Explore Supplements Subscribe
The VerdictHIGH CONVICTIONVerdict Score 80

Frozen shoulder heals itself in 1-2 years — but the right exercises cut months off that timeline.

Right now, let your arm hang at your side. Gently swing it in small circles — clockwise for 30 seconds, then counterclockwise for 30 seconds. That's a pendulum exercise. It's the safest starting point for a frozen shoulder, and it's what most rehab protocols begin with. Do it 3 times today.

  1. Here's what's really happening: the lining of your shoulder joint has thickened and scarred, physically locking the joint. It's not a muscle problem — it's a capsule problem.
  2. What most people get wrong: aggressive stretching in the early painful phase makes it worse. The capsule is inflamed — forcing it is like pulling on a fresh wound.
  3. Start here: gentle pendulum exercises daily, then progress to slow eccentric rotator cuff work as pain allows. The biggest study ever (503 people) showed structured physio gets the same 12-month results as surgery.

Your shoulder joint has a sleeve around it — like a shirt sleeve around your arm. That sleeve has shrunk in the wash. It's physically too tight for the ball to roll properly in the socket. The pain isn't from damage — it's your body's alarm system reacting to tissue being stretched beyond its current (shrunken) limit. The good news: unlike a shirt, your body can gradually re-stretch that sleeve. But yanking on it makes it shrink tighter. Gentle, progressive loading is what loosens it.

SH
Dr. Seth Holbrook, DPT — Doctor of Physical Therapy • Coach to 300+ clients
I built The Verdict to cut through recycled health advice and show what the evidence actually supports.

Frozen Shoulder

Optimal Rehab, Pain Management, Prognosis & Exercise Progression

Shoulder

HIGH CONVICTION

Right now, let your arm hang loose. Gently swing it in small circles — 30 seconds clockwise, 30 seconds counterclockwise. That's a pendulum exercise, and it's the safest first move for a frozen shoulder.

Pendulum swings use gravity to gently mobilize the joint without provoking the inflamed capsule. Every evidence-based rehab protocol starts here.

Takes less than 2 minutes. No equipment needed.

The Verdict

Frozen shoulder heals itself in 1-2 years — but the right exercises cut months off that timeline.

Your shoulder joint has a sleeve around it — like a shirt sleeve around your arm. That sleeve has shrunk in the wash. It's physically too tight for the ball to roll properly in the socket. The pain isn't from damage — it's your body's alarm system reacting to tissue being stretched beyond its current limit. Unlike a shirt, your body can gradually re-stretch that sleeve. But yanking on it makes it shrink tighter. Gentle, progressive loading is what loosens it.

  1. Here's what's really happening: the lining of your shoulder joint has thickened and scarred, physically locking the joint. It's not a muscle problem — it's a capsule problem.
  2. What most people get wrong: aggressive stretching in the early painful phase makes it worse. The capsule is inflamed — forcing it is like pulling on a fresh wound.
  3. Start here: gentle pendulum exercises daily, then progress to slow eccentric rotator cuff work as pain allows. The biggest study ever (503 people) showed structured physio gets the same 12-month results as surgery.

Want the full evidence? Keep scrolling

What Works

Shoulder rehabilitation — progressive loading and manual therapy

Tier 1 — Strong Evidence

Corticosteroid Injection + Structured Physical Therapy STRONG

UK FROST trial (n=503, Lancet 2020), multiple systematic reviews, APTA CPG

Injection suppresses inflammation inside the joint (most effective early). Structured PT maintains and restores movement while addressing motor control deficits.

Pain relief within 1-2 weeks of injection. Functional improvement over 6-12 weeks of PT.

Eccentric Loading + Scapular Stabilization + Graded Isotonics STRONG

Multiple RCTs (2022-2024), systematic reviews

Active loading addresses both the structural scarring and the muscle guarding component. Superior to passive stretching alone.

Measurable ROM gains within 4-8 weeks. Full benefit over 3-6 months.

See Tier 2 & 3 treatments

Tier 2 — Moderate Evidence

Manipulation Under Anesthesia (MUA) + Injection MODERATE

UK FROST trial — equivalent outcomes, highest cost-effectiveness

For patients failing 3-6 months of conservative care. Forcibly disrupts adhesions. Highest probability of cost-effectiveness in the FROST trial.

Immediate ROM gain post-procedure. 6-12 weeks rehab after.

High-Grade Joint Mobilization (Grade III/IV) + Exercise MODERATE

RCTs — 16% absolute risk difference for treatment success vs placebo

Best delivered in clinic where the therapist can immediately adjust based on tissue response. Less effective without concurrent pain management in the freezing phase.

Tier 3 — Emerging Evidence

Heavy Resistance Rotator Cuff Strengthening EMERGING

Limited RCTs for frozen shoulder specifically. Extrapolated from rotator cuff literature

3 sets of 6-8 reps at submaximal to maximal tolerance. For late-stage restoration when ROM allows full movement.

What Doesn't Work

  • Aggressive passive stretching in the freezing phase — provokes inflammatory flare-ups. The capsule is inflamed, not just tight.
  • Supervised neglect alone — frozen shoulder is self-limiting, but "do nothing" prolongs disability and causes secondary compensatory patterns.
  • Hydrodilatation as standalone — evidence remains inconclusive. No demonstrated benefit over standard injection (Challoumas et al. SR, 2020).
  • Ultrasound, TENS, laser therapy — no quality evidence supporting these passive modalities for frozen shoulder.

Exercise Prescription

Pendulum Swings

2 min each direction | 3x daily

Lean forward, let your arm hang. Gently swing in small circles. Let gravity do the work.

Should be pain-free. Smaller circles if sore.

Assisted External Rotation

3 x 10 | Daily

Hold a stick with both hands at waist height. Use your good arm to push the stick sideways, rotating the stiff arm outward. Hold 5 seconds.

Gentle stretch at end-range OK. Stop if sharp.

Wall Slides (Flexion)

3 x 10 | Daily

Face a wall. Walk your fingers up as high as you can, keeping elbow straight. Hold 5 seconds at top.

Mark your height and track progress weekly.

Eccentric External Rotation

3 x 10 | Every other day

Hold a light weight (1kg). Use your other hand to rotate outward, then slowly lower back in over 4 seconds. The slow lowering is the key.

Moderate effort, no sharp pain.

Scapular Squeezes

3 x 10 | Daily

Sit tall. Squeeze shoulder blades together and slightly down. Hold 5 seconds.

Effort in mid-back, not shoulder pain.

Red Flags

When to Seek Urgent Help

  • Sudden loss of motion after a fall or seizure — possible unreduced dislocation or fracture. Go to A&E immediately.
  • Fever, chills, severe redness, joint warmth — possible joint infection. Emergency referral needed.
  • Unexplained mass, severe unremitting night pain, unexplained weight loss — possible malignancy. See your GP urgently for investigation.
  • Rapid onset severe weakness or numbness down the arm — neurological assessment needed. Not explained by pain alone.

Refer to: A&E for suspected fracture/dislocation/infection. GP for suspected malignancy. Orthopedics for refractory cases (no progress after 6 months conservative care).

Return to Training

What's Actually Going On

Frozen shoulder mechanism — capsular thickening and fibrosis of the glenohumeral joint

The glenohumeral joint capsule — a thin tissue sleeve wrapping the ball-and-socket — becomes inflamed, then progressively scarred. The coracohumeral ligament and rotator interval thicken first, followed by the fold of capsule under the armpit (the axillary recess).

FREEZING
2-9 months
Pain dominates. ROM gradually decreases. Night pain common.
FROZEN
4-12 months
Stiffness dominates. Pain mainly at end-range. Inflammation settles.
THAWING
6-26 months
ROM slowly returns. Pain decreases. Fibrotic tissue remodels.

Important: some of the restriction is from muscle guarding — your muscles clamping down to protect the joint — not just structural scarring. This matters because active motor control training addresses the guarding component that passive stretching can't reach.

How to Identify It

Shoulder assessment — clinical examination of glenohumeral range of motion
?
Global restriction of BOTH active and passive ROM — the examiner can't push the arm further either. This is the hallmark.
?
Capsular pattern: external rotation most limited, then abduction, then internal rotation
?
Coracoid Pain Test Sn: 96-99% | Sp: 87-98%
Firm pressure on the coracoid process reproduces the concordant shoulder pain
?
Distension Test in Passive ER (DTPER) Sn: 100% | Sp: 90%
Sustained overpressure into external rotation at 0 degrees. True capsular restriction = hard end-feel with no give
?
Key differentiator: if someone else CAN move your arm further than you can, it's not frozen shoulder — look at the rotator cuff instead

The Debate

Passive Stretching vs Active Loading

APTA/JOSPT CPG, 2013

Pain-free stretching and supervised neglect. Match intensity to irritability. Favor passive modalities.

VS

Multiple RCTs, 2022-2024

Eccentric loading + scapular stabilization shows superior functional improvements compared to passive therapy.

Follow newer evidence. Active loading in the frozen/thawing phases. The guarding component responds to motor control training, not passive stretching. Pain-free stretching still appropriate in the highly irritable freezing phase.

PT vs Surgery — Does the Intervention Matter?

General orthopedic consensus

12 weeks intensive supervised PT as the standard conservative pathway.

VS

UK FROST Trial, Lancet 2020 (n=503)

Early structured PT showed no clinical superiority over MUA or arthroscopic release at 12 months.

PT is still the correct first line — same results, lowest risk, lowest cost. The self-limiting natural history normalizes outcomes across all pathways. But manage expectations: this is a marathon, not a sprint.

CPG Recency Check: The APTA/JOSPT 2013 CPG is older than the 5-year threshold. The 2020 FROST trial and 2022-2024 eccentric loading RCTs significantly update the evidence base.

Honest Limitations

Session Frequency Gap

The research: FROST trial used 12 supervised sessions over 12 weeks for optimal results.

The reality: Most healthcare systems cap at 6 sessions. Home exercise compliance in highly painful conditions is historically poor.

Adjustment: Front-load education in sessions 1-2. Teach 3-4 key exercises with clear pain-guided rules. Use virtual check-ins to maintain adherence.

Patient Expectation Mismatch

The research: Natural history is 12-24 months, sometimes up to 3 years.

The reality: Patients expect resolution in weeks. Frustration drives premature surgical referral before the thawing phase has even started.

Adjustment: Set timeline expectations at first visit with concrete milestones. "Pain improves significantly by month 3. Stiffness improves from month 6. Full resolution 12-24 months."

Diabetic Subgroup Under-Represented

The research: General population trials show favorable conservative outcomes.

The reality: Diabetics (10-20% of frozen shoulder patients) have worse prognosis, higher bilateral rates, and steroid injections complicate blood sugar control.

Adjustment: Expect longer timelines (up to 30 months). Coordinate injection timing with GP for blood sugar monitoring. Consider earlier MUA referral if conservative care stalls.

The Nuance

Shoulder anatomy — glenohumeral joint complexity and surgical considerations

Surgery isn't wrong — it's just rarely necessary. The UK FROST trial (503 patients, the largest ever) showed physiotherapy, manipulation under anesthesia, and arthroscopic capsular release all converge to the same outcomes at 12 months. MUA was the most cost-effective. Surgery had the lowest rate of needing further treatment. But structured PT carries the lowest risk and cost while achieving the same results.

Diabetic patients need separate expectations. They make up 10-20% of frozen shoulder cases but have worse prognosis, higher bilateral rates, and steroid injections complicate blood sugar control for 24-48 hours. Expect timelines closer to 24-30 months, not the standard 12-24.

Not all stiffness is structural. Part of the movement loss comes from active muscle guarding — your muscles clamping down to protect the joint. This is why eccentric loading and scapular stabilization work better than passive stretching. They retrain the nervous system to allow movement, not just stretch scarred tissue.

The 2013 guideline is outdated. It's the most recent dedicated CPG for frozen shoulder, but the FROST trial (2020) and multiple eccentric loading RCTs (2022-2024) have significantly shifted the evidence. The old "supervised neglect" approach leaves too much on the table.

What would change this protocol: a large RCT (n>400) comparing early heavy eccentric training vs passive stretching in the freezing phase, with MRI-verified capsular changes, stratified by diabetic status. If aggressive loading in the acute phase prevented structural fibrosis, it would completely shift early treatment away from the cautious approach.

Sources

HIGH CONVICTION
DM me on Instagram for guidance.

Verdict Score

How strong is the evidence for the claims in this review? Higher = more confidence the claims are supported. This does not measure how large the effect is or how important it is compared with other levers.

80 Strong evidence
80–100Strong evidence ◀
60–79Mixed but supportive
40–59Uncertain
0–39Weak support

Treatment Priority — Frozen Shoulder

Evidence-based treatment order for uncomplicated cases. Start at the top — most people don't need the bottom.

Red flags, progressive weakness, or bowel/bladder changes require immediate medical assessment and change this pathway.

1st Line
Education & Prognosis Setting
Self-limiting in most cases (12-30 months). Understanding the timeline reduces anxiety and overtreatment
Pain Management Within Tolerance
Ice/heat for comfort, activity modification. Pushing through sharp pain worsens the freezing stage
Gentle Range of Motion Within Tolerance
Pendulum exercises, assisted ROM. Do not force through pain barriers in the freezing stage
2nd Line
Progressive Stretching & Strengthening
Escalate as pain allows, typically in the thawing stage. End-range holds, rotator cuff loading
Corticosteroid Injection (Freezing Stage)
Short-term pain relief and ROM improvement in early freezing stage. Effect wanes by 6 months
Adjunct
Hydrodilatation
Capsule distension under imaging. Modest ROM gains, mainly in frozen stage
Manual Therapy
Adjunct to exercise in thawing/frozen stages. Not standalone
Limited Evidence
Manipulation Under Anaesthesia
Only after failure of 6+ months conservative management. Risk of humeral fracture
Arthroscopic Capsular Release
Last resort after failed conservative and injection management (>12 months)

Get weekly evidence-based rehab verdicts

Physio conditions reviewed against clinical evidence. What works, what doesn't, and what to do — from a practising physiotherapist.

Subscribe free

Want a coach, not just research?

The Verdict is built by the same team behind Precision Metrics — a physique and health coaching practice with 300+ clients coached. Dr. Seth Holbrook, DPT and Luke Holbrook lead the coaching.

Book a free consultation

Related free research

Pain & Rehab
Baxter's Nerve Entrapment — The Verdict
 Pain & Rehab
Heel Fat Pad Syndrome — The Verdict
 Pain & Rehab
Flexor Hallucis Longus Tendinopathy ("Dancer's Tendinitis") — The Verdict

There are 424 more inside

Conviction-scored verdicts on supplements, nutrition, training, physio, and recovery.

Explore all Get weekly verdicts