Despite the large population of patients with mental disorders and the rapid development of mental health services in China, there are few evaluations of Chinese mental health services from the patient perspective. Relevant instruments with robust psychometric properties are lacking. This study aimed to translate, adapt and validate the WHO responsiveness performance questionnaire for measuring the quality of hospital mental health services among Chinese patients. The adaption of the translated questionnaire incorporated experts' and patients' opinions. For psychometric testing, 193 outpatients and 168 inpatients completed outpatient and inpatient modules, respectively. The adapted questionnaire adhered to the WHO framework of responsiveness domains, and just four items had some wording changes. Item missing rates were below 6%. Both the outpatient and inpatient modules had acceptable internal reliability (Cronbach's α = 0.837 and 0.730) and most domains had desirable average inter-item correlation coefficients. The confirmatory factor analysis indicated an acceptable model fit for the inpatient module, while some goodness-of-fit indices for the outpatient module were a little outside of the recommended ranges. Except for 'talking privately' from the domain of confidentiality (both outpatient and inpatient modules) and 'waiting time' from the domain of prompt attention (the inpatient module), factor loadings of all other items were above 0.5. The Chinese version of the responsiveness performance questionnaire has acceptable feasibility, reliability, and validity in general and it can be used to measure, assess and improve the quality of mental health services in China.