Abstract
Moving in a manual wheelchair places significant strain on the upper extremities. The shoulder is particularly prone to overload owing to the joint’s high mobility and a relatively small muscle mass, which is not suited to repeatedly support and move the body’s weight. The purpose of our study was to evaluate the psychometric properties of the Wheelchair User’s Shoulder Pain Index (WUSPI‑Pol) in wheelchair athletes. This study was a cross‑sectional survey. To assess psychometric properties, we have examined 74 individuals participating in team wheelchair sports who were experiencing shoulder pain. We have assessed the reliability, internal structure, test–retest repeatability, and validity of the Polish version of the WUSPI. Scale reliability for the study sample was very good. Cronbach’s alpha for the total scale was 0.972. Test–retest reliability, assessed with the intraclass correlation coefficient (ICC), was very high (0.995), indicating near‑perfect agreement between the first and second administrations. Confirmatory factor analysis indicated a good model fit (CFI and TLI > 0.95; SRMR < 0.08), with RMSEA slightly higher than expected (0.076). External validity was evaluated by correlating WUSPI‑Pol scores with the QuickDASH and the Simple Shoulder Test; as expected, the correlations were positive. The Polish version (WUSPI‑Pol) is a reliable and valid instrument for assessing shoulder pain in wheelchair users.
Keywords: Psychometrics, Shoulder pain, Activities of daily living, The wheelchair user’s shoulder pain index
Subject terms: Neurological disorders, Diseases, Health care
Introduction
Moving in a manual wheelchair places significant strain on the upper extremities. The shoulder is most often overloaded because of the joint’s high mobility and a relatively small muscle mass that is ill‑suited to repeatedly support and move the body’s weight1,2. More than two‑thirds of wheelchair users report experiencing shoulder pain3.
Athletes who play wheelchair team sports are particularly prone to shoulder injuries and pain. In addition to activities of daily living, they place additional load on the shoulder complex during strenuous training sessions and competitions. Shoulder pain is the most frequently reported medical problem among wheelchair athletes, negatively impacting their athletic performance, daily functioning, and quality of life4,5. Appropriate protocols and risk‑monitoring strategies may reduce the occurrence of shoulder pain and the risk of shoulder injury6.
Early and accurate identification of shoulder pain—and of the activities that interfere with function—is important. A key instrument for this purpose is the Wheelchair User’s Shoulder Pain Index (WUSPI), designed to measure the severity of shoulder pain associated with functional activities in wheelchair users. The WUSPI comprises 15 items assessing shoulder pain during activities of daily living. During development, the authors selected the activities most likely to provoke shoulder pain in wheelchair users. The instrument is easy to administer and it takes about five minutes to complete7–9. The WUSPI is characterized by very good test–retest reliability, with an intraclass correlation coefficient (ICC) of 0.99 for the total score and for individual items ranging from 0.84 to 0.99. Cronbach’s alpha coefficient is 0.97, indicating excellent internal consistency9. Validation of the original version of the questionnaire also provided strong evidence of its construct validity. Curtis et al. demonstrated a significant relationship between the questionnaire results and objective clinical indicators: the total WUSPI score correlated negatively with the degree of shoulder range of motion, with moderately strong negative correlations noted between the WUSPI score and the range of shoulder abduction, flexion, and extension. This indicates that individuals with higher WUSPI scores (reflecting more severe shoulder pain during activity) generally exhibited more limited joint mobility, thereby confirming the validity of the measure—the questionnaire effectively captures the clinical relationship between pain and functional impairment7. Although the WUSPI is frequently used in cross‑sectional and clinical studies, it has thus far been translated and validated only in the United States7,8, Spain10, Denmark11, Norway12, and Korea13.
For individuals who use a wheelchair, understanding the nature of shoulder pain is essential2. Such knowledge may support the selection of orthopedic equipment, the design of exercise protocols and physical activity programs, patient education, as well as the provision of practical support in daily functioning. Despite the importance of regular medical assessments of the musculoskeletal condition of the shoulder among wheelchair athletes, systematic evaluation of the shoulder complex by medical personnel remains limited. In everyday clinical practice, there is a need for a simple and reliable tool to monitor shoulder status in this athletic population14. Standardization of shoulder‑pain assessment methods is also necessary to enable comparisons across studies and between countries15. The diversity of populations worldwide highlights the need for cross-culturally validated research tools and scales. Researchers and practitioners should have access to reliable and valid instruments that enable them to assess relevant constructs within their own cultural and linguistic contexts, thereby ensuring the highest quality of patient care16.
Clinicians in Poland continue to seek appropriate and validated instruments for measuring shoulder pain among wheelchair users. To the best of our knowledge, no reliable, translated, and validated tool for assessing shoulder pain in wheelchair users is currently available in Poland. The lack of a validated tool in Polish for assessing wheelchair-related shoulder pain constituted a significant gap in both clinical practice and research in Poland. Therefore, the aim of our study was to evaluate the psychometric properties of WUSPI‑Pol in wheelchair athletes.
Methods
Study design
The study was a cross‑sectional survey. The survey included athletes training in wheelchair rugby or basketball in Poland from 2021 to 2023. This study followed the Consolidated Criteria for Reporting Qualitative Research (COREQ) to ensure transparency and reproducibility of data collection17. Although the analyses were quantitative, COREQ domains were applied to reinforce methodological. The study presented in this manuscript focuses on the quantitative psychometric validation of the WUSPI-Pol questionnaire.
Procedures
Data were collected via face‑to‑face, pen‑and‑paper interviews conducted by trained researchers. The study took place at tournaments and club training camps across Poland among athletes who agreed to participate.
Participants
To assess psychometric properties, 74 wheelchair athletes were examined. Inclusion criteria were: using of a manual wheelchair for at least 6 months; participation in wheelchair rugby or basketball for at least 3 months; age ≥ 18 years; shoulder pain or dysfunction within the past week; and informed consent to participate. Exclusion criteria were: using of a wheelchair for < 6 months; participation in wheelchair rugby or basketball for < 3 months; age < 18 years; absence of shoulder pain or dysfunction; or lack of consent to participate.
Translation and cultural adaptation of WUSPI
The translation, cultural adaptation and validation of the WUSPI, as well as its use in research, were approved by the author of this tool, K.A. Curtis.
The first step was translation and cultural adaptation of the WUSPI questionnaire. This process, in accordance with standard guidelines16, included the following steps:
-
I
Forward translation from English into Polish by two independent translators (version A1 and A2).
-
II
Expert panel — a meeting between the two independent translators and the research team. The research team analyzed each item and response options in the questionnaire and produced a consensus translation (synthetic version A1–2).
-
III
Backward translation – English native speaker with fluent Polish, not familiar with the original version, performed back translation into English (version B). The research team compared the original English version with the back translation. Then, a team of experts in research methodology, linguistic translation, and rehabilitation evaluated the versions and formulated the preliminary translation version (version C).
-
IV
Harmonization – comparison of different translation versions (forward and backward) with the original and other existing translations to ensure terminological and conceptual consistency.
-
V
Pre‑testing and cognitive interviews — a study was conducted among 20 wheelchair users experiencing shoulder pain. Comprehension, item interpretation, and the cultural adequacy of the instrument were examined. The research team then compared and interpreted the pre‑test translation, resolved discrepancies, and corrected stylistic errors.
-
VI
Field testing — a pilot study in a group of 30 wheelchair users was carried out to assess translation stability, identify additional linguistic and cultural issues, and collect data for psychometric analysis (version D).
Ethics
In accordance with the Declaration of Helsinki, participants were informed about the purpose and course of the study and their right to withdraw at any stage. All participants provided written informed consent. The study was approved by the Bioethics Committee of the University of Rzeszow (Resolution No. 19/12/2019).
Measurement instruments
The primary instrument was the Wheelchair User’s Shoulder Pain Index (WUSPI). The Quick Disabilities of the Arm, Shoulder and Hand (QuickDASH) and the Simple Shoulder Test (SST) were also administered.
In addition, the following variables were collected: age, sex, place of residence, education level, time using a wheelchair, cause of disability, daily number of transfers, handedness, primary activity, employment, leisure‑time activities, and sport.
Wheelchair user’s shoulder pain index (WUSPI)
The WUSPI assesses shoulder pain in wheelchair users during functional activities. It comprises 15 items referring to pain experienced during the previous week while performing activities such as transfers, wheelchair mobility, sleep, and other activities of daily living. Each item has been scored from 0 (“no pain”) to 10 (“worst pain ever experienced”) using a 10‑cm visual analogue scale. Total scores range from 0 to 150; higher scores indicate greater shoulder pain. Prior studies report high internal consistency, test–retest reliability, and concurrent validity7–9.
Quick disabilities of the arm, shoulder and hand (QuickDASH)
The Quick Disabilities of the Arm, Shoulder and Hand (QuickDASH) is a measure of upper‑limb disability. It contains two groups of items. The first consists of six items regarding limitations in activities of daily living caused by upper‑limb pain. The second one consists of five items on limitations in social participation or activities of daily living caused by upper‑limb pain, sleep quality, and the severity of upper‑limb symptoms. Two optional modules — “Work” and “Sports/performing music” —may be omitted if inapplicable. Responses refer to the week preceding the survey and are scored from 1 to 5, where 1 indicates no difficulty or no pain and 5 indicates inability to perform the activity or unbearable pain. Scores were calculated according to published instructions18,19.
Simple shoulder test (SST)
The Simple Shoulder Test (SST) assesses functional limitations of a shoulder with dysfunction and pain. It consists of 12 items with dichotomous response options (Yes = 1; No = 0). Items can be grouped into domains: two of them assess function, seven assess muscle strength, and three others assess glenohumeral range of motion. The total score ranges from 0 (worst function) to 12 (best function) and refers to the past week19–21.
Statistical analysis
Qualitative variables were summarized as counts and percentages. Quantitative variables were summarized as the mean, standard deviation, median, and quartiles. Between‑group comparisons (retest vs. no retest) used the chi‑square test or Fisher’s exact test for qualitative variables and the Mann–Whitney U test for quantitative variables due to non‑normal distributions. Normality was assessed using the Shapiro–Wilk test. The significance level was set at 0.05. Analyses were performed in R, version 4.2.223.
The analysis was performed in R software, version 4.2.223.
Analysis of the reliability
The internal consistency of the scale was assessed using Cronbach’s alpha coefficient. According to Nunnally’s criterion, a scale was considered internally consistent if the measure was not less than 0.724.
Scale stability (reproducibility) was assessed by the test–retest method using the ICC. The following interpretive thresholds were applied: <0.50 — poor agreement; 0.50–0.75 — moderate; 0.75–0.90 — good; 0.90–1.00 — very good25,26.
The Standard Error of Measurement (SEM) and Minimum Detectable Change (MDC) were calculated to quantify measurement error. These analyses used data from 20 participants with two WUSPI assessments. The average interval between assessments, administered by two different assessors, was 7 days.
Confirmatory factor analysis (CFA) was used to evaluate the internal structure of WUSPI. Model fit was assessed using the RMSEA, CFI, TLI, and SRMR. According to Hu and Bentler, good fit is indicated by RMSEA < 0.06, CFI and TLI > 0.95, and SRMR < 0.0827.
Floor and ceiling effects
Floor and ceiling effects were calculated by determining the percentage of participants with the lowest or highest possible scores for each WUSPI question.
Analysis of the convergent validity
Convergent validity was assessed by correlating WUSPI scores with QuickDASH and SST scores. Since the distributions of the analyzed variables were non‑normal, Spearman’s rank correlation coefficient was used.
Results
Characteristics of the study group
The study involved 74 subjects, including 7 women and 67 men. The average age of the subjects was 34.5 years (SD = 9.6). Most of the subjects lived in the city (77.03%). Most of the subjects had secondary education (44.59%) or higher education (31.08%). The economically active were 75.68% of the respondents. The most common reason for disability was a spinal injury (59.46%). The average number of daily transfers in the study group was 11.66 (SD = 7.47). Most of the subjects were right-handed (87.84%) and played wheelchair rugby (56.76%).
The mean WUSPI score in the study group was 32.16 (SD = 28.2), QuickDASH − 22.9 (SD = 20.52), QuickDASH-Work − 18.84 (SD = 21.96), QuickDASH-Sport − 25.68 (SD = 23.24), SST − 8.34 (3.35).
Except for the type of sport played, there were no statistically significant differences between sociodemographic variables and WUSPI scores in groups with one and two (test-retest) tests (Table 1).
Table 1.
General socio-demographic characteristics of the study population (N = 74).
| Parameter | RETEST (N = 20) | No RETEST (N = 54) | Total (N = 74) | p | |
|---|---|---|---|---|---|
| Age [years] | Mean (SD) | 33.95 (8.34) | 34.70 (10.09) | 34.50 (9.60) | p = 0.826 |
| Median (quartiles) | 32.50 (29.00-40.50) | 35.50 (28.25-41.00) | 34.00 (29.00–41.00) | ||
| Range | 19.00–50.00 | 18.00–55.00 | 18.00–55.00 | ||
| Sex | Male | 18 (90.00%) | 49 (90.74%) | 67 (90.54%) | p = 1.000 |
| Female | 2 (10.00%) | 5 (9.26%) | 7 (9.46%) | ||
| Place of residence | City | 15 (75.00%) | 42 (77.78%) | 57 (77.03%) | p = 0.766 |
| Rural area | 5 (25.00%) | 12 (22.22%) | 17 (22.97%) | ||
| Education level | Primary or none | 1 (5.00%) | 2 (3.70%) | 3 (4.05%) | p = 0.052 |
| Middle school | 0 (0.00%) | 1 (1.85%) | 1 (1.35%) | ||
| Vocational | 5 (25.00%) | 9 (16.67%) | 14 (18.92%) | ||
| Secondary | 4 (20.00%) | 29 (53.70%) | 33 (44.59%) | ||
| University | 10 (50.00%) | 13 (24.07%) | 23 (31.08%) | ||
| Time on wheelchair [years] | Mean (SD) | 14.10 (7.94) | 15.56 (7.68) | 15.16 (7.73) | p = 0.491 |
| Median (quartiles) | 14.50 (7.75–21.25) | 14.50 (11.00-22.75) | 14.50 (10.00–22.00) | ||
| Range | 1.00–29.00 | 2.00–30.00 | 1.00–30.00 | ||
| Reason of disability | Spinal cord injury | 12 (60.00%) | 32 (59.26%) | 44 (59.46%) | p = 0.921 |
| Cerebral palsy | 1 (5.00%) | 5 (9.26%) | 6 (8.11%) | ||
| Amputation of lower limb(s) | 4 (20.00%) | 7 (12.96%) | 11 (14.86%) | ||
| Spina bifida/spinal cord hernia | 2 (10.00%) | 7 (12.96%) | 9 (12.16%) | ||
| Other | 1 (5.00%) | 3 (5.56%) | 4 (5.41%) | ||
| Daily number of transfers | Mean (SD) | 13 (7.81) | 11.17 (7.36) | 11.66 (7.47) | p = 0.194 |
| Median (quartiles) | 11 (8–16) | 10 (6–12) | 10 (6–14) | ||
| Range | 2–32 | 4–40 | 2–40 | ||
| Laterality | Left | 0 (0.00%) | 7 (12.96%) | 7 (9.46%) | p = 0.188 |
| Right | 19 (95.00%) | 46 (85.19%) | 65 (87.84%) | ||
| Both | 1 (5.00%) | 1 (1.85%) | 2 (2.70%) | ||
| Basic activity | Paid employment | 18 (90.00%) | 38 (70.37%) | 56 (75.68%) | p = 0.424 |
| School/University | 2 (10.00%) | 7 (12.96%) | 9 (12.16%) | ||
| Voluntary work | 0 (0.00%) | 1 (1.85%) | 1 (1.35%) | ||
| Retirement | 0 (0.00%) | 1 (1.85%) | 1 (1.35%) | ||
| Other | 0 (0.00%) | 7 (12.96%) | 7 (9.46%) | ||
| Work [h/week] | Mean (SD) | 33.00 (8.65) | 27.59 (14.12) | 29.05 (13.04) | p = 0.068 |
| Median (quartiles) | 35.00 (33.75-40.00) | 35.00 (15.75-35) | 35 (20.00–35.00) | ||
| Range | 10.00–40.00 | 0.00–60.00 | 0.00–60.00 | ||
| Free time activities [h/week] | Mean (SD) | 10.30 (4.84) | 10.08 (5.76) | 10.66 (5.50) | p = 0.825 |
| Median (quartiles) | 10.00 (6.75–10.50) | 10.00 (6.00-14.75) | 10.00 (6.00–14.00) | ||
| Range | 4.00–20.00 | 2.00–25.00 | 2.00–25.00 | ||
| Sports | Basketball | 13 (65.00%) | 19 (35.19%) | 32 (43.24%) | p < 0.042* |
| Rugby | 7 (35.00%) | 35 (64.81%) | 42 (56.76%) | ||
| WUSPI | Mean (SD) | 46.95 (36.86) | 26.69 (22.25) | 32.16 (28.20) | p = 0.156 |
| Median (quartiles) | 45.50 (9.75–83.25) | 19.00 (14-32.5) | 21 (13.25–40.50) | ||
| Range | 2.00–92.00 | 0.00–97.00 | 0.00–97.00 | ||
| QuickDASH | Mean (SD) | 31.70 (28.09) | 19.63 (16.03) | 22.90 (20.52) | p = 0.291 |
| Median (quartiles) | 23.86 (4.55–64.2) | 18.18 (6.82–28.98) | 18.18 (6.82–31.25) | ||
| Range | 0.00-68.18 | 0.00-68.18 | 0.00-68.18 | ||
| QuickDASH - Work | Mean (SD) | 27.19 (26.15) | 15.36 (19.21) | 18.84 (21.96) | p = 0.101 |
| Median (quartiles) | 21.88 (0.00-45.31) | 3.12 (0.00–25) | 6.25 (0.00-43.75) | ||
| Range | 0.00–75.00 | 0.00–56.25 | 0.00–75.00 | ||
| QuickDASH - Sport | Mean (SD) | 32.50 (30.12) | 23.15 (19.86) | 25.68 (23.24) | p = 0.336 |
| Median (quartiles) | 21.88 (4.69–56.25) | 25 (0.00-29.69) | 25.00 (0.00–50.00) | ||
| Range | 0.00–75.00 | 0.00-68.75 | 0.00–75.00 | ||
| SST | Mean (SD) | 6.95 (4.55) | 8.85 (2.65) | 8.34 (3.35) | p = 0.335 |
| Median (quartiles) | 7.50 (3.00–12.00) | 9.00 (8.00–11.00) | 9.00 (7.00–11.00) | ||
| Range | 1.00–12.00 | 1.00–12.00 | 1.00–12.00 | ||
p - Qualitative variables: chi-squared or Fisher’s exact test. Quantitative variables: Mann-Whitney test.
* Statistically significant (p < 0.05).
Analysis of the reliability
Internal consistency - Cronbach’s alpha
The internal consistency of the WUSPI was assessed using Cronbach’s alpha. Values from of 0.972 was obtained (Table 2).
Table 2.
Internal consistency and test-retest.
| Cronbach’s alpha | ICC |
|---|---|
| 0.972 | 0.995 (0.987–0.998) |
ICC – intraclass correlation coefficient.
Temporal consistency was estimated using the test-retest method. For this purpose, 20 respondents were asked to complete the WUSPI again after 7 days. The ICC coefficient was very high (0.995) indicating an almost perfect concordance between the results of the first and second application of the WUSPI and providing evidence of the temporal consistency of the tool (Table 2).
The sample size for the test-retest was calculated based on ICC. The sample size at which test-retest power reaches 0.8 (i.e., the level considered satisfactory) is 3 patients.
| Power | N needed for power > 0.8 |
|---|---|
| > 0.999 | 3 |
Discriminant power
All items of the scale have positive discriminant power and noticeably have exceed 0.4. Excluding any of the items does not increase Cronbach’s alpha coefficient, which means that the scale is well constructed (Table 3).
Table 3.
Discriminant power.
| Item | Discriminant power |
|---|---|
| WUSPI 1 | 0.815 |
| WUSPI 2 | 0.862 |
| WUSPI 3 | 0.840 |
| WUSPI 4 | 0.752 |
| WUSPI 5 | 0.673 |
| WUSPI 6 | 0.682 |
| WUSPI 7 | 0.821 |
| WUSPI 8 | 0.903 |
| WUSPI 9 | 0.933 |
| WUSPI 10 | 0.781 |
| WUSPI 11 | 0.902 |
| WUSPI 12 | 0.887 |
| WUSPI 13 | 0.918 |
| WUSPI 14 | 0.880 |
| WUSPI 15 | 0.834 |
Measurement errors
The SEM and MDC measurement error values for the overall result were 4.677 and 12.964, respectively (Table 4).
Table 4.
The measures of measurement error.
| SEM | MDC |
|---|---|
| 4.677 | 12.964 |
SEM - Standard Error of Measurement; MDC - Minimum Detectable Change.
Internal structure
Good Confirmatory Factor Analysis values were obtained. CFI and TLI indices (> 0.95), SRMR (< 0.08), slightly higher than expected RMSEA (0.076) (Table 5).
Table 5.
Confirmatory factor Analysis.
| RMSEA | CFI | TLI | SRMR |
|---|---|---|---|
| 0.076 | 0.979 | 0.969 | 0.032 |
RMSEA - Root Mean Square Error of Approximation; CFI - Comparative Fit Index; TLI - Tucker-Lewis Index; SRMR - Standardized Root Mean Residual.
Floor and ceiling effects
Floor and ceiling effects were generally below 50%; the highest floor effect was observed for items 8 and 9 (47.3%), with one item (13) exceeding 50% (51.4%) (Table 6).
Table 6.
Floor and ceiling effects.
| Item | Floor effect | Ceiling effect |
|---|---|---|
| WUSPI 1 | 23.0% | 0.0% |
| WUSPI 2 | 18.9% | 0.0% |
| WUSPI 3 | 21.6% | 0.0% |
| WUSPI 4 | 24.3% | 0.0% |
| WUSPI 5 | 17.6% | 0.0% |
| WUSPI 6 | 20.3% | 0.0% |
| WUSPI 7 | 44.6% | 0.0% |
| WUSPI 8 | 47.3% | 0.0% |
| WUSPI 9 | 47.3% | 0.0% |
| WUSPI 10 | 45.9% | 1.4% |
| WUSPI 11 | 33.8% | 0.0% |
| WUSPI 12 | 28.4% | 0.0% |
| WUSPI 13 | 51.4% | 0.0% |
| WUSPI 14 | 32.4% | 0.0% |
| WUSPI 15 | 23.0% | 0.0% |
Analysis of the validity
External validity was assessed by correlating the WUSPI-Pol scores with the scores of two tools: QuickDASH and SST. Positive correlations were expected between WUSPI and these tools. Analysis using the Spearman correlation coefficient showed that the expected correlations did indeed occur and were statistically significant (Table 7).
Table 7.
External validity.
| Parameter | WUSPI |
|---|---|
| Spearman’s correlation coefficient | |
| QuickDASH | r = 0.689, p < 0.001 * |
| QuickDASH - Work | r = 0.561, p < 0.001 * |
| QuickDASH - Sport | r = 0.589, p < 0.001 * |
| SST | r=-0.741, p < 0.001 * |
* Statistically significant (p < 0.05).
Discussion
Our evaluation of WUSPI‑Pol has addressed three aspects: internal consistency, temporal consistency, and external validity.
The internal consistency of WUSPI‑Pol was assessed using Cronbach’s alpha. A value of 0.972 was obtained, which exceeds the commonly accepted threshold of 0.70 for adequate internal consistency24. Curtis et al., while developing and validating the WUSPI, surveyed 64 athletes participating in a wheelchair sporting event for paralyzed veterans and found very high internal consistency (Cronbach’s alpha = 0.98)7,8. Ji‑Yeon et al. examined 64 patients with spinal cord injury who used wheelchairs and reported very high internal consistency for the Korean version (Cronbach’s alpha = 0.96)13. Arroyo‑Aljaro et al. also confirmed high internal consistency of the instrument in a group of 42 individuals with spinal cord injury (Cronbach’s alpha = 0.88)10.
Temporal consistency was estimated using the test–retest method. Twenty participants completed the WUSPI a second time after an average of 7 days. The ICC was very high (0.995), indicating almost perfect agreement between administrations and supporting the temporal consistency of the tool. Similarly, Curtis et al. reported high test–retest reliability (ICC = 0.99)7,8, and Arroyo‑Aljaro et al. registered very good repeatability (test–retest correlation 0.96; p = 0.01)10. Ji‑Yeon et al. recorded test–retest reliability for the total index score ranging from 0.88 to 0.99, indicating good to excellent reliability13.
To evaluate the internal structure of WUSPI‑Pol, we conducted additionally confirmatory factor analysis (CFA). We obtained satisfactory values: RMSEA = 0.076, CFI = 0.979, TLI = 0.969, and SRMR = 0.032. According to Hu and Bentler, a model can be considered a good fit when RMSEA < 0.06, CFI and TLI > 0.95, and SRMR < 0.08. As simultaneously meeting all four criteria may be overly restrictive, Hu and Bentler proposed a “two‑index” strategy whereby adequate fit is indicated when SRMR < 0.09 and at least one of the following holds: CFI > 0.96, TLI > 0.96, or RMSEA < 0.0627. Accordingly, our results can be considered to indicate acceptable fit.
In our study, ceiling/floor effects reaching or exceeding 40% were observed for several WUSPI items. This finding warrants cautious interpretation. On the one hand, it may reflect genuinely low levels of shoulder pain during everyday tasks among high‑functioning wheelchair athletes, who often display better physical conditioning, optimized wheelchair propulsion techniques, and effective compensatory strategies compared with non‑athlete users. On the other hand, such pronounced floor effects may indicate limited sensitivity of certain WUSPI items in this specific population. Therefore, future research should aim to distinguish these explanations, for example by comparing ceiling/floor effects in clinical and athletic samples and by adapting WUSPI items to sport‑specific demands to enhance sensitivity and ecological validity in athletic contexts.
In our study, external validity was assessed by correlating WUSPI‑Pol scores with those of two instruments previously validated in Poland: QuickDASH and SST. As expected, the correlations were statistically significant (p < 0.001).
Curtis et al. examined the validity of the WUSPI based on correlations with goniometric measurements of shoulder range of motion. Total WUSPI scores showed significant negative correlations with abduction (r = − 0.49), flexion (r = − 0.48), and extension (r = − 0.30). These moderate correlations have indicated that individuals with less shoulder range of motion report more pain during functional activities7. Ji‑Yeon et al. likewise reported correlations between total WUSPI score and shoulder abduction (r = − 0.59), flexion (r = − 0.58), extension (r = − 0.09), external rotation (r = − 0.07), and internal rotation (r = − 0.30), suggesting an association between higher WUSPI scores and loss of range of motion13.
Limitations of the study
The study has encompassed a specific and relatively homogeneous group. On the one hand, prior research highlights the high prevalence of shoulder pain in this population (38%–75%), stemming from substantial potential overload5. On the other hand, regular physical activity promotes protective adaptations—such as exercise‑induced hypoalgesia—and more economical performance of activities of daily living (e.g., wheelchair propulsion, transfers)28,29. These mechanisms may mitigate the pain reported during activities assessed by the WUSPI, despite the increased shoulder load during training and competition. Consequently, athletes may obtain lower WUSPI scores for activities of daily living than non‑athlete wheelchair users, in whom the absence of such adaptations may be associated with greater pain and higher WUSPI scores30.
Therefore, further research is needed in a larger sample, with greater participation of women and individuals who do not engage in sport.
Conclusion
The present study confirms that the Polish version of the Wheelchair User’s Shoulder Pain Index (WUSPI‑Pol) is a reliable and valid instrument for assessing shoulder pain in wheelchair athletes. The obtained results—high internal consistency, repeatability, and satisfactory values from factor analysis and external validity—indicate that WUSPI‑Pol can be successfully used in both clinical practice and research.
However, in light of the study’s limitations (a relatively small sample, predominance of men, and inclusion of athletes only), there is scope for further work. In particular, future studies should:
Encompass larger and more diverse samples, including women, non‑athletes, and individuals of different ages and with various causes of disability.
Analyze longitudinal variability—monitoring patients over longer periods would allow assessment of the usefulness of WUSPI‑Pol for tracking the progression of shoulder pain and the effectiveness of therapeutic interventions.
Implement the tool in clinical practice—for example, in rehabilitation and in evaluating the effects of physiotherapy, preventive training, and shoulder overload‑prevention programs for wheelchair users.
Combine WUSPI‑Pol with other clinical and biomechanical indicators—for example, range‑of‑motion measurements, muscle strength, or wheelchair propulsion technique—to develop comprehensive protocols for assessing the risk of shoulder pain.
Consider digital adaptation—creating an electronic version of the questionnaire would facilitate its use in telemedicine and multicenter studies.
Findings from the present study suggest that WUSPI‑Pol has great potential not only as a research instrument but also as a component of routine clinical practice in Poland. Further development and application may meaningfully improve quality of life for wheelchair users by enhancing the diagnosis, monitoring, and prevention of shoulder pain.
Author contributions
AS: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Validation, Visualization, Roles/Writing—original draft. Prepared Tables 1, 2, 3, 4, 5, 6 and 7, Writing—review & editing. BS: Validation, Formal analysis, Roles/Writing—original draft, Writing—review & editing. AWS: Data curation, Writing—review & editing. MC: Validation, Writing—review & editing. AWP: Supervision, Writing—review & editing. KC: Methodology, Supervision, Roles/Writing—original draft, Writing—review & editing. The authors read and approved the final manuscript.
Data availability
The datasets used and analyzed in the current study are available from the corresponding author on reasonable request. The Wheelchair User Shoulder Pain Instrument (WUSPI) is available, free of charge, to researchers and clinicians. To receive a copy of the instrument, please send an email to kacurtis@utep.edu with institutional affiliation and information about proposed use.
Declarations
Competing interests
The authors declare no competing interests.
Ethics approval
The study was conducted in accordance with the Declaration of Helsinki and approved by the Bioethics Committee of the University of Rzeszow (Resolution No. 19/12/2019).
Consent to participate
All participants provided written informed consent prior to enrollment in the study.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Vincent, C., Gagnon, D. H. & Dumont, F. ADMI group. Pain, fatigue, function and participation among long-term manual wheelchair users partnered with a mobility service dog. Disabi L Rehabil Assist. Technol.14 (2), 99–108 (2019). [DOI] [PubMed] [Google Scholar]
- 2.Mason, B., Warner, M., Briley, S., Goosey-Tolfrey, V. & Vegter, R. Managing shoulder pain in manual wheelchair users: a scoping review of Conservative treatment interventions. Clin. Rehabil. 34 (6), 741–753 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Curtis, K. A. et al. Shoulder pain in wheelchair users with tetraplegia and paraplegia. Arch. Phys. Med. Rehabil. 80 (4), 453–457 (1998). [DOI] [PubMed] [Google Scholar]
- 4.Riley, A. H. & Callahan, C. Shoulder rehabilitation protocol and equipment fit recommendations for the wheelchair sport athlete with shoulder pain. Sports Med. Arthrosc. Rev.27 (2), 67–72 (2019). [DOI] [PubMed] [Google Scholar]
- 5.Karasuyama, M., Oike, T., Okamatsu, S. & Kawakami, J. Shoulder pain in wheelchair basketball athletes: a scoping review. J. Spinal Cord Med.46 (5), 753–759 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Koerber, S. N., Wager, S. G., Zynda, A. J. & Santa Barbara, M. T. Scoping review: reducing musculoskeletal injury risk factors for adaptive sport athletes through prevention programs. Am. J. Phys. Med. Rehabil. 103 (11), 1045–1050 (2024). [DOI] [PubMed] [Google Scholar]
- 7.Curtis, K. A. et al. Reliability and validity of the wheelchair user’s shoulder pain index (WUSPI). Paraplegia33 (10), 595–601 (1995). [DOI] [PubMed] [Google Scholar]
- 8.Curtis, K. A. et al. Development of the wheelchair user’s shoulder pain index (WUSPI). Paraplegia33 (5), 290–293 (1995). [DOI] [PubMed] [Google Scholar]
- 9.Harvey, L. A. & Glinsky, J. V. Clinimetrics: the wheelchair user’s shoulder pain index (WUSPI). J. Physiother. 65 (1), 55 (2019). [DOI] [PubMed] [Google Scholar]
- 10.Arroyo-Aljaro, R. & Gonzale-Viejo, M. Validacione al Castellano Del wheelchair users shoulder pain index (WUSPI). Rehabilitacione43 (1), 2–9 (2009). [Google Scholar]
- 11.Larsen, C. M., Bruun, P., Hansen, S., Hansen, S. & Juul-Kristensen, L. H. B. Translation and Cross-Cultural Adaptation of the Danish Version: Wheelchair Users Shoulder Pain Index (WUSPI). (2015).
- 12.Leiulfsrud, A. S. & Lie, M. Translation, cross-cultural adaptation and reliability of a Norwegian version of the Wheelchair User’s Shoulder Pain Index (WUSPI). (NTNU 2019).
- 13.Ji-Yeon, P. & Sang-Hyun, C. The reliability and validity of Korean version of the wheelchair user’s shoulder pain index in wheelchair users. J. Korean Soc. Phys. Med.8 (4), 573–528 (2013). [Google Scholar]
- 14.Tsunoda, K. et al. Associations between wheelchair user’s shoulder pain index and tendinitis in the long head of the biceps tendon among female wheelchair basketball players from the Japanese National team. Asia Pac. J. Sports Med. Arthrosc. Rehabil Technol.24, 29–34 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hurd, W. et al. Self-reported shoulder pain assessment in adults with spinal cord injury who use a manual wheelchair: a scoping review. J. Spinal Cord Med. 1–15 (2025). [DOI] [PubMed]
- 16.Cruchinho, P. et al. Translation, cross-cultural adaptation, and validation of measurement instruments: a practical guideline for novice researchers. J. Multidiscip HealthcInt J. Qual. Health Care. 1719 (6), 2701–2728 (20242007). [DOI] [PMC free article] [PubMed]
- 17.Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. Int. J Qual. Health. Care. 19 (6), 349-57 (2007). [DOI] [PubMed] [Google Scholar]
- 18.Beaton, D. E., Wright, J. G. & Katz, J. N. Upper extremity collaborative Group. Development of the quickdash: comparison of three item-reduction approaches. J. Bone Joint Surg. Am.87 (5), 1038–1046 (2005). [DOI] [PubMed] [Google Scholar]
- 19.Golicki, D., Krzysiak, M. & Strzelczyk, P. Translation and cultural adaptation of the Polish version of the disabilities of the Arm, shoulder and hand (DASH) and quickdash questionnaires. Orthop. Traumatol. Rehabil. 16 (4), 387–395 (2014). [DOI] [PubMed] [Google Scholar]
- 20.Lippitt, S. B. et al. American Academy of Orthopaedic Surgeons,. A practical tool for evaluating function: the simple shoulder test. The shoulder: a balance of mobility and stability. Rosemont (IL) 501–518 (1993).
- 21.Van Kampen, D. A., van Beers, L. W., Scholtes, V. A., Terwee, C. B. & Willems, W. J. Validation of the Dutch version of the simple shoulder test. J. Shoulder Elb. Surg.21 (6), 808–814 (2012). [DOI] [PubMed] [Google Scholar]
- 22.Ślęzak, M. et al. Polish cultural adaptation of general shoulder assessment scores in use for painful shoulder: ASES, UCLA, constant Score, SST (Part I). Preliminary study. Rehabil Orthop. Neurophysiol. Sport Promot. 17, 7–27 (2016). [Google Scholar]
- 23.Core Team R: A Language and Environment for Statistical Computing. https://www.r-project.org (R Foundation for Statistical Computing, 2020).
- 24.Nunnally, J. C. Psychometric Theory 2nd edn (McGraw-Hill, 1978).
- 25.Streiner, D. L. & Norman, G. R. Health Measurement Scales (Oxford University Press, 1995).
- 26.Koo, T. K. & Li, M. Y. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J. Chiropr. Med.15 (2), 155–163 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hu, L. T. & Bentler, P. M. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct. Equ Model.6 (1), 1–55 (1999). [Google Scholar]
- 28.Vaegter, H. B. & Jones, M. D. Exercise-induced hypoalgesia after acute and regular exercise: experimental and clinical manifestations and possible mechanisms in individuals with and without pain. Pain Rep.5(5), e823 (2020). [DOI] [PMC free article] [PubMed]
- 29.Briley, S. J., Vegter, R. J. K., Goosey-Tolfrey, V. L. & Mason, B. S. Alterations in shoulder kinematics are associated with shoulder pain during wheelchair propulsion sprints. Scand. J. Med. Sci. Sports. 32 (8), 1213–1223 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Soo Hoo, J. A. et al. Shoulder pain and ultrasound findings: a comparison study of wheelchair athletes, nonathletic wheelchair users, and nonwheelchair users. PMR14 (5), 551–560 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and analyzed in the current study are available from the corresponding author on reasonable request. The Wheelchair User Shoulder Pain Instrument (WUSPI) is available, free of charge, to researchers and clinicians. To receive a copy of the instrument, please send an email to kacurtis@utep.edu with institutional affiliation and information about proposed use.