Skip to main content
. 2010 Apr 30;19(5):657–666. doi: 10.1007/s00520-010-0878-x

Table 2.

Feasibility, score distributions, and reliability of the Perform Questionnaire (N = 437)

Physical limitations Activities of daily living Beliefs and attitudes Global scorea
Items (N) 3 3 3 12
Mean 11.5 11.7 11.3 34.8
SD 4.7 3.9 4.5 12
Time taken for administration, mean (SD) 8.8 (8.9)
Easy or very easy to answer the questionnaire, N (%) 343 (81.1)
Completion rateb, N (%) 419 (95.88) 417 (95.42) 358 (81.92) 351 (80.32)
Range of missing answers, N (%) 2.5% (item 1)–3.7% (item 2) 2.3% (item 6)–16.9% (item 8) 2.3% (item 12)–3.7% (item 9) 2.3% (items 6,12)–16.9% (item 8)
Theoretical rangea 4–20 4–20 4–20 12–60
Observed rangea 4–20 4–20 4–20 12–60
Floorc (%) 8.6 2.1 6.9 1.7
Ceilingd (%) 4.3 0.9 1.6 0.3
Internal consistency, CA 0.90 0.80 0.85 0.94
Test–retest reliability among stable patients (N = 64)e
Mean score (SD) at visit 1 11.9 (4.6) 11.7 (3.8) 11.7 (4.4) 35.5 (12.1)
Mean score (SD) at visit 2 11.5 (4.3) 11.5 (3.5) 11.0 (44.4) 34.3 (11.5)
ICC 0.77 0.76 0.84 0.83

CA Cronbach’s alpha, ICC intraclass correlation coefficient, SD standard deviation

aLow scores indicate worse patient perception of CRF

bPercentage of respondents with no missing data in any of the 12 items

cPercentage of patients with the worst possible score

dPercentage of patients with the best possible score

eStability defined as a change <5 mm in the VAS fatigue between study visits