Contextual and relational models of music therapy suggest the relevance of music-related outcomes in clients’ everyday life, but no standardized instruments exist for that purpose. We aimed to develop and test the reliability and validity of a self-report questionnaire to measure interest in music (IiM). Twelve Likert-scaled items reflecting pro-social and socially avoidant aspects of interest in music were developed and translated into three languages (English, Norwegian, German). A total of 144 adults with a mental disorder and low therapy motivation completed the IiM questionnaire and adjacent self-report measures (self-esteem, social relationships). Interest in music was also rated by a mental health professional. Completion of the IiM was repeated after one and three months. Dimensionality, internal consistency, test–retest reliability and concurrent validity were examined. Analyses indicated two dimensions, Musical Activity and Emotional Engagement with Music (10 items) and Social Avoidance through Music (2 items). Internal consistency (Cronbach’s alpha 0.89 and 0.77), test–retest reliability (1- and 3-month intra-class correlation coefficients (ICCs) ranging from 0.61 to 0.85), and concurrent validity were demonstrated. The IiM scale fills an important gap in the tools available for music therapy assessment and outcome research in mental health.