Is a picture worth a thousand medical words? A randomized trial of reporting formats for medical research data

Abstract

Monitoring data that vary over time is an essential component of medical practice. This is doubly true in clinical trials in which the overall safety and efficacy of investigational treatments in populations must be monitored in addition to the status of the individual patients who receive them. We report the results of a randomized trial of four reporting methods for time-dependent information derived from clinical trials; narrative text, table, pie chart and icon. Multivariate analysis of variance with a repeated measures design was used to analyze the efficiency of subjects' (physicians, research nurses and laboratory personnel) assimilation of information. Icons were found to be superior to the other reporting formats tested in both speed (p less than 0.0001) and accuracy (p = 0.02). The differences were most pronounced in subjects' first exposure to the data, suggesting that icons reduce the time needed for training. We conclude that icons are a valuable method for presentation of time-dependent information in medical settings.