The study was carried out to validate objectively the relationship between clinical impressions of parents and ECG evidence of life-threatening arrhythmias in infants considered to be at high risk. Twenty infants considered to be at high risk were provided with a hospital grade home monitor with a hard copy-recording capability (n = 12) and/or an ECG telephone transmission system (n = 8). There were 93 alarms considered by the parents as true alarms. All of the ECG tracings recorded during the 93 episodes were normal. Infant stimulation was provided for 11 infants and was not administered during 82 of these alarms. None of the infants died and all are well at 1 year of age. Our data raises questions regarding parental clinical ability to correctly perceive a true near miss episode in most cases and may explain, in part, the lack of consistency noted in previously published physiologic studies of "high risk" populations. Finally, the study raises further questions regarding the use of parental impression alone without supportive objective data as an indication for therapy in the majority of such patients. Research efforts and application of therapeutic measures will benefit greatly by development of objective data to support and supplement parental clinical observations.