Revisiting the Factor Structure of the Insomnia Severity Index Among Survivors of the 2013 Super Typhoon Haiyan

Objective: The insomnia severity index (ISI) is often used as a screening tool used with its recommended clinical cutoff scores. However, this practice presumes a single-factor structure, which has received little theoretical or empirical support. This study examined the factor structure of the ISI with the aim to validate its clinical application in postdisaster contexts. Method: One hundred sixty-eight Filipino adult survivors of the 2013 Super Typhoon Haiyan completed two waves of questionnaires that were 7 months apart. Confirmatory factor analyses were first conducted to compare models with different factor structures, followed by a test of longitudinal measurement invariance. The longitudinal interrelations among factors of the ISI as well as between insomnia and symptoms of psychological distress were then examined with path analysis. Results: Results supported a two-factor model, composed of severity and impact of insomnia, which achieved strong longitudinal invariance. The change in impact of insomnia at T2 was predicted by severity of insomnia at T1. Impact of insomnia was significantly associated with symptoms of stress but not depression and anxiety. Conclusion: We preliminarily propose a model in which the impact of insomnia mediates the influence of severity of insomnia on subsequent stress. Clinical Impact Statement The Insomnia Severity Index is a popular assessment tool. This study suggests that the scale is better construed as a two-factor measure: severity of insomnia and impact of insomnia. The two factors are associated with different longer-term clinical outcomes; severity of insomnia predicts impact of insomnia, which, in turn, is associated with longer-term symptoms of stress.

Automated summary

This study examined the factor structure and longitudinal relationships of the Insomnia Severity Index (ISI) in a postdisaster context. The results supported a two-factor model, with severity and impact of insomnia, and found that severity of insomnia predicted the impact of insomnia, which in turn was associated with longer-term symptoms of stress.