Since it is suggested that only effective cardiopulmonary resuscitation (CPR) improves survival rates, quality control of training outcomes is important and comparisons between different training methods are desirable. The aim of this study was to test a model of quality assurance, consisting of a computer program combined with the Brennan et al. checklist, for evaluation of CPR performance. A small group of trained medical professionals (cardiac care unit nurses) (n = 10) was used in this pilot study. The result points out several points of concern: half of the participants did not open the airway prior to breathing control. Over 90% of all inflations were ‘too fast’ and 71% were ‘too much’. Only 6.5% of the inflations were correct. On average, the participants made 5.4 inflations per minute. Concerning chest compressions, 40% were ‘too deep’ while only 4% were ‘too shallow’. In spite of the fact that the participants had an average rate at 95 compressions per minute the number of compressions varied between 32 and 51 during 1 minute. When new guidelines are discussed, it would be beneficial if they were tested by a number of people to investigate if following the guidelines is at all possible. © 2001 Lippincott Williams & Wilkins, Inc.