Pain is a common and distressing symptom in ICU patients. Effective management can only be achieved with accurate pain assessment. However, this is difficult in the critically ill as patients are often unable to communicate verbally due to the presence of endotracheal/tracheostomy tubes and sedation. The aim of this study was to validate the Chinese version of the Critical-Care Pain Observation Tool (CPOT) in non-verbally communicating patients. Purposive sampling was used. A total of 120 conscious and unconscious patients in the intensive care units of a medical center participated in the study. Patients were assessed before, during, and 20 minutes after the 2 following procedures: (1) nociceptive procedure: performing suction, and (2) non-nociceptive procedure: taking noninvasive blood pressure (NIBP). Consciously ventilated patients were also asked to provide a self-report level of pain. Discriminant validity was supported with increases of the CPOT during suction, but remaining stable during NIBP. Discriminant validity was also supported by higher scores during suction versus at rest. For criterion validity, the CPOT scores were correlated to the patients’ self-reports of pain. Using a CPOT cutoff score of 2 yielded a sensitivity of 60% and a specificity of 100%. These results support that CPOT is a valid tool to assess pain in critically ill patients. Further studies are needed to determine if CPOT can be used in other critical ill patients for assessing changes in severity of pain over time.