A user-centered framework for redesigning health care interfacesIntroductionHuman ndash computer interface design methodsUser/environmental analysisTask analysisRepresentational analysisTypes of interface designsFunctional analysisHuman ndash computer interface evaluation methodsInspection methodsHeuristic evaluationComparison of the users rsquo and designers rsquo conceptual modelKeystroke level modelCognitive walkthroughComparative analysisSmall-scale usability studiesComparison of the redesigned application with the original applicationAn approach to redesigning a health care application: a case studyStep 1: Analysis of the original applicationStep 2: User analysis for the redesigned applicationStep 3: Comparative analysesStep 4: Functional analysisStep 5: Representational analysisStep 6: Creation of the prototypeStep 7: Small-scale usability studiesStep 8: Modify the prototypeStep 9: Compare new and old systemsTask completion timeTask successUser satisfactionStep 10: Make final modifications for program releaseDiscussionConclusionAcknowledgmentsReferencesMethodological ReviewA user-centered framework for redesigning health care interfacesConstance M. Johnsona,*, Todd R. Johnsonb, Jiajie ZhangbaDepartment of Biostatistics and Applied Mathematics, The University of Texas, M. D. Anderson Cancer Center,1515 Holcombe Boulevard, 447, Houston, TX 77030, USAbThe University of Texas Health Science, Center at Houston, School of Health Information Sciences, Houston, TX, USAReceived 7 October 2004Available online 30 November 2004AbstractNumerous health care systems are designed without consideration of user-centered design guidelines. Consequently, systems arecreated ad hoc, users are dissatisfied and often systems are abandoned. This is not only a waste of human resources, but economicresources as well. In order to salvage such systems, we have combined different methods from the area of computer science, cognitivescience, psychology, and human–computer interaction to formulate a framework for guiding the redesign process. The paper pro-vides a review of the different methods involved in this process and presents a life cycle of our redesign approach. Following thedescription of the methods, we present a case study, which shows a successfully applied example of the use of this framework. Acomparison between the original and redesigned interfaces showed improvements in system usefulness, information quality, andinterface quality.Ó 2004 Elsevier Inc. All rights reserved.Keywords: User-computer interface; Human engineering; Data display; User-centered design; Usability assessment1. IntroductionThe goal in the design of user-centered software is tocreate systems that are modeled after the characteristicsand tasks of the users. Employing cardinal axioms ofgood design early and throughout the design life cyclegives rise to systems that are easy to learn, increase userproductivity and satisfaction, increase user acceptance,decrease user errors, and decrease user training time.In converse, not doing so often requires the redesignof a system. Redesigning interfaces is not only time con-suming, but costly and frustrating for both the users anddesigners.Health care software developers often overlook rele-vant user characteristics, user tasks, user preferences,and usability issues, resulting in systems that decreaseproductivity or simply remain unusable. The US Gen-eral Accounting Offic e, a major supporter of softwareengineering, found that 98% of software designed forthe US government was ‘‘unusable as delivered’’ [1]. Sit-tig and Stead [2] additi onally found this same problemin clinical information systems. In one system reviewedsystem implementation took three years longer than in-tended and the cost was approximately three timesgreater than the original budget [2]. Several factorscould be ascribed to poor systems development suchas cost and time restrictions and/or developer lack ofuser-centered design knowledge. Only 61% of informa-tion system projects meet the requirements of thecustomerÕs specifications [3]. Furthermore, 63% of pro-jects go over their estimated budgets with the top citedreasons related to initial inadequate user analysis [4].Simply put, not enough resources are be ing allocatedto basic design principles, especially in the beginningphase of a project. Fixing a problem in the developmentphase is estimated to cost 10 times more than fixing aproblem in the design phase. Fixing a problem after1532-0464/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved.doi:10.1016/j.jbi.2004.11.005*Corresponding author. Fax: +1 713 563 4242.E-mail address: [email protected] (C.M. Johnson).www.elsevier.com/locate/yjbinJournal of Biomedical Informatics 38 (2005) 75–87shipping a system costs 100 times more than fixing aproblem in the design phase [5]. Incorporating good de-sign principles in the be ginning phase of a project notonly saves time and money, but also decreases designchanges late in the development process [6].In the current information age, health care providersare challenged with an increasing amount of informa-tion, and therefore have a greater need to utilize technol-ogies to efficiently manage such information. Theirability to easily adopt and implemen t these technologiesdepends upon the ease of use of these tools. Inform ationtechnology is changing the way patient information isobtained and gathered and can impact the decision-making processes of clinicians. It is well documentedthat poor information displays can lead to inefficientcare which may include redundant ordering of tests ormissing information important to the diagn osis of thepatient [7–9]. The key is having the right informationin the right place for the right clinician. Clinicians needa concise conceptualization and representation of com-plex clinical data for accurate problem solving and deci-sion making. Therefore, health care applications mustbe carefully crafted, considering the different back-grounds and tasks of the health care providers, to ensurenot only that these programs meet the standards andmodels outlined by the profession, but are intuitiveand easy to use.There are several valid user- centered design method-ologies, however, none address the methods required inthe process of redesign. This paper reviews the methodsrequired in redesigning user-centered interfaces and pre-sents a fram ework for the redesign process. These meth-ods are based on