Clarifying the Factor structure of the Intolerance of Uncertainty Inventory: Measurement Invariance and Validity in Predicting Clinical Outcomes

The Intolerance of Uncertainty Inventory (IUI) is a two-part inventory assessing the general unacceptability of uncertainty in terms of generalized maladaptive beliefs (IUI-A) and specific individuals’ reactions to uncertainty, such as avoidance, doubt, overestimation of threat, worry, control of uncertainty, and seeking reassurance (IUI-B). Previous research has examined the factor structure of both parts with mixed results, especially for IUI-A. The present study took advantage of the Farsi-language version to clarify the IUI factor structure, assess measurement invariance between clinical and nonclinical samples, and examine the criterion-related and incremental validity of IUI-A and IUI-B. We administered the IUI to a mixed psychiatric sample of patients who met DSM-IV criteria for affective disorders (N = 198) and community participants in relatively good health (N = 427). Exploratory structural equation modeling analyses were used to test alternative factor models and determine whether IUI factors predicted depression and anxiety symptoms. A bifactor model was found to be the best fit for IUI-A and IUI-B, and reliability analyses supported the use of the total scores for both parts. The IUI-A and IUI-B were scalar invariant. The general factors of each part and the IUI-B worry factor were higher in patients with affective disorders than in community participants. Validity analyses showed that the two general factors tapped into the same variance in depression and anxiety symptoms as the competing instruments, although not superior in predictive performance to the Intolerance of Uncertainty scale (IUS-12) and the Disorder Specific Intolerance of Uncertainty (DSIU) scale. The present study clarified the factorial structure of the IUI and provided evidence that IUI total scores are reliable and valid for assessing the Intolerance of Uncertainty construct and predicting clinical outcomes.