Stigma against people who have alcohol and drug problems severely affects their health and well-being. An instrument based on stigma theory assessing individual-level stigma is essential for a comprehensive understanding of their stigma. We evaluated the validity and reliability of the Japanese version of the Substance Use Stigma Mechanism Scale (SU-SMS-J) among a population who had alcohol or drug use problems. Adults with experience in substance use disorders from psychiatry outpatient departments and rehabilitation facilities participated in the self-administered questionnaire survey. Confirmatory factor analysis was conducted to test the structural validity of the 5-factor model proposed in other language versions, and factor loadings and correlation between the subscales were confirmed. The correlations between the SU-SMS-J and psychometric properties related to substance use (e.g., severity of substance use, motivation to change) were investigated to assess concurrent validity. Internal consistency was assessed using Cronbach's alpha coefficients. Data from 126 participants were analyzed. The 5-factor model was acceptable with good or reasonable model fit indices. The correlations between subscales were weak to moderate, and this result suggested the SU-SMS-J assessed different but related components of stigma: enacted, anticipated, and internalized stigma from different stigma sources (family and healthcare workers). The SU-SMS-J and subscales showed moderate concurrent validity. Internal consistency was mostly sufficient, with Cronbach's alpha coefficients of 0.86 for all items and 0.66-0.93 for subscales. The SU-SMS-J is valid and reliable for use among populations with substance use problems in various settings in Japan.