The purpose of this study was to examine the effect of supportive nursing on stress reaction of breast cancer patients undergoing chemotherapy. The nonequivalent control group pre-test/post-test design was used for this experimental study. The subjects were 32 patients who were receiving chemotherapy after mastectomies at K hospital in Taegu from June, 1994 to June 1995. Among 32 subjects, 16 were placed in the experimental group and 16 in the control group. The experimental and control groups were tested for general characteristics, trait anxiety, health locus of control, family support, state anxiety, hopelessness, physical stress, and anxiety behavior. Collected data was analized by means of a chisquare test and a t-test for the comparative analysis of the general characteristics and homogeneity of subjects. ANOVA, and MANOVA were used for testing the hypothesis. Reliability of the tools were analyzed using the Pearson Correlation coefficient. The results of this study were as follows : 1. The hypothesis : The stress reaction of the experimental group which took supportive nursing was lower than the stress reaction of the control group : this was supported statistically. The main variable influenced in stress reaction was hopelessness. Supportive nursing for breast cancer patients, who are receiving chemotherapy, was especially effective in the reduction of hopelessness compared to state anxiety, physical stress, and anxiety behavior. 2. An analysis of the difference on stress reaction, according to the frequency of supportive nursing between the control and experimental group, showed the level of hopelessness of the experimental group was lower than the control group after four supportive meeting sessions. But there was no statistical difference in state anxiety, physical stress, and anxiety behavior. In conclusion, this study supported utilization of supportive care as well as demonstrating the effectiveness of the System-Developmental Stress Model developed by Chrisman and Riehl-Sisca.