Rett syndrome is a severe neurodevelopmental disorder that affects about 1 in 10,000 females. Clinical trials of disease modifying therapies are on the rise, but there are few psychometrically sound caregiver-reported outcome measures available to assess treatment benefit. We report on a new caregiver-reported outcome measure, the Rett Caregiver Assessment of Symptom Severity (RCASS). Using data from the Rett Natural History Study (n = 649), we examined the factor structure, using both exploratory and confirmatory factor analysis, and the reliability and validity of the RCASS. The four-factor model had the best overall fit, which covered movement, communication, behavior, and Rett-specific symptoms. The RCASS had moderate internal consistency. Strong face validity was found with age and mutation type, and convergent validity was established with other similar measures, including the Revised Motor-Behavior Assessment Scale, Clinical Severity Scale, Clinical Global Impression Scale, and the Child Health Questionnaire. These data provide initial evidence that the RCASS is a viable caregiver-outcome measure for use in clinical trials in Rett syndrome. Future work to assess sensitivity to change and other measures of reliability, such as test--retest and inter-rater agreement, are needed.