The purpose of this study was to identify factors influencing patient satisfaction and to evaluate the utility of patient satisfaction as an outcome indicator. The study was conducted by mailed questionnaire. The subjects were 900 patients discharged from adult nursing units in a tertiary teaching hospital. On the discharge date, questionnaires were distributed by two trained research assistants. The questionnaire developed by the researchers was based on Larson(1996)'s study, and consisted of 71 items with the following components: overall satisfaction, domain-specific satisfaction(administration process, hospital facility and environment, nurses, and doctors), patients' loyalty(intention to use the health care service of the hospital in the future), recommendation to others, health benefits, and demographic characteristics. Each item was rated using a five point Likert scale ranging from '1=strongly disagree' to '5= strongly agree'. The response rate was 43%(387/900). The satisfaction level with the health care service was generally high. Perceived health status was the only significant factor influencing satisfaction level. Satisfaction with doctors contributed the most to explaining overall satisfaction. Overall satisfaction was significantly correlated with patient loyalty, recommendation, to other and perceived health benefit. It was found that the score of satisfaction was positively correlated with the score of loyalty, recommendation, and health benefit. Therefore, patient satisfaction seems to be a good outcome indicator.