Rheumatoid Arthritis-Work Instability Scale

Rheumatoid Arthritis (RA): (Gilworth et al., 2003; n = 31 subjects who returned the original questionnaire, were employed, and who agreed to participate in a full vocational assessment; assessors were 2 professionals who were both chartered physiotherapists and registered ergonomists)

• A score of 10 or more on the 23-item WI scale was shown to have 82% sensitivity to the need for workplace modifications, while a score of 17 or more gave 95% specificity. The RA-WIS can therefore be scored in 3 bands:
  • <10 indicates low Work Instability (WI) and low risk of work disability (WD)
  • 10 to 17 indicates moderate WI and medium risk of WD
  • >17 indicates high WI and high risk of WD

Rheumatoid Arthritis (RA): (Gilworth, et al., 2009; n = 306 confirmed RA patients either in work or temporarily on sick leave (less than 6 months): French (n = 75), mean age = 46 years, female = 79%;  Dutch (n = 85), mean age = 45 years, female = 73%; German (n = 73), mean age = 43 years, female = 72%; plus a randomly selected matching set (n = 73) from the original developmental study (Gilworth, et al., 2003); age = between 18 and 60 years)

• Two cut points were identified for the RA-WIS: one at 10 indicating the transition from low to medium risk and the other at 17 indicating the transition from medium to high risk

Rheumatoid Arthritis (RA) or Osteoarthritis (OA): (Tang et al., 2010; n = 250 (120 with RA and 130 with OA); female:male ratio of 5:1; mean age of subgroups: OA = 54.0 (6.7), RA = 46.6 (10.2), p <0.0001; participants required to be 1) attending an outpatient rheumatology clinic with a diagnosis of OA or RA, or attending an arthritis treatment program within the past two years with a physician diagnosis of OA or RA, and 2) gainfully employed at the time or recruitment; patients followed at 3, 6, and 12 months after initial baseline; patient sampling from two tertiary-level rheumatology clinics in urban teaching hospitals (n = 142) and an outpatient arthritis treatment program providing multidisciplinary services (n = 108))

• >13 is most accurate RA-WIS cut point for any form of work transition (sensitivity 51%; specificity 83%)

Rheumatoid Arthritis (RA): (Schmidt, et al., 2020; n = 140 employed patients with RA, female = 111 (79.3%), mean (SD) age = 48.7 (11.0), disease duration > 5 years from diagnosis = 114 (81.4%); physician completed Part 1 of anonymized form based on patient medical history/lab results and patient completed Part 2 of form that included an anonymous questionnaire; Polish translation of RA-WIS)

• <10 indicates low Work Instability (WI)
• 10 to 17 indicates moderate WI
• >17 indicates high WI

Rheumatoid Arthritis (RA): (Macedo, et al., 2009; n = 90 employed patients attending routine specialized inflammatory clinics, 70 (78%) female, mean age = 48.3 years (range = 20-69), mean disease duration (n = 86) = 9.7 (8.8) years (range = 0-38), occupational classification (n = 77) = blue collar, 10 (13%); white collar, 67 (87%))

• Mean score on the Rheumatoid Arthritis-Work Instability Scale (RA-WIS, n = 90): 8.97 (6.58, range 0-22)
• Mean score on the Disease Activity Score 28-joint count (DAS28, n  = 80): 3.67 (1.54, range 0.8-7.00)
• Mean score on the Health Assessment Questionnaire (HAQ, n = 79): 0.90 (0.74, range 0-2.63)

Rheumatoid Arthritis (RA) or Osteoarthritis (OA): (Beaton et al., 2010; n = 250 (120 with RA and 130 with OA), female = 82.7%, mean age = 50.6 (9.2), 54.7% diagnosed with arthritis for >5 years 9.8% diagnosed for <1 year, patients followed at 3, 6, and 12 months after initial baseline, patient sampling from two tertiary-level rheumatology clinics in urban teaching hospitals (n = 142) and an outpatient arthritis treatment program providing multidisciplinary services (n = 108)

• Mean score on the RA-WIS (n = 223): 8.3 (6.4)

 Early Rheumatoid Arthritis (RA): (Revicki et al., 2015; n = 148, mean age = 46.8 years, female = 83 (56.1%), data from 56-week, randomized control trial of patients with early RA with data collected at baseline and at weeks 12, 16, 24, and 56)

• Mean score on the RA-WIS (n = 148): 15.3 (5.5)

Rheumatoid Arthritis (RA): (Schmidt, et al., 2020)

• Mean score on the RA-WIS (n = 140) = 13.45 (6.14)

Rheumatoid Arthritis (RA): (Gilworth et al., 2003; n = 51 (41% of the 123 subjects who returned both questionnaires sent at 2-week intervals to 229 potential subjects)

• Acceptable test-retest reliability: (Spearman’s rho = 0.89) 

Early Rheumatoid Arthritis (RA): (Revicki et al., 2015; n = 148)

• Excellent test-retest reliability: (ICC = 0.91)

Rheumatoid Arthritis (RA): (Gilworth, et al., 2009; n = 306)

• Excellent overall Person Separation Index (PSI) on RA-WIS for all countries = 0.914
  • Excellent PSI values on RA-WIS by country: UK (0.928), Dutch (0.930), German (0.917), and French (0.898)
  • The consistently high PSI values throughout the analyses of the data for the RA-WIS resulted in the determination that the measure was suitable for individual use.

Rheumatoid Arthritis (RA) or Osteoarthritis (OA): (Beaton et al., 2010) 

• Excellent: Cronbach’s alpha (n = 223) = 0.92*

Early Rheumatoid Arthritis (RA): (Revicki et al., 2015; n = 148)

• Excellent: Cronbach’s alpha = 0.89

*Scores higher than 0.9 may indicate redundancy in the scale questions. 

Predictive validity

Rheumatoid Arthritis (RA) or Osteoarthritis (OA): (Tang et al., 2010; n = 250 (120 with RA and 130 with OA); female:male ratio of 5:1; mean age of subgroups: OA = 54.0 (6.7), RA = 46.6 (10.2), p <0.0001; participants required to be 1) attending an outpatient rheumatology clinic with a diagnosis of OA or RA, or attending an arthritis treatment program within the past two years with a physician diagnosis of OA or RA, and 2) gainfully employed at the time or recruitment; patients followed at 3, 6, and 12 months after initial baseline; patient sampling from two tertiary-level rheumatology clinics in urban teaching hospitals (n = 142) and an outpatient arthritis treatment program providing multidisciplinary services (n = 108))

• Higher baseline RA-WIS (RR = 1.05 [95% CI 1.00-1.11] per unit increase in score) was predictive of impending arthritis-related work transitions (i.e. permanent reductions in work hours, disability {sick} leaves of absence, changes in occupation or job, or temporary unemployment {became unemployed but looking for work}) after controlling for pain intensity and arthritis diagnosis as the only covariates retained.
  • A significant interaction effect (p = 0.04) was found  between RA-WIS and pain intensity, with the RA-WIS having significant associations with the outcome at higher levels of pain, but not at lower levels of pain.
• The RA-WIS was not found to be a predictor or covariate for any of the secondary outcomes relating to decreases in perceived work capacity

Rheumatoid Arthritis (RA): (Macedo, et al., 2009)

• A Disease Activity Score 28-joint count (DAS28) of greater than 3.81 combined with a Health Assessment Questionnaire (HAQ) score greater than 0.55 characterized the moderate-high RA-WIS (>= 10) group with high sensitivity (0.90) and specificity (0.76) and had the strongest overall association with moderate-high RA-WIS (OR = 29.87)

Rheumatoid Arthritis (RA): (Schmidt, et al., 2020)

• Significant differences among employed patients with low-to-moderate and high RA-WIS scores on various clinical characteristics

Clinical characteristics of employed patients according to work instability risk, mean (SD)

Characteristic

Low-to-moderate RA-WIS (0-17)

High RA-WIS (>17)

ESR (mm/h)

18.3 (16)

28.2 (21.9)*

CRP (mg/dl)

0.6 (1.0)

1.1 (1.0)*

Tender joint count

4.5 (4.0)

7.6 (6.3)*

Swollen joint count

2.7 (3.0)

5.2 (5.5)*

HAQ-DI

0.7 (0.5)

1.1 (0.5)*

Pain – VAS 10 cm

3.9 (1.9)

6.5 (2.2)*

DAS28

3.5 (1.2)

4.2 (1.4)*

DAS28-CRP

3.9 (1.0)

4.3 (1.2)*

ESR – erythrocyte sedimentation rate, CRP – C-reactive protein, HAQ-DI – Health Assessment Questionnaire Disability Index, VAS – Visual Analogue Scale, DAS28 – Disease Activity Score in 28 joints.

*p < 0.05, Mann-Whitney U test

Convergent validity

Early Rheumatoid Arthritis (RA): (Revicki et al., 2015; n = 148)

• Adequate to Excellent convergent validity between RA-WIS and Duration of morning stiffness, the Health Assessment Questionnaire-Disability Index (HAQ-DI), the 28-Joint Disease Activity Scale (DAS28), the Patient Global Assessment of Disease Activity, the Patient Assessment of Pain, the RA Quality of Life Questionnaire (RAQoL), the Swollen joint count (SJC28), the Tender joint count (TJC28), the Number of hours lost per week because of RA, the Number of hours lost per week, the Number of hours worked per week, and the effect of RA on work performance

Measure

RA-WIS at Week 0

RA-WIS at Week 24

Duration of morning stiffness

0.31a

0.36b

HAQ-DI

0.55c

0.76c

DAS28

0.37c

0.55c

Patient Global Assessment of Disease Activity

0.47c

0.66c

Patient Assessment of Pain

0.55c

0.62c

Physician Global Assessment of Disease Activity

0.31a

0.52c

RAQoL

0.77c

0.81c

SJC28

0.07

0.46c

TJC28

0.35c

0.55c

No. of hours lost per week because of RA 

0.37c

0.27b

No. of hours lost per week

0.30a

0.09

No. of hours worked per week

-0.09

-0.12

Effect of RA on work performance

0.37c

0.64c

ap < 0.001

bp < 0.05

cp < 0.0001

Rheumatoid Arthritis (RA): (Gilworth et al., 2003; n = 31 subjects who returned the original questionnaire, were employed, and who agreed to participate in a full vocational assessment; assessors were 2 professionals who were both chartered physiotherapists and registered ergonomists) 

• “Rasch model, which identified 23 items on a single construct of WI that were free from item bias for age and sex. Fit to the model was confirmed by excellent item fit (mean = 0.056, SD = 0.092) and person fit (mean = −0.062, SD = 0.595) statistics. Item trait interaction chi‐square of 34.1, (degrees of freedom (df) = 46; P = 0.90) showed the classic property of invariance for the scale.”

Rheumatoid Arthritis (RA): (Gilworth, et al., 2009; n = 306)

• The data from the four countries (UK, Dutch, German, French) pooled together showed good fit to the Rasch model (item fit mean = -0.168, SD = 1.277, person fit mean = −0.197, SD = 0.739, interaction chi-square = 133.5 (df = 92), p < 0.003) and demonstrated strict unidimensionality.
• Individual country data also showed good fit to the Rasch model, with the exception of France, where one item—“Au travail, j’ai des jours avec et des jours sans” (I get good days and bad days at work)—showed significant misfit to the expectation of the model. Fit to the model expectation was good following the removal of this item.
• Differential item functioning (DIF) by country was found to be present for six of the items. However, the DIF appeared to both cancel out at the test level in the pooled data, and impact the test score itself only marginally. 

Rheumatoid Arthritis (RA): (Macedo, et al., 2009)

• Adequate convergent validity with DAS28 (r = 0.53)
• Excellent convergent validity with HAQ (r = 0.66)
• DAS, HAQ, and sex explained 54% of the variability in RA-WIS scores (R2 = 0.57; adjusted R2 = 0.54)

Rheumatoid Arthritis (RA) or Osteoarthritis (OA): (Beaton et al., 2010) 

Correlations of RA-WIS with various theoretical constructs (n=223)

Theoretical constructs (a priori hypothesized relationship)

RA-WIS

Work-oriented constructs

Self-rated work productivity (r < -0.5)

-0.54a

Perceived impact of health problems on work (r > 0.5)

0.73a

Self-rated difficulty doing work (r > 0.75)

0.71

Satisfaction with occupational performance (r < -0.75)

-0.68

Self-rated ability to work (r < -0.75)

-0.67

Intrusion of arthritis on work ability (r > 0.75)

0.74

Self-rated job performance in past week (r < -0.5)

-0.56a

Arthritis hindrance on work performanceb

105.1c

Disease-oriented constructs

General perceived disability (r > 0.5)

0.66a

Arthritis severity (r > 0.5)

0.62a

Pain intensity over past week (r > 0.5)

0.67a

aA priori hypothesis met.

bF values from analysis of variance between known groups {[0, no] vs. [1, to a degree] vs. [2, very much] were reported.

cTukey P < 0.05 for all t-test comparisons (i.e. 0 vs. 1, 1 vs. 2, 0 vs. 2) and evidence of logical gradient.

• RA-WIS correlated to the largest number of constructs (7 out of 11) above the hypothesized level compared to 8 other scales
• RA-WIS also exhibited the strongest known-group validity (F = 105.1 vs. F = 9.5 – 68.9 for the other scales) in differentiating workers experiencing varying levels of arthritis-related hindrance on work performance
• Adequate to Excellent convergent validity of the RA-WIS with other measures of at-work productivity in workers with arthritis (|r| = 0.41-0.77)

Rheumatoid Arthritis (RA): (Schmidt, et al., 2020)

• Adequate convergent validity of the RA-WIS with the Visual Analogue Scale (r = 0.59, p < 0.001), the Health Assessment Questionnaire Disability Index (r = 0.52, p < 0.001), and the Disease Activity Score in 28 joints (r = 0.31, p < 0.001)
• Poor convergent validity of the RA-WIS with the DAS28-C-Reactive Protein (r = 0.28, p < 0.001)

Rheumatoid Arthritis (RA): (Gilworth, et al., 2009; n = 306)

• The RA-WIS questionnaires in the UK, Dutch, German, & French versions were field tested for face validity
  • The questionnaires were largely judged to be understandable and patients stated that the RA-WIS covered notions that were important for them
  • The clarity and wording of the instructions were also deemed to be easily understood and there were no modifications of instruction or question phrasing at this stage

Rheumatoid Arthritis (RA) or Osteoarthritis (OA): (Beaton et al., 2010) 

• The RA-WIS ranked second overall (to the Workplace Activity Limitations Scale or WALS) and no worse than second in 5 of 7 criteria related to the responsiveness to change in work ability
• The RA-WIS ranked last overall in terms of responsiveness to change in work productivity, especially for deteriorations where it ranked last for effect size, standardized response mean, and Spearman’s r 

Early Rheumatoid Arthritis (RA): (Revicki et al., 2015; n = 148)

• Significantly different RA-WIS total change scores among American College of Rheumatology (ACR) classification groups from mean baseline to week 24 and from baseline to week 56 (p < 0.0001)
  • Greatest differences at week 24 observed between responder groups with improvements of less than 20% and those with 20% to less than 50% (5.01 points)
  • Greatest differences at week 56 observed between responder groups with improvements of 20% to less than 50% and those with 50% to less than 70% (4.35 points)
• Significantly different RA-WIS total change scores based upon DAS28 remission status, with larger changes from baseline to week 24 observed for patients in remission (DAS28 < 2.6) than in those not in remission (DAS28 => 2.6, p < 0.001), with similar findings in RA-WIS total change scores from baseline to week 56 (p < 0.0001)
  • Differences between remission and non-remission groups were 5.14 points at week 24 and 3.48 points at week 56 

Fibromyalgia: (Hammond et al., 2023; Phase 2: n  = 156 persons with primary diagnosis of fibromyalgia; female = 146 (94%); mean age (years (SD)) = 45.71 (10.05); SEM = Standard Deviation from the first test x (square root of (1-ICC))

• SEM = 1.20

Fibromyalgia: (Hammond et al., 2023; Phase 2: n  = 156 persons with primary diagnosis of fibromyalgia; MDC = SEM x 1.96 x square root of 2)

• MDC = 3.38