Contents lists available at ScienceDirect International Journal of Medical Informatics journal homepage: www.elsevier.com/locate/ijmedinf Improving prognosis communication for patients facing complex medical treatment: A user-centered design approach Sunyoung Kim*, Beatrice Trinidad, Lisa Mikesell, Mark Aakhus School of Communication and Information, Rutgers University, United States ARTICLE INFO Keywords: User-centered design Prognosis Risk communication Visual aid Electronic decision aid Decision making ABSTRACT Background: The understanding and processing of numerical prognostic information can be challenging for patients who suffer from disease and the stress of a diagnosis. Objective: This paper investigates how patients diagnosed with Leukemia respond to different graph re- presentations of prognosis information. Methods: We conducted a user-centered design process, for which three experimental prototypes (vertical, horizontal, and pie charts) with and without animation were developed. Twelve patients diagnosed with Leukemia were recruited to evaluate the prototypes using a think-aloud interview protocol. Results: The results showed a preference for vertical bar charts over horizontal and pie charts. In addition, we found that animating the charts to “fill-up” generally conveyed a subtle sense of positivity even when diagnosis information was negative. The value of explicitly indicating numeric values and scale varied but the results suggest that what matters to participants is having control over when such details would be seen. The results also point out that making sense of prognostic information involves balancing the tension between information utility and patient judgments about authenticity and credibility of prognosis information. Conclusion: Our findings are important for the design and implementation of representations of prognostic in- formation. They suggest that an appropriate visual format can reduce potential negative effects in conveying prognosis information, as well as helping patients stay positive and motivated for cure in the delivery of prognosis information. 1. Introduction The meaningful communication of numeric prognostic information is essential for establishing informed decision-making in clinical set- tings [10]. However, even for patients with high numeracy, the un- derstanding and processing of numerical information can be challenged when suffering from disease and the stress of a diagnosis [1,28,40]. To facilitate the communication of numeric prognostic information elec- tronic decision aids adopt a variety of visuals (e.g., icons, charts, pic- tograms) [9,14]. In particular, graphs are widely used in decision aids because they support human processing of quantitative information by reducing the amount of mental computation required to interpret in- formation [41]. However, there is little empirical guidance regarding how to use different visual aids to better support patients’ under- standing of risk information, especially when such risks may be life- threatening or reduce life expectancy [45]. For distressed patients who face a complex medical treatment, it is as important to minimize emotional stress and promote patients’ optimism as to deliver realistic information for effective risk communication [13]. In the design of electronic decision aids, therefore, it is essential to understand how different types of visuals for presenting numeric information influence patients’ comprehension of prognosis both informationally and emo- tionally [32] and subsequent psychological adjustment of decision- making [43]. The objective of this study, therefore, is to investigate how patients facing a complex medical treatment perceive and respond to different types of visual representations presenting numeric data of projected treatment outcomes. Acute myeloid leukemia (AML) is a type of malignant blood cancer, characterized by the rapid growth of abnormal cells that build up in the bone marrow and blood and interfere with normal blood cells [39]. In AML, a major therapeutic challenge is deciding whether a patient should receive allogeneic hematopoietic stem cell transplant or proceed with chemotherapy. This treatment decision requires patients to un- derstand projections of survivability and the likelihood of relapse after https://doi.org/10.1016/j.ijmedinf.2020.104147 Received 11 December 2019; Received in revised form 31 March 2020; Accepted 16 April 2020 ⁎ Corresponding author at: Rutgers University, School of Communication and Information, 4 Huntington Street, New Brunswick 08901, United States. E-mail addresses: sunyoung.kim@rutgers.edu (S. Kim), beatrice.trinidad@gmail.com (B. Trinidad), lisa.mikesell@rutgers.edu (L. Mikesell), aakhus@rutgers.edu (M. Aakhus). International Journal of Medical Informatics 141 (2020) 104147 1386-5056/ © 2020 Elsevier B.V. All rights reserved. T