A torn rotator cuff usually announces itself with pain on the outer side of the shoulder that worsens when you lift your arm overhead or reach behind your back, along with weakness that makes it hard to hold things at arm’s length or raise your arm past shoulder height. Several at-home checks can help you gauge whether your rotator cuff is involved, but the reality is that no single physical test, even when performed by a trained clinician, is accurate enough to confirm or rule out a tear on its own. Understanding which signs point toward a rotator cuff problem and which ones should send you straight to a doctor can save you both time and unnecessary worry.
What a Torn Rotator Cuff Typically Feels Like
The rotator cuff is a group of four muscles and their tendons that wrap around the ball of the shoulder joint. Together, they keep the ball centered in its socket while you move your arm. When one or more of those tendons tear, the most common symptom is a deep, aching pain in the outer shoulder that often radiates partway down the upper arm. It tends to spike during specific motions: reaching overhead to grab something from a high shelf, pulling a seatbelt across your body, or pushing yourself up from a chair.
Weakness is the other hallmark. You might notice that lifting a gallon of milk to pour it feels harder than it should, or that your arm gives out partway through raising it to the side. In studies comparing different ways to read shoulder tests, muscle weakness turns out to be a better indicator of a torn tendon than pain alone.
Night pain deserves special attention. Many people with rotator cuff problems report that the shoulder hurts most when they try to sleep on the affected side, or even just lying on their back. A prospective study found that nighttime pain scores correlated with the severity of tendon damage, particularly in the subscapularis tendon, and also tracked with disability and reduced physical function.
Self-Checks for a Possible Tear
There are several clinical shoulder tests that have been studied extensively. Some of them are simple enough to attempt at home, though they were designed to be performed by a clinician who can feel for subtle weakness or compare sides. Here are the most commonly referenced ones, organized by which part of the rotator cuff they target.
The Empty Can and Full Can Tests (Supraspinatus)
The supraspinatus is the tendon that tears most often. To check it with the “empty can” test, extend your arm straight out in front of you at about a 45-degree angle from your body, turn your thumb down as if pouring out a can, and try to resist downward pressure. In the “full can” version, you do the same thing but with your thumb pointing up. Pain or, more importantly, weakness during either test suggests supraspinatus involvement. Research comparing the two found that both are roughly equivalent in accuracy when weakness is the criterion. One study measured about 75% accuracy for the full can and 70% for the empty can test when weakness was the indicator, with no significant difference between them.
A larger cohort study reported that the empty can test (also called Jobe’s test) had a sensitivity of about 88% for supraspinatus tears but a specificity of only 62%, meaning it catches most tears but also flags a fair number of shoulders that turn out to be fine. The full can test was less sensitive at around 70% but more specific at 81%.
The Drop Arm and Lag Sign Tests (Infraspinatus and Teres Minor)
The infraspinatus and teres minor are the tendons on the back side of the rotator cuff. They control external rotation, the motion of rotating your arm outward like you’re swinging a door open. The external rotation lag sign is tested by having someone place your arm at your side with the elbow bent to 90 degrees, then rotating your hand outward as far as it goes. If you cannot hold that position and your hand drops back inward, the test is positive. This lag sign was found to be less sensitive but more specific than the empty can test for diagnosing supraspinatus and infraspinatus tears, meaning it misses some tears but is more reliable when it does flag one.
A biomechanical study confirmed that lag sign tests are highly sensitive specifically for loss of infraspinatus-teres minor function, reaching 100% sensitivity in cadaveric specimens when those muscles had no force at all. In a living person with a partial tear, the results are less clear-cut.
The drop arm test is simpler: raise your arm fully overhead, then slowly lower it. If you cannot control the descent and the arm drops suddenly, it suggests a significant tear. A systematic review found that a positive drop arm test had a likelihood ratio of about 3.3, which modestly increases the probability that rotator cuff disease is present.
Belly Press and Lift-Off Tests (Subscapularis)
The subscapularis sits on the front of the shoulder and controls internal rotation, the motion of reaching behind your back. The belly press test involves pressing your palm into your stomach and pushing inward while trying to keep your elbow forward. If your elbow drifts backward, the subscapularis may be compromised. The lift-off test requires placing the back of your hand against your lower back and trying to push it away from your body; inability to do so points to a subscapularis tear.
A meta-analysis of subscapularis tests found that the bear-hug, belly-press, and lift-off tests all achieved high pooled specificity above 90%, meaning a positive result is fairly trustworthy. However, sensitivity was much lower. One surgical study found the belly-press test was only about 28% sensitive for detecting any subscapularis tear, though it improved to roughly 57% for full-thickness tears. The lift-off test hit 100% specificity in the same study, but its sensitivity was similarly low for partial tears.
Why Self-Checks Have Real Limits
If you read the numbers above and felt reassured that these tests sound pretty good, there is an important caveat. A systematic review of shoulder examination reliability found that the majority of studies showed poor reliability for all physical examination procedures used in shoulder assessments, and that no test has consistent evidence of acceptable reliability. Higher-quality studies were actually less likely to meet pre-agreed reliability thresholds.
Part of the problem is that these tests depend heavily on technique. A clinician stabilizes your shoulder blade, applies force at a precise angle, and compares the affected side to the healthy one. When you try them on yourself using your other hand, or have a family member push on your arm, you lose much of that control. A study comparing telehealth shoulder examinations (where a clinician watched via video and guided the patient) to in-person exams found that the diagnostic effectiveness of stand-alone tests was poor regardless of the setting, with the exception of a few highly specific tests. None achieved greater than 70% accuracy.
Combinations of tests perform better than any single one. Clinical examination findings paired with imaging can help confirm the presence of a shoulder problem, though even combined findings are better at ruling a tear in than ruling one out. In other words, if multiple self-checks point to a problem, take it seriously. But a negative result on one or two tests does not mean your rotator cuff is fine.
Tears Without Symptoms Are Surprisingly Common
One of the most counterintuitive findings in shoulder research is how many people walk around with a torn rotator cuff and feel nothing. A study using ultrasound on people who had no shoulder complaints found rotator cuff tears in 23% of them overall. The numbers climbed steeply with age: about 13% of people in their 50s, 20% in their 60s, 31% in their 70s, and 51% of those over 80 had a tear on the painless side. The authors concluded that rotator cuff tears must, to a certain extent, be regarded as normal degenerative attrition, not necessarily causing pain or functional impairment.
A Japanese population screening study found similar patterns, with tear prevalence rising from about 11% in the 50s to 37% in the 80s. Strikingly, asymptomatic tears accounted for about 65% of all tears detected. In people over 60, roughly two-thirds of all tears caused no symptoms at all. A UK-based study placed the population prevalence of at least one full-thickness tear at about 22%, with the difference between age groups highly significant statistically.
This matters for you because it means that finding a tear on an MRI does not automatically mean the tear is what is causing your pain. And it means that some degree of tendon wear is a normal part of aging, much like gray hair or wrinkles in the skin.
Traumatic Versus Gradual-Onset Tears
People often assume a rotator cuff tear requires a dramatic injury, like falling on an outstretched arm or wrenching the shoulder during a sport. That does happen, and traumatic tears tend to show up in younger patients (average age around 53 versus 57 for degenerative tears) and are more common in men. People with traumatic tears typically present sooner, averaging about three months from injury to seeking care, compared to over a year for degenerative tears.
But the distinction between “traumatic” and “degenerative” may be less clean than it sounds. A study comparing tendon tissue from acute trauma-related tears to chronic non-traumatic tears found no significant difference in the overall degree of degenerative change under the microscope. That suggests many so-called traumatic tears happen in tendons that were already weakened by age-related wear, and the injury was simply the final straw. The practical takeaway is that if you felt a pop or sudden onset of weakness after a fall, a sporting collision, or a heavy lift, a traumatic tear is likely. But if your shoulder has been gradually getting worse over months without a clear injury, a degenerative tear is just as plausible.
Jobs, Sports, and Other Risk Factors
Certain activities load the rotator cuff in ways that accelerate wear. A systematic review and meta-analysis identified working with the shoulder above 90 degrees as a significant risk factor for rotator cuff tendinopathy, roughly doubling the odds. Age over 50 and diabetes were also associated with increased risk.
A large study using the UK Biobank found that workplace physical demands are an important risk factor for rotator cuff disease requiring surgery, especially for workers with more than a decade of exposure. Jobs requiring high static strength were associated with roughly twice the hazard of needing surgery when performed for 11 to 20 years, compared to workers without that exposure. Painters, mechanics, warehouse workers, and construction laborers fall into this category, along with anyone whose daily work involves repetitive overhead reaching.
In sports, overhead athletes like swimmers, baseball pitchers, tennis players, and volleyball players put constant stress on the supraspinatus and infraspinatus tendons. Weekend athletes who suddenly ramp up throwing or overhead activity after a long layoff are at particular risk because their tendons have not adapted to the load.
What Else Could Be Causing Your Shoulder Pain
Not all shoulder pain comes from the rotator cuff. A pinched nerve in the neck (cervical radiculopathy, especially at the C5-C6 level) can produce pain that radiates into the shoulder and mimics a rotator cuff tear almost perfectly. One study noted that it is difficult to diagnose the exact origin of pain in patients with neck pain radiating to the shoulder based on MRI alone, because both conditions often coexist. A clinical review emphasized that the anatomic region a patient reports as painful can mislead the clinician, and that differentiating between cervical spine and shoulder pathology requires a careful history and targeted provocative maneuvers for both areas.
Frozen shoulder (adhesive capsulitis) is another common mimic. It produces pain and stiffness that overlap with rotator cuff symptoms, but the hallmark is a global loss of motion in all directions, not just weakness in specific movements. Calcific tendinitis, where calcium deposits form inside the tendon, can cause intense flare-ups that feel like an acute tear. Shoulder arthritis, labral tears, and even referred pain from the gallbladder or heart can land in the same neighborhood. If your symptoms do not fit the classic rotator cuff pattern of pain plus weakness with specific overhead or rotational movements, consider these alternatives.
Red Flags That Mean You Should See a Doctor Soon
Some presentations warrant prompt medical evaluation rather than watchful waiting. A clinical guideline published in the BMJ identified shoulder pain with an inability to raise the arm above 90 degrees after trauma as a red flag warranting same-day plain X-ray assessment. The same guideline recommends urgent specialist referral if you cannot raise your arm above 90 degrees more than two weeks after a shoulder injury, or if you are over 40 and have dislocated the shoulder joint. A confirmed acute tear should be discussed with a surgeon promptly because delays can allow the tendon to retract and the muscle to develop fatty changes that make later repair more difficult.
Other situations that justify a call to your doctor sooner rather than later include sudden severe weakness after a specific injury (especially if you heard or felt something pop), pain that is steadily worsening over weeks despite rest and over-the-counter anti-inflammatory medication, and night pain severe enough to regularly wake you from sleep.
How the Diagnosis Gets Confirmed
If your self-checks and symptoms raise suspicion, imaging is the next step. MRI is the most commonly ordered study and gives detailed views of the tendons, muscles, and surrounding structures. Ultrasound is a faster, cheaper alternative that can be done in the office during your appointment. A systematic review and meta-analysis comparing the two found no statistically significant difference in diagnostic sensitivity, specificity, or accuracy between ultrasound and MRI for supraspinatus tears of any thickness.
Ultrasound has the advantage of being dynamic: the examiner can watch the tendon move in real time and test specific positions. Its downside is that it is more operator-dependent, meaning the quality of the exam depends heavily on who is holding the probe. MRI gives a more comprehensive look at the entire shoulder, including the labrum, bone, and surrounding muscles, which is helpful when the diagnosis is uncertain or surgery is being considered.
Plain X-rays do not show tendons but can reveal indirect signs of a large chronic tear, such as upward migration of the ball toward the acromion, and can rule out fractures after trauma.
Living With a Tear and When Surgery Enters the Picture
Finding out you have a rotator cuff tear does not automatically put you on a surgical track. Conservative management is typically the first approach for degenerative tears, and evidence suggests that patients who commit to physical therapy often report high satisfaction, improved function, and success in avoiding surgery. The focus of rehab is strengthening the remaining intact rotator cuff muscles and the deltoid to compensate for the torn tendon, restoring range of motion, and retraining movement patterns.
Research on people with massive rotator cuff tears has shown that the body tries to adapt on its own. Electromyography studies found that patients with large tears reorganize their muscle activation patterns, increasing use of muscles like the latissimus dorsi and teres major to compensate for the deficient cuff. This compensation is not perfect, though. Kinematic studies found that older adults with rotator cuff tears use more internal rotation and altered movement patterns to perform everyday tasks like combing their hair, reaching up, and washing under their arms. Some people with advanced tears resort to leaning their trunk to the opposite side to get their arm overhead, a strategy that often falls short.
Clinical guidelines suggest that surgical repair of degenerative full-thickness tears is best suited for active, motivated patients with a non-stiff shoulder and muscles that have not yet developed significant fatty replacement. Age, arm dominance, activity level, tear thickness, and tear location all factor into the decision. For traumatic tears in younger patients, the timeline may be shorter because early repair can prevent retraction and muscle deterioration. Studies have shown that both traumatic and degenerative tears heal at comparable rates after surgical repair, with over 90% demonstrating good tendon integrity at six months.
Why the Same Tear Hurts in Some People and Not in Others
The disconnect between structural damage and symptoms remains one of the puzzles in shoulder research. As the asymptomatic prevalence data shows, a torn tendon does not guarantee pain. Several factors seem to tip the balance. Tears that involve the subscapularis, which is the workhorse of internal rotation, appear more likely to cause night pain. Larger tears and those with more retraction tend to produce greater functional loss. Inflammation, which can flare independently of tear size, plays a role in acute pain episodes. And psychosocial factors like stress, sleep quality, and pain catastrophizing influence how much a person suffers from a given structural problem, a pattern seen across musculoskeletal conditions.
This is why two people with identical-looking tears on MRI can have completely different experiences: one is unable to lift a coffee cup, the other plays recreational tennis without thinking about their shoulder. It also means that the question “is my rotator cuff torn” may matter less than “is whatever is going on in my shoulder limiting my life?” If self-checks point toward a rotator cuff issue and your symptoms are interfering with sleep, work, or activities you care about, get it evaluated. If your shoulder is a little achy but fully functional, the answer may already be that it does not matter whether there is a small tear in there.