Purpose: The purpose of this study is to evaluate the effect of rehabilitation education on the reduction of the stress of family members who have patients suffering from stroke and to find a new way to nurse the patients and their family. Subjects & Methods: The subjects were sixty-one family members with the patients who had been hospotalized in K hospital of oriental medicine from september the 9th, 1996 to september the fourteen, 1996. This study was performed by simulated control group pretest-posttest design; pretest was done on the control group through a questionnaire, counselling and observation while posttest was done on the experimental group 1-2 days after systemic rehabilitation education. To teach the patients and their family, the amended version of a book written by Lee Hae-jin was used as a tool for systemic rehabilitation education. As a method to estimate ADL score, modified Kang's method was applied and ADL score was measured by well-trained technician. As for the tool to estimate the degree of family stress, Choi's method adjusted to this study was applied. In the analysis of the data, social property of the patient and the characteristic of the disease were surveyed in $X^2$ examination to confirm the consistency between the experimental group and the control group. The diffrence in the degree of the stress, which is a dependent factor, was examined by t-test. The difference in ADL score between the experimental group and the control group was examined by t-test. The difference in the degree of the stress according to the general feature of the family with stroke patient, social property of the patients and the characteristic of the disease were surveyed by F examination. The difference in family stress according to the degree of ADL was surveyed by F examination. RESULTS: 1. After hypothetically-examined systemic rehabilitation education, the total of the score of family stress surveyed in 34 items of three domains was compared between the experimental group and the control group. There was no statistically significant difference between two groups; mean score of experimental group=2.28, that of control group=2.93(t=.17, df=59, p=. 66). 2. In the survey on family stress in 34 items, the items over mean score 3.0 were firstly the anxiety of possible disability and relapse of the disease and secondly to watch the patient's suffering without doing anything in the domain of the change of the disease and the difficulty in caring. And the items of the lowest stress with less than mean 2.0 score were little chance to meet the relative and friends, inconsistent treatment and attitude of the medical workers and the change of the attitude of the relative due to the patient orderly in the domain of social and personal relation and the responsibility as the family. The items which showed the difference between two groups were aggravation of neighboring patient(t=3.36, df= 59, p=.001) and the possibility of patient's death(t=2.19, df=58.38, p=.033) in the domain of the change of the disease and the difficulty in caring. 3. In the study on the stress difference according to general features of the family with the stroke patient, the score of family stress with the occupation was higher with mean 2.49 than that of the family stress without occupation with mean 2.16, but there was no significant difference. (F=5.21, df=1/59, p=.026). 4. In the study on the stress difference according to social property of the patient and the characteristic of the disease, there was significant difference in the age of the patients (F=2.98, df=3/57, p=.039). These results show that even if there is no statistically significant difference between two groups, sixteen of the experimental group are less than 3.0 in ADL score(standard 6 score)while eight of the control group are less than 3.0 and that ten of the experimental group are in the year range of 39-49 while four of the control group are in the year range of 39-49 which showed significant difference in family stress. These imply that there is a possibility that the experimental group have serious and fundamental stress resulting in high pretest stress compared with the control group. It might be due to the characteristic of simulated control group pretest-posttest design that the psychologic-supportive effect by the education was not observed. On the basis of these results, the followings are suggested. 1) A study on the nursing-mediated method to reduce the stress in the items which are not resolved by rehabilitation education, a study on nursing according to the patient's age and a study on the supportive nursing toward the family with occupation are required. 2) More than two times consecutive nursing-mediated rehabilitation education to measure the family stress is required. 3) Comprehensive and multilateral systemic education program including the instruction on western-eastern medicine, physical therapy, exercise and diet through collaboration of the experts in each field is required. 4) Family stress at home as well as in the hospital needs to be estimated and home rehabilitation and home-nursing needs to be continued.