Nurses frequently care for sleepless elderly patients on bed rest in a hospital environment. Our previous study with young adults showed that bright light exposure during the daytime affected the induction of nocturnal deep sleep. The purpose of this study is aimed at finding whether similar research could be observed with hospitalized elderly patients. Seven patients (mean age 67; range 57-77 yrs, males 3: females 4) served as participants and their informed written consent was obtained. A fluorescent lamp fixed in the bed frame near the head of the patient was turned on at 10:00 h and off at 15:00 h each day for 1 week (BL). Moreover, each patient was required to stay near this light during this period. The patients lived in a room facing north, where the ambient light intensities ranged from 50 to 300 lx during the daytime. Their activities were continuously measured using an Actiwatch (model-AWL, Mini-Mitter, USA). Salivary samples were collected at midnight for the measurement of melatonin. The findings were compared between 2 days before BL exposure (baseline) and the last 2 days during BL exposure, respectively. The bright light exposure during the daytime prolonged "Time in Bed" (p<0.05), increased "Immobile Minutes" (p<0.05), and delayed "Get up Time"(p<0.01). The average melatonin secretion at midnight in four patients increased from 7.5 ± 2.6 pg/ml to 13.3 ± 9.2 pg/ml. These findings suggest that diurnal bright light exposure for hospitalized elderly patients lying in bed under dark condition during the daytime may favor clinically the induction of nocturnal deep sleep. Attention should be given to the illumination conditions for elderly patients in hospitals to improve their impaired sleep.