Patients readmitted to the intensive care unit (ICU) have a significantly higher mortality rate. The role of intensivists in judging when to discharge patients from the ICU is very important. We undertook this study to evaluate the effect of the intensivists' discharge decision-making on readmission to ICU. The intensivists actively participated in the discharge decision-making, with the discharge guideline taken into consideration, in respect of group 1 patients, but not in respect of group 2. The readmission rate in group 1 was lower than that in group 2. The readmission in patients in each group was associated with higher mortality rates and longer lengths of stay at the ICU. Respiratory failure was the major cause of readmission. In the non-survivors out of the readmitted patients, the Acute Physiology and Chronic Health Evaluation (APACHE) III scores on the initial discharge and readmission, the multiple organ dysfunction syndrome (MODS) scores on the initial admission, discharge and readmission were higher than the corresponding indices in the survivors. We conclude that the readmission rate was lower when intensivists participated in the discharge decision-making, and that APACHE III and MODS scores on the first discharge and readmission were significant prognostic factors in respect of the readmitted patients.