SSRI – Shoulder Database
Established 2014 - Collecting information about shoulder symptomsEthics
The ethics concerning establishment and maintenance of this Database was considered by the Human Research Ethics Committee of North Shore Private Hospital and approved on 21 November, 2014.
Aim
The aim of the SSRI Shoulder Database is to collect information regarding shoulder symptoms experienced by the general population irrespective of differentiating factors or whether surgery is required.
Of particular interest is how shoulder symptoms impact daily living activities (e.g. eating, bathing and recreation), sleep, mobility, range of motion, strength and pain. Also of interest are symptom cause (if known), duration, treatments sought and whether patients concurrently experience symptoms with any other joints.
Over time, the Database should be able to observe trends in shoulder pathology and, hopefully, contribute positively to clinical outcomes for patients.
Method
A shoulder symptoms form and 2 shoulder questionnaires are completed by patients who are seeking specialist opinion for their shoulder. Data is de-identified to protect patient privacy and confidentiality1, entered into the SSRI Shoulder Database and then analysed using Socrates Orthopaedic Outcomes2 software.
Shoulder Symptoms Form
The shoulder symptoms form was designed in house for use by Sydney Shoulder Specialists surgeons and is continually updated and improved:
Shoulder Survey 1
The Oxford Shoulder Score3 is a patient-based questionnaire used to assess shoulder symptoms. It contains a mixture of pain and function questions, derived from over 200 initial question models based on in-depth patient and clinician interviews. It has been validated against clinician-based and general health status measures.
The score is simple to complete, sensitive to clinical change, and has proven consistently reliable in determining outcomes following shoulder surgery.
The total score ranges from 0 to 48 points, with higher scores indicating better function. Categorical ratings are assigned as follows: 37 to 48 points is excellent; 25 to 36 is good; 13 to 24 points is a fair outcome; and a score of less than 12 points is considered poor.
Sydney Shoulder Research Institute uses the score under licence from ISIS Innovation which is a subsidiary of the University of Oxford in England.
Shoulder Survey 2
The American Shoulder & Elbow Society4 Rating Scale (Shoulder) was published by the Research Committee of the American Shoulder and Elbow Surgeons. The scale contains both a patient-derived subjective assessment and physician-derived objective assessment.
For the purposes of this database, we are using only the patient-derived subjective assessment. This part of the scale consists of 2 equally weighted domains – function and pain – and has been widely utilised for assessing outcomes in patients with rotator cuff disease and shoulder instability. The total score ranges from 0 to 100 points, with higher scores indicating better function. Categorical ratings are assigned as follows: 76 to 100 points is excellent; 51 to 75 is good; 26 to 50 points is a fair outcome; and a score of less than 25 points is considered poor.
We have made 6 modifications to the scale which we believe do not affect its validity or reliability. Modifications were made after hand scoring 500 scales and observing consistent patterns of omissions, errors and corrections made by patients and confusions reported by same. Following the introduction of our modified form, omissions, errors and confusion in the next 500 scored scales significantly decreased:
- The question “do you take pain medication (aspirin, Tylenol, Advil, etc.)?” was changed to “do you take pain medication (Panadol, Nurofen, Aspirin, etc.)?” to make it more relevant to Australian patients;
- The question “how many pills do you take each day (average)?” was changed to “how many tablets do you take each day (on average) for your shoulder?” as patients were noting ALL medications artificially inflating the count;
- The question “circle the number in the box that indicates….your ability to comb hair” was changed to “circle the number in the box that indicates….your ability to comb hair (or if bald/other handed do that action)” as patients were writing “I am bald” or “I am not left/right handed” then ignoring the question;
- The ranking boxes were colourized with “0” scores made red to reinforce the meaning of this score as an inability to do an action and “3” scores made green to reinforce the meaning of the score as an ability to do an action without difficulty;
- The ranking box question was changed with “Affected side” becoming “Left shoulder” and “Opposite side” becoming “Right shoulder” and the addition of “we need both (sides) for comparison” as patients were only completing one side or writing “left” or “right” above the side, often when they had changed their mind as to which side represented which shoulder;
- The order of questions was changed with Questions 6, 9, 1, 2, 3, 4, 5, 7 and 8 presented in our version as Questions 1, 2, 3, 4, 5, 6, 7, 8 and 9, respectively as we felt this facilitated ease of and therefore encouraged completion.
You may have noticed that The American Shoulder & Elbow Society Rating Scale (Shoulder) and American Shoulder & Elbow Society Rating Scale (Elbow) are essentially the same forms with the terms “shoulder” and “elbow” interchanged.
The American Shoulder & Elbow Society Elbow Rating Scale (Shoulder) does not require a licence therefore Sydney Shoulder Research Institute and other health professionals are freely able to use it for research or clinical purposes.
To date, we have collected 1811 sets of data. Preliminary results are presented below for interest:
Results
Distribution of Gender
About 3/5 of people presenting with shoulder symptoms were male:
Distribution of Current Age
About 1/3 of patients presenting with shoulder symptoms were aged 53 – 69 years, 1/4 were aged 37 – 52 years and 1/5 were aged 21 – 36 years:
Distribution of Cause of Injury
A little over 1/3 of people reported having no known injury while the highest known cause of symptoms was sport followed by falls:
Duration of Symptoms
Nearly 1/2 of people reported having experienced shoulder symptoms for over 12 months
Duration of Symptoms in Years
Of people who reported experiencing shoulder symptoms for more than a year, the majority experienced them for between 2 – 19 years. Some reported experiencing symptoms between 20 – 40 years while an even smaller group reported experiencing symptoms for more than 40 years (!):
Right or Left Handed?
The majority of people with shoulder symptoms were right handed which reflects the general population:
And YET….
The difference between people who presented with shoulder symptoms on their dominant side compared to their non-dominant side was not as great:
Other Joint Problems
Nearly 1/3 of people reported symptoms in the other shoulder and another 1/3 reported knee symptoms:
Notes:
1The difference between the two terms is that confidentiality relates to an ethical duty, whilst privacy is a common law or statutory right, which is reflected via legislative instruments such as the Privacy Act 1988 and the National Privacy Principles (NPP).
2Learn more about Socrates Orthopaedic Outcomes Software
3Learn more about the Oxford Shoulder Score
4Learn more about the American Shoulder & Elbow Society