The main purposes of this study were to determine the time interval between the onset of symptoms of myocardial infarction and treatment-seeking time and to identify the factors related to the interval time. This study used a retrospective design. The sample consisted of 45 patients aged over 30 who were diagnosed with an acute myocardial infarction at two large university affiliated medical center from September 1, 1997 to June 30, 1998. Data was collected by using questionnaries, which included demographic data, permonitory clinical signs and symtoms of myocardial infarction, and a measure of the severity of the signs and symptoms. Also semi-structured interviews and chart reviews were used to obtain information related to treatment-seeking time. The results of this study are summarized as follows ; 1. The most frequent premonitory clinical symptom was chest pain(92.9%), the second, was perspiration(81.0%), and the next were nausea(40.5%) and dyspnea(38.1%). Thirty two patients reported having more than four premonitory signs and symtoms. Patients described the characteristics of chest pain as “somethings very heavy pressing down”(26.2%), “felt like my chest would burst”(24.4%), or “sharp pain”(16.7%), Over 95% of the sample reported having chest pain. 2. Twenty two (52.4%) patients reported to have “very severe” premonitory pain. 3. The mean time interval between the onset of signs and symptoms and the arrival at the medical center was 6.39$pm$10.80 hours in 42 samples, the mean time from the onset to arrival at a local hospital was 3.27$pm$5.39 hours and for transfer from a local hospital to the medical center was 4.75$pm$9.87 hours in patients who had arrived at medical center via local hospital. 4. The severity of premonitory signs and symptoms did not differ significantly according to existence of premonitory signs and symptoms. 5. There was no significant relationship between treatment-seeking time and age, gender, marital status, economic status, occupation, or residence. But education had significant relationship(r＝－0.51, p＝0.01). Analysis of difference of the time interval according to the premonitory signs and symptoms showed that the time was shorter in patients who experienced nausea or dyspnea(U＝115.50, p＝0.01, U＝132.00, p＝0.04), however the severity of premonitory signs and symptoms did not have statistical significance.