How long does frozen shoulder last?

How long does frozen shoulder actually last, month by month?

In a 2- to 27-year follow-up, untreated primary frozen shoulder lasted a mean of 15 months, range 4 to 36 months, while shoulders already treated non-operatively before referral averaged 20 months, range 6 to 60 (Vastamäki et al., 2012). The wide range is the honest answer.

The number people want is a single figure. The number the evidence supports is a distribution. In the largest untreated series available, 51 patients with 52 shoulders were managed by observation alone, and their illness averaged 15 months from onset to resolution, with individual cases running from 4 months to 3 full years (Vastamäki et al., 2012). A second group in the same study, who had already received some form of non-operative treatment before their specialist consultation, averaged 20 months with a range stretching to 60 months. That second group was not slower because treatment harmed them; they were the harder cases who sought more care.

Why does it take so much longer for some people?

All four were identified as risk factors for persisting complaints in a consecutive series of 71 shoulders followed for a mean of 32 months (Sedlinsch et al., 2020).

Diabetes is the single factor clinicians watch most closely, and the mechanism is thought to be glycation of collagen making the capsule less willing to remodel. The effect shows up as slower recovery rather than a worse endpoint. When arthroscopic capsular release was compared in 17 diabetic and 20 idiopathic shoulders, the diabetic group had significantly lower forward flexion at 3 months (140.0° versus 151.5°) and at 6 months (152.2° versus 161.8°), lower external rotation at 6 months (43.3° versus 55.0°), and a lower American Shoulder and Elbow Surgeons score at 12 months (77.7 versus 88.8). Crucially, there were no significant differences between the two groups at final follow-up, a mean of 48 months out (Cho et al., 2016).

Is it really frozen shoulder, or something else that stiffens a shoulder?

In one surgical series of 71 stiff shoulders, only 8 (11%) were classified as primary frozen shoulder and 63 (89%) as secondary (Sedlinsch et al., 2020). That was a refractory surgical cohort, so it overstates the secondary share in the general population, but the direction is right: a stiff shoulder deserves a diagnosis before it gets a timeline.

The reason this matters for your question is arithmetic. Every duration figure quoted above belongs to a specific diagnosis. If your shoulder is stiff because of a cuff tear, glenohumeral osteoarthritis, calcific tendinopathy or scarring after an operation, then the 15-month figure is not your figure and neither is the recovery pattern. Adhesive capsulitis is estimated to affect up to 5% of the general population (Anjum et al., 2020), which is common enough to be the first guess and rare enough that the first guess is often wrong.

What decides when you can go back to normal use of your arm?

Four criteria, not a date. One: sleeping through the night on that side for two consecutive weeks. Two: passive external rotation within roughly 20 degrees of the other arm. Three: hand behind the back within about one vertebral level of the other side. Four: a full working day with no pain rebound the next morning.

Dates are a poor discharge criterion because the population range is enormous and yours is a sample of one. Criteria are better because they are observable and they are yours. The four above are the ones used in this clinic, and they are ordered deliberately: sleep recovers first, then range, then load tolerance. If someone tells you the shoulder is ready but you are still being woken at 3 a.m. by it, the shoulder is not ready.

Does physiotherapy shorten frozen shoulder, or just make it more bearable?

The honest answer is that the measured effect is on range and function rather than on total illness duration. In a randomised trial, adding joint mobilisation to stretching raised abduction from 91.9° to 172.8° and external rotation from 28.1° to 77.7° at one-year follow-up, while the stretching-only group did not show those changes (Çelik and Kaya Mutlu, 2016).

Nobody has run the trial that would answer the duration question cleanly: randomise a few hundred people with confirmed primary adhesive capsulitis to physiotherapy or to nothing, and measure months to resolution. The trials that exist measure range, pain and function at fixed time points, which is a different question. So a claim that physiotherapy cuts a frozen shoulder from 18 months to 9 is not supported by the literature, and you should be sceptical of any clinic that makes it.

What happens in a first physiotherapy visit for a stiff shoulder in Tel Aviv?

480 ILS flat, 50 to 60 minutes, one-to-one, VAT included, no deposit and no referral needed to book. The first session is measurement and explanation rather than machines: goniometer readings of both shoulders, a differential examination separating capsular restriction from cuff or neck referral, and a home programme you leave with the same day.

The measurement matters more than it sounds. Passive external rotation, abduction and internal rotation are recorded on both shoulders, so that every later session has a number to compare against instead of an impression. When you ask in month five whether anything is improving, the answer should be a degree figure and a date, not reassurance. The published trials track exactly these measures, which is what makes your own numbers interpretable against them.

When should you see a doctor instead of a physiotherapist?

Five findings send you to a physician first. Fever with a hot, swollen shoulder. Stiffness that began after a fall, a dislocation or a direct blow. Unexplained weight loss or any history of cancer. New weakness, numbness or pins and needles down the arm.

Those five are not a hedge. Septic arthritis of the shoulder is a same-day medical problem and no amount of mobilisation helps it. Stiffness after trauma can be a missed fracture or an unreduced posterior dislocation, both of which need imaging before anyone stretches anything. A shoulder that hurts more at night each week, in someone with a cancer history, is a shoulder that needs a doctor to look at it before a physiotherapist does.

Frequently Asked Questions