You reach for the coffee mugs on the top shelf and stop halfway. Something catches. You bring the arm down, shake it out, and try again with a little hop.
Or it is your hair in the morning. Or the second sleeve of a jacket. Or fastening something behind your back, which used to take two seconds and now takes a small negotiation with your own body.
Nobody sat you down and told you this was starting. You just noticed one day that you had begun using the other arm for everything above shoulder height.
Here is the useful news. "I cannot lift my arm overhead" is not one problem. It is at least four different problems that happen to produce the same complaint, and they need four different plans. The good part is that you can narrow it down a long way at home, in about two minutes, with one other person and a kitchen chair.
Most overhead shoulder trouble after 50 comes on slowly and is safe to investigate calmly. A small number of presentations are not. Please do not run the self-test below if any of the following apply to you.
Get urgent medical assessment if you have:
None of that is meant to alarm you. It is meant to be checked and cleared first, so the rest of this article is safe to use.
Nothing dramatic happens to the shoulder at 50. What happens is that decades of ordinary use start to show.
The rotator cuff is a set of four small muscles and their tendons that hold the ball of the arm bone centered in a shallow socket. Those tendons wear, and the wear shows up on scans in a rising share of people every decade.
A 2014 pooled analysis in the Journal of Shoulder and Elbow Surgery combined 30 studies covering 6,112 shoulders. Rotator cuff abnormalities were present in 9.7% of people aged 20 and under and in 62% of people aged 80 and over.
This matters more than the headline numbers, so read it twice.
The authors of that pooled analysis concluded that the prevalence of rotator cuff abnormalities in people without symptoms is high enough to suggest that this degeneration is part of normal aging.
A separate study screened 664 residents of one village with ultrasound on both shoulders. Full-thickness rotator cuff tears turned up in 22.1% of them. Of all the tears found, 65.3% caused no symptoms at all. Over the age of 60, roughly two-thirds of tears were silent.
So a scan showing a tear does not automatically explain your pain. Plenty of people your age have one and have never noticed. That is exactly why what you can and cannot do matters more than what an image shows.
Shoulder arthritis follows a similar curve. Osteoarthritis usually affects people over the age of 50, according to the American Academy of Orthopaedic Surgeons.
Frozen shoulder is different. It picks a narrower window, with peak incidence between 40 and 60 years of age, and it affects women more often than men.
And the neck ages too. The discs and joints in the lower neck stiffen and narrow with time, which is why a shoulder complaint in a 60 year old sometimes turns out to be a neck complaint.
So the fifties are not when your shoulder breaks. They are when four separate slow processes reach the point where you notice them.
Read this first. This test sorts likelihood. It does not diagnose. That is not modesty, it is the published evidence.
A 2012 systematic review with meta-analysis in the British Journal of Sports Medicine pooled the shoulder tests clinicians use every day. The authors concluded that "use of any single physical examination test to make a pathognomonic diagnosis cannot be unequivocally recommended," and that combinations of tests "provide better accuracy but marginally so."
Read that carefully in both directions. It does not mean shoulder tests are useless. It means no single one of them settles the question, and stacking them up helps less than you would hope.
Use this to decide what to ask about. Do not use it to decide what you have.
What you need: a helper, a straight-backed chair, and a wall or doorway you can face.
Safety rules, please follow all four:
Stand facing forward with your arms at your sides. Without help, raise the sore arm out to the side and up toward the ceiling, as far as it will go. Note how high you get. Use a landmark: waist height, chest height, shoulder height, ear height, or fully overhead.
Now do the same with the good arm and compare.
Sit down and let the sore arm hang completely loose. Tell your helper the arm is "theirs" now. Your job is to do nothing at all.
Your helper takes the wrist and elbow and lifts the arm slowly out to the side and upward. Slowly. They stop at pain or at a firm block, whichever comes first.
The question is simple: does the arm go further when someone else lifts it than when you lift it yourself?
This single comparison does more sorting than anything else on the list.
The American Academy of Physical Medicine and Rehabilitation puts it about as plainly as it can be put. Loss of passive external rotation, they write, "is the single most important finding on examination and helps to differentiate the diagnosis from rotator cuff pathology, since the latter does not result in loss of passive ROM."
Translated: if a cuff tendon is the problem, someone else can usually still move your arm through the range. If the joint capsule has tightened, they cannot, no matter how gently they try.
One honest qualifier. Moderate to severe arthritis in the ball-and-socket joint also restricts passive motion. So a passive lift that will not go narrows things to two candidates, not one.
Keep the elbow tucked against your side, bent to a right angle, forearm pointing straight ahead. Now swing the forearm outward, away from your body, like opening a gate. That is external rotation.
Compare sides. Then have your helper do it for you, gently, with the elbow still tucked in.
External rotation is the direction frozen shoulder tends to lose first. Pay attention to how it feels when your helper does it, not just how far it goes.
Raise the arm out to the side again, slowly, and pay attention to where in the movement it hurts, not just whether it hurts. A painful arc means pain appears roughly between 70 and 120 degrees of raising and then eases off above that.
Be aware of what that finding is worth. In the meta-analysis above, the painful arc had about 53% sensitivity and 76% specificity. In plain terms: a lot of people with subacromial trouble will not have a painful arc, but if you do have one, it is a reasonable pointer.
Reach behind your back as if fastening a coat, and note how high your thumb gets up your spine. Compare sides.
Then bring the sore arm straight out in front to shoulder height and draw it across your chest toward the opposite shoulder. Pain right on the top of the shoulder, at the bony bump where the collarbone meets the shoulder blade, points at the small AC joint rather than the main shoulder joint.
| What you found | Points toward | What to notice next |
|---|---|---|
| You cannot lift it, but your helper can move it most of the way up | Rotator cuff problem | Weakness more than stiffness. Night pain lying on that side. |
| Your helper cannot move it either, and outward rotation went early | Frozen shoulder, or arthritis in the main joint | Stiffness in every direction, including ones that do not hurt |
| Deep ache in the side or back of the shoulder, grinding or clicking, gradual over years | Shoulder osteoarthritis | Crepitus. Night pain. Stiffness that is there every single day. |
| Pain right on top of the shoulder with the cross-body reach | AC joint arthritis | Overhead reaching and behind-the-back motions also provoke it |
| Pain or tingling travels past the elbow into the forearm or hand | Neck-referred pain | Symptoms change when you turn or tip your head |
| Pain appears only between about 70 and 120 degrees, then eases | Subacromial problem, often cuff-related | Overhead work makes it worse. Arms down makes it better. |
Now here is the honest part, again. Two of those rows overlap. Restricted passive motion narrows things to two possibilities, not one, because severe arthritis in the ball-and-socket joint restricts passive motion too. A self-test cannot separate those. An examination and, when needed, an X-ray can.
What it is. Wear, irritation, or tearing in the tendons that steer and stabilize the shoulder. It can be a partial tear, a full tear, or tendon irritation without a tear at all.
Who gets it. Anyone, but the odds climb steeply with age. Degenerative tears build slowly, usually with no single moment you can point to.
What it feels like. Pain when lifting and lowering the arm. Weakness lifting or rotating. Pain at rest and at night, especially lying on that shoulder. Sometimes a crackling sensation. Many people describe it as "the arm just will not go," which is the giveaway, because a helper can usually still move it.
At the extreme end of that pattern is what clinicians call pseudoparalysis: generally defined as being unable to actively lift the arm above 90 degrees while passive motion stays free and there is no nerve injury. Researchers actively debate where that cutoff should sit, so treat the term as a description, not a verdict.
What usually helps. Loaded, progressive exercise for the cuff and the shoulder blade muscles, plus temporary changes to how you do overhead tasks. AAOS reports that in about 80 to 85% of patients, nonsurgical treatment relieves pain and improves function in the shoulder.
How long. One review notes that patients who are going to respond to non-surgical care generally do so within 6 to 12 weeks. Physical therapy has been described as an effective primary treatment in most patients, even with full-thickness tears followed over two years.
If this row sounds like you, our deeper articles on rotator cuff tears and shoulder impingement go further than this page does.
What it is. The capsule around the joint thickens and contracts. The joint physically will not travel, no matter who is moving it.
Who gets it. Peak incidence falls between 40 and 60 years of age, and women are affected more often than men. It is uncommon overall, with a reported prevalence of around 3% to 5% of the general population.
Two medical conditions raise the risk noticeably. Citing a meta-analysis, AAPM&R reports that people with diabetes were about five times more likely to develop frozen shoulder. It separately reports an increased risk of 2.69 times for people with thyroid disease. Longer duration of diabetes is also associated with poorer outcomes.
If you have either condition and your shoulder has started stiffening, mention it early. It changes how seriously a clinician takes the stiffness.
What it feels like. A painful phase first, then a stiff phase. The tell is that stiffness shows up in directions that have nothing to do with your daily complaint. Reaching for a seatbelt, turning a steering wheel, and putting a hand behind your head all become awkward at once.
The order is fairly predictable. External rotation goes first, then raising the arm out to the side, then reaching behind the back, then reaching forward.
How long, honestly. This is the cause where expectation management matters most, and where you should be suspicious of any tidy number.
Two respected sources give different figures. StatPearls puts the freezing phase at 2 to 9 months and the frozen phase at 4 to 12 months. AAPM&R puts freezing at 2 to 6 months, frozen at 4 to 12 months, and thawing at 6 to 26 months.
We are not going to pick the friendlier set for you. Take the honest headline instead: frozen shoulder is measured in many months to a couple of years, not weeks. Anyone who tells you otherwise is guessing.
What usually helps. Time is genuinely part of it, but "wait it out" is not a plan. Guided range of motion work, graded stretching into the restricted directions, and keeping the rest of the arm and shoulder blade strong all matter, because the shoulder you get back should be a working one. Around 80% of patients regain near-normal or normal function with proper treatment, though roughly 10% to 20% have some residual symptoms.
Our frozen shoulder deep dive covers the phases in more detail.
What it is. Cartilage wear, in either of two different joints. The big ball-and-socket joint is the glenohumeral joint. The small joint on top, where the collarbone meets the shoulder blade, is the acromioclavicular or AC joint. Osteoarthritis is actually more common in the AC joint.
Who gets it. Osteoarthritis usually affects people over the age of 50.
What it feels like. Location is the clue. Glenohumeral arthritis tends to be centered in the side or back of the shoulder. AC joint arthritis is focused on the top of the shoulder. Both bring grinding, clicking, or snapping sounds, and night pain is common enough that sleeping becomes difficult.
AC joint arthritis has a distinctive fingerprint. People often have a broadly intact range of motion, with the exception of cross-body reaching, behind-the-back motions, and overhead reaching, all of which produce pain right over the AC joint.
What usually helps. AAOS lists activity modification and rest, physical therapy exercises, anti-inflammatory medication, injections, and heat and ice among nonsurgical options. Set honest goals here. Exercise does not rebuild cartilage. What it can do is improve how the joint is loaded and controlled, which is often what determines whether you can reach the shelf.
This is the cause that gets missed, and it is worth understanding even if it turns out not to be yours.
The nerves that supply your arm exit the spine in your neck. When a disc bulge or an arthritic spur narrows the space around one of those nerve roots, the brain can interpret the signal as coming from anywhere along that nerve's territory. Cervical spine problems commonly refer pain into the shoulder, upper back, chest, and arm. The shoulder itself can be perfectly healthy.
What makes someone suspect the neck:
Here is why the neck gets missed so often. Cervical radiculopathy also tends to flare with overhead arm activity. That is the exact movement that provokes a cuff problem. So the trigger looks identical from the outside, and everyone reasonably assumes the shoulder is at fault.
The tie-breakers are the ones above: the electric quality of the pain, symptoms traveling below the elbow, and symptoms that respond to what your neck does rather than what your arm does.
In clinic, one tool for this is the Spurling test. The examiner extends and rotates the neck toward the painful side and applies gentle downward compression. It is described as highly specific but only moderately sensitive, which makes it useful for confirming a suspicion and poor at ruling one out.
Do not try that one at home, and do not try it on a parent. It deliberately compresses an extended, rotated neck, and it belongs in the hands of a therapist or physician. The self-test earlier in this article was designed to stay safe without it.
The encouraging part. More than 85% of acute cervical radiculopathy resolves without any specific treatment within 8 to 12 weeks, and at three years around 83% of people have regained satisfactory function.
The part to take seriously. If you notice clumsy hands, unsteadiness walking, or new bladder urgency, that is a different situation and needs prompt medical assessment.
You deserve the real picture rather than a sales pitch, so here it is with its limits attached.
A 2016 Cochrane review of manual therapy and exercise for rotator cuff disease identified 60 eligible trials. Only one of those compared a combination of manual therapy and exercise reflective of common current practice against placebo, and that trial found no clinically important differences between groups in any outcome. Comparisons against other treatments were rated low quality.
A 2020 systematic review and meta-analysis in Clinical Rehabilitation pooled 7 trials with 468 participants and separated exercise by type. Progressive, resisted exercise produced a 15 point improvement in a composite pain and function score on a 100-point scale, with a 95% confidence interval of 9 to 21.
Non-progressive, non-resisted exercise produced 4 points, with a confidence interval of -2 to 9, which is not a significant benefit. The authors classified all findings as low certainty and said they should be interpreted with caution.
Two honest conclusions follow from that.
First, how the exercise is done appears to matter more than whether you exercise. Loaded and progressed beats gentle and unchanging in the pooled data. That is why a generic printout of arm circles is not a plan.
Second, the certainty is low and nobody should pretend otherwise. Anyone promising you a guaranteed outcome for a shoulder is going beyond what the research supports.
Meanwhile, StatPearls reports that roughly 60% of patients with impingement report satisfactory outcomes within two years using a conservative approach of rehabilitation and activity modification. That is a real number of real people, and it is also not everyone.
HWY Physical Therapy is built specifically around adults over 50. Dr. Raj Pusuluri, PT, DPT, treats that population all day, which matters for a complaint like this one, because the four causes above overlap heavily in this age group.
Here is the sequence.
Oregon law lets a physical therapist evaluate and treat a self-referred patient, so you do not need a physician referral to start. We are direct-pay, no insurance required. If a screen turns up something that belongs with a physician, we say so and point you there. That is a professional duty, not a favor.
If you want to know what a first appointment involves in detail, we wrote that out: what to expect at your first physical therapy visit after 50.
One note on technology, since people ask. We use the Neubie, a device cleared by the FDA that applies direct current and is not a TENS unit, for indications including neuromuscular re-education and pain management. It is available in clinic only, and no other physical therapy clinic in Salem offers it. It is not the answer to a mechanical shoulder problem, and it never replaces loaded exercise.
You are the reason a lot of shoulders finally get looked at, so this section is for you.
Parents minimize shoulder problems more than almost any other complaint. A knee that gives way is frightening. A shoulder that will not reach the shelf just gets worked around, silently, for a year or more.
Signs the workaround has started:
How to raise it without a fight. Do not lead with "you need to see someone." Lead with a task. "Can you show me how you get the mugs down?" One demonstration tells you more than ten questions, and it lets them notice the limitation themselves rather than being told about it.
How you can genuinely help. Be the helper for the two-minute test above. You are the one who can move the arm while they relax, and that active versus passive comparison is the single most useful thing you can bring to an appointment.
Two things to write down and bring: which specific tasks have become impossible, and roughly when each one stopped. A timeline separates a shoulder that has been stiffening for eight months from one that changed in a week, and those are different conversations.
If surgery has already been mentioned to them, our guide on deciding between surgery and physical therapy after 50 walks through the trial evidence and the questions to ask a surgeon.
It depends entirely on which of the four causes you have, which is exactly why the sorting matters.
For rotator cuff problems, patients who respond to non-surgical care generally do so within 6 to 12 weeks. Frozen shoulder runs on its own clock: a freezing phase of 2 to 9 months, a frozen phase of 4 to 12 months, then thawing. Cervical radiculopathy often settles inside 8 to 12 weeks. Arthritis is managed rather than resolved, so the plan is about capability, not a finish line.
Rather than a promised number of visits, ask your therapist for a stated goal and a stated review date. If you are not moving toward the goal by that date, the plan should change.
AAOS lists physical therapy exercises among the nonsurgical treatments for shoulder arthritis, alongside activity modification, medication, injections, and heat and ice.
Set the goal correctly, though. Exercise does not regrow cartilage and no honest therapist will tell you it does. What it can change is strength, range, and how the joint is loaded during the movements you care about, which is often the difference between reaching a shelf and not reaching it.
Then you escalate, and you should. A plan that is not working after a fair trial needs changing, not repeating.
Worth knowing before you assume surgery is the next step: a large placebo-controlled trial published in The Lancet in 2018 randomized 313 people to arthroscopic decompression, placebo surgery, or no treatment. Everyone enrolled had already completed a non-operative program with exercise therapy and at least one steroid injection. Decompression did not beat placebo surgery, with a difference of -1.3 points on the Oxford Shoulder Score against a pre-specified threshold of 4.5 points for a meaningful difference.
That does not mean shoulder surgery never helps. It means "physical therapy did not fix it" is a reason for a careful conversation, not an automatic operation.
Some soreness after new loading is common, in the way that soreness after any new exercise is common. It should settle within a day or so and should not keep climbing session after session.
There is a relevant finding here. In the 2020 review, adverse events, meaning short-term increases in pain, appeared higher with non-resisted or non-progressive exercise than with progressive resisted exercise. The certainty of that evidence is low. Still, it is a reason not to assume that gentler automatically means safer.
Tell your therapist what you felt and when. Soreness is information, and the plan should adjust to it.
Night pain lying on the affected shoulder is one of the most reported symptoms in both rotator cuff problems and shoulder arthritis, so most people answer this question themselves within a week.
Practical things that help: sleep on the other side with a pillow hugged in front of you so the sore arm rests supported rather than hanging; if you sleep on your back, put a small pillow or folded towel under the upper arm so the shoulder is not pulled backward.
Sleep position is a comfort measure, not a treatment. If night pain is the main thing wrecking your week, that is worth mentioning explicitly at your evaluation, because it changes how a plan gets built.
Rest is a reasonable short-term response to a flare-up. It is a poor medium-term strategy, and for frozen shoulder in particular it can cost you range you then have to fight to get back.
A better framing: reduce the specific movements that provoke the problem, keep everything else moving, and get the thing evaluated so you know which of the four causes you are actually managing.
If you have gotten to the bottom of this page, you have probably already run the test in your head. You have a rough idea which row you fall into.
The next move is to have someone measure it properly, rule out the overlaps a home test cannot separate, and put a plan against it with a date on it.
Book a Wellness Screening Call with Dr. Raj: Schedule your call
Or call us at (971) 202-1979. Walk-ins are welcome at both clinics.
North Salem HWY Physical Therapy 2615 Portland Rd NE, Salem, OR 97301 (located at Center 50+)
South Salem HWY Physical Therapy Clinic South Salem 180 Ramsgate Square S, Salem, OR 97302
Phone (both locations): (971) 202-1979 Hours: Monday to Friday, 8:00 AM to 5:00 PM. Walk-ins welcome.
Book online:
https://meetings.hubspot.com/raj-pusuluri/hwy-pt-schedule
This article is general information about a common problem. It is not a diagnosis and it is not medical advice for your specific situation. If your shoulder pain follows an injury, comes with fever or swelling, or involves new weakness or numbness, seek medical assessment rather than starting an exercise program.