Objective: To investigated the concept and clinical practice of patient-controlled analgesia (PCA) in the treatment of cancer pain. Methods: Doctors, nurses, pharmacists from the oncology department, pain department, or hospice department were investigated using an electronic questionnaire from December 1 to December 31, 2021. In addition to the basic information, there were 26 questions were collected, including the current situation of cancer pain treatment, the concept of medical staff on PCA treatment of cancer pain and the clinical practice of PCA. Results: Questionnaires from 2 872 medical staff were collected from 993 hospitals in 30 provincial administrative units. Only 34.8% (955/2 748) of medical staff considered that the satisfaction rate of cancer pain control was over 75%, and 27.9% (548/1 968) of medical staff convinced that the satisfaction rate of breakthrough pain control was less than 50%. 97.1% (2 439/2 513) of medical staff considered that PCA could be effectively used for cancer pain treatment. The proportion of medical staff in secondary and tertiary hospitals who thought that PCA was applicable to cancer pain that could not be effectively alleviated by standardized non-invasive drug administration was 64.6% (319/494) and 69.1% (1 262/1 826) respectively, which was higher than that in primary hospitals [57.0% (110/193)] (P=0.002). In different occupations, the proportion of nurses who convinced PCA treatment of cancer pain increased the risk of addiction and drug overdose was 62.8% (431/686) and 76.1% (522/686), respectively, which was higher than doctors [39.2% (670/1709) and 58.2% (995/1709), respectively] and pharmacists [49.2% (58/118) and 65.3% (77/118), respectively] (all P<0.001). There was no significant difference in type of pump, route of administration, mode of infusion, protocol for PCA administration and selection of common medication in PCA treatment of cancer pain among different hospitals (all P>0.05). The calculation of continuous infusion dose and rescue dose of PCA was not uniform among different hospitals. After initiation of PCA, 71.7% (1 226/1 709) of hospitals had insufficient analgesia and most of them needed to be adjusted for 1-3 times to achieve satisfactory analgesia. Conclusion: Medical staff have insufficient cognition of PCA treatment of cancer pain and there is a lack of unified guidance in clinical practice. Therefore, it is an urgent need to develop an expert consensus on PCA treatment of cancer pain.