraying but not completely snapped. And 11% had full tears, where the tendon had ripped all the way through.

Age drove the numbers up. Nobody under 45 had a complete tear, but by age 70 and beyond, 28% did. Partial tears started showing up in people’s 40s and became the norm by their mid-50s. Men and women had tears at basically the same rates.

Among more than 1,000 pain-free shoulders, 96% showed some kind of rotator cuff problem on MRI. Among painful shoulders, 98% did. Early wear and partial tears appeared at identical rates whether the shoulder hurt or not.

Complete tears did seem more common in painful shoulders at first glance: 15% versus 7% in pain-free shoulders. But when researchers accounted for other factors like arthritis in the shoulder joint or problems with nearby tendons, even that difference disappeared. The final analysis showed less than a 1% gap, basically statistical noise.

Perhaps the most telling finding: of 96 complete tears identified in the study, 75 were in people who felt no shoulder pain whatsoever.

When Everyone Has It, What Does It Mean?

Shoulder pain ranks as the third most common musculoskeletal complaint in primary care, affecting up to 31% of people in any given month. When patients show up hurting, about half get sent for MRI scans, and up to 82% of doctors consider imaging essential for diagnosis.

Once the MRI comes back showing tears or degeneration, the treatment machine kicks into gear. Three-quarters of patients get referred to physical therapy. Imaging-guided cortisone injections have exploded 46-fold since 2000 in some places. Rotator cuff repair surgeries have doubled to septupled across high-income countries in recent decades.

But there’s a math problem with diagnosing based on findings that appear in virtually everyone. When 99% of people over 50 have rotator cuff changes, spotting one on MRI tells doctors almost nothing useful. A positive scan doesn’t prove what’s causing the pain unless something else points to it, like a recent injury, sudden loss of strength, or specific exam findings that fit.

Ibounig and his colleagues argue that the medical field needs to rethink its language. Calling these findings “tears” implies something broke that needs fixing. It suggests trauma, damage, disease. But if 96% of pain-free middle-aged shoulders have them, maybe they’re just normal aging, like gray hair or wrinkles, but for tendons.

The researchers suggest different terms: fraying, wear, changes, degeneration. Words that describe what’s happening without the alarm bells. Less dramatic language might ease patient anxiety about their bodies falling apart. It could also help doctors and patients have more honest conversations about whether imaging or aggressive treatment actually makes sense.

The research isn’t the first to find a disconnect between shoulder scans and symptoms, but it’s the most rigorous. Previous studies were smaller, often focused on people already in pain, or used less advanced imaging. A 2018 review of 56 studies found no consistent link between imaging findings and symptoms, but most of that research had quality problems.

Three earlier population studies found similar patterns. A small Australian study of 30 older adults found 43% had partial tears and 20% had complete tears. British researchers scanning 463 women in their 60s and 70s found 34% with partial tears and 11% with complete tears. A Japanese study of 683 village residents found 17% with complete tears. But none matched the scale or rigor of the Finnish work.

For anyone over 40 dealing with shoulder pain, the message is both reassuring and frustrating. Reassuring because that “tear” on your MRI probably isn’t as serious as it sounds: most people have them and feel fine. Frustrating because it means figuring out what’s actually causing the pain becomes harder. The scan that was supposed to provide answers might just be showing normal aging that has nothing to do with why you’re hurting.

The bigger question is what medicine does with this information. If rotator cuff findings are essentially normal after middle age, why are we still scanning so many shoulders and operating based on what we find? Ibounig and his team argue their findings should prompt doctors to question whether routine imaging helps patients or just leads to overtreating incidental findings that would never cause problems.

At minimum, the study suggests patients and doctors should view shoulder MRI results with healthy skepticism. That tear might be there. It might look impressive on the scan. But odds are good it’s been there for years, causing zero problems, and treating it won’t fix whatever’s actually wrong.

---

Disclaimer: This article is for informational purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition. Never disregard professional medical advice or delay in seeking it because of something you have read in this article. If you think you may have a medical emergency, call your doctor or 911 immediately.

---