To investigate factors that influence length of stay (LOS) in patients hospitalized for transient ischemic attacks (TIAs). This may help to reduce unnecessary days of hospitalization. Retrospective study. A large, non-academic general hospital. All patients hospitalized for TIAs during the years 1996-1998 were retrospectively studied. We analyzed the relationship between demographic, clinical and management features and LOS. We studied 157 patients. Median LOS was 8 days (range, 1 to 21 days). Statistical analysis showed that female sex, being hospitalized in 1996, hospitalization from Wednesday through Saturday, and second-level cardiological, neuroradiological and miscellaneous investigations significantly increased LOS. The LOS of patients with TIAs may be reduced through measures that accelerate the diagnostic work-up and by scheduling the admission to the first days of the week, when this is possibile.