Effect size and experimental power analysis in a paediatric cardiology telemedicine system

A telemedicine system was installed between the University of North Carolina Hospitals and the New Hanover Regional Medical Center. It allowed the transmission of neonatal echocardiograms for immediate reporting. During a six-month study period the system was used for the interpretation of 110 echocardiograms from 48 babies. There were 38 babies studied in a retrospective control period. Hospital length of stay decreased by an average of six days in the telemedicine group, representing an annual saving of some $1.3 million. However, these apparent differences were not significant (P = 0.2) and a power analysis suggested that a sample size of some 600 would have been necessary.