Participants
The study population consisted of patients with neck pain with or without arm pain. They were recruited from outpatients and inpatients of the orthopedic and neurosurgery departments and the pain clinic of Kaohsiung Medical University Hospital in Taiwan. Patients were identified by physicians on the basis of symptoms, physical signs, and imaging study results. The diagnoses included degenerative joint disease, herniated intervertebral disc, strain and sprain, and non-specific neck pain. Participants were excluded if the neck pain had been attributed to shoulder disease, inflammatory rheumatic disease, or pain caused by cancer. The study was approved by the hospital’s institutional review board (KMUH-IRB-980589 and KMUH-IRB-E(I)-20210363), and written informed consent was obtained from all the participants.
Procedure
Each participant was asked to complete a questionnaire booklet, which contained the Taiwan version of the QuickDASH, the Taiwanese version of the NDI, the Medical Outcomes Survey Short Form 36 (SF-36), and two types of Visual Analogue Scale (VAS), namely, the VAS-N (assessing neck and shoulder pain) and the VAS-A (assessing arm and hand pain) for determination of the convergent validity. The total sample size of the study was 189. To measure the test–retest reliability, 16 patients from this sample with chronic neck pain of at least 3 months and without recent changes in pain or function were invited to participate in the study. The retest was conducted at intervals of 7–14 days.
Instruments
QuickDASH
The QuickDASH, which contains 11 items, was developed from the DASH through three item-reduction approaches by Beaton et al. and the Upper Extremity Collaborative Group [13]. The 11 items ask about the following: opening a jar (Q1); doing household chores (Q2); carrying a shopping bag (Q3); washing one’s back (Q4); using a knife (Q5); recreational activities with force or impacts (Q6); interference with social activities (Q7); limitations in work or daily activities (Q8); shoulder, arm, or hand pain (Q9); tingling (Q10); and difficulty sleeping (Q11). The two optional modules (work and sports/performing arts) were not used in this study. Each item is scored on a 5-point scale, with 1 representing no limitation and 5 representing maximal limitation. At least 10 of the 11 items must be completed for a score to be calculated. The scores are transformed to a scale of 0 (no disability) to 100 (severe disability).
Neck Disability Index
The NDI contains 10 items developed by Vernon and Mior. The questions were designed to understand how neck pain affects the ability to manage activities in everyday life [8]. The 10 items ask about pain intensity, the level of disability in personal care, lifting, work, headache, concentration, sleeping, driving, reading, and recreation. Each item is scored on a 6-point scale, with 0 representing no limitation and 5 representing maximal limitation. The scores of the 10 items are summed to obtain the NDI score (0–50 points). The Taiwanese version of the NDI is a well-developed and validated questionnaire [15].
Medical Outcomes Survey Short Form 36
The SF-36 measures eight subscales of health-related quality of life: physical functioning (PF), role limitations due to physical health problems (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role limitation due to emotional problems (RE), and mental health (MH) [16]. The subscale scores are transformed to a scale of 0 to 100, and higher scores indicate better quality of life. Two summary measures, the physical component summary (PCS) and the mental component summary (MCS), are calculated from the eight health subscales, and higher scores (0–100) indicate better physical and mental function, respectively. The Taiwanese version of the SF-36 has demonstrated good reliability and validity [17].
Data management and statistical analyses
The α level was set at 0.05. All statistical analyses were performed in IBM SPSS Statistics version 21; CFA was performed in IBM SPSS AMOS version 21 (IBM Corp., Armonk, NY).
Acceptability
To assess the acceptability of the QuickDASH, the missing data, observed score versus possible score range, and floor and ceiling effects were analyzed. Missing data of less than 5% is considered to be acceptable. Floor and ceiling effects greater than 15% are considered to be significant. The acceptable skewness range is from − 1 to 1 [18, 19].
Reliability
Internal consistency
The internal consistency of the 11 items was assessed by Cronbach’s α. The corrected item–total correlations and the Cronbach’s α with each item deleted were assessed. A Cronbach’s α ≥ 0.7 is acceptable [18, 19]. For item–total correlation, the correlations should be moderate to high. For Cronbach’s α with an item deleted, an increase of more than 0.1 indicates that the item is not correlated very well with the scale [20].
Test–retest reliability
Both relative and absolute reliability were used to assess the test–retest reliability. Intraclass correlation coefficients (ICCs) were used to estimate the relative reliability. An ICC value ≥ 0.75 indicates excellent reliability [21]. The standard error of measurement (SEM) and minimal detectable change (MDC) were used to estimate the absolute reliability [22]. Measurement error was evaluated by SEM, which was calculated by using the formula (standard deviation of all test scores) X√ (1-ICC).
The MDC is used as the threshold to indicate whether the change score of an individual subject is real and not within the measurement error for an individual at the 0.95 confidence level. The formula for MDC is 1.96 X SEM X√2. A good outcome measure should have low SEM and MDC values that are sensitive to change resulting from clinical intervention.
Construct validity
Convergent and divergent validity
The convergent validity of the QuickDASH was assessed with the correlations between the QuickDASH and the NDI, the VASs (VAS-N and VAS-A), and the SF-36 (PCS/MCS and 8 subscales). Spearman rank correlation was used to examine the correlations. We hypothesized that the QuickDASH and the NDI would have strong correlation, and that the correlation of the QuickDASH would be higher with the VAS-A than with the VAS-N. We also hypothesized that the QuickDASH would have a higher correlation with the PCS than with the MCS, and that its correlation would be higher with the PF, RP, BP, and GH than with the VT, SF, RE and MH subscales. The divergent validity of the QuickDASH was assessed with the correlations between the QuickDASH and age and disease duration. We hypothesized that the QuickDASH would have low or no correlation with age and disease duration.
Exploratory factor analysis (EFA)
EFA was used to investigate the dimensionality of the scale to identify the underlying factor structure of the items. Half the patients (n = 94) were randomly selected from the whole sample of patients for EFA. The Kaiser–Meyer–Olkin measure of sampling adequacy and Bartlett’s test of sphericity were used to assess the data fit for factor analysis. High values of the Kaiser–Meyer–Olkin (greater than 0.5) and small values of Bartlett’s test of sphericity (significance level less than 0.05) generally indicate that a factor analysis is useful with the data [23]. We hypothesized that the items could be extracted into one, two or more factors. All eleven items of the QuickDASH were used for the EFA. The principal axis factoring method was used for factor analysis extraction, and the direct oblimin method of oblique rotation was used for factor analysis rotation.
Confirmatory factor analysis (CFA)
CFA is used to test the fit of an a priori hypothesized structure of a scale [24]. The patients (n = 95) included in the CFA were those not used for the EFA. We tested the hypothesized model after the results of the EFA for the QuickDASH were obtained. In the parameter estimates, a factor loading of each section of higher than 0.5 was used to confirm satisfactory fit. Several fit indices were used to assess the goodness of fit of the models. [24–26] The indices included the χ2 divided degree of freedom (χ2/df), Root Mean Square Error of Approximation (RMSEA), Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), and Normed Fit Index (NFI).