Advertisement

Development, Internal Construct, and Unidimensionality of the Tetraplegia Upper Limb Activities Questionnaire, TUAQ. Part 1

  • Johanna Wangdell
    Correspondence
    Corresponding author Johanna Wangdell, PhD, Centre for Advanced Reconstruction of Extremities, CARE, Mölndals sjukhus/ Sahlgrenska University Hospital, 431 80 Mölndal, Sweden.
    Affiliations
    Centre for Advanced Reconstruction of Extremities, Sahlgrenska University Hospital/Mölndal, Mölndal, Sweden

    Department of Hand Surgery, Institute of Clinical Sciences, University of Gothenburg, Gothenburg, Sweden
    Search for articles by this author
  • Bridget Hill
    Affiliations
    Department of Plastic and Reconstructive Surgery, Austin Health, Melbourne, Australia

    Epworth Monash Rehabilitation Medicine Unit, Melbourne, Victoria, Australia
    Search for articles by this author
  • Jennifer A. Dunn
    Affiliations
    Department of Orthopaedic Surgery and Musculoskeletal Medicine, University of Otago, Christchurch, New Zealand
    Search for articles by this author
Open AccessPublished:May 17, 2022DOI:https://doi.org/10.1016/j.apmr.2022.02.024

      Highlights

      • The Tetraplegia Upper Limb Activity Questionnaire (TUAQ) is based on 708 activities identified by persons living with tetraplegia.
      • The questionnaire comprises 10 items.
      • Items are scored for both performance and satisfaction on a 5-point response scale.
      • Results supports the internal construct and unidimensionality.
      • The TUAQ is well targeted regardless of age, sex, spasticity, or time.

      Abstract

      Objective

      This study aims to develop a patient-reported outcome measure that focuses on relevant daily activities relying on upper extremity for individuals with tetraplegia.

      Design

      Cross-sectional study.

      Setting

      Spinal cord injury units in 2 countries.

      Participants

      Ninety-nine individuals (N=99) with C2-C8, American Spinal Injury Association Impairment Scale A-D tetraplegia, mean age 46 years, 1- 43 years post injury.

      Interventions

      Thirteen items included in the initial testing were chosen from 708 activity limitations identified by individuals with tetraplegia. Items were pilot tested for wording, response options, and relevance for both performance and satisfaction. Items were analyzed and reselected using exploratory factor analysis and Rasch analysis for local dependency, dimensionality, differential item functioning (DIF), threshold response, and targeting.

      Main Outcome Measures

      Not applicable.

      Results

      Exploratory factor analysis supported a 2-factor solution for both performance and satisfaction. While data fit the Rasch model, there was evidence of local dependency and multiple disordered thresholds. Three items were removed because of high interitem correlation and DIF and the scale rescored to 5 response options. The remaining 10 items demonstrated fit to the Rasch model, with no local dependency, no multidimensionality, no item or person misfit, and minimal disordered thresholds.

      Conclusions

      Results support the internal construct validity and unidimensionality of the Tetraplegia Upper Limb Activities Questionnaire (TUAQ), a 10-item, 5-response patient-reported outcome measure assessing performance and satisfaction with activities targeted to the upper extremity for individuals with tetraplegia. Further testing is required and ongoing to evaluate reliability and responsiveness of the TUAQ.

      Keywords

      List of abbreviations:

      ADL (activities of daily living), AIS (American Spinal Injury Association Impairment Scale), COPM (Canadian Occupational Performance Measure), DIF (differential item functioning), ICF (International Classification of Functioning, Disability, and Health), PROM (patient-reported outcome measure), SCI (spinal cord injury), TUAQ (Tetraplegia Upper Limb Activities Questionnaire)
      After a cervical spinal cord injury (SCI), people with tetraplegia demonstrate a wide variety of activity restrictions in their daily living, primarily based on the severity and level of injury. Historically, outcome measures have focused on impairments such as range of movement and strength. Recently outcome measures specifically related to upper extremity activity in people with tetraplegia have been identified as an important area of development.
      • Sinnott KA
      • Dunn JA
      • Wangdell J
      • Johanson ME
      • Hall AS
      • Post MW.
      Measurement of outcomes of upper limb reconstructive surgery for tetraplegia.
      • Fawcett J
      • Curt A
      • Steeves J
      • et al.
      Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials.
      • Jones LAT
      • Bryden A
      • Wheeler TL
      • et al.
      Considerations and recommendations for selection and utilization of upper extremity clinical outcome assessments in human spinal cord injury trials.
      While impairment tests provide information on an individual's hand function in terms of strength and movement, they do not provide any information as to how this is translated into activities of daily living (ADL). The ability to perform daily activities is challenging to measure because it is multifaceted, and it is difficult to determine whether improvement is because of neurologic improvement or development of adaptive skills. In addition, relevant daily activities differ between individuals and can change over time. The upper limb is used in the majority of daily activities and even more so for an individual with tetraplegia. To measure changes in upper limb use in this population because of an intervention, to date most studies have used impairment tests. This is primarily because of the difficulties finding an ADL measure that is specific and sensitive enough to detect the small changes in this population that can affect hand use.
      • Sinnott KA
      • Dunn JA
      • Wangdell J
      • Johanson ME
      • Hall AS
      • Post MW.
      Measurement of outcomes of upper limb reconstructive surgery for tetraplegia.
      Several authors have reviewed the existing outcome measures highlighting the issues with the these tools and the need to develop a reliable, valid, and responsive outcome measures that shows change in ADL function in this population.
      • Sinnott KA
      • Dunn JA
      • Wangdell J
      • Johanson ME
      • Hall AS
      • Post MW.
      Measurement of outcomes of upper limb reconstructive surgery for tetraplegia.
      ,
      • Velstra IM
      • Ballert CS
      • Cieza A.
      A systematic literature review of outcome measures for upper extremity function using the international classification of functioning, disability, and health as reference.
      ,
      • van Tuijl JH
      • Janssen-Potten YJ
      • Seelen HA.
      Evaluation of upper extremity motor function tests in tetraplegics.
      There are a number of hand function tests available, some developed specifically for the population with tetraplegia and others for a generic population with upper limb impairment.
      • Sinnott KA
      • Dunn JA
      • Wangdell J
      • Johanson ME
      • Hall AS
      • Post MW.
      Measurement of outcomes of upper limb reconstructive surgery for tetraplegia.
      The tetraplegia-specific tests include the Graded and Redefined Assessment of Strength, Sensibility, and Prehension,
      • Kalsi-Ryan S
      • Curt A
      • Verrier MC
      • Fehlings MG
      Development of the Graded Redefined Assessment of Strength, Sensibility and Prehension (GRASSP): reviewing measurement specific to the upper limb in tetraplegia.
      Grasp Release Test,
      • Wuolle KS
      • Van Doren CL
      • Thrope GB
      • Keith MW
      • Peckham PH.
      Development of a quantitative hand grasp and release test for patients with tetraplegia using a hand neuroprosthesis.
      Van Lieshout Test,
      • Spooren AI
      • Janssen-Potten YJ
      • Post MW
      • Kerckhofs E
      • Nene A
      • Seelen HA.
      Measuring change in arm hand skilled performance in persons with a cervical spinal cord injury: responsiveness of the Van Lieshout Test.
      Motor Capabilities Scale,
      • Fattal C.
      Motor capacities of upper limbs in tetraplegics: a new scale for the assessment of the results of functional surgery on upper limbs.
      Capabilities of the Upper Extremity-Test.
      • Marino RJ
      • Goin JE.
      Development of a short-form Quadriplegia Index of Function scale.
      Generic hand function tests include the Sollerman test,
      • Sollerman C
      • Ejeskar A.
      Sollerman hand function test. A standardised method and its use in tetraplegic patients.
      Jebsen-Taylor Hand Function Test,
      • Jebsen RH
      • Taylor N
      • Trieschmann RB
      • Trotter MJ
      • Howard LA.
      An objective and standardized test of hand function.
      and Action Research Arm Test.
      • Carroll D.
      A quantitative test of upper extremity function.
      They were all developed to evaluate hand function and/or grip and pinch but may not translate into the ability to use the hand in daily activities. There are hand ability questionnaires that are not specific for SCI, for example, the Disabilities of Arm, Shoulder, and Hand
      • Hudak PL
      • Amadio PC
      • Bombardier C.
      Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand) [corrected]. The Upper Extremity Collaborative Group (UECG).
      and AbilHand.
      • Penta M
      • Thonnard JL
      • Tesio L.
      ABILHAND: a Rasch-built measure of manual ability.
      However these measures include several items not relevant for persons with tetraplegia.
      Most ADL tests are designed to reflect daily life in general or to calculate degrees of dependence.
      • Keith RA
      • Granger CV
      • Hamilton BB
      • Sherwin FS.
      The functional independence measure: a new tool for rehabilitation.
      Also, diagnosis-specific ADL instruments, both observer and questionnaires, cover many activities that do not solely assess the upper limb, for example transfers, walking, or bowel and bladder function.
      • Marino RJ
      • Goin JE.
      Development of a short-form Quadriplegia Index of Function scale.
      ,
      • Catz A
      • Itzkovich M
      • Agranov E
      • Ring H
      • Tamir A.
      SCIM-spinal cord independence measure: a new disability scale for patients with spinal cord lesions.
      These broad ADL instruments are therefore not sensitive enough to capture clinically relevant changes in interventions specifically targeting the upper limb.
      The most common outcome measure to evaluate change in everyday activities after upper limb reconstruction in the population with tetraplegia is the Canadian Occupational Performance Measure (COPM).
      • Koch-Borner S
      • Dunn JA
      • Friden J
      • Wangdell J.
      Rehabilitation after posterior deltoid to triceps transfer in tetraplegia.
      • Wangdell J
      • Fridén J.
      Satisfaction and performance in patient selected goals after grip reconstruction in tetraplegia.
      • Wangdell J
      • Fridén J.
      Activity gains after reconstructions of elbow extension in patients with tetraplegia.
      • Wangdell J
      • Bunketorp-Käll L
      • Koch-Borner S
      • Fridén J.
      Early active rehabilitation after grip reconstructive surgery in tetraplegia.
      • Wangdell J
      • Reinholdt C
      • Fridén J.
      Activity gains after upper limb surgery for spasticity in patients with spinal cord injury.
      • Reinholdt C
      • Friden J.
      Outcomes of single-stage grip-release reconstruction in tetraplegia.
      The strength of the COPM is that it encourages the individual to determine their limitations in their daily life that is of highest importance for them. It is therefore highly patient-centered, which is the overall aim of the new instrument described in this article. While the patient-centered approach is a strength, it also a limitation. Because it is so highly individualized, group analyses and comparisons between settings are difficult because the activities evaluated differ between individuals.
      In conclusion, there is a lack of a reliable, valid, and responsive patient-reported outcome measure (PROM) that is clinically feasible and assesses daily upper limb activities relevant to people with tetraplegia.

      Aim

      We aimed to develop and evaluate the internal construct validity and dimensionality of a new PROM specific to daily upper limb activities relevant to people with tetraplegia.
      The predefined criteria of the new measure were the following: (1) a patient-reported activity measure, explicitly measuring activities using upper limb function for individuals with tetraplegia, typically C5- C8, but also higher incomplete injuries with preserved function in upper limbs; (2) able to be administered as a questionnaire (paper and electronic), thus enabling remote follow-up outside clinic appointment visits; (3) fast and easily administered for the individual and clinicians; and (4) stable psychometric properties, to be used for clinical and research purposes.

      Methods

      The Tetraplegia Upper Limb Activities Questionnaire (TUAQ) was developed in a collaboration between universities in 2 countries. The study protocols were approved by ethics committee of Göteborg (D-nr: 099-16) and University of Otago Ethics Committee (Health) ref 18/055).

      Development of the TUAQ

      The development and initial psychometric testing of the TUAQ involved 3 phases (fig 1).

      Phase 1. Item generation

      Prioritized daily activities dependent on hand function were collected from COPM assessments completed by people with tetraplegia considering undergoing upper limb reconstructive surgery for grip and/or pinch reconstruction at 2 specialized units. The 708 identified activities were independently classified into the International Classification of Functioning, Disability, and Health (ICF) domains by 2 of the authors (J.D., J.W.) following recommended linking rules.
      • Cieza A
      • Geyh S
      • Chatterji S
      • Kostanjsek N
      • Ustun B
      • Stucki G.
      ICF linking rules: an update based on lessons learned.
      The activities (regardless of sex, age, and level of injury) were ranked, and the 12 most common activities were included for pilot testing (table 1). Selection of activities was performed by 2 of the authors (J.D., J.W.), both with more than 15 years clinical experience in SCI rehabilitation.
      Table 1The 12 most common activities mentioned from the 708 identified activity limitations, linked to ICF
      Goal GroupFrequency (n)%ICF ChapterICF Link
      Eating659.15 Self-care550
      Writing with a pen598.33 Communication345
      Drinking375.25 Self-care560
      Manage buttons and zippers283.95 Self-care540
      Cut food when eating263.65 Self-care550
      Handle money (wallet, coins, credit card)212.98 Major life areas860
      Pick up items from flat surface192.74 Mobility
      Shaving/put on makeup182.55 Self-care520
      Open bottles and jars152.16 Domestic life630
      Handle a book/paper152.19 Communication, social and civic life; 8 Major life areas920; 859
      Adjust clothing101.45 Self-care540
      Buttering a sandwich101.46 Domestic life630
      Total32345.2

      Scoring system

      To generate an appropriate measurement scale for the TUAQ existing outcome measures were reviewed. The scale used in COPM (rated performance and satisfaction 1-10) and the suggested scale in the ICF (rated difficulty 1-5) were selected for trial.
      • Law M
      • Baptiste S
      • McColl M
      • Opzoomer A
      • Polatajko H
      • Pollock N.
      The Canadian Occupational Performance Measure: an outcome measure for occupational therapy.
      ,
      World Health Organization
      International Classification of functioning, disability and health (ICF).

      Phase 2. Pilot testing of items and scoring system

      To ensure items were relevant and to determine face validity, the 12 preliminary items were pilot tested by 6 people with tetraplegia. Each individual rated their ability to perform the activities using both scales: (1) the COPM scale (1-10 for performance [1=impossible and 10=extremely well] and 1-10 for satisfaction with the performance [1=not satisfied at all and 10=satisfied]) and (2) the ICF experienced problems 1-5 (1=no problem, 5=complete problem). Additionally, they were asked if the wording of items was understandable and relevant and if any important activities were missing. After feedback, items were subsequently tested by an additional 15 people with tetraplegia. Four clinicians, both occupational therapists and physiotherapists, experienced in SCI upper limb rehabilitation provided additional written feedback on definitions, understanding, and relevance of the items for people with tetraplegia and on the 10-point scale. Items were subsequently forward and back translated between Swedish and English and cross-culturally adapted (the use of specific items) according to standardized guidelines.
      • Beaton DE
      • Bombardier C
      • Guillemin F
      • Ferraz MB.
      Guidelines for the process of cross-cultural adaptation of self-report measures.
      The measure was developed using the 2 language versions in parallel throughout development.

      Phase 3. Initial psychometric testing for internal construct validity and dimensionality

      Participants and procedures

      Individuals with cervical SCI with no other injuries or limitations in their upper limb and ability to use their hand in ADL to some extent (typically C5-8 but also incomplete higher injuries) were included. They were recruited from 3 clinical settings and included individuals with long-standing tetraplegia in routine SCI follow-up, pre- and post upper limb reconstructive surgery, and a registry. Data were psychometrically evaluated using exploratory factor analysis and Rasch analysis guided by the Consensus-Based Standards for the Selection of Health Measurement Instruments checklist.
      • Mokkink LB
      • Terwee CB
      • Patrick DL
      • et al.
      The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes.

      Data collection

      Demographic data of sex, age, motor level of injury according to International Standards for Neurological Classification of Spinal Cord Injury,
      • Kirshblum SC
      • Burns SP
      • Biering-Sorensen F
      • et al.
      International standards for neurological classification of spinal cord injury (revised 2011).
      time post injury, whether they experienced limiting spasticity in upper limb (yes/no), and surgical interventions to the upper limb were collected. Participants filled out the TUAQ independently or with assistance if they were unable to write or type. The Swedish version was completed on paper. and the English version was electronic.

      Data analysis

      Data analyses were performed using SPSS for Mac Version 25.a For Rasch analyses the RUMM 2030 partial credit model softwareb was used to allow the thresholds to vary for each item rather than restricting the model by assuming that all response options are the same. Raw and logit scores were reported as mean, SD, and range. Data were analyzed separately for performance and satisfaction results. First, exploratory factor analysis was used to assess dimensionality and identify the presence of any subset of items for both the performance and satisfactions scales. Factorability of the data were assessed using the Kaiser-Meyer-Olkin measure of sampling adequacy and Bartlett test of sphericity and principal component analysis in SPSS.
      • Tabachnick BG
      • Fidell LS.
      Using multivariate statistics.
      Second, data were then analyzed using Rasch analysis following published criteria.
      • Tennant A
      • Conaghan PG.
      The Rasch measurement model in rheumatology: what is it and why use it? When should it be applied, and what should one look for in a Rasch paper?.
      ,
      • Pallant JF
      • Tennant A.
      An introduction to the Rasch measurement model: an example using the Hospital Anxiety and Depression Scale (HADS).
      The overall fit of the scale was evaluated using a chi-square statistic with a nonsignificant P value indicative of adequate fit to the Rasch model. Fit of the individual items and persons were assessed using 2 indicators: a nonsignificant chi-square statistic and a fit residual value within the range ±2.5. Person separation reliability was evaluated, and values can be interpreted in the same way as Cronbach α coefficients, with values >0.7 considered the minimally accepted value for distinguishing among groups of individuals.
      Differential item functioning (DIF) was conducted for age, sex, time post injury, and country, fixed variables that would not be expected to change if people with the same level of the underlying trait respond to the items in the same way. Response dependency among items was investigated by inspecting the residual correlation matrix for values >0.2. Dimensionality testing involved conducting a series of t tests to compare Rasch derived scores from 2 subsets of items identified from principal components analysis of the residuals. Less than 5% of tests should be significantly different (or the lower bound of the binomial confidence interval should overlap by 5%) for the scale to be considered unidimensional.
      Items were systematically evaluated, and where data did not fit the model items were sequentially removed and data reanalyzed in an iterative process using the following decision rules: (1) importance to people with tetraplegia, (2) maintenance of ICF categories, (3) maintenance of a variety of different grip styles, and (4) maintenance of the fit to the Rasch model for both the performance and satisfaction scales while limiting local dependency and maintaining dimensionality. Sample size for Rasch analysis is influenced by scale targeting. A planned sample size of 108 participants provided accurate estimations of person and item locations within 0.5 logits (99% confidence interval).

      Results

      Phase 1. Identification of items and scale

      From the 708 prioritized activities identified by people with tetraplegia in 2 different countries, the 12 most highest ranked activities relevant to most people with tetraplegia independent of sex, age, and level of injury were chosen for inclusion in the test version of the TUAQ. These 12 activities represented 45% of all initially identified 708 prioritized activities and covered 6 different ICF chapters (see table 1). The activities appear to capture a variety of ADL using different grip styles with varied complexity and were relevant to most people with tetraplegia.

      Phase 2. Face validity and feasibility: pilot testing

      Participants in the first pilot round (n=6) reported that the 1-10 scale for performance and satisfaction best reflected their ability and was easier to score than the 5-point score. Two changes were made: (1) drinking was separated into drinking from a bottle and drinking from a mug because each requires a different type of grip and (2) a separate column was included for each item to indicate the use of splints or aids for clinical purposes not included in the score. After the second pilot test (n=15 persons with tetraplegia and n=4 clinicians), minor changes to were made to clarify wording. Ongoing translation and cross-cultural adaptation performed at this point did not raise any concerns.

      Phase 3. Exploratory factor analyze and Rasch analysis

      Ninety-nine people with tetraplegia were recruited to this phase (71 in Sweden and 28 in New Zealand). The mean age of participants was 45 years (range, 16-80 years), and 69 (70%) were male. Time post injury ranged from 0.5-43 years and level of injury between C2-C8 with mostly American Spinal Injury Association Impairment Scale (AIS) A (55%) but also AIS B, C, and D. Sixty-two participants indicated they experienced limiting spasticity (table 2).
      Table 2Demographics. (N=99)
      CharacteristicsMean (SD)Range
      Age (yrs)45.3 (16.1)16-80
      Time post injury (yrs)10.5 (11.3)0.5-43
      CharacteristicsVariablesN (%)
      CountryNZ28 (28%)
      Sweden71 (72%)
      SexFemale30 (30%)
      Male69 (70%)
      ISNSCIC21 (1%)
      C33 (3%)
      C46 (7%)
      C522 (23%)
      C624 (24%)
      C733 (34%)
      C81 (1%)
      missing9 (10%)
      AISA51 (51%)
      B17 (18%)
      C7 (8%)
      D17 (17%)
      missing7 (8%)
      Limiting spasticityYes62 (63%)
      No36 (36%)
      missing1 (1%)
      The 13 items were analyzed using exploratory factor analysis because the Kaiser-Meyer-Olkin measures of sampling adequacy were 0.932 and 0.922, and the Bartlett test of sphericity was highly significant (P<.001) for the performance and satisfaction scales. This indicated the factorability of the data file. Principal component analysis revealed 2 eigenvalues >1, suggesting a 2-factor solution that explained 69% and 68% of the variation for the performance and satisfaction scales, respectively. The scree plot and parallel analysis identified a possible 2-factor solution for both scales.

      Rasch analysis

      Overall fit statistics of the 2 scales fit the Rasch model, (table 3, analyses 1 and 2) with nonsignificant chi-square values (P=.74 and P=.76 Bonferroni adjusted), item and person fit residual SDs close to 1, and person separation indices of 0.94 and 0.91, respectively. No item or person was misfitting; however, 7 pairs of items correlated >0.2 (drinking from a bottle, drinking from a mug, and eating with a fork and/or spoon for both performance and satisfaction scales, cutting food and buttering bread for satisfaction scale only). Thresholds were disordered for 22 of the total 26 items (13 for performance and 13 for satisfaction) and the target map identified multiple items that were measuring the same level of ability. Both of these factors indicate local dependency, which may artificially inflate the person separation index. There was uniform DIF for “adjust clothing” by country for performance and satisfaction, uniform DIF for “zipping clothing” performance by country, and nonuniform DIF for “drink from a bottle” by sex for performance and satisfaction. The examination of DIF is only exploratory at this stage, and further testing is required to confirm these results.
      Table 3Summary of fit statistics
      Item Fit ResidualPerson Fit ResidualItem Trait InteractionUnidimensionality
      ActionAnalysisItems (n)MeanSDMeanSDχ
      • Fawcett J
      • Curt A
      • Steeves J
      • et al.
      Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials.
      (df)
      P Value95% CIPSI
      Initial Performance1130.240.90−0.240.7321.03 (26).741.0-1.50.94
      Delete 3 items2100.260.91−0.320.7613.68 (20).850.2-1.00.92
      Rescore3100.120.84−0.211.0221.96 (20).340.3-1.00.92
      Initial satisfaction4130.170.670.171.0020.7 (26).760.5-1.50.92
      Delete 3 items5100.170.850.181.0714.46 (20).810.2-1.00.90
      Rescore6100.140.86−0.191.0918.03 (20).580.1-0.70.9
      Abbreviations: CI, confidence interval; PSI, person separation index.
      Three items were removed: drinking from a mug (because of correlation >0.2 with 2 other items), buttering bread (because of correlation with cutting food), and zipping clothing (because of DIF issues by country and difficulties in standardizing the activity on a day-to-day basis [ie, size or type of zipper]). This resulted in good fit to the Rasch model (see table 3, analyses 2 and 5) for both scales; however, there were still multiple disordered thresholds. All remaining items across the 2 scales were rescored (1:1,2,3,4; 2:5,6,7; 3:8; 4:9; and 5:10). The final 10-item, 5-response option performance and satisfaction scales showed adequate fit to the Rasch model with no evidence of local dependency or dimensionality (see table 3, analyses 3 and 6). Items were well targeted across the spectrum of ability with no floor or ceiling effect (fig 2). The logit value for persons was −0.53 for performance and −0.62 for satisfaction, indicating most had a lower level of ability than the average of item difficulty and lower satisfaction with their ability to perform them. This is supported by fig 2, which shows that while there is a good spread of items, the person's scores are congregated more to the left of the mean 0 location score.
      Fig 2
      Fig 2Target map for person and item thresholds for the final 10-item version. (A) Performance scale; (B) Satisfaction scale. The upper section of each panel shows the distribution for the participants. The lower section of each panel shows the distribution of the items.
      Items generally shared a similar hierarchical order for both performance and satisfaction scales (table 4). There were 2 points of difference. Participants rated their satisfaction with writing more highly than their performance of the task, while conversely they found adjusting clothing more difficult to perform than some other activities but were more satisfied with this outcome.
      Table 4Individual item difficulty
      Performance ItemsLocationSEFitResddfχ
      • Fawcett J
      • Curt A
      • Steeves J
      • et al.
      Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials.
      dfProbability
      Eating with a fork or spoon−1.2940.1480.84784.943.78920.150361
      Drinking from a bottle−0.8560.1320.41078.84.51420.104672
      Grasp and reposition book/tablet−0.8460.1471.13684.940.75220.686722
      Shaving/put on makeup−0.4640.134−0.94777.221.58620.452582
      Writing with a pen (with your best hand)−0.1670.1521.3484.961.69720.428076
      Pick up items form a flat surface0.0050.1370.28284.960.54620.761024
      Adjust upper half clothing pulling down back to waist level0.0740.138−1.12884.964.51620.104576
      Handle banknotes credit card in/out of wallet0.7500.145−0.02984.960.50120.778556
      Open previously opened jars0.9020.139−0.19184.960.81920.664016
      Cut food when eating (with your best hand)1.8950.162−0.55384.093.23720.19819
      Satisfaction ScaleLocationSEFitResddfχ
      • Fawcett J
      • Curt A
      • Steeves J
      • et al.
      Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials.
      dfProbability
      Eating with a fork or spoon−1.2380.1350.27884.060.62320.732297
      Drinking from a bottle−0.8660.1231.17677.202.34420.309747
      Grasp and reposition book/tablet−0.8090.1380.50984.060.06920.965882
      Shaving/put on makeup−0.6020.125−0.71776.341.48920.47502
      Adjust upper half clothing pulling down back to waist level−0.0240.124−0.61684.065.07320.079132
      Pick up items form a flat surface−0.0230.129−0.19384.061.5520.460737
      Writing with a pen (with your best hand)0.1000.1321.90784.061.16520.558634
      Handle banknotes credit card in/out of wallet0.5540.133−0.61384.062.6320.268426
      Open previously opened jars0.6380.128−0.47684.062.49520.287292
      Cut food when eating (with your best hand)2.270.1540.16584.060.59320.743437
      Abbreviation: FitResd, fit residual.
      A conversion chart showing the Rasch-derived score for each raw score adjusted to a 0-100 interval scale was calculated and appended to the TUAQ questionnaire (appendix 1). The Rasch derived scores only apply to respondents with a complete data.

      Discussion

      The initial psychometric testing of the TUAQ shows promising results in the process of developing a PROM that focuses on relevant daily activities that use the upper extremity for individuals with tetraplegia. The TUAQ has the potential to fill an identified gap in outcome measurement for the population with tetraplegia.
      • Sinnott KA
      • Dunn JA
      • Wangdell J
      • Johanson ME
      • Hall AS
      • Post MW.
      Measurement of outcomes of upper limb reconstructive surgery for tetraplegia.
      ,
      • Velstra IM
      • Ballert CS
      • Cieza A.
      A systematic literature review of outcome measures for upper extremity function using the international classification of functioning, disability, and health as reference.
      ,
      • van Tuijl JH
      • Janssen-Potten YJ
      • Seelen HA.
      Evaluation of upper extremity motor function tests in tetraplegics.
      With further testing, this assessment has the potential to be used to follow individuals over time or groups across different management interventions.
      Results from the preliminary validation of the TUAQ suggest it is a 10-item, 5-response unidimensional PROM designed to measure activity performance and satisfaction in persons with tetraplegia regardless of age, sex, time post injury, or limiting spasticity. The TUAQ is scored by totaling the performance scores and satisfaction scores separately. This gives a raw score from 1-50 for each scale. For all statistical analyses the raw score needs to be converted to an interval score using the relevant table on the scoring sheet. This then gives a score from 0-100 for each scale, where a lower scores represents lower performance or satisfaction. Results should be reported TUAQ Performance xx/100; TUAQ Satisfaction xx/100 to demonstrate the conversion into the interval score. The 10-item, 5-point response scale and conversion table included in the scoring sheet meet the our predefined criteria of being easy to administer for both the individual and health professional.
      Face validity and relevance of the measure by the target group is critical for any outcome measure. Therefore, items were derived from analysis of 708 activity limitations identified by individuals with C5- C8 tetraplegia from 2 countries. The test versions were reviewed by 21 individuals with tetraplegia and 4 experienced therapists with no substantial changes. Face validity is further supported by the targeting of the 2 scales, which have a good spread of items across the ability level of a wide cross section of people with tetraplegia. While the original items for the TUAQ were derived from people with injury levels C5-C8, during the recruitment phase in the clinical setting we realized that the questionnaire would be relevant to all individuals with tetraplegia. The increasing number of incomplete higher injuries (C2-4 AIS C and D) were therefore also included. The inclusion of this group did not lead to any outliers in the results. This suggests that the TUAQ can be relevant to individuals with tetraplegia, but a determined cutoff for recommended use with higher levels of incomplete tetraplegia is still to be defined. Limiting spasticity was self-reported by 62 participants but it did not appear to have influenced the overall fit to the Rasch model. Moreover, there were no misfitting participants, which suggests that the presence or absence of spasticity does not influence how people with tetraplegia use their hands in the items assessed by this measure.
      The items demonstrated a good variation in difficulty for both performance and satisfaction. The main difference between the 2 scales was with writing, which participants were less likely to rate high on satisfaction than performance. We postulate this may be because while the actual performance of writing might not be difficult, people may be less satisfied with the actual outcome (eg, legibility, form, speed, personal touch, etc). This reinforces the importance of measuring both performance and satisfaction.
      The 3 excluded items in the final version of the TUAQ—drinking from mug, buttering bread, and zipping clothing—were considered misfitting possibly because of the variation of object design and grasp patterns or compensatory techniques that can be used to perform the activity. As an example, individuals with tetraplegia often adapt their zippers with rings to enable a finger slipped into the ring rather than using a pincher-type grasp, or they may chose clothing without zippers, so performance can vary greatly depending on garment. After making these changes the final TUAQ fit the Rasch model and the retained activities still covers 6 of the 9 chapters in the ICF. In addition, the activities in the TUAQ requires a range of grasp patterns and demonstrates different levels of complexity of upper limb ability, fulfilling the predefined objectives as suitable for individuals with tetraplegia.
      The performance for “drinking from a bottle” was the only nonuniform DIF item by sex for the performance scale only. The rationale for keeping the item in the TUAQ was that DIF for the satisfaction scale was uniform and we felt that this activity was essential. Drinking was one of the most common activity limitations identified by people with tetraplegia (37%) in our item identification (see table 1). We felt that drinking from a bottle was more standardized activity than drinking from a mug in terms of the items used for the activity. Often people with tetraplegia do not grasp the handle of a mug, rather they hook their thumb through the handle resulting in balancing the mug using the thumb and the ulnar side of the hand with the wrist extended. In comparison, an individual is required to grasp a bottle drink with 1 hand.
      One of the predefined aims for the measure was that it needed to be quick and easily administered for individuals with tetraplegia and clinicians. The fact that the TUAQ includes 10 items and fits onto 1-page makes it convenient for individuals with tetraplegia and easy for clinicians or researchers to calculate the score. Another aspect of clinical relevance was the inclusion of the column for use of aids after response from clinicians during stage 2. From a clinical point of view, the use of aids is important information that needs to be obtained in the discussion with the individual. It will, however, not be calculated from a scientific point of view because of difficulties in differentiating between the use of aids (such as a writing splint) vs adaptive strategies such as weaving a pen between fingers or holding a pen using pinch grip. The perceived difference between satisfaction and performance might also be capturing the use of aids to some extent, but that is yet to be proven. The aim of this measure is to capture the performance and satisfaction with daily tasks, independent of technique. If the aim is to investigate grip ability, there are many other tests available that cover this aspect.
      • Kalsi-Ryan S
      • Curt A
      • Verrier MC
      • Fehlings MG
      Development of the Graded Redefined Assessment of Strength, Sensibility and Prehension (GRASSP): reviewing measurement specific to the upper limb in tetraplegia.
      • Wuolle KS
      • Van Doren CL
      • Thrope GB
      • Keith MW
      • Peckham PH.
      Development of a quantitative hand grasp and release test for patients with tetraplegia using a hand neuroprosthesis.
      • Spooren AI
      • Janssen-Potten YJ
      • Post MW
      • Kerckhofs E
      • Nene A
      • Seelen HA.
      Measuring change in arm hand skilled performance in persons with a cervical spinal cord injury: responsiveness of the Van Lieshout Test.
      • Fattal C.
      Motor capacities of upper limbs in tetraplegics: a new scale for the assessment of the results of functional surgery on upper limbs.
      • Marino RJ
      • Goin JE.
      Development of a short-form Quadriplegia Index of Function scale.
      • Sollerman C
      • Ejeskar A.
      Sollerman hand function test. A standardised method and its use in tetraplegic patients.
      • Jebsen RH
      • Taylor N
      • Trieschmann RB
      • Trotter MJ
      • Howard LA.
      An objective and standardized test of hand function.
      • Carroll D.
      A quantitative test of upper extremity function.
      The final version of the TUAQ is shown in appendix 1.

      Study strengths and limitations

      A strength of this study is that we have developed an outcome measure to use in the population with tetraplegia that is based on the self-identified COPM goals of people with tetraplegia. There are a number of limitations. Participants were recruited from units that specialize in reconstructive upper limb surgery for tetraplegia, which might have led to selection bias and may limit applicability to the general population with tetraplegia, and the sample size is relatively small, in the lower bound, for using Rasch analysis. The results must therefore be interpreted with caution because all data are preliminary, and additional testing is required to confirm these findings. While not a limitation, it is acknowledged that the Rasch analysis for this measure was performed on the 10-item scale. The scale was thereafter reduced to 5-level scoring to increase stability.

      Conclusions

      The initial psychometric testing shows that the TUAQ has the potential to fill an identified gap in available outcome measures targeting daily activities relying on upper limb use in persons with tetraplegia. The results support the internal construct validity and unidimensionality of the TUAQ, a 10-item, 5-response PROM that assesses activity performance and satisfaction with activities that specifically require use of the upper limb in persons with tetraplegia. The results indicates that the TUAQ is well targeted regardless of age, sex, time after injury, or limiting spasticity. The short administration time for both the individual and the clinician has the potential for clinical use. Further testing is required and ongoing to evaluate reliability and responsiveness of the TUAQ.

      Suppliers

      • a.
        SPSS for Mac Version 25; SPSS Inc.
      • b.
        RUMM 2030 partial credit model software; Rumm Laboratory Pty Ltd.

      Acknowledgments

      We thank Birgitta Rustner and Charlotte Orrby Johnsson at Linköpings University Hospital Sweden for their valuable contribution in data collection and Professor Jan Fridén for his encouragement to develop this measure.

      Appendix. Supplementary materials

      References

        • Sinnott KA
        • Dunn JA
        • Wangdell J
        • Johanson ME
        • Hall AS
        • Post MW.
        Measurement of outcomes of upper limb reconstructive surgery for tetraplegia.
        Arch Phys Med Rehabil. 2016; 97: S169-S181
        • Fawcett J
        • Curt A
        • Steeves J
        • et al.
        Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials.
        Spinal Cord. 2007; 45: 190-205
        • Jones LAT
        • Bryden A
        • Wheeler TL
        • et al.
        Considerations and recommendations for selection and utilization of upper extremity clinical outcome assessments in human spinal cord injury trials.
        Spinal Cord. 2018; 56: 414-425
        • Velstra IM
        • Ballert CS
        • Cieza A.
        A systematic literature review of outcome measures for upper extremity function using the international classification of functioning, disability, and health as reference.
        PM R. 2011; 3: 846-860
        • van Tuijl JH
        • Janssen-Potten YJ
        • Seelen HA.
        Evaluation of upper extremity motor function tests in tetraplegics.
        Spinal Cord. 2002; 40: 51-64
        • Kalsi-Ryan S
        • Curt A
        • Verrier MC
        • Fehlings MG
        Development of the Graded Redefined Assessment of Strength, Sensibility and Prehension (GRASSP): reviewing measurement specific to the upper limb in tetraplegia.
        J Neurosurg Spine. 2012; 17: 65-76
        • Wuolle KS
        • Van Doren CL
        • Thrope GB
        • Keith MW
        • Peckham PH.
        Development of a quantitative hand grasp and release test for patients with tetraplegia using a hand neuroprosthesis.
        J Hand Surg Am. 1994; 19: 209-218
        • Spooren AI
        • Janssen-Potten YJ
        • Post MW
        • Kerckhofs E
        • Nene A
        • Seelen HA.
        Measuring change in arm hand skilled performance in persons with a cervical spinal cord injury: responsiveness of the Van Lieshout Test.
        Spinal Cord. 2006; 44: 772-779
        • Fattal C.
        Motor capacities of upper limbs in tetraplegics: a new scale for the assessment of the results of functional surgery on upper limbs.
        Spinal Cord. 2004; 42: 80-90
        • Marino RJ
        • Goin JE.
        Development of a short-form Quadriplegia Index of Function scale.
        Spinal Cord. 1999; 37: 289-296
        • Sollerman C
        • Ejeskar A.
        Sollerman hand function test. A standardised method and its use in tetraplegic patients.
        Scand J Plast Reconstr Surg Hand Surg. 1995; 29: 167-176
        • Jebsen RH
        • Taylor N
        • Trieschmann RB
        • Trotter MJ
        • Howard LA.
        An objective and standardized test of hand function.
        Arch Phys Med Rehabil. 1969; 50: 311-319
        • Carroll D.
        A quantitative test of upper extremity function.
        J Chronic Dis. 1965; 18: 479-491
        • Hudak PL
        • Amadio PC
        • Bombardier C.
        Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand) [corrected]. The Upper Extremity Collaborative Group (UECG).
        Am J Ind Med. 1996; 29: 602-608
        • Penta M
        • Thonnard JL
        • Tesio L.
        ABILHAND: a Rasch-built measure of manual ability.
        Arch Phys Med Rehabil. 1998; 79: 1038-1042
        • Keith RA
        • Granger CV
        • Hamilton BB
        • Sherwin FS.
        The functional independence measure: a new tool for rehabilitation.
        Adv Clin Rehabil. 1987; 1: 6-18
        • Catz A
        • Itzkovich M
        • Agranov E
        • Ring H
        • Tamir A.
        SCIM-spinal cord independence measure: a new disability scale for patients with spinal cord lesions.
        Spinal Cord. 1997; 35: 850-856
        • Koch-Borner S
        • Dunn JA
        • Friden J
        • Wangdell J.
        Rehabilitation after posterior deltoid to triceps transfer in tetraplegia.
        Arch Phys Med Rehabil. 2016; 97: S126-S135
        • Wangdell J
        • Fridén J.
        Satisfaction and performance in patient selected goals after grip reconstruction in tetraplegia.
        J Hand Surg Eur Vol. 2010; 35: 563-568
        • Wangdell J
        • Fridén J.
        Activity gains after reconstructions of elbow extension in patients with tetraplegia.
        J Hand Surg Am. 2012; 37: 1003-1010
        • Wangdell J
        • Bunketorp-Käll L
        • Koch-Borner S
        • Fridén J.
        Early active rehabilitation after grip reconstructive surgery in tetraplegia.
        Arch Phys Med Rehabil. 2016; 97: S117-S125
        • Wangdell J
        • Reinholdt C
        • Fridén J.
        Activity gains after upper limb surgery for spasticity in patients with spinal cord injury.
        J Hand Surg Eur Vol. 2018; 43: 613-620
        • Reinholdt C
        • Friden J.
        Outcomes of single-stage grip-release reconstruction in tetraplegia.
        J Hand Surg Am. 2013; 38: 1137-1144
        • Cieza A
        • Geyh S
        • Chatterji S
        • Kostanjsek N
        • Ustun B
        • Stucki G.
        ICF linking rules: an update based on lessons learned.
        J Rehabil Med. 2005; 37: 212-218
        • Law M
        • Baptiste S
        • McColl M
        • Opzoomer A
        • Polatajko H
        • Pollock N.
        The Canadian Occupational Performance Measure: an outcome measure for occupational therapy.
        Can J Occup Ther. 1990; 57: 82-87
        • World Health Organization
        International Classification of functioning, disability and health (ICF).
        World Health Organization, Geneva2001: 31 (Short version)
        • Beaton DE
        • Bombardier C
        • Guillemin F
        • Ferraz MB.
        Guidelines for the process of cross-cultural adaptation of self-report measures.
        Spine (Phila Pa 1976). 2000; 25: 3186-3191
        • Mokkink LB
        • Terwee CB
        • Patrick DL
        • et al.
        The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes.
        J Clin Epidemiol. 2010; 63: 737-745
        • Kirshblum SC
        • Burns SP
        • Biering-Sorensen F
        • et al.
        International standards for neurological classification of spinal cord injury (revised 2011).
        J Spinal Cord Med. 2011; 34: 535-546
        • Tabachnick BG
        • Fidell LS.
        Using multivariate statistics.
        Pearson/Allyn & Bacon, Boston2007
        • Tennant A
        • Conaghan PG.
        The Rasch measurement model in rheumatology: what is it and why use it? When should it be applied, and what should one look for in a Rasch paper?.
        Arthritis Rheum. 2007; 57: 1358-1362
        • Pallant JF
        • Tennant A.
        An introduction to the Rasch measurement model: an example using the Hospital Anxiety and Depression Scale (HADS).
        Br J Clin Psychol. 2007; 46: 1-18

      Linked Article