The aim of the study was to assess the usefulness of three different morbidity scores in predicting unsuccessful resuscitation. We reviewed the records of adult patients who underwent CPR between September 1994 and June 1996 in The Royal Hampshire County Hospital, Winchester. Demographic data and enough clinical data to calculate the Pre-Arrest Morbidity score (PAM), the Prognosis After Resuscitation score (PAR) and the Modified PAM Index (MPI) were collected. During the study period 264 consecutive adult patients underwent inpatient CPR. Twenty-eight (11%) of the patients survived to discharge from hospital. Patients who died had significantly higher morbidity scores than those who survived. No patient with a PAM score greater than 6/25, PAR greater than 7/28 or MPI greater than 6/24 survived. There were 47/264 patients who scored above this threshold for the PAM score giving a sensitivity for predicting unsuccessful CPR of 20%. The sensitivity of the PAR was 29% and MPI was 22%. Each score identified a different group of patients for whom CPR was unsuccessful. Using all three scores in combination identified 42% of the unsuccessful CPR attempts. Morbidity scores are likely to need further refinement in order to be a useful bedside tool for predicting success for individual patient resuscitation attempts.