Measuring participation after traumatic brain injury: Construct validity and clinical utility of the sydney psychosocial reintegration scale

Abstract

Objective: Participation is an important domain for measuring the level of adaptation attained by people with traumatic brain injury (TBI). Yet, the construct of participation is difficult to operationalise and scales of participation are generally limited because they use Likert-type rating scales with ordinal-level measurement. This study aims to improve the Sydney Psychosocial Reintegration Scale version 2 (SPRS-2), a reliable and valid 12-item instrument with two forms (Form A: {\dq}change since injury{\dq}; Form B: {\dq}current status{\dq}) that measures occupational, interpersonal and living skills functioning. Construct validity of the SPRS-2 was examined with Rasch analysis and a reliable change index score was generated. Participants and Methods: Participants were convenience samples (n=351, pooled from seven studies) of people with severe TBI, at least 12 months post-trauma and living in the community. Rasch analyses were conducted on Form A (n=201) and Form B (n=150). Results: Both Form A and Form B were a good fit to the Rasch model, for both person (3.36 and 3.03 respectively) and item (7.78 and 7.25 respectively) separation and reliability coefficients were high (all {\textgreater}0.90). Average infit statistics met standard criteria. The reliable change index value was calculated using logit scores. Conclusions: The SPRS-2 demonstrates strong psychometric qualities as a measure of participation after TBI. In particular, logit scores derived from Rasch analyses are valuable in that they provide interval-level data, and the availability of a reliable change index score means that the clinician can readily ascertain whether significant change has occurred in individual patients.