Graph Format Effects in Processing Health Outcome Information

Abstract

Decision support tools that incorporate predictive risk estimates can be used to assist patients and their families in making better-informed choices about treatment options. The format utilized to present predictive risk estimates can influence risk perception and treatment decisions. The study reported here investigated the influence of graph format on information processing and decision-making in relation to rt-PA therapy for stroke. Forty-five older adults were asked to make a hypothetical decision about rt-PA while viewing rt-PA risk information presented in one of three graph formats. Eye tracking, scan path, and transition analysis were used to investigate differences in information processing by graph format. Graph format did not affect whether or not study participants said yes to rt-PA treatment. There was an effect of graph format on decisional uncertainty, study time, and memory accuracy. Mean fixation densities and common transitions were significantly different by information area, graph format. and time epoch. Whether graph format alone can influence decision strategies enough to affect choice remains an open question. However, using fixation density and transition probabilities together appears to be a viable means of inferring information processing and discerning information processing differences.