To quantify the impact of headache of daily living, we developed a 25-item headache disability inventory (HDI). The alpha version of the HDI (alpha-HDI) consisted of 40 items, each requiring a "yes" (four points), "sometimes" (two points), or "no" (zero points) response based on items derived empirically from case history responses of subjects with headache. From the alpha-HDI, we derived a 25-item beta version (beta-HDI) with the items subgrouped into functional and emotional subscales. The internal consistency/reliability was strong, as was construct validity. The test-retest reliability for the beta-HDI was acceptable for the total score and functional and emotional subscale scores. A 29-point change (95% confidence interval) or greater in the total score from test-retest must occur before the changes can be attributed to treatment effects. The HDI is useful in assessing the impact of headache, and its treatment, on daily living.