Application of Second-Order Growth Mixture Modeling to Longitudinal Traumatic Brain Injury Outcome Research: 15-Year Trajectories of Life Satisfaction in Adolescents and Young Adults as an Example

Abstract

ObjectiveTo demonstrate the application of second-order growth mixture modeling (SO-GMM) using life satisfaction among adolescents and young adults with traumatic brain injury (TBI) up to 15 years post injury. DesignSO-GMM, a data-driven modeling approach that accounts for measurement errors, was adopted to uncover distinct growth trajectories of life satisfaction during 15 years post injury. Membership in growth trajectories was then linked with baseline characteristics to understand the contributing factors to distinct growth over time. SettingTraumatic Brain Injury Model System National Database. ParticipantsA total of 3756 adolescents and young adults (AYAs) with TBI aged 16-25 years (mean age, 20.49±2.66 years; 27.24% female). InterventionsNot applicable. Main Outcome MeasuresSatisfaction With Life Scale. ResultsFour quadratic growth trajectories were identified: low-stable (16.6%), which had low initial life satisfaction and remained low over time; high-stable (49.3%), which had high life satisfaction at the baseline and stayed high over time; high-decreasing (15.8%), which started with high life satisfaction but decreased over time; and low-increasing (18.2%), which started with low life satisfaction but increased over time. Sex, race, preinjury employment status, age, and FIM cognition were associated with group assignment. ConclusionsThis study applied SO-GMM to a national TBI database and identified 4 longitudinal trajectories of life satisfaction among AYAs with TBI. Findings provided data-driven evidence for development of future interventions that are tailored at both temporal and personalized levels for improved health outcomes among AYAs with TBI.