Standardized translated instruction versus spontaneously translated instruction: Test-retest and interrater reliability of a hand function test

STUDY DESIGN: Cross-sectional study.

INTRODUCTION: Adhering to test administration and standardized instructions is important for attainment of accurate and reliable results in performance-based tests.

PURPOSE OF THE STUDY: To determine test-retest and interrater reliability of standardized translated instruction (St-TI) and spontaneously translated instruction (Sp-TI) of a hand function test.

METHODS: Four raters and seventy-two subjects were divided into 2 groups: St-TI group, direct administration of the Hong Kong Chinese version of the Jebsen Hand Function Test to subjects by raters; and Sp-TI group, spontaneously translating the Jebsen-Taylor Hand Function Test from English into Chinese by raters. Test-retest and interrater reliability were calculated based on instruction time by the rater and performance time by the subject.

RESULTS: Test-retest and interrater reliability of instruction time by rater for St-TI has intraclass correlation coefficient of 0.35 to 0.70 and 0.24 to 0.55, respectively, whereas that for Sp-TI was -0.50 to 0.18 and -0.09 to 0.51, respectively. Test-retest and interrater reliability of performance time by subject for St-TI was 0.56 to 0.84 and 0.33 to 0.78, respectively, whereas that for Sp-TI was 0.54 to 0.87 and 0.35 to 0.77, respectively. Sp-TI had two test-retest minimal detectable change percent values that fell within the acceptable range (subtest 3 = 21.9% and subtest 6 = 25.7%).

CONCLUSION: Instruction time by rater for Jebsen-Taylor Hand Function Test subtests had generally poor to moderate test-retest and interrater reliability for both St-TI and Sp-TI. Performance time by subject generally had moderate to good reliability, except for St-TI with poor to good interrater reliability.