Assessing Bilateral Upper Extremity Integration using the Bilateral Task Assessment

Research Objectives: To investigate the validity and reliability of the Bilateral Task Assessment (BTA) in the inpatient rehabilitation (IP rehab) setting.

Design: Study on psychometric properties.

Setting: An urban, inpatient rehabilitation facility.

Participants: Inpatient rehabilitation consumers assessment results.

Interventions: The BTA was administered by occupational therapists to 3,962 individuals in IP rehab. Psychometric analysis consisted of Cronbach’s alpha and item-to-total correlations to determine internal consistency, a correlation analysis to assess concurrent validity, factor analysis to determine dimensionality, and item response theory to compute scores and assess test difficulty.

Main Outcome Measures: Validity, inter-rater/test-retest reliability, and internal consistency of the Bilateral Task Assessment.

Results: The scoring standard (.93) and times score (.86) indicates that both variables have good internal consistency. Factor analysis indicates a large effect size for BTA task scores and moderate effect size for BTA times. The additional analyses of the rating scale data confirmed the high internal consistency (.93) and high ITT correlations ranging from .86 - .90. The concurrent validity analysis found significant correlations with seven other measures of upper extremity function. Confirmatory factor analysis indicated that the BTA is a unidimensional measure (1-factor RMSEA = .06, CFI = .99). IRT analysis determined that the measure is most accurate for logit scores around -.40.

Conclusions: This study determined that the BTA demonstrated strong psychometric qualities and that it can be more widely used within occupational therapy practice to assess bilateral integration in stroke survivors. The BTA is a valid and consistent measure that can support BUE integration in occupational performance through assessment in a more functional context.

Author(s) Disclosures: The authors have nothing to disclose.