Menopause and Frozen Shoulder: What the Evidence Actually Shows
Start with the shoulder exam
Frozen shoulder needs a hands-on assessment first. Find My HRT Path can organize a separate menopause-care decision for symptoms HRT is established to treat, without substituting for shoulder care.
Menopause and frozen shoulder overlap in timing, but menopause has not been shown to cause adhesive capsulitis. In the 2026 Duke pilot, diagnoses were 3.95% with hormone therapy versus 7.65% without, but the result was not statistically significant. The Menopause Society says hormone therapy is not recommended for frozen shoulder. Get the shoulder assessed on its own clock.
This page is most useful for you if:
- Your shoulder stiffened over weeks or months, with no clear injury you can name.
- You cannot fasten a bra, reach a seatbelt, wash your hair normally, or sleep on that side.
- You are in the menopause transition and want to know whether the hormone connection is real before your next appointment.
- Someone told you to “give it a year,” and you want to know what acting earlier could change.
Do not start here if:
- Your shoulder pain began after a fall, impact, dislocation, or recent surgery.
- Your shoulder is hot, red, visibly deformed, or badly swollen, or you have a fever.
- You have new numbness, tingling, loss of feeling, or weakness in your hand or arm.
- You have chest pressure, shortness of breath, sweating, nausea, or pain spreading to your jaw, back, or arm.
Those need prompt in-person care. Chest-pressure or breathing symptoms can be an emergency — call emergency services.
Here is what matters more than winning an argument about estrogen: frozen shoulder can take a long time, recovery is not always complete, and the treatment with the strongest short-term comparative evidence works earlier than most women are told. You do not need to settle the hormone question before you act on the shoulder.
| The question | What the verified evidence says | The next decision |
|---|---|---|
| Did menopause cause this? | The timing overlap is real. Causation is not proved. | Rule out look-alikes and stronger risk factors. |
| Will hormone therapy treat it? | No current guideline recommends hormone therapy for frozen shoulder. | Do not start, stop, or change hormone therapy to chase the shoulder. |
| Is research underway? | Yes. A 60-person UCSF trial is recruiting and has no results yet. | Treat the shoulder now; do not wait for trial results. |
| What helps most? | An intra-articular corticosteroid injection has the strongest short-term comparative evidence; rehabilitation and home exercise remain central. | Match treatment intensity to pain and irritability. |
| Can video care diagnose it? | Video can triage, but it cannot directly test passive range of motion. | Arrange an in-person shoulder examination when frozen shoulder is suspected. |
The HRT Index is the independent decision resource for online menopause and HRT care — comparing telehealth providers on clinical legitimacy, care quality, medication fit, price transparency, and access, with every claim verified and dated, so women can choose the path that fits their situation before their first consult.
This page contains clearly labeled affiliate links to Midi Health and Sesame. We may earn a commission at no extra cost to you. The shoulder recommendation comes first either way: an online HRT visit does not replace a hands-on shoulder examination.
What we actually verified for this page
We checked the original studies, current guidance, live trial record, and current provider terms rather than repeating the most dramatic headline. The core medical conclusion survived that process: menopause may be part of the context, but hormone therapy is not current treatment for frozen shoulder. Several numbers and commercial claims in the draft did not survive.
On August 4, 2026, we checked:
- The final peer-reviewed Duke pilot: sample size, age range, raw percentages, odds ratio, confidence interval, p-value, title, publication date, and conclusion.
- The 2026 AOSSM meeting poster involving 13,946 self-selected survey respondents: its published numbers and the limits built into its design.
- The live UCSF trial record: recruitment status, treatment arms, eligibility, exclusions, outcomes, sample size, and estimated completion.
- The Menopause Society’s explicit guidance on hormone therapy and frozen shoulder.
- The 2026 systematic review of 93,021 women and the shoulder-diagnosis study buried inside the broader evidence base.
- Current orthopedic guidance on diagnosis, staging, injections, rehabilitation, hydrodilatation, imaging, and surgery.
- Current Midi and Sesame pricing, coverage, visit format, and eligibility language directly from their sites.
For provider facts, we applied The HRT Index Verification Standard across the five pillars of clinical legitimacy, care quality, medication fit, price transparency, access. This was a documentation review, not a mystery-shopper consult and not a firsthand treatment experience.
This page has not been medically reviewed by a clinician. Every clinical conclusion is tied to a named primary or authoritative source. Where a study can support association but not causation, the page says so.
What is the connection between menopause and frozen shoulder?
Frozen shoulder peaks in the same midlife window as the menopause transition and affects women more often than men. That is a real association in timing, not proof that falling estrogen causes the condition. The cleanest current answer is: menopause may change the musculoskeletal environment, but adhesive capsulitis still has to be diagnosed and treated as a shoulder disorder.
Frozen shoulder is the everyday name for adhesive capsulitis. The shoulder joint sits inside a capsule of connective tissue. In adhesive capsulitis, that capsule becomes painful, thickened, and contracted, restricting movement.
The important correction is this: frozen shoulder can be primary or idiopathic, meaning no clear trigger is found, or secondary, following an injury, surgery, prolonged immobilization, or another medical condition. Frozen shoulder itself is not a rotator cuff tear or a fracture, but tears, arthritis, and other problems can coexist or mimic it. That is why “nothing is torn” and “there was nothing to injure” are not safe conclusions before an examination.
What is solidly established
The most consistent facts are not controversial:
- Frozen shoulder is most common in midlife, often between about 40 and 60 or 65.
- Women are affected more often than men.
- Diabetes is the strongest established medical association.
- Thyroid disease is also associated.
- Both active and passive shoulder motion become restricted.
- External rotation is usually especially limited.
You will still see “women get frozen shoulder four times more often than men” presented as a settled statistic. It is not. Published estimates vary substantially by setting and study design. The responsible conclusion is that female predominance is real; a universal four-times figure is not.
The result that complicates the menopause story
This is the part that deserves to be near the top, not buried after the sales pitch.
A study of women who already had shoulder pain compared 197 perimenopausal women with 113 premenopausal women. Adhesive capsulitis was the most common diagnosis in both groups: 35.4% versus 31.8%, a difference that was not statistically significant. Synovitis — inflammation of the joint lining — was significantly more common in the perimenopausal group: 25.1% versus 6.2%.
That does not prove hormones caused the synovitis. It does show that, in this shoulder-pain sample, the significant menopause-stage difference appeared in inflammation rather than frozen shoulder specifically.
Does that kill the connection? No. It sharpens it. The broader midlife musculoskeletal signal may be real without adhesive capsulitis being a direct estrogen-deficiency diagnosis. That distinction is the difference between a useful hypothesis and a treatment claim.
We could have left this out. It does not help sell anything. But if the one study that pushes back on the clean story disappears from the page, you have no reason to trust the rest of it.
Why can shoulder pain show up during perimenopause or menopause?
Muscle and joint symptoms become more common across the menopause transition, but the evidence does not separate hormonal change cleanly from aging, activity, sleep, body composition, diabetes, thyroid disease, and other causes. Laboratory work gives estrogen a plausible role in inflammation and fibrosis. It has not shown that estrogen treatment prevents or reverses frozen shoulder in women.
In 2024, researchers proposed the term “musculoskeletal syndrome of menopause” for a broad cluster that includes joint pain, loss of muscle mass, loss of bone density, and tendon or cartilage changes. Their review reported that more than 70% of women experience musculoskeletal symptoms and about 25% are disabled by them.
That phrase is a clinical framing, not a formal diagnosis and not proof that every new ache is hormonal. Frozen shoulder should not become the label that absorbs every shoulder problem in a woman over 40.
Two reviews, two very different-looking numbers
| Review | What it included | What it found |
|---|---|---|
| Lu et al., 2020 | 16 studies; 5,836 women | Pooled musculoskeletal pain prevalence of about 71% in perimenopause |
| Kruse et al., 2026 | 37 studies; 22 countries; 93,021 women | 40% before menopause, 57% during perimenopause, 59% after menopause |
Both reviews point in the same direction: musculoskeletal pain is common and rises around the transition. They do not give the same estimate, because the included populations, definitions, and methods differ. The 2026 authors also reported high heterogeneity and limited reporting of specific diagnoses.
Small correction, large consequence: the 71% figure belongs to perimenopausal women in that review. Repeating it as “71% of postmenopausal women” changes the group and overstates what was measured.
What the estrogen-and-fibrosis study actually did
A 2025 study is behind many versions of the headline “estrogen protects the shoulder.” Researchers used an immobilized-rat model and synovial-derived fibroblasts from people with frozen shoulder. Estradiol reduced fibrotic markers through signaling involving the G-protein-coupled estrogen receptor, or GPER.
Here is what they did not do: treat a single woman’s frozen shoulder with estrogen.
That is not a criticism. It is useful mechanism research and a reason to run a human trial. A result in rats and cultured cells is not proof of clinical benefit, a dosing instruction, or permission to improvise with a hormone product.
Should I start with hormone therapy or a shoulder appointment?
Start with the shoulder appointment. A suspected frozen shoulder needs an examination that compares how far you can move the arm with how far another person can move it. If you also have hot flashes, night sweats, vaginal symptoms, or other menopause concerns, handle that as a separate decision — not as a substitute for the shoulder workup.
The right online HRT provider isn't the same for every woman — it depends on your symptoms, your age and whether you have a uterus, your medication route preference (patch, pill, gel, or vaginal estrogen), your risk history, your insurance or cash-pay situation, and your state. Some situations belong with an in-person clinician first. Because a general answer can't resolve those for you, use The HRT Index's Find My HRT Path tool to match your situation to the right provider — and to flag when online care isn't the right starting point — before your first consult.
On this page, that safety flag matters more than usual. A shoulder that may be frozen belongs with an in-person examiner first. A video clinician can take a history, watch you move, and help triage. They cannot directly move your relaxed arm and feel whether the joint hits the same mechanical wall.
Find My HRT Path takes about 90 seconds, does not require an email, and keeps FDA-approved and compounded options clearly separated. It is an education and routing tool, not a diagnosis; a licensed clinician makes treatment decisions.
→ Find the HRT route that fits — including when online care is not the right start
Is it frozen shoulder, a rotator cuff problem, or something else?
The defining pattern is restriction of both active and passive movement: you cannot move the arm normally, and an examiner cannot move it normally for you. A rotator cuff problem often limits active movement more than passive movement. That distinction is powerful, but it is not a safe do-it-yourself diagnosis; do not force the arm.
Two terms make the rest easier:
- Active range of motion: how far you move the arm yourself.
- Passive range of motion: how far an examiner moves the relaxed arm for you.
In adhesive capsulitis, both are restricted because the capsule itself has contracted. External rotation — rotating the forearm outward while the elbow stays near your side — is often especially restricted. Reaching behind your back is mainly an internal-rotation task; it is commonly limited too, but it is not the definition of external rotation.
What else can look like frozen shoulder?
| Condition | Typical pattern | Why it gets confused with frozen shoulder | What usually resolves the uncertainty |
|---|---|---|---|
| Adhesive capsulitis | Gradual pain and stiffness; active and passive motion restricted; external rotation especially limited | Night pain and loss of overhead reach overlap with other disorders | Hands-on range-of-motion examination; X-ray if another cause must be excluded |
| Rotator cuff tendinopathy or tear | Pain and weakness with lifting; passive motion may be less restricted | Night pain and painful overhead movement | Strength testing; ultrasound or MRI when results would change management |
| Glenohumeral or AC-joint arthritis | Stiffness, pain, sometimes grinding | Similar age range and motion loss | X-ray |
| Calcific tendinopathy | Often sudden, severe pain | Pain can make the shoulder appear “frozen” | Examination and imaging |
| Cervical radiculopathy | Neck-to-arm pain, numbness, tingling, or weakness; may travel below the elbow | The brain reads nerve pain as shoulder pain | Neck and neurologic examination |
| Inflammatory arthritis or polymyalgia rheumatica | Often both shoulders; prolonged morning stiffness; hips or systemic symptoms may join | Midlife timing overlaps | Medical assessment and targeted testing |
| Referred cardiac or other internal-organ pain | May not track with shoulder movement; can come with chest or breathing symptoms | It can be felt in the shoulder or arm | Urgent medical evaluation when red flags are present |
Imaging is not automatically required to diagnose frozen shoulder. Plain X-rays may be used to look for arthritis or another bony cause. Ultrasound or MRI can help when a rotator cuff tear or another soft-tissue problem is suspected. The useful question is not “Can I get a scan?” It is “What decision would the scan change?”
The two medical associations that matter more than a hormone theory
Diabetes is the strongest established association. A 2023 systematic review and meta-analysis found that people with diabetes had about 3.69 times the odds of developing frozen shoulder compared with people without diabetes, although the included studies varied in quality.
Thyroid disease is also associated. A meta-analysis of 10 case-control studies involving 127,967 people found higher odds with thyroid disease overall (OR 1.87) and hypothyroidism specifically (OR 1.92). Association does not mean thyroid disease caused an individual case, but it is stronger evidence than the current menopause-specific evidence.
A clinician may review your diabetes and thyroid history and decide whether any testing is appropriate. That is different from telling every reader to order a panel.
Do not let a compelling estrogen story crowd out the conditions with the firmer evidence behind them.
When this is not routine shoulder pain
Get prompt care if:
- The pain followed trauma, dislocation, or surgery.
- The shoulder is deformed, hot, red, or severely swollen.
- You have fever or feel acutely unwell.
- You have new numbness, loss of sensation, or weakness.
- You have chest pressure, shortness of breath, sweating, nausea, or spreading jaw, back, or arm pain.
Night pain by itself is common in the painful stage of frozen shoulder. Gradual one-sided stiffness without fever, redness, swelling, trauma, or nerve symptoms is not automatically an emergency. It still deserves an appointment.
Which frozen shoulder stage am I in, and why does it change treatment?
Frozen shoulder usually moves through overlapping freezing, frozen, and thawing phases. The dates are not exact, but the treatment emphasis changes: an angry, painful shoulder needs symptom control and gentle movement; a less irritable, stiff shoulder can tolerate progressively stronger stretching and mobilization. The wrong intensity at the wrong time can flare pain and reduce your ability to keep going.
| Stage | What it often feels like | Commonly described duration | What usually matters now | What can backfire now |
|---|---|---|---|---|
| Freezing / painful | Pain is building, often worst at night; motion begins to disappear | About 6 weeks to 9 months | Assessment, pain control, consideration of an intra-articular steroid injection, gentle motion within tolerance | Aggressive end-range stretching through high pain |
| Frozen / stiff | Pain may settle while stiffness dominates | Roughly 4 to 12 months | Progressive stretching, joint mobilization, function-focused rehabilitation | Abandoning movement because the pain is quieter |
| Thawing / recovery | Motion returns slowly and unevenly | About 6 months to 2 years, sometimes longer | Progressive mobility and strengthening as tolerated | Stopping the plan when the shoulder is “mostly” back |
These stages overlap. A shoulder can be painful and stiff at the same time, and calendar time alone does not tell a therapist how hard to push.
Why the stretches you found online may be hurting
A rehabilitation model published in the Journal of Orthopaedic & Sports Physical Therapy uses irritability — how easily pain is provoked and how long it lasts — to guide intensity.
| Irritability pattern | What it can look like | Stretching approach described in the rehabilitation literature |
|---|---|---|
| High | Constant or night pain; pain begins before the end of range; symptoms linger after movement | Very short, relatively pain-free holds — often 1 to 5 seconds — with low intensity |
| Moderate | Pain mainly at the end of the available range | Short holds, often 5 to 15 seconds, with gradual progression |
| Low | Minimal resting pain; stiffness is the main limit | Longer end-range work and progressive loading as tolerated |
Those hold times are a published framework, not a prescription for every shoulder. Your clinician or physical therapist should set the dose for your examination findings and response.
Now think about the “push through it” video you followed at 2 a.m. It may have been written for a low-irritability shoulder in the later stage. Using that intensity on a high-irritability shoulder can flare pain and guarding.
If you have been pushing until you hurt and feeling guilty for not pushing harder, you can stop feeling guilty. You may have had the right idea at the wrong dose.
Build a Shoulder Record before the appointment
This replaces guesswork with a page your clinician can use. Print it, copy it into your phone, or fill it out before you go.
| What to record | Your answer |
|---|---|
| When did the pain or stiffness begin? | |
| Was the onset gradual, sudden, or after injury/surgery? | |
| Which tasks disappeared first: outward rotation, overhead reach, dressing, seatbelt, washing hair, sleeping on that side? | |
| Is pain or stiffness the bigger problem today? | |
| Does the pain wake you at night? | |
| What movements or stretches made it better or worse? | |
| Do you have diabetes, thyroid disease, inflammatory disease, or a history of shoulder immobilization? | |
| Any fever, redness, swelling, numbness, tingling, weakness, chest symptoms, or recent trauma? | |
| What have you already tried, and what happened afterward? |
This is not a diagnostic score. Its job is to keep the ten-minute appointment from disappearing into “my shoulder hurts.”
Does HRT help frozen shoulder?
Hormone therapy is not a treatment for frozen shoulder in current guidance. The 2026 Duke pilot found a large-looking raw difference but no statistically significant association. A much larger 2026 survey reported a stronger association, but its self-selected, cross-sectional design cannot prove prevention or treatment. The UCSF randomized trial is recruiting; it has no results.
This is the section we could have soft-pedaled. We are not going to.
The evidence is moving. That does not mean the answer has already moved with it.
Stage 1: the Duke result behind years of headlines
Duke researchers reviewed records for 1,952 postmenopausal women ages 40 to 60 seen in one health system for shoulder pain, stiffness, or adhesive capsulitis. Only 152 were using hormone therapy; 1,800 were not.
| Group | Adhesive capsulitis diagnosed |
|---|---|
| Hormone therapy, n=152 | 3.95% |
| No hormone therapy, n=1,800 | 7.65% |
That looks like roughly half. It is also where the confident headline stops and the statistics begin.
The adjusted result was OR 1.99, 95% CI 0.86–4.58, p=0.10 for the no-hormone-therapy group relative to the hormone-therapy group. The confidence interval crossed the no-difference line, and the p-value did not meet the study’s significance threshold.
In plain English: the study was compatible with a protective association, a smaller effect, or no effect. It was not large enough to settle the question.
The authors finally published it online on January 30, 2026 under the title:
“A preliminary pilot study to address design issues related to research on potential association of hormone therapy and adhesive capsulitis.”
That title is not a retreat. It is accurate science. It is also not permission to write “HRT halves frozen shoulder risk.”
Stage 2: the 2026 survey that is too large to ignore and too weak to treat on
At the 2026 American Orthopaedic Society for Sports Medicine annual meeting, researchers presented an anonymous, social-media-distributed survey of 13,946 postmenopausal women ages 45 to 65. Fifty-three percent reported using menopausal hormone therapy, and 15% reported a previous adhesive capsulitis diagnosis.
Among complete responses used for that analysis, adhesive capsulitis after starting hormone therapy was reported by 3.8% of 5,392 respondents, compared with 25.0% of 6,184 respondents not using hormone therapy. The poster reported p<0.0001 and an odds ratio of 0.1192. Among 486 respondents who reported being diagnosed first and then starting hormone therapy, 58.2% felt their symptoms improved.
Those are public numbers. The earlier draft’s claim that “no numbers are public” was wrong.
They are also not treatment proof. The survey was:
- Cross-sectional, not randomized.
- Distributed through social media accounts of specialists.
- Anonymous and self-reported.
- Dependent on respondents accurately reporting diagnosis, hormone use, and timing.
- Vulnerable to selection, recall, and healthy-user effects.
- A meeting poster, not a full peer-reviewed clinical report.
- Unable to separate a treatment effect from differences between women who did and did not use hormone therapy.
The 58.2% improvement result has no untreated comparison group and measures what respondents felt, not a standardized examination.
So yes: this is a signal worth following. No: it is not a reason to start hormone therapy for a shoulder.
Stage 3: the UCSF trial that has not answered anything yet
UCSF is running a Phase 4 randomized pilot trial, NCT07278323, titled Hormone Replacement Therapy as an Adjunct Treatment for Adhesive Capsulitis of the Shoulder (Frozen Shoulder) in Peri- and Postmenopausal Women. The UCSF page listed it as accepting new patients when checked on August 4, 2026.
| Trial detail | Verified August 4, 2026 |
|---|---|
| Location | UCSF Orthopaedic Institute, San Francisco |
| Investigator | Stephanie E. Wong, MD |
| Planned enrollment | 60 women |
| Age | 40 and older |
| Design | Randomized, interventional, Phase 4 pilot |
| What everyone receives | Physical therapy plus a glenohumeral joint steroid injection |
| What the experimental group adds | A skin patch plus a daily oral hormone-therapy pill |
| Primary outcome | American Shoulder and Elbow Surgeons score at 6 months |
| Other outcomes | Range of motion and functional workspace |
| Key exclusions | Hormone-sensitive cancer history, bilateral adhesive capsulitis, or current hormonal treatment including contraceptives |
| Estimated completion | December 2026 |
| Results | None posted |
Now look at the row that answers the real question.
Every participant receives the trial protocol’s shoulder care: physical therapy and a joint injection. Hormone therapy is being tested on top of that care, not instead of it.
The researchers testing hormones are not asking participants to wait on hormones while the capsule tightens. They are testing an add-on.
If you have been quietly hoping that starting hormone therapy means you can skip the shoulder appointment, the trial design gives you the clearest available answer: do not skip it.
The thing we cannot honestly give you
Here is the damaging admission, plainly: hormone therapy is not a frozen shoulder treatment, and we are not going to imply otherwise.
The Menopause Society says estrogen-containing hormone therapy is not recommended for the management of musculoskeletal conditions outside osteoporosis risk reduction, explicitly listing arthritis, joint pain, and frozen shoulder.
That should make the next decision easier, not harder.
If hormone therapy were the established answer, it would still require a prescription, individualized risk assessment, and follow-up. Meanwhile, your shoulder would still need to be examined.
Because it is not the shoulder treatment, you can separate the two clocks:
- The shoulder clock: assessment, pain control, and stage-matched rehabilitation now.
- The menopause clock: a separate conversation about hot flashes, night sweats, genitourinary symptoms, bone protection, and whether hormone therapy fits your history.
What does not work is waiting on one decision to solve the other — or dismissing the menopause conversation because it will not unfreeze the shoulder.
What about the Women’s Health Initiative joint-pain result?
The Women’s Health Initiative estrogen-alone randomized trial included 10,739 postmenopausal women with prior hysterectomy and tested oral conjugated equine estrogen, not an estradiol patch and not frozen-shoulder treatment. At one year, joint pain was reported by 76.3% in the estrogen group and 79.2% with placebo — a statistically significant but modest difference of about three percentage points. Joint swelling was reported more often with estrogen: 42.1% versus 39.7%.
That trial supports a small average effect on general joint symptoms in a specific population. It does not show that estrogen treats adhesive capsulitis.
The HRT Index 2026 Menopause–Frozen Shoulder Evidence Ledger
| Claim you may see | What the source actually shows | What you should not conclude |
|---|---|---|
| 1. “Menopause causes frozen shoulder.” | Frozen shoulder and menopause overlap in age; causation has not been demonstrated. | That a midlife diagnosis is automatically hormonal. |
| 2. “Women get it four times more often.” | Women are affected more often, but estimates vary and four-times is not a universal consensus figure. | That the highest reported estimate is the settled one. |
| 3. “Musculoskeletal pain rises at menopause.” | The 2026 review found 40% before, 57% during, and 59% after menopause, with high heterogeneity. | That every pain condition is caused by estrogen loss. |
| 4. “Perimenopause specifically means more frozen shoulder.” | In one shoulder-pain study, adhesive capsulitis was 35.4% vs 31.8%, not significant; synovitis was 25.1% vs 6.2%, significant. | That the diagnosis-specific link is proved. |
| 5. “Estrogen prevented shoulder scarring.” | Estradiol reduced fibrotic markers in a rat model and cultured fibroblasts. | That a woman’s frozen shoulder was treated or prevented. |
| 6. “The Duke study proved HRT halves risk.” | 3.95% vs 7.65%, but OR 1.99, 95% CI 0.86–4.58, p=0.10; final paper calls itself a pilot. | That the raw percentage difference proves protection. |
| 7. “The 2026 survey proves it now.” | A 13,946-person self-reported cross-sectional survey found a large association. | That self-selected survey data prove causation or treatment. |
| 8. “A trial proved hormones work.” | UCSF is recruiting 60 women; every participant receives PT plus injection; no results exist. | That registration, recruitment, or study design equals a result. |
| 9. “Estrogen fixes joint pain.” | WHI found about a three-point reduction in general joint pain and more reported swelling in a specific estrogen-alone trial. | That it treats frozen shoulder. |
| 10. “Menopause experts recommend HRT for joints.” | The Menopause Society explicitly says estrogen-containing HT is not recommended for frozen shoulder. | That this is an accepted indication. |
| 11. “Hormones are the biggest risk factor.” | Diabetes has an OR around 3.69; thyroid disease around 1.87 in meta-analyses. | That diabetes and thyroid history are side notes. |
| 12. “Put estrogen cream on the shoulder.” | No FDA-approved estrogen product is labeled to treat adhesive capsulitis or for application to the shoulder. | That a vaginal or compounded hormone product can be repurposed from an anecdote. |
The number most likely to be misread: 3.95% versus 7.65% is a raw difference from one health system, with only 152 hormone-therapy users. It is not proof that hormone therapy cut an individual woman’s risk in half.
What actually helps a frozen shoulder?
An intra-articular corticosteroid injection has the strongest short-term comparative evidence, especially for pain and function in the first weeks and months. Rehabilitation and a home-exercise plan remain central, with intensity matched to irritability. Hydrodilatation may add temporary improvement in disability and passive external rotation. Surgery is reserved for persistent disabling cases.
There is no instant fix. There is a better-supported sequence.
The steroid injection, honestly
A 2020 systematic review and network meta-analysis of 65 studies and 4,097 participants found intra-articular corticosteroid injection to be the most beneficial nonsurgical treatment for short-term pain and function. The advantage narrowed over time, which is why an injection is better understood as a window than a cure.
Its job is to calm the shoulder enough for sleep, daily function, and movement to become possible. Pairing an injection with a rehabilitation or home-exercise plan can add early benefit for some outcomes, but the best combination depends on stage, irritability, medical history, and access.
If you have diabetes, say so before an injection. Corticosteroids can temporarily raise blood glucose, and the decision may require a monitoring plan.
Treatment options, without the sales gloss
| Option | What the evidence supports | Where it fits |
|---|---|---|
| Stage-matched rehabilitation and home exercise | Core treatment; dose and intensity should change with irritability | All stages |
| Intra-articular corticosteroid injection | Strongest comparative support for short-term pain and function | Often most useful earlier, when pain blocks movement and sleep |
| Pain-relief medication | Can support sleep and participation; suitability depends on health history and other medicines | Supporting role, decided with a clinician |
| Hydrodilatation | May produce transient improvements in disability and passive external rotation; durability and clinical importance remain uncertain | Selected patients after discussion of alternatives |
| Manipulation under anesthesia | An escalation option when substantial disability persists despite nonsurgical care | Refractory cases |
| Arthroscopic capsular release | Surgical release can restore motion in selected persistent cases | Last-line option after nonsurgical treatment fails |
| Hormone therapy | Not recommended as frozen shoulder management; under study only as an add-on | Separate decision for established menopause indications |
| Estrogen cream on the shoulder | No approved indication and no clinical evidence | Do not improvise this route or body site |
The access problem nobody can solve with a better paragraph
The bottleneck is often not information. It is getting an appointment with someone who can examine the shoulder.
A video visit can be useful for triage, history, visible movement, referral, and deciding whether imaging or in-person care is needed. It cannot directly test passive range of motion. Sesame’s current orthopedic service is explicitly a video service, and Sesame itself says some patients will still need an in-person examination or imaging for a definitive diagnosis.
Sesame shows the exact cash price for each available clinician before booking; pricing and state availability vary. It does not accept Medicare, Medicaid, or other third-party insurance, and its terms require users to certify that they are not beneficiaries of a federal healthcare program such as Medicare, Medicaid, or TRICARE.
Best use here: rapid video triage when local access is blocked — not a substitute for the hands-on examination this page has told you to get.
→ See current online orthopedic availability and the exact price shown for your state Affiliate link. Sesame is video-based, cash-pay, and not available to Medicare, Medicaid, or TRICARE beneficiaries under its current terms. An in-person follow-up may still be required.
Prefer insurance or need the physical examination first? Start with primary care, sports medicine, orthopedics, or a physical therapist where direct access rules allow. Bring the Shoulder Record and ask for active-versus-passive range of motion to be compared.
How long does frozen shoulder last, and does it fully go away?
Frozen shoulder commonly runs for one to three years, sometimes longer. Most people improve substantially, but “self-limiting” does not mean fast, complete, or harmless. Long-term studies found 59% of shoulders normal or near-normal at about four years, 35% with mild ongoing symptoms, and 6% with severe persistent symptoms. Measurable motion loss can outlast major disability.
This is where “give it a year” comes from. It is also why that sentence can be so expensive.
Waiting is not the same as informed watchful care. A gradual condition can still deserve early confirmation, pain control, and a plan.
What long-term follow-up found
| Study or review | Long-term finding | What it means |
|---|---|---|
| Hand et al., 269 shoulders | 59% normal or near-normal; 35% mild residual pain or functional loss; 6% severe symptoms at a mean 4.4 years | Most improved, but a large minority was not symptom-free |
| Shaffer et al. | 60% had some measurable motion restriction at a mean seven years; functional limitation was usually mild | A shoulder can remain objectively restricted even when life is much better |
| 2026 American Journal of Medicine review | The simple “self-limited” description is increasingly challenged by prolonged symptoms and incomplete recovery | Counsel in years, not weeks, and do not promise perfect resolution |
Those results are not a reason to panic. They are a reason to stop treating “eventually” as a care plan.
What this condition actually steals
Clinical words do not capture the daily humiliation of it.
It is fastening the bra before putting it on. Sliding into a coat one arm at a time. Reaching for the seatbelt with the other hand. Waking every time you roll toward that side. Planning a shower around whether you can reach your own hair.
It is not dramatic enough for other people to understand and relentless enough to rearrange your day.
Common reader language is remarkably consistent: it came out of nowhere · I did not injure it · nobody mentioned menopause · is this permanent · will HRT fix it · I cannot sleep on that side anymore.
If those words are in your head, you are not imagining the pattern. You still need the diagnosis.
My hormone therapy fixed everything else. Why not my shoulder?
Hormone therapy relieving hot flashes, night sweats, sleep disruption, or genitourinary symptoms while the shoulder remains frozen does not mean the regimen failed. Those are different outcomes with different evidence. A persistent shoulder needs shoulder-specific care; hormone-therapy changes should be based on its established indication, your response, side effects, and your prescriber’s assessment.
The reasoning is understandable: the hot flashes stopped, sleep improved, and the shoulder did not move — so perhaps the dose is too low, the route is wrong, or the product needs changing.
That is the wrong tree.
Do not raise, stop, or switch prescribed hormone therapy to chase a shoulder. Bring hormone symptoms and side effects to the prescriber who knows your history. Bring the shoulder to someone who can examine it.
The menopause conversation can still be worth having
Hormone therapy remains first-line treatment for bothersome vasomotor symptoms and is indicated for genitourinary syndrome of menopause, primary ovarian insufficiency, and prevention of bone loss and fracture in appropriate patients. That is why the conversation can be worthwhile even when it does nothing to the capsule.
Better sleep can make a brutal year more manageable. Relief from hot flashes is real relief. Neither claim needs to be inflated into “HRT treats frozen shoulder.”
The point is that it is a separate decision, made for its own reasons.
Midi Health offers video care for menopause in all 50 states and says it is in-network with most PPO plans. Current self-pay pricing is $250 for an initial visit and $150 for continued-care visits. Coverage varies by plan; deductibles, coinsurance, and copays may apply.
Midi’s own frozen-shoulder article says hormone therapy is not indicated or validated as treatment for the condition. That matters here. A menopause service willing to state where its treatment does not belong is more useful than one that turns every midlife symptom into the same prescription.
The limitations matter too:
- A Midi video visit does not replace an in-person passive-range-of-motion examination.
- Midi cannot treat Medicaid or Medi-Cal patients, even as self-pay patients.
- Midi is not covered by Medicare or Medicare-related insurance plans. Medicare beneficiaries may use Midi as self-pay patients, but Midi says claims related to visits, medications, or associated services cannot be submitted to Medicare.
- Prescription and pharmacy costs are separate from the visit price and depend on the treatment selected and coverage.
→ Check Midi’s current coverage and self-pay visit price Affiliate link. Prices and coverage verified August 4, 2026; confirm your exact network status and patient responsibility before booking.
Why this is not a hormone-provider roundup
We could have filled this section with more affiliate buttons. We did not.
No online HRT platform can directly confirm passive shoulder restriction over video, and no compounded hormone product should be positioned as frozen-shoulder treatment. Compounded drugs are not FDA-approved; the FDA does not review them for safety, effectiveness, or quality before marketing.
That does not make every compounded prescription inappropriate. It means compounded and FDA-approved products must remain clearly separated — and neither category belongs on this page as a shoulder cure.
What if I have frozen shoulder symptoms after breast cancer or while taking an aromatase inhibitor?
Aromatase inhibitors commonly cause joint and muscle symptoms, but shoulder pain on an aromatase inhibitor is not automatically frozen shoulder. It still needs the same examination and differential diagnosis. Systemic hormone therapy is generally not recommended after breast cancer, with rare exceptions requiring shared decision-making with the oncology team. The nonhormonal shoulder-treatment sections of this page still apply.
Anastrozole, letrozole, and exemestane lower estrogen as part of breast-cancer treatment. Musculoskeletal symptoms are common and can become severe enough to threaten adherence. That makes the pain real. It does not make every case adhesive capsulitis.
A randomized trial of 121 breast-cancer survivors with aromatase-inhibitor arthralgia found that a year-long exercise program reduced worst joint pain compared with usual care. That is evidence for a nonhormonal way to address aromatase-inhibitor joint symptoms; it is not a frozen-shoulder-specific cure.
Do not stop an aromatase inhibitor because of joint pain without speaking to your oncology team. The team may consider exercise programming, symptom treatment, assessment for another diagnosis, or changes within the cancer-treatment plan.
The UCSF frozen-shoulder hormone trial excludes anyone with a history of hormone-sensitive cancer. Current Menopause Society guidance says systemic hormone therapy is not recommended for breast-cancer survivors, while acknowledging rare exceptions that require collaborative shared decision-making with the healthcare team.
You are the group the menopause internet often serves worst because there is no easy hormone button to put underneath the paragraph. We are not going to put one here.
For the broader nonhormonal menopause decision, see menopause treatment when you cannot take estrogen.
Will frozen shoulder happen to my other shoulder?
It can. Published reviews commonly place second-shoulder involvement at about 6% to 17%, often within five years of the first shoulder, and diabetes increases concern. The practical advantage of having been through it once is not guaranteed prevention. It is recognizing the pain-and-stiffness pattern earlier and getting assessed before major motion loss is established.
The early pattern is usually:
- Pain builds, often at night.
- External rotation begins to disappear.
- Active and passive movement become restricted.
- The change progresses over weeks or months rather than resolving after a few days.
Reaching behind the back may become difficult too, but that movement mainly reflects internal rotation and combined shoulder motion. The cleaner early clue is progressive loss of external rotation in more than one direction of movement.
What we can and cannot say about prevention
There is no established medication, supplement, cream, or stretching protocol proven to prevent frozen shoulder.
What is supported:
- Diabetes belongs in the clinical picture because it is the strongest established association.
- Staying active and resistance training support broader midlife bone, muscle, and metabolic health.
- Early attention to progressive stiffness can open access to short-term treatment before months of severe restriction pass.
What is not supported: marketing a hormone, supplement, or topical product as frozen-shoulder prevention.
If someone is selling certainty here, they are ahead of the evidence.
What should I do next for menopause and frozen shoulder?
Book an in-person shoulder assessment and bring a written record of onset, lost movements, night pain, risk factors, and what happened after each treatment or stretch. Ask whether passive motion is restricted and what irritability level is being treated. Handle any menopause consultation alongside that plan, for the symptoms hormone therapy is established to treat — not as a delay.
This week or this month
1. Arrange a hands-on shoulder examination. Primary care, sports medicine, orthopedics, or a physical therapist can be reasonable starting points depending on local access and direct-access rules.
2. Use the Shoulder Record. Write down the onset, whether there was trauma or surgery, the movements and daily tasks you lost, night pain, diabetes or thyroid history, nerve symptoms, and every treatment you tried.
3. Ask for active and passive range of motion to be compared. That sentence moves the visit further than “my shoulder hurts.”
4. Ask what irritability level the plan is treating. You want to know why the exercise dose fits the shoulder you have today, not the shoulder in a generic video.
5. Discuss whether an intra-articular corticosteroid injection fits your stage and medical history. If you have diabetes, include glucose effects in that conversation.
6. Keep the hormone decision separate. If you also have hot flashes, night sweats, sleep disruption, vaginal symptoms, or other menopause concerns, discuss those on their own evidence and timeline.
Ten questions worth bringing to the appointment
- Is my passive range restricted too, or mainly my active range?
- Which diagnosis best explains the pattern, and what are the main look-alikes?
- Is external rotation especially restricted?
- What stage or irritability level do you think I am in?
- Do I need an X-ray or other imaging, and what would it change?
- Does my diabetes, thyroid, cancer-treatment, or surgery history change the plan?
- Is a joint injection appropriate now, later, or not at all?
- What should I stop doing because it is flaring the shoulder?
- What should improvement look like over the next six to twelve weeks?
- When do we reassess, and what finding would make you change course?
Frequently asked questions about menopause and frozen shoulder
The questions below close the common gaps: whether frozen shoulder is an official menopause symptom, whether HRT can cure it, how to distinguish it from a cuff problem, how hard to stretch, whether imaging is needed, how long it lasts, and what online care can and cannot do. The direct answer stays the same: examine and treat the shoulder separately.
Is frozen shoulder an official menopause symptom?
No. It is not a recognized indication for hormone therapy. It is a shoulder condition that peaks in the same midlife window, and researchers are investigating whether hormone changes contribute to risk or severity.
Can low estrogen cause shoulder pain?
Estrogen biology is relevant to connective tissue and inflammation, and musculoskeletal symptoms rise around the menopause transition. No human study has shown that low estrogen directly causes frozen shoulder in an individual woman.
Will HRT cure my frozen shoulder?
There is no evidence that it will. The Menopause Society does not recommend estrogen-containing hormone therapy for frozen shoulder. The only active randomized trial is testing hormone therapy as an add-on to physical therapy and a joint injection, and it has no results.
Is the UCSF frozen-shoulder hormone trial recruiting?
The UCSF study page listed NCT07278323 as accepting new patients when checked August 4, 2026. It is a San Francisco trial for women 40 and older with diagnosed adhesive capsulitis, with important exclusions including current hormonal treatment, bilateral disease, and hormone-sensitive cancer history.
How do I know whether it is frozen shoulder or a rotator cuff tear?
The main clue is passive motion. In frozen shoulder, both active and passive movement are restricted. With a rotator cuff problem, active motion and strength may be worse than passive motion. A clinician should test this; do not force your own arm.
Why is frozen shoulder worse at night?
Night pain is common in the painful or freezing phase. Position, pressure, and an irritable joint can repeatedly wake you. Night pain alone is not proof of frozen shoulder because rotator cuff disorders and other shoulder conditions can do it too.
Should I keep stretching if it hurts?
Do not assume more pain means more progress. High-irritability shoulders are commonly managed with short, low-intensity, relatively pain-free movement, while stronger end-range work belongs later. A physical therapist should set the dose for your stage and response.
Is exercise safe with frozen shoulder?
A shoulder diagnosis does not automatically mean stopping every form of exercise. Shoulder-specific intensity should match the diagnosis and irritability. Stop and seek guidance when exercise produces a major or prolonged flare, new weakness, numbness, swelling, redness, fever, or trauma-related pain.
Do I need an MRI?
Not automatically. Frozen shoulder is mainly a clinical diagnosis. X-rays can help exclude arthritis or another bony cause; ultrasound or MRI may be useful when a rotator cuff tear or another soft-tissue problem would change treatment.
Can frozen shoulder affect both shoulders?
Yes. The second shoulder is reported in about 6% to 17% of cases, often within five years. Simultaneous bilateral disease is less common and should increase attention to diabetes, thyroid disease, and other systemic context.
Can an online menopause clinician diagnose frozen shoulder?
A video clinician can take a history, observe movement, and triage. They cannot directly test passive range of motion, which is central to confirming the pattern and separating it from common look-alikes. Suspected frozen shoulder belongs with an in-person examination.
Can I put estrogen cream on my shoulder?
No FDA-approved estrogen product is labeled to treat adhesive capsulitis or for shoulder application. Vaginal estrogen is labeled for specific genitourinary uses, and compounded products are not FDA-approved. Do not improvise a route, dose, or body site from an anecdote.
Does frozen shoulder always go away?
Most people improve substantially, but complete recovery is not guaranteed and the timeline is often one to three years or longer. Long-term cohorts found mild residual symptoms in a substantial minority and severe persistent symptoms in a smaller group.
What is the fastest realistic way to improve it?
There is no instant cure. The fastest evidence-aligned route is an early correct diagnosis, consideration of an intra-articular corticosteroid injection when appropriate, and rehabilitation or home exercise matched to irritability. Waiting without assessment and forcing an angry shoulder can both cost time.
Still not sure which problem to solve first?
Solve the unexamined shoulder first and use the HRT tool for the separate menopause decision. The two paths can run together, but neither should impersonate the other. Find My HRT Path takes about 90 seconds and flags when online care is not the right starting point.
Use an in-person shoulder assessment for the shoulder decision. Use Find My HRT Path for the menopause decision.
Related reading
- Perimenopause symptoms checklist — what else belongs in the wider pattern
- HRT benefits and risks — what hormone therapy is established to do
- Menopause treatment when you cannot take estrogen — nonhormonal routes, including after breast cancer
- Nonhormonal options — what is available and what the evidence supports
- How long do menopause symptoms last? — the wider timeline
The HRT Index is an independent editorial resource and is not a clinic, pharmacy, or telehealth service. This page contains labeled affiliate links to Midi Health and Sesame. A commission may be earned at no extra cost to the reader; it does not change the shoulder-first recommendation or the limitations stated before each link. See the full affiliate disclosure.
Educational research only. Not medical advice. Not a substitute for examination, diagnosis, or treatment by a qualified clinician.