||This study was made to determine (a) if there is a significant difference between anxiety levels of psychiatric patients at a midway point in their hospitalization and at the time of their discharge, and (b) is extra-hospital factors such as illness, employment status, or follow-up care influence anxiety levels as the patient is being released from the hospital. This was accomplished by administration of the Taylor Manifest Anxiety Scale (MAS) and a structured interview. Every new patient admitted to the psychiatric unit of Salt Lake General Hospital who met the following criteria was selected as a subject: (1) he must have been in the hospital for ten day before the initial MAS was given and (2) he must have been in the hospital for a period of two weeks before the second test and the structured interview were give. Information concerning independent variables (age, occupation, sex, race, medication, source of referral to the hospital, and method of payment) was recorded. Each patient was individually administered the tests. For each scale (MAS, L, and K) a t test was used to determine the significance of the difference between the initial and retest. A Pearson product moment correlation was completed from the test-retest record of the 30 patients. There were no significant differences in means of standard deviations on the three scales. The positive correlations were all significant at the .01 level of confidence. The hypothesis that there would be an increase in anxiety at the time of discharge was not confirmed. There was a significant tendency for patients scoring low on the Taylor scale to be less candid (as inferred from their scores on the L scale) than individual scoring high. The structured interview divulged the present methods of discharging patients and information relating to the extra-hospital situations at the time of discharge. Since extra-hospital factors are highly variable and personal, an interview seemed to be the best method to determine their effect. Although not statistically significant the patterns of responses indicate that follow-up care and illness in the family are situational factors affecting anxiety. Two by two contingency tables were used to see if there was a reliable relationship between the MAS and the items in the interview as well as the independent variables. One question was significant "" How do you feel about going home? "" and was confirmed at the .05 level of significance (X = 6.42). The patients who desired to return home had a lower anxiety rate than the ones who desired to remain hospitalized. This study is consistent with past studies and indicates the MAS measures basic anxiety. Basic anxiety levels have not changed and the patient returning to the community meets stress situations which create situational anxiety that is manifest. The test is not refined enough to pick up small changes that the individual experiences during the course of hospitalization. The L and K scales show defensive reactions to anxiety. The defense responses may be effective in reducing anxiety and thus the anxiety level may appear low on testing. The interview revealed extra-hospital factors that affect the patient at the time of discharge and shows a need to deal with the patient"™s environment and his response to it. Results of the structured interview could be improved if the author had established rapport with the patients and also a sense of trust had been developed. It is the conclusion of the writer that further care should be implemented as the patient leave the hospital and that nursing should assume continuity of patient care in the community. The nurse, functioning as a member of the health team, can insure that problems are dealt with as they arise, decrease anxiety, and create a therapeutic environment for the family.