Clinical synthesis based on the most recent meta-analyses and CPGs: JOSPT 2025 Desmeules (rotator cuff CPG), JOSPT 2022 Lafrance CPG (return to work), BMJ Rapid Recommendation Vandvik 2019 (decompression), meta-analyses Longo 2021, Ryösä 2017, MOON cohort Kuhn 2013/Kuhn 2024 (10-yr).

Rotator cuff tear is a clinical condition of essentially degenerative origin in middle-aged and older people, distinct from acute traumatic tears in young individuals. It involves loss of integrity of one or more of the four cuff tendons: supraspinatus (most often), infraspinatus, teres minor and subscapularis.¹ The symptomatic presentation combines pain (often at night and worsened in abduction), functional weakness and progressive loss of active mobility.²,³ The fundamental nuance that should guide the clinician: a tear on imaging does not equate to a diagnosis of the cause of symptoms. The prevalence of asymptomatic tears is so high that it is essential to combine clinical examination and imaging to confirm that the lesion is responsible. 🦴

How is this condition defined, who does it affect and what are the risk factors?

The contemporary nomenclature, harmonised by the JOSPT 2025 Desmeules CPG on rotator cuff tendinopathy⁴ and the JOSPT 2022 Lafrance CPG on return to work⁵, distinguishes several entities within a single continuum:

• Rotator cuff-related shoulder pain (RCRSP) : umbrella term covering tendinopathy, partial tears and symptomatic degenerative complete tears (Lewis 2018).⁶
• Partial tear : intratendinous, bursal-side or articular-side lesion, classified according to Ellman.
• Full-thickness tear (transfixing): the loss of continuity extends through the whole thickness of the tendon, from the articular side to the bursal side.
• Massive tear : involves at least 2 tendons or a retraction > 5 cm (DeOrio and Cofield criteria).

📈 The age-dependent prevalence is the key epidemiological finding. The systematic meta-analysis Teunis 2014 (J Shoulder Elbow Surg, 30 studies, n=6,112 shoulders, PMID 25441568) establishes an ascending prevalence curve: 9.7% before age 30, 19.1% between 30 and 39, 35.4% between 60 and 69, 62% after age 80.⁷ The Japanese mass-screening study Yamamoto 2010 (J Shoulder Elbow Surg, n=683 participants, 1,366 shoulders, PMID 19540777) finds an overall prevalence of 20.7% of shoulders in the general population, and 16.9% of subjects without symptoms carry a tear.⁸ A second Japanese mass screening, Minagawa 2013 (a village in Akita Prefecture, n=664 residents examined, PMID 24403741), finds 22.1% full-thickness tears, of which 34.7% were symptomatic and 65.3% asymptomatic : asymptomatic tears are twice as frequent there as symptomatic tears.⁹

The risk factors of degenerative tear are now well characterised. Age is the dominant non-modifiable factor. Among metabolic factors, the Giri 2023 meta-analysis (Ann Phys Rehabil Med, SR/MA, PMID 35257948) is the modern reference source: it finds an OR 1.49 [95% CI 1.43-1.55] for diabetes, OR 1.40 [1.19-1.65] for hypertension and OR 1.48 [1.42-1.55] for dyslipidaemia in the association with tear.¹⁰ Smoking remains a major factor: Carbone 2012 (J Shoulder Elbow Surg, n=408, PMID 21524922) demonstrated a dose-response relationship between the intensity/duration of smoking and tear size.¹¹ The twin study Gwilym 2009 (J Bone Joint Surg Br) establishes a genetic component with an increased risk in first-degree relatives.¹²

What happens in the cuff and how does it evolve naturally?

🧬 The pathophysiology of degenerative tear brings together factors that are intrinsic (biological alteration of the tendon: collagen disorganisation, reduced tenocytes, increased matrix metalloproteinases) and extrinsic (repeated mechanical stresses, accessory subacromial impingement).¹³ The supraspinatus tendon has a critical zone of reduced vascularisation, described by Rathbun & Macnab 1970 and Lohr & Uhthoff 1990, located about 1 cm in front of its insertion on the greater tuberosity: it is the preferred site of degenerative tears.¹⁴ The role of subacromial impingement as a primary cause is today largely questioned: it is rather seen as a secondary aggravating factor on an already weakened tendon.¹⁵

The natural history was clarified by several longitudinal cohorts and recent meta-analyses. Two sources shed light on distinct aspects: the prospective cohort Keener 2015 (J Bone Joint Surg Am, 224 subjects followed with annual ultrasound, median follow-up 5.1 years, prognostic level II) for the structural side, and the SR/MA Khatri 2019 (Am J Sports Med, 57 randomised trials, PMID 29963905) for the clinical side.¹⁶,¹⁷

• 61% of full-thickness tears enlarge in Keener 2015, versus 44% of partial tears and 14% of control shoulders; the median time to enlargement is 2.8 years.
• In this same cohort, made up of initially asymptomatic tears, enlargement is associated with the onset of new pain : it occurs in 50% of shoulders with a full-thickness tear, versus 46% of partial tears and 28% of controls. The association is measured in this specific population and does not transfer as such to the worsening of an already symptomatic tear.
• Enlargement is accompanied by progression of the muscle degeneration of the supraspinatus: it worsens in 30% of shoulders whose tear enlarged, versus 4% of those whose tear remained stable (Keener 2015).
• On the clinical side, Khatri 2019 does not measure lesion enlargement: its conclusion is that full-thickness tears improve over time on the Constant score, whether they are treated surgically or not.
• Fatty infiltration of Goutallier ≥ 3 is considered irreversible and constitutes a point of no return for effective surgical repair.

In a multicentre cohort of 127 patients aged 50 to 75 treated without surgery from the outset, among the 94 and then 70 patients reviewed with MRI, muscle degradation (54% at 2 years, 71% at 5 years) was not correlated with the Constant score, and most tears remained stable (70% at 2 years, 60% at 5 years):¹⁸ in a patient whose function is stable, more fat on MRI does not by itself signal worsening. The patient is informed of the risk of tear progression and followed;¹⁹ in adults under 70, surgical referral is considered if function does not improve at 3 to 6 months.

The assessment of a degenerative tear rests on the convergence of several sources of information. No single test and no imaging sign is sufficient to confirm the clinical diagnosis of symptomatic tear: the nuance is crucial, since the prevalence of asymptomatic tears is high (Yamamoto 2010; Tempelhof 1999; Sher 1995).¹,² Modern diagnostic logic consists of: (1) establishing a pre-test probability from the history, (2) modulating it with a cluster of clinical tests, (3) requesting imaging only to confirm or clarify after an initial period or to guide a therapeutic decision.³

Which questions should you ask to understand the patient and their history?

🤔 History remains the most discriminating diagnostic tool. The relevant elements are:

• Age : high pre-test probability beyond 60 years, low before 40 in the absence of trauma.⁴
• Location : the pain is typically anterolateral in the arm, often radiating towards the deltoid V, rarely below the elbow. Radiation below the elbow should raise suspicion of a cervical origin.⁵
• Character of the pain : dull and deep at rest; triggered by movements of elevation above the head, carrying loads, dressing, hair care.⁶
• Night pain : extremely characteristic. Lying on the affected side typically wakes the patient. Its presence markedly increases the pre-test probability of a symptomatic tear.⁷
• Onset : insidious and progressive for degenerative tear; sudden after an ordinary movement, it may signal the decompensation of a pre-existing asymptomatic tear. An overt traumatic episode (violent fall, sudden traction) should lead to reclassification as an acute traumatic tear, for which management is different (often surgical, especially in young people).⁸
• Functional impact : assess with a valid PROM (SPADI, DASH, Constant-Murley, Oxford Shoulder Score). The SPADI is particularly relevant in physiotherapy (MCID 8-13 points according to Roy 2009; Tveita 2018).⁹,¹⁰
• Systemic comorbidities : diabetes, dyslipidaemia, hypertension, active smoking (see Giri 2023 SR/MA, OR 1.40-1.49).¹¹
• Psychosocial and occupational factors : kinesiophobia, self-efficacy, expectations (Chester 2018 BJSM), working conditions, constraint of overhead movements (screening with the Lafrance 2022 CPG).¹²,¹³

Which clinical tests should be performed, and which other conditions must be ruled out?

🎯 The physical examination should look for four pieces of information: (1) integrity of the individual cuff tendons, (2) presence of subacromial impingement, (3) involvement of neighbouring structures (acromioclavicular joint, biceps), (4) signs of cervical or neurological origin.

Inspection & palpation : look for atrophy of the fossae (supraspinous and infraspinous), a sign that is both sensitive and highly specific (> 90%) of an extensive chronic tear (Walch; Codman).¹⁴ Palpation of the deltopectoral groove may reveal swelling of the long head of biceps in the event of an intra-articular rupture.

Specific cuff tests and their diagnostic properties : the meta-analyses Hegedus 2012 (BJSM, PMID 22773322)¹⁵ and Lädermann 2020-2021 (KSSTA, PMID 32725446)¹⁶ summarise the pooled performances:

The combination of tests (for example painful arc + drop arm + ER resistance) is more discriminating than any isolated test.¹⁵,¹⁶ The drop arm and the lift-off are SpIN tests (Specific In = “Specific, Positive → IN”), very useful for confirming a suspicion; the painful arc and the Jobe are SnNOUT tests (Sensitive Negative → OUT), more useful for ruling out.

Essential differential diagnosis :

• C5-C6 cervical radiculopathy : Spurling test, neurological examination of dermatomes/myotomes, biceps tendon reflexes. May mimic a Jobe deficit or a positive lift-off through neurogenic weakness.¹⁷
• Acromioclavicular arthropathy : localised palpation of the AC joint, cross-body adduction test, pain localised to the lateral end of the clavicle (Hermans 2013).¹⁸
• Adhesive capsulitis (frozen shoulder) : global loss of range of motion, both active AND passive, particularly external rotation: a limitation typically absent in isolated rotator cuff tear.¹⁹
• Parsonage-Turner syndrome (acute brachial neuritis): intense acute pain rapidly followed by profound weakness and rapid atrophy; EMG guides (see Cureus 2020 case, PMID 32742859).²⁰
• Calcific tendinitis, bursitis, tumour (rare).

Should patients with symptomatic rotator cuff tear be classified, and for what benefits?

💡 Yes: classification has prognostic and therapeutic value. The two reference systems, validated for more than 30 years, remain the standard:

• Patte 1990 classification (PMID 2323151): assesses the tendon retraction on MRI/CT arthrography: Stage I (little retraction, up to the humeral apex), Stage II (up to the top of the humerus), Stage III (beyond, massive retraction).²¹
• Goutallier 1994 / Fuchs 1999 classification : assesses the fatty degeneration of the muscle (initially on CT, adapted to MRI by Fuchs): Stage 0 (normal), Stage 1 (some fatty streaks), Stage 2 (< 50% fatty muscle), Stage 3 (≈ 50%), Stage 4 (> 50% fat).²² A stage ≥ 3 is considered a point of no return for surgical repair.
• The Ellman classification for partial tears: grade 1 (< 3 mm depth), 2 (3-6 mm), 3 (> 6 mm), with specification of the side (bursal/articular/intratendinous).²³
• The size of the full-thickness tear (DeOrio & Cofield 1984, taken up by Snyder): small (< 1 cm), medium (1-3 cm), large (3-5 cm), massive (> 5 cm or ≥ 2 tendons).²⁴

Clinical benefits of classification: (1) it informs the conservative vs surgical decision, (2) it defines realistic expectations to share with the patient, (3) it guides rehabilitation (precautions vs rapid progression), (4) it standardises communication between physiotherapist, physician, surgeon and radiologist.

The contemporary management of symptomatic degenerative rotator cuff tear has undergone a major paradigm shift since 2013: the prospective multicentre cohort MOON (Multicenter Orthopaedic Outcomes Network) led by Kuhn et al. established that 75% of patients with atraumatic full-thickness tear avoid surgery at 2 years with a structured physiotherapy programme (n=433, JSES 2013, PMID 23540577).¹ The 10-year follow-up (Kuhn 2024, JBJS) confirms the durability of the conservative response, with maintained functional improvement in most patients who were not initially operated on.² 🦴

Where do you start? What is the recommended hierarchy of interventions?

The hierarchy of evidence-based interventions for symptomatic degenerative tear rests on the closest recommendations: the JOSPT 2025 Desmeules CPG³ (by clinical continuity, see below), the JOSPT 2022 Lafrance CPG (return to work)⁴, the BMJ Rapid Recommendation Vandvik 2019 on subacromial surgery⁵ and the 2026 Dutch update, which addresses isolated supraspinatus tear.²³

A clarification of scope is needed from the outset: the JOSPT 2025 CPG itself delimits its scope to rotator cuff tendinopathy, calcific or not, and to partial tears; full-thickness tears do not fall within its declared scope.³ Its recommendations on education, progressive exercise and the absence of superiority of one exercise format are transposed here to degenerative full-thickness tear by clinical continuity, the MOON cohort (Kuhn 2013, Kuhn 2024) remaining the direct source for this population.¹,²

1. Patient education + therapeutic exercise (first line).
2. Adjuvant manual therapy : may be added for short-term pain management (moderate recommendation).
3. Physical modalities and shockwave therapy : optional adjuvants depending on the clinical profile (weak to moderate evidence).
4. Injections (corticosteroids, PRP, etc.) : to be discussed case by case, never first line (variable evidence, transient effects).
5. Surgical referral : considered after failure of at least 12 weeks (and more generally 3-6 months) of well-conducted conservative management, or from the outset in specific situations (acute traumatic tear in a young active person, major functional deficit that is not improving).

The meta-analyses comparing surgery vs conservative treatment (Ryösä 2017 Disabil Rehabil, Longo 2021 BMC Musculoskelet Disord) do not demonstrate any clinically important superiority of surgery for degenerative tear at 1 and 2 years: function does not differ, and the slight advantage of surgery for pain at 1 year remains below the threshold of clinical importance.⁶,⁷ The randomised FINO trial (Kukkonen 2014/2015) compares physiotherapy, acromioplasty + PT, and repair + acromioplasty + PT in patients > 55 years with non-traumatic tear: no significant difference between the three groups at 1 and 2 years.⁸,⁹ A single-centre randomised trial (Moosmayer: 103 patients, full-thickness tears of 3 cm at most, 57% of them with acute onset on a chronic background; primary repair versus physiotherapy with possible secondary repair) finds an advantage of primary repair, small at 5 years according to its authors (Constant +5.3 points),²⁵ which grows at 10 and then 15 years (+11.8 points; 15 of the 51 patients in the physiotherapy arm operated on secondarily), the 26 unrepaired tears going on average from 16.2 to 31.6 mm.²² The 2026 Dutch update judges the comparative evidence very uncertain; for adults under 70 with an isolated non-traumatic supraspinatus tear, it proposes starting with non-surgical treatment and considering repair if there is no improvement at 3 to 6 months, and asks that the patient be informed of the risk of tear progression and followed.²³ The CSAW trial (Beard 2018, Lancet) of isolated subacromial decompression versus placebo (diagnostic investigative surgery without any procedure) demonstrated no clinically significant difference, supporting the BMJ Rapid Recommendation Vandvik 2019 against isolated decompression for chronic shoulder pain.¹⁰,⁵

What is the place of exercise, and is there a superior approach? 💪

Therapeutic exercise is the cornerstone of management. Its superiority over no treatment is established; the contemporary question is: is one type of exercise superior to another?

In rotator cuff tendinopathy, no exercise format stands out : recent reviews (Naunton 2020, Dominguez-Romero 2021, Pieters 2020 JOSPT update) do not distinguish between the formats (eccentric, concentric, isometric, slow heavy loading, functional or scapula-focused exercises),¹¹,¹²,¹³ and the Lafrance 2024 meta-analysis, which compares programmes by their content, finds motor control probably slightly superior to a non-specific programme for disability (moderate certainty, small effect).²⁴ Full-thickness tear is included there only by clinical continuity. The principles of progression and individualisation matter more than the choice of a particular format.

The RoCTEx trial by Ingwersen 2017 (Orthopaedic Journal of Sports Medicine, and not BJSM 2021 contrary to some erroneous citations) compares progressive heavy-load exercise vs traditional low-load exercise in 100 patients with rotator cuff tendinopathy: it does not demonstrate any superiority of the heavy-load protocol.¹⁴

Several practical principles emerge:

• Start with exercises below the pain threshold, favouring the quality of scapulohumeral movement.
• Include exercises targeting the scapular stabilisers (serratus anterior, middle and lower trapezius).
• Progress in load, complexity and range according to tolerance (Mertens 2022 narrative review).¹⁵
• Encourage simple home programmes (2-4 exercises) to promote long-term adherence (Littlewood 2014, Edwards 2016).¹⁶,¹⁷
• Pain ≤ 3-4/10 during or after exercise, which does not worsen within 24 h, is acceptable.

Adjuvant therapies: what are manual therapy, shockwave therapy and injections worth?

Concerning therapeutic ultrasound, low-level laser and heat/cold modalities: no demonstrated added benefit over physiotherapy, on low-quality evidence, that would justify recommending them routinely (Page 2016 Cochrane).¹⁸

Beyond the physical: how do you educate the patient and address psychological factors? 🧠

Patient education is a major determinant of therapeutic success, and is no longer regarded as a “supplement” but as a central intervention.¹⁹ The key messages to convey:

• The presence of a tear on imaging does not necessarily mean that the tear is the cause of the pain (corollary of the data of Yamamoto 2010, Sher 1995, Teunis 2014).
• “Hurt does not equal harm”: moderate pain during exercise is not a signal of additional tissue damage.
• Function can be restored without structural healing of the tear: this is the explicit aim of conservative treatment.
• Movement is safe and beneficial; inactivity is harmful.

The psychological factors (kinesiophobia, catastrophising, self-efficacy, expectations) are powerful predictors of outcome, independent of tear size or biomechanical properties:

• Chester 2018 BJSM (PMID 27445360): in a multicentre cohort of n=1,030 patients with shoulder pain, psychological factors (kinesiophobia, low expectations, catastrophising) are the most powerful predictors of outcome at 6 months.²⁰
• Chester 2019 BJSM (PMID 30626599): CART analysis identifies self-efficacy (PSEQ questionnaire) as the main predictor of persistent pain at 6 months.²¹
• Lafrance 2022 CPG : explicitly recommends screening for psychosocial factors as an integral part of the initial assessment.⁴

The prevention of recurrences and the sustaining of results rest on two pillars: (1) the empowerment of the patient through a well-structured self-management programme, (2) a progressive, criteria-based return to activities and sport. The JOSPT 2022 Lafrance and JOSPT 2025 Desmeules CPGs converge on these lines, the latter for tendinopathy and partial tears, the only scope it declares it covers.¹,² The aim is not structural healing (rarely obtained without surgery), but lasting functional restoration with the capacity for self-management in the event of a future pain flare-up.³

How do you make the patient an active participant in their recovery through self-management?

👍 Self-management is the process by which the patient, guided by the clinician, becomes the main actor in their rehabilitation and in the future management of their condition. The key elements:

• Structured therapeutic education : nature of the condition, role of loading, dissociation between imaging and symptoms, safety of movement.
• Simple home programme (2-4 exercises) : Littlewood 2014 showed an efficacy comparable to a complex supervised programme.⁴ Edwards 2016 and Pieters 2020 confirm that long-term adherence is inversely correlated with the complexity of the programme.⁵,⁶
• Load progression according to tolerance (pain ≤ 3-4/10 acceptable during exercise provided there is no worsening at 24 h).
• Self-monitoring : teach the patient to identify the signs of improvement (strength, function, sleep quality) and the signs of worsening that warrant reassessment.
• Strategies for pain flare-ups : temporary reduction of load without complete cessation, sleep management, occasional anti-inflammatory technique.

Self-efficacy (measured by the Pain Self-Efficacy Questionnaire, PSEQ) is one of the most powerful predictors of long-term outcome (Chester 2019).⁷ Strengthening self-efficacy is therefore an explicit objective for the clinician.

When and how should a safe return to activities and sport be planned?

The planning of return to activities and sport is based on functional and psychological criteria rather than on a fixed timetable.⁸,⁹ The systematic review Migliorini 2023 (and, previously, the SR of Klouche 2016) established an overall return-to-sport rate of the order of 60-85 % depending on the sport, the level of practice and the type of management (conservative or surgical).¹⁰

The recommended sequence:

1. Prerequisite : full active mobility without pain, strength ≥ 80% relative to the healthy side (measured with an isometric dynamometer).
2. Functional tests : wall/incline/floor push-ups according to progression, progressive throwing (overhead sports), scapulohumeral endurance, scapular control in movement.
3. Progressive resumption : crescendo intensity and volume, sport-specific movement in a controlled and then open environment.
4. Psychological assessment : confidence in movement, absence of residual kinesiophobia (PSEQ, Tampa Scale for Kinesiophobia).
5. Communicating realistic expectations : for degenerative tear, return to the same pre-symptom level is not systematic; the main aim is satisfactory function in daily life and in the activities that are meaningful to the patient.

The MOON cohort followed at 10 years (Kuhn 2024) shows that among patients who succeeded with the initial conservative treatment, the majority remain satisfied at 10 years without surgery, although a minority are converted to surgery during follow-up (typically for symptomatic worsening or loss of function).¹¹

🧐 The analysis of published clinical cases to complement randomised trials offers a contextualised perspective. This section presents three types of situation: (1) a “classic” course from the MOON cohort, (2) cases of diagnostic mimics taken from the PMC literature, (3) a complex case of massive tear with advanced degeneration.

Analysis of a “classic” case: the MOON cohort and its lessons

The MOON cohort (Multicenter Orthopaedic Outcomes Network, Kuhn 2013, J Shoulder Elbow Surg, n=433, PMID 23540577) is the archetypal case-mix whose characteristics are shared by every patient with symptomatic degenerative tear:¹

• Mean age 62 years, with no identified acute trauma.
• Full-thickness tear of the supraspinatus ± infraspinatus, mean size 1.5-2 cm.
• Insidious pain, night pain, weakness in elevation.
• Structured programme of 6-12 weeks: phase 1 pain control + recovery of range of motion; phase 2 progressive strengthening of the stabilisers; phase 3 functional reintegration.
• Result at 2 years : 75% avoid surgery, significant improvement in ASES and SST.
• Result at 10 years (Kuhn 2024): most initial responders remain satisfied with maintained function.²

The key lesson: structural healing of the tear is not necessary to obtain an excellent functional result. The proposed mechanism is neuromuscular optimisation of the intact muscles (deltoid, scapular muscles, healthy parts of the cuff).

The diagnostic challenge: when a symptomatic rotator cuff tear mimics another condition

🕵️‍♂️ Published cases of diagnostic “mimics” are valuable reminders. Several verified PMC publications are representative:

• Parsonage-Turner syndrome initially presenting as a rotator cuff tear: Ibrahim 2020 (Cureus, PMID 32742859, PMC7388810) describes a case of atypical presentation 10 weeks after surgical repair of a rotator cuff tear: EMG finds denervation of the biceps, deltoid, supraspinatus and infraspinatus without cervical involvement, the electrophysiological signature of the syndrome.³
• Atypical Presentation Confounded by Surgical Rotator Cuff Injury (PMC5623205): rotator cuff surgery preceded the onset of PTS, delaying the diagnosis and compromising rehabilitation.⁴
• C5-C6 cervical radiculopathy mimicking a supraspinatus deficit: neurogenic weakness can be clinically indistinguishable from a full-thickness tear without targeted EMG/cervical MRI (Slaven 2010 J Man Manip Ther, PMID 22131792).⁵
• Adhesive capsulitis at the early painful stage: the loss of passive mobility in external rotation (typically absent in isolated rotator cuff tear) should raise suspicion (Hand 2008, JSES).⁶

Study of a complex case: massive tear with Goutallier 3-4

Massive tears with advanced fatty degeneration (Goutallier ≥ 3) represent a therapeutic challenge. Several studies have documented that these patients can obtain a significant functional improvement with conservative treatment, although the prognosis is more guarded than with a Goutallier ≤ 2.

• Collin 2014 (J Shoulder Elbow Surg, n=66 patients with massive tear not treated surgically) : retrospective analysis of the predictive factors of pseudoparalysis. The loss of active shoulder function is not determined by size alone, but by the combination of massive tear + advanced degeneration.⁷
• Levy 2008 (Tendon transfers in irreparable rotator cuff tears, review) : even with a massive tear, optimisation of the remaining muscles (deltoid, pectoralis major, latissimus dorsi) can restore useful function.⁸
• The rehabilitation programme prioritises: glenohumeral centring, scapular control, strengthening of the anterior and middle deltoid (which becomes the functional pivot), endurance training.

Refractory cases despite well-conducted conservative management may be referred for salvage surgery : partial repair, tendon transfer (latissimus dorsi, pectoralis major, lower trapezius), reverse shoulder arthroplasty in older patients. These options are to be discussed with a shoulder surgeon case by case.

🧐 Applying the recommendations in everyday practice requires a combination of clinical judgement, standardised measurement tools and an ability to navigate the individual and systemic barriers to implementation.

When, and to which other health professionals, should patients be referred?

🧭 The referral decision depends on the following criteria:

1. Red flags (see diagnosis section): immediate medical referral (general practitioner, emergency department, specialist depending on the context).
2. Failure of conservative treatment after at least 12 weeks (and generally 3-6 months) of well-conducted management, without significant functional improvement: referral to a shoulder surgeon to discuss the imaging work-up and the options.
3. Young patients with acute traumatic tear (clear mechanism, MRI showing a recent full-thickness tear with little retraction and good-quality muscle): early surgical referral is often recommended.
4. Dominant psychosocial factors (severe kinesiophobia, catastrophising, depression, anxiety): referral to a pain psychologist or psychotherapist (Coronado 2020).¹
5. Adaptation of the workstation or of activities of daily living (ADL): referral to an occupational therapist or occupational physician.²
6. Pharmacology : if NSAIDs, step 1-2 analgesics or an injection are under discussion, collaboration with the general practitioner or the rheumatologist.

How do you measure outcomes and overcome barriers to implementation?

📊 Objective measurement of outcomes is essential. The validated tools recommended for the shoulder:

• SPADI (Shoulder Pain and Disability Index) : 13 items, 5 pain + 8 function, MCID 8-13 points. Pivotal tool in physiotherapy.³
• DASH (Disabilities of the Arm, Shoulder and Hand) or its short version QuickDASH : 30 or 11 items, DASH MCID ≈ 10 points.⁴
• Constant-Murley Score : combines pain (15 pts), ADL (20 pts), mobility (40 pts), strength (25 pts). MCID about 10-15 points depending on the studies.⁵
• Oxford Shoulder Score (OSS) : 12 items, MCID about 4-5 points (Tveita 2018).⁶
• ASES (American Shoulder and Elbow Surgeons Score) : used in research and surgery.
• PSEQ (Pain Self-Efficacy Questionnaire) for screening psychological factors.

Assessment at baseline, at 6 weeks, at 12 weeks and at 6 months makes it possible to quantify the trajectory and objectify progress.

Overcoming barriers to implementation

🚧 The main barriers to implementing EBP in physiotherapy practice have been identified in several systematic reviews:

• Scurlock-Evans 2014 (J Contin Educ Health Prof, SR): lack of time, limited critical appraisal skills, lack of access to databases, lack of institutional support.⁷
• Bork 2023 (BMC Health Serv Res, qualitative study of German physiotherapists): confirms the persistence of these barriers, and adds the lack of mentoring and clinical leadership.⁸
• Tools to overcome them: pre-appraised resources (Cochrane Summaries, JOSPT-type CPGs), continuing education, mentoring, journal clubs, a work environment that supports learning.

Successful implementation of EBP is not only an individual matter but a systemic issue : funding models, resources allocated to continuing education, access to scientific resources.⁹