The purpose of this study was to investigate the impact of depression, discomfort, spirituality, physical care, and opioid use on pain with terminally ill cancer patients residing in hospice units. The convenient sample of this study consisted of 41 terminally ill cancer patients at three hospice units in university affiliated hospitals. Patients were interviewed with structured questionnaires three times at predetermined intervals: admission to the hospice unit (Time 1), one week later (Time 2), and two weeks later (Time 3). The data was collected from January 1998 to January 1999 and was analyzed using ANOVA, Pearson correlation coefficient, and multivariate multiple regression. The results of this study were as follows: 1. The mean age of the participants was approximately 55 years old. In terms of diagnosis, lung cancer showed the highest frequency (19.5%), followed by stomach cancer and rectal cancer (17.1%). The motive of seeking hospice unit admission was control (72. 2%), followed by spiritual care (50%), and symptom relief (38.9%). 2. Regarding the type of pain felt, the highest pain frequency the participants experienced was deep pain (55%), followed by multiple pain (25%), intestinal pain (10%), then superficial (5%) and neurogenic pain (5%). For the level of pain measured by VAS, there was no significant difference among the three time points; Time 1 (5.04 +/-2.21), Time 2 (4.82+/-2.58) and Time 3(4.73+/-2.51). 3. There was significant change seen in spirituality and physical care in each time interval. Namely, the longer the length of admission at the hospice unit, the higher the importance of spirituality (p=0.0001) and the more the physical care the participants received (p=0.01). The opioid use at the three time points showed the following frequencies : Time 1 (75.6%), Time 2 (85.4%) and Time 3 (75.6%). 4. Regarding factors influencing pain, the pain level was significantly affected by the depression level (p <0.01) and the opioid use (p <0.1). These results were the most significant at the two time points (Time 1 and Time 2). At Time 3 (two weeks later), the pain level was significantly affected by the depression level (p <0.05) and the amount of physical care the participants received (p <0.1). In conclusion, the terminally ill cancer patients had moderate pain, were generally depressed, and were treated with opioid analgesics. As approaching death, the patients received more physical care due to increased physical symptoms experienced and they had a higher perception of the importance of spirituality. Thus, health care professionals need to provide continuous care for each of them to die comfortably physically, psycho- logically, and spiritually.