Aim. This paper is a report of psychometric testing of the Swedish version of the Child Drawing: Hospital Manual.Background. Drawings have shown to be useful in assessing emotional status and anxiety in children because they generally speak to us more clearly and openly through their drawings than they are willing or able to verbally.Method. The Child Drawing: Hospital Manual was translated into Swedish according to World Health Organization guidelines (a routine procedure for translation of English instruments) in order to assess anxiety by analysing the drawings of 59 children (5–11 years), of whom nine were girls and 50 boys undergoing day surgery during 2007–2009.Results. Inter-rater reliability (five independent scorers) was high and internal consistency reliability was good (coefficient alpha = 0.77). Parts A and C, as well as the total scale score of the Child Drawing: Hospital Manual, discriminated anxiety significantly between the group of children undergoing day surgery and a comparison group of school children, indicating adequate construct validity.Conclusion. For the Swedish version of the Child Drawing: Hospital Manual, our study demonstrates evidence for adequate construct validity in Parts A and C (and total scale score), high inter-rater reliability and acceptable internal consistency reliability. However, some improvements are needed before the instrument will be a clinically useful assessment of anxiety in children undergoing day surgery.