The shoulder joint is complex, and pain affecting its use in daily activities and sport is common. This article series will review common shoulder diagnoses, critically assess the associated narratives provided to patients, and describe their impact on real-world outcomes.
We will begin with the assessment of the scapula and its movement. The interaction between the scapula, thorax, and humerus during shoulder movement has been termed “scapulohumeral rhythm” or “glenohumeral rhythm”. The underlying assumption in this context is that normal scapulohumeral rhythm is well-defined in humans. Scapular dyskinesis, then, refers to any abnormality or deviation from “normal” kinematics of the scapula.
Let’s consider what this “normal” scapulohumeral rhythm entails. In order to avoid pathology of the rotator cuff musculature and its associated tendons, it is thought that raising the arm overhead requires the scapula to:
1) upwardly rotate
2) posteriorly tilt
3) externally rotate
Presumably, if the scapula fails to move “correctly” (i.e., dyskinesis) or is otherwise “unstable”, the soft tissues underlying the acromion process can become impinged, painful, and can progressively degenerate or tear — particularly with overuse.
This all sounds superficially plausible from a mechanical perspective. However, despite the frequency with which this issue is confidently diagnosed and treated in practice, this narrative deserves a critical appraisal. This will be the topic of our first discussion.
What is “Normal”? What is “Abnormal”?
The first issue to address is the difficulty in precisely defining “normal” scapular positioning and movement. This has been attempted using various quantitative and qualitative assessments in the research literature ranging from plain visual inspection, to insertion of bone pins with motion tracking devices, to x-ray fluoroscopy [1,2]. There is substantial complexity involved in measuring 3-dimensional scapular movement about multiple axes when it is buried deep under skin and several muscular layers, combined with the wide range of anthropometric variation observed across the population. Additionally, these studies often involve very small sample sizes, making it difficult to confidently generalize normative values for how everyone’s scapulae “should” move.
From a qualitative perspective, most studies on healthy individuals tend to show the general expected pattern of upward rotation, posterior tilting and external rotation with arm elevation. However, quantitative data have predictably been quite mixed, leaving us with difficulty in defining exactly how much the scapula “should” move in a certain way vs. how much is “abnormal”, as well as the extent to which “abnormal” movement is correlative or predictive for pain or functional impairment [1-3].
For example, there is no consensus on a “normal” resting scapular position across the population, and significant postural asymmetries and variation are widely prevalent in asymptomatic individuals as well. If we are to pathologize pain-free, normally functioning individuals solely based on apparent asymmetry or postural/movement preferences that we deem “incorrect”, we automatically diagnose the entire population as abnormal, and the label suddenly loses all meaning and utility.
This reductionist thinking fails to account for adaptations related to occupation, sport, and aging that would be considered “normal” for each individual cohort. For example, the expectation of glenohumeral movement of a baseball player being equivalent to a pianist’s does not account for the specificity of each task, and the long-term adaptations that these tasks induce. A more optimistic argument holds that motor variability reflects the flexibility and adaptability of the neuromuscular system in exploring new or alternative movement solutions, rather than reflecting “dysfunction”.
There is similarly no accepted definition or measure of scapular “stability”, and the term has increasingly become conflated with dyskinesis in clinical practice — in other words, a “dyskinetic” scapula is considered “unstable”, and thus requires targeted stabilization exercises. In this context, McQuade et al.  argue:
“Joint instability is characterized by mobility that exceeds physiological limits without adequate control. When applied to the [scapulothoracic joint], this suggests that the orientation or trajectory of the articulation would not be recoverable following a perturbation. Because this foundational definition is difficult to apply directly to the scapula, the term scapula stability has come to imply “normal” scapula movement on the thorax during upper extremity motions. What becomes obvious is that this clinical definition is not objective or quantifiable, presenting a major challenge for discussing, evaluating, and studying scapula stability.”
The broad variation and inconsistency between individuals is even observed in the setting of diagnosed shoulder pathology [1-4]. For example, among 9 studies comparing upward rotation of the scapula in the setting of diagnosed impingement syndrome versus in healthy controls:
- Four studies showed decreased rotation
- One study showed increased rotation
- The rest showed no difference.
For posterior tilting:
- Four studies showed increased tilt
- Two studies showed decreased tilt
- One study showed no difference.
For internal rotation:
- Two studies studies showed increased rotation
- Five studies showed no difference.
These studies are also limited by the above-described difficulties in accurately measuring scapular positioning and motion, as well as by typical sample size and other methodological concerns.
Additional confounders emerge from evidence that acute fatigue and increases in training load can induce transient “dyskinesis” without inducing shoulder pain. For example, an observational trial of 78 competitive swimmers without shoulder pain examined them for the presence of scapular dyskinesis before a single swim training session and at four subsequent time points throughout the session . Prior to training, examiners reported zero swimmers demonstrating dyskinesis, a finding that is somewhat surprising and suspicious on its own. However, 25 minutes into the session, 37% of swimmers showed evidence of dyskinesis; at 50 minutes, 68%; at 75 minutes, 73%, and at the end of the training session, 82% showed dyskinesis. Despite this rapid increase in apparent shoulder “dysfunction,” none of the participants reported shoulder pain.
Another study of 20 healthy tennis players showed that a fatiguing session of tennis serves induced decreases in scapular upward rotation that spontaneously resolved after 24 hours, compared to no changes observed in a control group . Researchers interpreted this decrease in upward rotation as a potential risk for injury (presumably due to increasing risk for subacromial impingement), and suggested that clinicians therefore consider monitoring athletes’ scapular upward rotation for the first 24 hours after training, though it is unclear what one would do with this information.
In contrast, a separate trial subjected 29 healthy athletes to a fatiguing protocol designed to simulate repeated throwing, and showed evidence of increases in scapular upward rotation, compared to no change in the non-throwing arm . As you might expect, researchers hypothesized that this represented a compensatory “impingement-sparing” response of the shoulder – precisely the opposite conclusion of the tennis study above.
If you find it difficult to reconcile all these findings, you’re not alone. Of course, it would not be surprising if painful or fatigued shoulders might appear to move differently than pain-free shoulders. But these data should make us seriously question the idea that any observed “abnormalities” in scapular motion directly cause pain or pathology, and therefore need to be “fixed”.
Instead, it seems more plausible that these observations, if true, might result from pain or fatigue – and potentially even represent adaptive strategies in certain contexts. Consider someone limping due to a knee injury, which represents an adaptive strategy to protect the knee from further insult. If a clinician mistakenly attributes the patient’s knee pain to the limp, seeing it as a “dysfunctional movement pattern” to be corrected, they are likely to prescribe an inappropriate treatment.
Bottom line: the overwhelming range of inter-individual variability in posture and movement strategies should make us reconsider the utility (or feasibility) of trying to dichotomize “normal” versus “abnormal” movement in a meaningful/useful way.
Despite what we just finished discussing, for now let’s pretend that we actually could clearly define “normal” and “abnormal” scapular positioning and movement. Theoretically, we’d like a way to assess this that is:
1) reliable (i.e., consistent between assessments and/or between examiners – if it’s not reliable, it’s automatically useless)
2) accurate (i.e., correct)
3) valid (i.e., actually measures what it is supposed to measure)
Several studies of visual movement analysis have shown poor reliability for the diagnosis of scapular movement abnormalities or “dysfunction” [1-4]. For example, in one study 11 experienced physiotherapists visually assessed scapular motion in a set of subjects and were asked to classify them as 1) asymptomatic, 2) symptomatic in the left shoulder, 3) symptomatic in the right shoulder, or 4) symptomatic in both shoulders . Therapists were able to accurately classify patients’ symptom status just 58% of the time, with a kappa score of 0.23 (i.e., poor agreement between examiners), and only five subjects had two or more therapists agree on the actual type of movement abnormality observed.
Of course, clinicians may object that we don’t just evaluate patients visually – we can touch and feel and move them around, and these methods may provide better results. Clinical tests such as the lateral scapular slide test (LSST), the scapular dyskinesis test (SDT), and several others have been developed and assessed for clinical use.
Fortunately, we have a 2013 article in the British Journal of Sports Medicine titled “Diagnostic accuracy of scapular physical examination tests for shoulder disorders: a systematic review” that includes eight studies examining the clinical performance and utility of these scapular tests . After analysis, they found:
“None of the studies included in this systematic review reported an ability to discriminate between those with and those without shoulder pain or a specific pathology based on findings from scapular physical examination tests. These findings suggest that scapular asymmetry or motion alterations do not provide any additional clinical examination benefit with regard to diagnosing shoulder pain or pathology.”
This led them to conclude that “Overall, no physical examination test of the scapula was found to be useful in differentially diagnosing pathologies of the shoulder”.
To give a specific example from the paper rather than just quoting the conclusion, one of the included studies by Tate et al. looked at 142 collegiate athletes participating in sports requiring overhead use of the arm . Two therapists applied the SDT and judged subjects to have “no dyskinesis” or “obvious dyskinesis”; subsequent 3-dimensional kinematic analysis generally confirmed that subjects deemed to have “obvious” dyskinesis did, in fact, show less scapular upward rotation, less clavicular elevation, and greater clavicular protraction. However, they also found no difference in shoulder symptoms between the no-dyskinesis and obvious-dyskinesis groups. This should raise the question: how, exactly, did the test help us here? If the test can’t differentiate people with or without symptoms, why do we care about these findings?
Similar findings were reported for the other studies included in this systematic review. It appears that even if tests can detect apparent “differences” in movement with reasonable accuracy and reliability (which is often not the case), they often don’t tell us anything about symptoms or pathology.
In perhaps the most telling study on the topic, when the Scapular Dyskinesis Test was evaluated in 135 subjects with clinicians blinded to the presence of symptoms (i.e., they didn’t know whether or not the subject had shoulder pain), no difference was detected in the prevalence of dyskinesia between the subjects with shoulder pain compared to the healthy controls . Conversely, when clinicians were made aware of patient’s symptoms, they detected dyskinesia more frequently in patients with shoulder pain. In other words, the examiners found a problem more frequently when they knew what they were looking for. This is not particularly surprising given what we know about the role of cognitive biases in clinical practice, and suggests that observer bias significantly influences SDT results, further diminishing its clinical utility.
So it appears that even if these scapular tests are reliable, accurate, and valid (which is typically not the case), they are still not useful in differentiating patients with or without symptoms to rule in or rule out a particular diagnosis [9-13].
A potential counter-argument could be that while detecting scapular dyskinesis right now doesn’t differentiate individuals with or without symptoms, perhaps it might predict future risk of developing shoulder pain, and thus screening might help us intervene sooner. A 2018 meta-analysis of five studies by Hickey et al. initially appears to support this idea . The included studies looked at a total of 419 athletes and found that 56 of the 160 (35%) athletes deemed to have scapular dyskinesis experienced shoulder pain over the subsequent 9-24 months, compared to 65 of 259 (25%) without dyskinesis subsequently experiencing shoulder pain.
Based on these numbers, baseline screening would provide a diagnostic accuracy of 54% for predicting future shoulder pain – that is, equivalent to flipping a coin to determine whether someone would develop shoulder pain in the future. The authors themselves argue that screening for scapular dyskinesis is therefore not helpful, and this sentiment (as well as additional criticisms) are levied by Littlewood et al. in their editorial response to this paper . Given what we know about 1) the effect of fatigue and training load on scapular dyskinesis (see above), and 2) the established role of fatigue and training load on injury risk , another plausible explanation for these findings may be that the slightly higher risk of shoulder pain in the dyskinesis group is reflective of a higher level of fatigue or training load, and the observed “dyskinesis” is simply a surrogate marker for this.
So again, despite what we’ve covered so far, for the sake of further discussion let’s pretend that:
1) we can clearly define “normal” vs. “abnormal” (even though we can’t),
2) that we can accurately and reliably differentiate these using the clinical exam (even though we can’t),
3) that “abnormalities” do reliably correlate with pain or impairments in function (even though they don’t).
We would then expect that correcting “abnormal” scapular positioning or movement would reliably restore normal pain-free function, right? So, what happens if we intervene with treatments targeted at improving scapular dyskinesis or scapular “stabilization”?
A study on 59 patients diagnosed with shoulder impingement syndrome and scapular dyskinesis showed that a 6-week exercise intervention produced significant improvements in shoulder pain and function that were maintained at 6-month follow-up, without any significant changes in scapular mechanics .
A similar trial randomized patients with shoulder impingement syndrome to “scapular-focused treatment” versus “control” therapy, and found that both groups improved, although the “scapular-focused” treatment group improved to a greater extent … despite no actual changes in any metric of scapular mechanics over the course of the 3-month trial 
Although research on “scapular-focused rehabilitation” tends to show good outcomes in terms of pain and function, these findings (and others ) suggest that the mechanism of improvement in shoulder pain and function is probably independent of changes in scapular mechanics — and should make use question whether we can (or should) deliberately modify apparent dyskinesis at all [3,20,21]. Furthermore, if we can’t, we must consider the potential nocebo effects of delivering this narrative to our patients and inducing fear of pain or injury due to such ill-defined “abnormalities” or moving in a way that we arbitrarily deem “wrong”.
And finally, to the extent that we do observe a change in scapular mechanics over the course of rehabilitation — which could be attributed to the numerous problems with accurate, reliable, and non-blinded outcome measurements — it seems more plausible that it would be the result of pain relief and functional improvement, rather than the cause … Kind of like your limp going away and starting to walk normally again as your knee or foot injury heals. We observe the same phenomenon occurring in research on stabilization exercise or other “corrective” exercise intervention for movement abnormalities in the setting of low back pain and other pain conditions.
All of this should make us doubt whether scapular dyskinesis is 1) useful to “measure” or assess or 2) a modifiable target for specific rehabilitation at all. Indeed, McQuade’s 2016 paper on scapular stabilization states :
“… a theoretical connection between dyskinesis, scapula instability, and shoulder pathology has become widely accepted without clear evidence that such connections exist. The concern here is that the presence of dyskinesis may lead to quick conclusions about pathomechanisms and interventions that stifle clinical reasoning and decision-making. The notion that scapula dyskinesis is an indicator of instability and linked to pathology is not supported by recent independent literature reviews.”
Instead, we should consider whether there be something else going on here than “moving wrong” or simple mechanical irritation in the subacromial space. For this next step, we’ll dig into the Impingement Syndrome in more detail in the next article.
Thanks to Michael Ray, MS, DC, Derek Miles, DPT, and Thomas Campitelli, DTFP for their assistance in editing this article.
 Kibler et al. Clinical implications of scapular dyskinesis in shoulder injury: the 2013 consensus statement from the ‘Scapular Summit’. Br J Sports Med. 2013 Sep;47(14):877-85.
 Willmore et al. “Scapular Dyskinesia: Evolution towards a Systems-Based Approach.” Shoulder & Elbow 8.1 (2016): 61–70.
 McQuade et al. Critical and Theoretical Perspective on Scapular Stabilization: What Does It Really Mean, and Are We on the Right Track? Phys Ther. 2016 Aug;96(8):1162-9.
 Ludewig PM, Reynolds JF. The association of scapular kinematics and glenohumeral joint pathologies. J Orthop Sports Phys Ther 2009;39:90–104.
 Madsen et al. Training Induces Scapular Dyskinesis in Pain-Free Competitive Swimmers: A Reliability and Observational Study. Clinical Journal of Sport Medicine, 21(2), 109–113.
 Rich et al. “Scapular Upward-Rotation Deficits After Acute Fatigue in Tennis Players.” Journal of Athletic Training 51.6 (2016): 474–479.
 Maenhout et al. “Acromiohumeral Distance and 3-Dimensional Scapular Position Change After Overhead Muscle Fatigue.” Journal of Athletic Training 50.3 (2015): 281–288.
 Hickey et al. Accuracy and reliability of observational motion analysis in identifying shoulder symptoms. Man Ther. 2007 Aug;12(3):263-70.
 Wright et al. Diagnostic accuracy of scapular physical examination tests for shoulder disorders: a systematic review. Br J Sports Med. 2013 Sep;47(14):886-92.
 Tate et al. A clinical method for identifying scapular dyskinesis, part 2: validity. J Athl Train 2009;44:165–73.
 Plummer et al. Observational Scapular Dyskinesis: Known-Groups Validity in Patients With and Without Shoulder Pain. J Orthop Sports Phys Ther. 2017 Aug;47(8):530-537.
 Lange et al. The reliability of physical examination tests for the clinical assessment of scapular dyskinesis in subjects with shoulder complaints: A systematic review. Physical Therapy in Sport, 26, 64–89.
 Nijs J, Roussel N, Vermeulen K and Souvereyns G. Scapular positioning in patients with shoulder pain: a study examining the reliability and clinical importance of 3 clinical tests. Arch Phys Med Rehabil 2005; 86: 1349–55. Hickey et al. Scapular dyskinesis increases the risk of future shoulder pain by 43% in asymptomatic athletes: a systematic review and meta-analysis. Br J Sports Med. 2018 Jan;52(2):102-110.
 Littlewood et al. Scapular dyskinesis and shoulder pain: the devil is in the detail. Br J Sports Med 2018;52:72-73.
 Jones et al. Training Load and Fatigue Marker Associations with Injury and Illness: A Systematic Review of Longitudinal Studies Sports Med (2017) 47: 943.
 McClure et al. Shoulder function and 3-dimensional kinematics in people with shoulder impingement syndrome before and after a 6-week exercise program. Phys Ther. 2004 Sep;84(9):832-48.
 Struyf et al. Scapular-focused treatment in patients with shoulder impingement syndrome: a randomized clinical trial. Clinical Rheumatology, 32(1), 73–85.
 Turgut et al. Effects of Scapular Stabilization Exercise Training on Scapular Kinematics, Disability, and Pain in Subacromial Impingement: A Randomized Controlled Trial. Archives of Physical Medicine and Rehabilitation. 2017; 98(10), 1915–1923.e3.
 Bury et al. Effectiveness of scapula-focused approaches in patients with rotator cuff related shoulder pain: A systematic review and meta-analysis. Man Ther. 2016 Sep;25:35-42.
 Reijneveld et al. Clinical outcomes of a scapular-focused treatment in patients with subacromial pain syndrome: a systematic review. Br J Sports Med. 2017 Mar;51(5):436-441.