4 - (IOP 20) Improving Insomnia Assessment in Veterans: A Moderated Nonlinear Factor Analysis on Insomnia Severity Index

Introduction: Insomnia is common among veterans, particularly those with mental health conditions like depression and anxiety, and can lead to significant health complications. Routine screening in healthcare settings is crucial to prevent chronic insomnia. The Insomnia Severity Index (ISI), a widely used and validated tool, has been adapted for diverse populations, but its measurement invariance (MI) and differential item functioning (DIF) remain underexplored. This study uses moderated nonlinear factor analysis (MNLFA) to address this gap, leveraging its flexibility to model item variability based on individual characteristics and improve accuracy with continuous and categorical predictors.

Methods: Veterans (N = 1029) from the Miami VA sleep center completed a baseline psychosocial assessment, and diagnosis data were extracted from medical records. MNLFA was used to model nighttime (items 1a,b,c) and daytime symptoms (items 4–7) separately, examining the effects of age, gender, race/ethnicity, depression, anxiety, posttraumatic stress disorder (PTSD), and chronic pain on latent means, variances, and item invariance. DIF-adjusted factor scores, confirmatory factor analysis (CFA) factor scores, and sum scores were compared.

Results: The veteran sample (N=1029) was middle-aged (M=52, SD=14.5), predominantly male (83.5%) and White (57.3%), with 50% diagnosed with chronic pain and clinical depression. DIF analysis showed ISI1 had intercept DIF for age, Hispanic/White identity, chronic pain, and depression, as well as loading DIF for age. ISI3 had intercept DIF for depression. ISI4 exhibited intercept and loading DIF for male gender. ISI5 and ISI6 showed intercept DIF for age, and ISI7 showed both intercept and loading DIF for PTSD. Factor scores derived from MNLFA, CFA, and sum scores were highly correlated across both factors.

Conclusion: This study examined the MI and DIF of ISI by investigating how an array of psychosocial factors influences insomnia severity ratings. Findings indicate that group differences in items without DIF likely reflect genuine disparities. MNLFA demonstrated methodological advantages by allowing for the simultaneous modeling of mean and variance effects while also accounting for DIF across both continuous and categorical predictors. While the covariates selected align with characteristics commonly seen in veteran populations, future research could expand on this by incorporating additional factors, such as the apnea-hypopnea index, and examining interactions between covariates. Although difficult to implement in primary care, MNLFA-based factor scores hold promise for secondary predictive models.