Does menopause actually cause frozen shoulder?

What frozen shoulder actually is

Frozen shoulder, or adhesive capsulitis, is not a vague midlife ache. The capsule of connective tissue around the shoulder joint thickens and tightens, and the joint slowly loses movement in every direction. Reaching a back pocket, doing up a bra, sleeping on that side: each becomes hard, then sometimes impossible.

It is also more common than most people assume. A 2020 review in Psychopharmacology Bulletin puts prevalence at 2 to 5% of the general population, notes that it affects women more than men, and names diabetes as the most striking risk factor, with prevalence climbing to about 20% in people with diabetes. Thyroid disease is linked too. The NHS is refreshingly blunt about the rest: "It's often not clear why people get a frozen shoulder."

Why menopause keeps coming up

The age pattern is hard to ignore. Frozen shoulder peaks in midlife, and shoulder clinics see a lot of women in their 40s and 50s, the same years as perimenopause and menopause. Duke orthopaedic surgeon Dr. Jocelyn Wittstein has made this her research focus, and Duke's own write-up of her work notes that women with menopausal symptoms such as hot flushes and night sweats present with frozen shoulder at notably high rates.

The proposed mechanism is biologically reasonable. "Estrogen plays a role in stimulating bone growth, reducing inflammation, and promoting connective tissue integrity," Wittstein said in a Duke announcement of her team's research. If estrogen helps keep connective tissue supple and inflammation in check, losing it could plausibly tip a vulnerable shoulder capsule toward stiffening. It fits a wider pattern too: menopause shows up in muscle, bone and joints, not only in hot flushes, which is part of why interest in things like creatine for women has grown.

Plausible is not the same as proven, though. No study has yet shown that falling estrogen directly causes the capsule to thicken.

What the key study actually found

The study most often cited comes from Wittstein's group at Duke. They reviewed the records of 1,952 menopausal women aged 45 to 60, of whom 152 were on hormone therapy. According to the published abstract, frozen shoulder showed up in 4.0% of women on HRT and 7.7% of those not on it.

Framed as odds, women not on HRT had 99% greater odds of frozen shoulder, which sounds dramatic. Read one line further. The association did not reach statistical significance (p = 0.11), and the confidence interval was wide enough to include no effect at all. The design was retrospective and single centre, and women who take HRT can differ from those who do not in access to care, overall health and much else. The authors themselves called for larger, prospective, multi-centre studies.

So the honest reading is a signal worth chasing, not a finding to build a treatment plan on. When later coverage describes estradiol as roughly halving risk, it is leaning on preliminary data that has not yet been confirmed.

So does menopause cause it?

The defensible answer is: not proven, but possibly a contributor. Menopause cannot be the whole story, since men get frozen shoulder and diabetes is a far stronger, better documented risk factor. What the evidence supports is that midlife women are a high risk group, that hormonal change is a leading hypothesis for part of that risk, and that the research is only just getting started.

How long it lasts

Frozen shoulder is slow. It typically moves through a painful phase, a stiff or frozen phase and a gradual thaw. Cochrane describes it as lasting up to two to three years, and the 2020 review notes most people's symptoms resolve on their own within one to three years. The NHS says the pain and stiffness usually go away eventually. Usually is doing real work in that sentence, and early pain control matters during a long wait.

What treatment evidence supports

The strongest short-term evidence is for steroid injection. A Cochrane review of 32 trials and 1,836 people found that at around seven weeks, 77% of people given a glucocorticoid injection reported treatment success, versus 46% with manual therapy and exercise. That was moderate quality evidence, and side effects were similar between groups.

For people who do not settle, the UK FROST trial in The Lancet compared early structured physiotherapy with a steroid injection, manipulation under anaesthesia and keyhole capsular release in 503 patients. At 12 months all three groups had improved substantially and none was clinically superior, but capsular release carried more serious adverse events and higher costs.

What has no good evidence: HRT as a treatment for an established frozen shoulder, and the supplements, oils and "protocols" sold as menopause shoulder fixes.

Our take

The menopause link to frozen shoulder is real enough to take seriously and too thin to treat as settled. Midlife women deserve faster recognition of a condition that can steal years of easy movement, and steroid injection plus guided rehab has the best evidence. Hormone therapy may one day earn a place in prevention, but right now that is a research question, not a prescription.

This is general information, not medical advice. Talk to a clinician before changing anything, especially if you are pregnant, nursing, or managing a condition.