Targeted Box and Blocks Test: Evidence of Convergent Validity in Upper Limb Prosthesis User Population

Research Objectives: To conduct an initial psychometric evaluation in the upper limb prosthesis user population of a novel performance-based measure called the targeted Box and Blocks Test (tBBT) requiring controlled grasping, transport, and release of objects that can be used to evaluate upper limb functional ability.

Design: Nonrandomized controlled trial

Setting: Academic research lab

Participants: A convenience sample of n = 8 transradial myoelectric prosthesis users (3F/5M; mean age 46.3 ± 14.3 years). Nearly all prosthesis users used their myoelectric prosthetic device >8 hours/day and >5 days/week. Time since amputation for all but two prosthesis users was less than 6 years.

Interventions: Not applicable

Main Outcome Measures: Participants were timed as they performed two trials each of the tBBT (in both a standing and seated position); the fork, reach overhead, and door knob tasks from the Activities Measure for Upper Limb Amputees (AM-ULA); and the picture frame, bottle squeeze, and pick up dice tasks from the Capacity Assessment of Prosthesis Performance for the Upper Limb (CAPPFUL). Average time of trials was used to calculate the Pearson correlation coefficients between tBBT and each task performed from the CAPPFUL and AM-ULA.

Results: Nearly all correlations between the tBBT (standing and seated) and select tasks from the AM-ULA and CAPPFUL were moderate to strong, ranging from 0.48 to 0.90. The correlation between tBBT in a seated position and the reach overhead AM-ULA task was weaker (R = 0.38). Interestingly, there was a significant strong negative correlation between tBBT score and the CAPPFUL task of squeezing water from a bottle (R = -0.89 for tBBT standing and R = -.76 for tBBT seated).

Conclusions: Preliminary results demonstrate the tBBT in both seated and standing positions has acceptable convergent validity in the upper limb prosthesis user population. Additional psychometric properties (e.g., known-group validity, test-retest reliability) should be determined to fully characterize this novel measure.

Author(s) Disclosures: None