To evaluate the factors affecting the time between symptom onset and hospital arrival in patients with acute myocardial infarction (AMI), we gave a detailed questionnaire to all who were admitted or transferred with AMI from January 1988 to February 1989. In these 126 patients (94 men, 32 women) the mean prehospital time was 5.9 +/- 11.0 hours (median 2.0, range 0.4 to 69.0). The time between symptom onset and reaching a decision that medical care should be sought was 62% of the mean prehospital time. In 100 (79%) patients, the prehospital time was less than or equal to 6 hours; of these, 61 (61%) were retrospectively judged to have been optimal candidates for lytic therapy. Stepwise multiple regression selected the following 4 variables as independent predictors of prehospital time: slow symptom progression; low income; female gender; and advanced age. All of these variables are predictive (p less than 0.03) of increased prehospital time; absence of prior AMI was of borderline additional significance (p = 0.053). Similarly, logistic regression analysis selected slow symptom progression, female gender and low income as significant (p less than or equal to 0.02) independent predictors of prehospital time greater than 6 hours. The logistic regression model incorporating these 3 variables had a sensitivity of 54%, a specificity of 95% and a positive predictive value of 72% in identifying patients with prehospital time greater than 6 hours. Thus, these data indicate it is possible to characterize patients likely to experience undue prehospital delay during AMI, which may be of importance to future public education efforts.