Spanish version of SPADI (shoulder pain and disability index) in musculoskeletal shoulder pain: a new 10-items version after confirmatory factor analysis

Shoulder pain is one of the most common musculoskeletal conditions seen in primary care [1], after low back and neck pain. It affects one in three adults [2, 3], accounting 1 % of General Practice consultations in primary care [4]. In working-age populations, the prevalence of shoulder pain associated with musculoskeletal disorders is even higher [5], and increases with age [6].

Shoulder disorders are frequently accompanied by pain and restricted shoulder movement leading to difficulties in performing certain activities. Recent research suggests that shoulder pain not only affects function during work and leisure time activities, but also may interfere with psychological and social wellbeing [7]. Additionally, environmental factors, such as psychological distress, may contribute to the development of chronic shoulder problems [8].

A variety of musculoskeletal pathologies can cause shoulder pain including subacromial syndrome, frozen shoulder, rotator cuff tendonitis and tear, calcyfying tendonitis, biceps large portion tendonitis, and tear and gleno-humeral instability [9].

The impact of shoulder disorders can be assessed in different ways. Traditionally, the assessment has focused on the impairments associated with shoulder pathology by evaluating the range of motion, strength, or pain [10]. However, patients are more concerned with the activity limitations that result from these impairments. This has lead to an increasing emphasis on patient reported outcome (PRO) measures.

The Shoulder Pain and Disability Index (SPADI) is a PRO measure that was developed for use in an outpatient setting. It was designed to measure the impact of shoulder pathology in terms of pain and disability, for both current status, and change on status over time [11]. The SPADI is a self-administered questionnaire that consists of two dimensions, one for pain and the other for functional activities [11]. The pain dimension consists of five questions regarding the severity of an individual’s pain. Functional activities are assessed with eight questions. The SPADI takes 5 to 10 min for a patient to be completed. To answer the questions, subjects place a mark on a “0 to 10” numbered scale for each question. Verbal anchors for the pain dimension are ‘no pain at all’ and ‘worst pain imaginable’, and those for the functional activities are ‘no difficulty’ and ‘so difficult it required help’. The scores from both dimensions are averaged to produce a total score [11].

The SPADI has been used for measuring the outcomes in different studies and shoulder conditions, such as shoulder pain, various upper extremity diagnoses, various shoulder diagnoses, adhesive capsulitis, rotator cuff, after shoulder arthroplasty, total shoulder arthroplasty, various shoulder surgery, and in different populations and clinical settings, as orthopedic practice, outpatient physiotherapy and community volunteers [12].

The SPADI has been shown to be valid as a measure of pain and disability in community-based patients reporting shoulder pain due to musculoskeletal pathology. The SPADI has good internal consistency with a Cronbach’s alpha of 0.95 for the total score, 0.92 for the pain subscale and 0.93 for the disability subscale. The SPADI has also shown ability to detect change over time [13]. In comparing a number of shoulder specific questionnaires in primary care, the SPADI and SRQ (shoulder questionnaire rating) were found to be the most sensitive to detecting change and the SPADI required the least time to complete [14]. Another study reviewing the clinimetric properties of several shoulder questionnaires concluded that SPADI had good construct validity (>0.74) [15].

The increased number of international research projects as well as the diversity of populations and cultures living in a same region, has created the need to validate PRO measures in groups different from those originally used to develop the measure. This requires both translation into a new language and accommodation for differences in cultural characteristics [16].

To our knowledge, only one study has attempted to validate a Spanish language version of the SPADI and this study was conducted in women with shoulder pain following breast cancer surgery rather than shoulder pain due to musculoskeletal pathology [17]. Furthermore, no confirmatory factor analysis has been carried out for any of the multiple versions of SPADI. There is a need of constant update for any PRO measure to guarantee that the mentioned PRO retains all its psychometric properties and its equivalence between original and translated versions, as well as to evaluate its performance in other contexts [18]. This process should be carried out with robust methods, such as confirmatory factor analysis, to test the hypothesis of its original dimensionality. Moreover, the validity of a PRO need to be tested to assess whether its validity is dependent on the population in which the instrument was originally validated, as clinimetric properties many times depend on situational circumstances [18]. Therefore, additional testing in specific populations, i.e., musculoskeletal shoulder pain disorders, need to be carried out.

Hence, the aim of this study is twofold: 1) to translate and validate a Spanish version of the SPADI for Spanish population with a broad range of shoulder disorders and, 2) to determine the psychometric properties of this instrument, using confirmatory methods.

This study aimed to carry out the cross-cultural adaptation of SPADI for Spanish population and, secondly, to determine the psychometric properties of this version, in a population of patients suffering from shoulder pain. Regarding the first objective, the cross-cultural adaptation, our study encountered a few problems during the translation process, primarily in related to time period reference in the instructions. These problems were solved by the research team and the original author. The participants had no difficulty comprehending the questionnaire during the pilot study, and the translated version demonstrated good results in readability tests. We modified the original 13 item English language version of the SPADI by deleting those items with low correlations (items 1, 11 and 12). The confirmatory factor analysis for this new version attained the best fit. There was a strong correlation between the new Spanish Language SPADI and the Quick-Dash demonstrating good convergent validity. One of the strengths of this study was the scientific rigour of the methods used during the cross-cultural adaption phase. This included involving the original author to ensure the accuracy of the translation process. A previous study to adapt a new Spanish version of SPADI has been done recently [17]. These authors did not find any significant problems and/or difficulties in the cross-cultural adaptation process. The lack of difficulties in comprehension for the participants during the cross-cultural process is consistent with other studies [17, 28, 29]. Moreover, our new Spanish language version achieved good readability. The time required to complete the new Spanish Language SPADI are similar to the original version [11], although other cross-cultural versions have reported even less time to complete [29].

Regarding the psychometric properties, this is the first study of a Spanish language version of SPADI to utilize a factor structure analysis, with confirmatory methods. As a result of our structure factor analysis, items 1 (“How severe is your pain at your worst”), 11 (“Placing an object on a high shelf”) and 12 (“How much difficult do you have carrying a heavy object of 10 pounds”) were eliminated to achieve the best adjustment for the model. As a result, the new Spanish version for SPADI has different pain and disability subscales than the original English language version. Other studies have reported a unidimensional structure, without differences between pain and disability subscales [30, 31], while others were in consistent with the original version (two subscales) [32]. Based on our results, we recommend the use of this new 10-item Spanish SPADI-version.

The internal consistency of our version of the SPADI was good (Cronbach’s alpha of 0.90) was consistent with the findings of Torres-Lacomba et al [17], who also reported good values in internal consistency (0.96). It is important to note that the sample for Torres-Lacomba study was very different from the sample for our study in that it was drawn from women with shoulder pain after breast cancer surgery. Similar to other cross-cultural studies on SPADI [30–32], we found a strong correlation between the SPADI and the Quick-DASH indicating good convergent validity. Alsanawi et al [33] found a correlation between the SPADI and DASH of 0.84 (Spearman coefficient), while Ebrahimzadeh et al [28] reported a correlation of 0.61.

Our results imply that a new 10-items Spanish version for SPADI could be used by researchers and clinicians as a self-reported disability measure in patients suffering from shoulder pain in both routine clinical practice or in clinical research trials, This version keeps the properties of the original version, with two subscales (pain and disability). Spanish language is one of the most spoken languages around the world. Even though different Spanish speaker countries present some differences, semantic and grammatical rules are homogeneus, so that the version presented in this study could be of reference for other cross-cultural studies on SPADI, requiring only minor changes to adapt wording to accommodate the specific terms used in informal language. This new index could permit comparisons between other countries when studies on shoulder pain will be carried out.

However, there are some limitations in this study that should need to be recognized. Firstly, psychometric properties for Spanish-SPADI such as test-retest reliability, sensibility to change, as well as divergent validity have not been determined. The variety of different shoulder pain conditions in the participant sample could mean a risk of bias. Nevertheless, recent literature [34] recommends to avoid the use of subgroups in shoulder pain due to the lack of a gold standard for each of the diagnostic labels. The present study did no distinguish between participants with acute versus chronic shoulder pain and this might have influenced our findings so they must be taken with caution. Future investigations should be conducted to determine cutpoints in the score of this new version that could be used to classify patients with mild, moderate and/or severe shoulder pain.

This study carried out a cross-cultural adaptation and validation of Spanish language version of the SPADI for Spanish population, and an examination of the psychometric properties of this new version. This study validated a new 10-items version of SPADI for Spaniards providing a patient reported outcome measure for use in this population in both clinical practice and research.