A Network Analysis of the PART-O at 1 and 2 Years After TBI: A Veterans Affairs Model Systems Study

Abstract

Objective: The construct of participation after traumatic brain injury (TBI) can be difficult to operationalize. Psychometric network analysis offers an empirical approach to visualizing and quantifying the associations between activities that comprise participation, elucidating the relations among the construct's components without assuming the presence of a latent common cause and generating a model to inform future measurement methods. The current research applied psychometric network analysis to the Participation Assessment with Recombined Tools-Objective (PART-O) within a sample of service members and veterans (SM/Vs) with a history of TBI at 1 and 2 years (T 1 and T 2) postinjury. Participants: Participants (N = 663) were SM/Vs with a history of TBI who completed comprehensive inpatient rehabilitation services at a Department of Veterans Affairs (VA) Polytrauma Rehabilitation Center (PRC). Setting: Five VA PRCs. Design: Cross-sectional, retrospective analysis of data from the VA TBI Model Systems study. Main Measures: PART-O. Results: Network analysis demonstrated that the PART-O structure was generally consistent over time, but some differences emerged. The greatest difference observed was the association between “spending time with friends” and “giving emotional support” to others. This association was more than twice as strong at T 2 as at T 1. The “out of the house” item was most central, as demonstrated by dense connections within its own subscale (Out and About) and items in other subscales (ie, Social Relations and Productivity). When examining items connecting the 3 subscales, the items related to giving emotional support, internet use, and getting out of the house emerged as the strongest connectors at T 1, and the internet was the strongest connector at T 2. Conclusion: Providing emotional support to others is associated with greater participation across multiple domains and is an important indicator of recovery. Being out and about, internet use, and engagement in productive activities such as school and work shared strong associations with Social Relations. Network analysis permits visual conceptualization of the dynamic constructs that comprise participation and has the potential to inform approaches to measurement and treatment.