Sports Medicine - The Downward Dog: An Age Old Yoga Exercise With Modern Day Shoulder Rehabilitation Application ׀ by John O'Halloran DPT, OCS, ATC, CSCS, Cert MDT


Dr. John W. O'Halloran, DPT, PT, OCS, ATC, CSCS, cert MDT, is a licensed Physical Therapist and Athletic Trainer with over 24 years of experience in the field of rehabilitation. He is a board certified orthopaedic clinical specialist by the American Physical Therapy Association. He earned his post-professional Doctor of Physical Therapy from Temple University. Currently, Dr. O'Halloran is a director of physical therapy/sports medicine at Southeastern Orthopedics in Greensboro, North Carolina, co-owner of GOSMC Properties, LLC and owner of O'Halloran Consulting, LLC in Summerfield, North Carolina. He has also worked in a variety of settings including universities and hospitals. Dr. O'Halloran is also a former orthopedic instructor at the physical therapy assistant program at Guilford Technical Community College in Greensboro, North Carolina. In 1999, he became credentialed with the McKenzie Institute in the mechanical diagnosis and treatment of the spine. Dr. O'Halloran is also a certified functional capacity evaluator in the Blankenship Method. His unique evaluation and treatment skills make him a sought after clinical instructor for physical therapy and athletic training topics. He has spoken both locally and internationally on topics such as sports specific rehabilitation of the shoulder, spine rehabilitation and treatment of foot and ankle injuries. Dr. O'Halloran has studied orthopedic and sports therapy abroad in Australia and New Zealand. His post graduate work has included manipulative therapy. His wide range of clinical experience combined with his extensive knowledge in the fiscal management of physical therapy outpatient rehabilitation services make him a very knowledgeable and skilled instructor. For a complete listing of upcoming courses for John O’Halloran, click here.

The Downward Dog: An Age Old Yoga Exercise With Modern Day Shoulder Rehabilitation Application

Today’s rehabilitation environment involves providing services that ensure quality care that is designed to meet the needs of the patient, payer and provider. Practicing in this environment is quite a change from the days of yesteryear when you could treat a rotator cuff repair for 30 visits and no one would blink an eye. You were paid by performing a whole lot of intervention and reimbursement was determined by the adding up of units of CPT codes and procedures. Contract negotiations were based on a “fee schedule” vs. today’s trend of outcome based contracting and selected providers who can document results that skilled intervention is necessary. Back in those days your outcome may have been attributed to the natural history of the disease (simply the passage of time) vs. your specific treatment intervention. This obviously can still be the case and just because you did not get a good outcome does not mean your intervention was not effective (1). Also there was little to no discussion of documenting a treatment effect and utilizing outcome measures such as the DASH –Disabilities of the Arm Shoulder and Hand (2).

Rehabilitation today is now being performed in the age of accountability. New frontier lingo like classification systems, clinical predictor rules and regional interdependence is hopefully rolling off the tongues of new grads and clinician’s who stay current. The age of accountability is driving us in the orthopedic rehab settings to be more precise and efficient. This mindset is to enhance our outcomes that parallel the advances in orthopedic surgery.

There are many influences of an outcome (1). The actual intervention that we provide needs to be directed at restoring function earlier and earlier in the care plan. We also need to create an environment that integrates the whole kinetic chain (3). This approach will open the door to facilitate neuromuscular control and re-education thus allowing the underlying dysfunction to be addressed or corrected. This thought process will enable the clinician the ability to design “corrective” therapeutic exercises vs. a series of single joint isolation therapeutic exercises. I have always said that there should be a separate CPT code for “corrective “exercise 97110-C vs. just having your patient go over to a corner and pull on a rubber band and follow “protocol”. Why should that provider be reimbursed the same as the provider who is able to evaluate and identify movement dysfunction and integrates correction into the whole chain? In those “old days” our training involved the previously mentioned series of single joint isolation exercises that were based on the latest EMG article. We would have our shoulder patient perform 10 separate exercises for weeks before implementing functional exercises later. It always would be a point of curiosity to me as to why we “ortho” clinicians would think this way. It would be weeks before we put the “part into the whole” (“neuro” principle) and stimulated the sensorimotor system (4).

A clinical example of this concept is the patient with pain to arm elevation. These patients are typically 45 yrs old and have pain with overhead movements. Radiographs show A/C DJD and a slight curved acromion. These patients have a diagnosis of impingement. The clinician’s exam reveals weak scapular and rotator cuff muscles and a tight posterior capsule. They pull out the bottom staff office drawer prescription plan is typically 7-8 individual “isolated” rotator cuff and scapular stabilization exercises and some posterior shoulder stretching with local modalities as needed. After 3-4 weeks the patient is still complaining of pain. This standard approach failed to look at WHY the patient was “impinging “with arm elevation. The exam typically looks at individual muscles and movements rather than the entire upper quadrant chain. The patient thus returns to the orthopedist and subsequently undergoes a arthroscopic shoulder decompression. The patient returns to PT after surgery and goes through 6 weeks of forced conservative care, ROM and reconditioning. The patient reports much less pain with arm elevation and is discharged. The question that should be asked is, was the disappearance of pain the result of surgery, therapy intervention or just the passage of time by removing the shoulder from aggravating factors? How often do these same patients return to us with a “chronic “diagnosis? We must begin asking ourselves were the musculoskeletal impairments causing the functional deficit of raising the arm overhead with ADL’s and/or sport activities ever identified?

The approach of form and function would have assessed this patient quite differently. All the required components of arm elevation would have been assessed and a clinical hypothesis would have been formed as to why the patient has pain to overhead movements. The exam would have included: looking proximally to the ground up, at the trunk/lower extremity and thoracic spine as we know we need thoracic extension for efficient arm elevation (5). Attention would have been given to the scapular restrictors such as the pectorals minor and levator scapula. An assessment of the over dominance of the upper force couples that are known to inhibit the lower force couples would have been performed as this affects scapular humeral rhythm and decreases the subacrominion space, leading to pain and inhibition of the rotator cuff. This sequenced approach is another example of the orthopedic clinician incorporating a fundamental of neurological sensorimotor system rehabilitation (7).

Yoga participants are familiar with the fundamentals of the downward facing dog technique. With a few adjustments to the traditional technique, a sequence to facilitate and inhibit muscle timing required in arm elevation can be implemented. Incidentally, this would be an interesting EMG study to see if the suprascapular muscles are indeed inhibited when the lower scapula muscles are facilitated. The adjustments to the downward facing dog are designed to facilitate the lower force couples of the scapulahumeral rhythm complex all the while inhibiting the often over dominant upper force couples. An example of this is what Sahrmann describes as movement impairment during the act of arm elevation (6). This movement impairment syndrome results in downward rotation of the scapular when the rhomboids and levator scapulae are over dominating the action of the lower force couples (serratus anterior, lower trapezius). Vladimir Janda noted in 1979 predictable muscle patterns of tightness (levator/upper trapezius and pectoralis) would inhibit phasic muscles such as the serratus anterior and lower trapezius. Janda stressed that this leads to movement dysfunction. These patterns are the result of chronic pain or disuse neural drive. Janda clearly identified this as The Upper Cross Syndrome (7).

• When I teach the downward “therapy” dog, I instruct my patients to really emphasize the pushing of the hands into the floor as the buttock is raised. This pushing movement is creating activation of the serratus anterior similar to the push up with a plus however the clinically significant difference is that the scapula is functionally upwardly rotating with the serratus activation vs. the wall push-up plus exercise activates the serratus in a non-functional movement pattern of horizontal adduction and protraction vs. your desired “corrective” movement which is upward rotation. The other key advantage here is that as the body is being elevated via the pushing action of the trunk and hip extension as the lower trapezius will be facilitated. As the hips go into extension and you instruct the patient to “grow” into the movement by accentuating the PSIS ‘s to the sky , you instruct the patient to inhale with a strong diaphragmatic breath which creates this strong thoracic extension action further facilitating the smooth upward rotation of the scapular. I also instruct my patients to exhale at the top of the movement as they emphasize the pushing movement of the hands into the floor further getting those last few degrees of upward rotation. Another tip is to have your patient tuck their chin down and away from the tight levator side at the top of the movement. This tip is an extremely functional way to stretch the tight levator as the scapula is upperly rotated maximizing the elongation of the levator compare this to how the levator is traditionally stretched by having the patient side bend and rotate away with the arm down at the side similar to looking into the arm pit. This traditional manner is not functional at all and will not get the same timing sequence required of the scapula force couples being facilitated at the correct degree of scapulahumeral rhythm (8). The real beauty of this exercise pattern is that it is creating an inhibitory effect on the often over dominant and tight levator scapulae by the stimulus to the previous phasic muscles (serratus anterior/lower trapezius) (9). An overly tight levator will result in downward rotation of the scapula when arm elevation is attempted during traditional arm elevation therapy exercises. The downward facing “therapy “ dog naturally creates the environment of the entire kinetic chain of arm elevation: THE LEGS DRIVING THE TRUNK WHICH ENABLES THE SCAPULA TO BE PROPERLY POSITIONED TO HOUSE THE HUMERAL HEAD WHICH STIMULATES THE ROTATOR CUFF AND DELTOID MUSCULATURE TO CENTER THE BALL IN THE SOCKET (3).

To the best of my knowledge this exercise has not been researched with EMG studies. I am basing my analysis on the extensively studied mechanics of upper rotation of the scapulae and force coupling. I am clinically expressing a corrective therapeutic exercise based on my extensive clinical application. I have observed countless patients who have had difficulty regaining that smooth overhead movement and who have had struggled with putting all the little “parts” into the “whole” movement. Having said that I would like to offer this rationale to anyone who would like to take it on and perform first some EMG studies to put it out there and see if this exercise is facilitating and inhibiting the described patterns. I would at least start with this experiment basically because it is what the current method of investigation in our professions is. I would then like to see it be part of randomized controlled trial to truly measure the effectiveness of the intervention being described.

I will describe a common therapeutic session that I employ in a patient I often see who comes to me after being involved in a therapy program doing the “cookbook” approach such as described in my introduction. This patient will often seek another opinion because of lack of progress with the dreaded shoulder hike secondary to impairments such as restriction of soft tissue and weakness. I will first assess the movement pattern and based on what restrictors to arm elevation occurs, whether it is soft tissue or accessory joint mobility, I will address that. I will then quickly integrate the downward dog exercise to get everything working in balance. The exercise will quickly kick in muscle groups that have been dormant for a long period of time and you will see an amazing “freeing up” of previously tight motions. I will then follow the patient’s response to movement model so typical of the McKenzie Method and reassess their movement. I will document movement patterns of slowed velocity, postural change and look for less and less of this in subsequent sessions. Manual Therapy combined with exercise is effective in shoulder dysfunction. It should be stressed that the sequencing of the therapeutic session that is vital. One key point needs to be that you have to employ your manual therapy skills directed to the pectoralis minor, subscapulairis, levator scapular and the infraspinatus due to the chronicity of these cases. A progression of forces on the soft tissue will augment the corrective exercise.

As stated above traditional isolated therapy exercises work the parts with the hope that all the proper neuromuscular timing will occur when the patient is asked to perform arm elevation. When the timing is off often because of the long term lack of neural drive, the patient will often perform that faulty arm elevation movement described as the SHOULDER HIKE. I strongly encourage clinicians to incorporate therapeutic exercises that have stood the test of time. I believe the downward facing “therapy” dog gets the job done.


Last revised: February 11, 2010
John O'Halloran DPT, OCS, ATC, CSCS, Cert MDT


1. Herbert RD et al. Outcomes Measures Measure Outcomes not Effects of Intervention. Australian Journal of Physiotherapy .2005: 51:3-4.
2. Solway S, Beaton DE, McConnell S, Bombardier C. The DASH Outcome Measure User Manual, Second Ed. Toronto: Institute for Work and Health, 2002.
3. McMullen J, Uhl T. A Kinetic Chain Approach for Shoulder Rehabilitation. Journal of Athletic Training. 2000:35 (3):329-337.
4. Panjabi MM. The Stabilizing system of the spine. Part 1. Function, dysfunction, adaptation, and enhancement. J Spinal Disord.1992: 5(4);383-389.
5. Kebaetse M, McClure P, Pratt N. Thoracic Position Effects on Shoulder Range of Motion, Strength and Three-Dimensional Scapular Kinematics. Arch Phys Med Rehabil. 1999: Vol 80; 945-950.
6. Sahrmann S. Diagnosis and Treatment of Movement Impairment Syndromes.2002: St Louis, Mosby; p 219-220.
7. Janda V, Identification of the Upper and Lower Cross Syndromes. 1979
8. Hoppenfeld S. Physical Examination of the Spine and Extremities.1976: Conn, Appleton-Century-Crofts; p 23.
9. Sherrington CS. On reciprocal innervation of antagonistic muscles.1907: Proc R Soc Lond. 79B; 337.

Terms & Conditions

Please review our terms and conditions carefully before utilization of the Site. The information on this Site is for informational purposes only and should in no way replace a conventional visit to an actual live physical therapist or other healthcare professional. It is recommended that you seek professional and medical advise from your physical therapist or physician prior to any form of self treatment.