OBJECTIVE: The objective of this analysis was to evaluate the psychometric properties of the 24 Hour Headache Disability Questionnaire (Disq-24) using confirmatory factor analysis. Disq24 is a 14-item questionnaire with a 6-point Likert-type scale that measures disability in the 24 hours following onset of headache pain. METHODS: We utilized data from an acute migraine treatment clinical trial in which the Disq-24 was administered 24-hours after onset of headache pain (N=323). The Disq24 was originally hypothesized to measure the impact of headache on Family/Social Activities, Work Activities and Emotions/Feelings. It is uncertain whether the Family/Social/Work Activities domain should be combined to a single factor. We tested the hypotheses by estimating 2-factor and 3-factor confirmatory factor-analytic models using maximum likelihood fitting function in AMOS Version 4.0. To assess data fit, we used the chi-square test and fit indices including Tucker-Lewis Index (TLI), Incremental Fit Index (IFI) and Root Mean Square Error of Approximation (RMSEA). RESULTS: The chi-square statistic indicated that the 2-factor model had a poor fit.(p=0.000). However given the sensitivity of this test to sample size, the other fit indices TLI (0.923), IFI (0.936) and RMSEA (0.107) indicated a reasonably good fit. Item 8 (0.434) and Item 10 (0.449) had low squared multiple correlations, indicating only a small proportion of variance in these items was explained by this model. Data fit improved marginally in the 3-factor model but correlation between the Work and Social Factor was 0.943 indicating that both were driven by a single factor. Deleting Items 8 and 10 resulted in a model with TLI (0.947), IFI (0.958) and RMSEA (0.098) indicating a significant improvement in fit. CONCLUSIONS: Results from CFA suggested better data fit with a 2-factor structure. The psychometric properties of the instrument can be improved by revising/deleting Items 8 and 10.