We describe the development of a Chinese version of the Brief Pain Inventory (BPI-C) and demonstrate its reliability and validity. We also report the use of the BPI-C in a three hospital study of cancer pain and its treatment. As with other language versions of the BPI, factor analysis of the BPI-C items results in a two factor solution that satisfies the criteria of reproducibility, interpretability and fit in a confirmatory setting. The first factor consists of the four pain severity scales, while the seven pain interference scales comprised the second factor. The BPI-C proved to be a reliable measure of both the severity and impact of pain in patients with cancer. Coefficient alpha for the pain severity and pain interference items were 0.894 and 0.915, respectively. The sample (N = 147) was gathered at three cancer treatment hospitals in Beijing. The patients from these hospitals reported higher levels of pain severity and pain interference compared with patients in similar studies done at the time (1991-1992) in the United States and France. This was in keeping with the finding that a larger proportion (67%) of the cancer patients in these Beijing hospitals were judged to have inadequate analgesia as assessed by the Pain Management Index (PMI), an estimate of adherence to the World Health Organization (WHO) guidelines for cancer pain management.