Design guidelines for a multi-screen user interface: Redesigning a Hospital Information System

(1)

Redesigning a Hospital Information System

Nawishca Clemence Mook

University of Amsterdam

(2)

2

Design guidelines for a multi-screen user interface

Redesigning a Hospital Information System

Student N.C. Mook Student number 10669477 nawizz@gmail.com Mentor Erik Olierook

User Interface Designer Usability team ChipSoft TT Melissaweg 23-25 1033 SP Amsterdam EOlierook@chipsoft.nl Tutor Tom Broens Assistant Professor

Department of Medical Informatics Academic Medical Center

Meibergdreef 15 1105 AZ Amsterdam J1b-114

t.h.broens@amc.uva.nl

Location of the Scientific Research Project

ChipSoft

TT Melissaweg 23-25 1033 SP Amsterdam

Practice teaching period

(3)

3

Preface

This thesis is part of the final phase of the master Medical Informatics. By means of this thesis, I close an important journey of various studies and working experiences. Because of previous experience in IT services, I wanted to pursue this in the medical sector. During this master, I learned the need of medical informatics and became more eager to combine IT and health care, particularly the usability of health care applications. Fortunately, the University of Amsterdam also introduced master students to companies in the medical informatics sector. Because of this, I came into contact with ChipSoft, which develops Hospital Information Systems; they had an interesting assignment which combines IT with health care.

I would like to thank everyone who helped and provided support during this last phase. First, a special thanks to my tutor Tom Broens who guided me throughout the process, his feedback and encouragement helped me a lot. I also want to thank ChipSoft for providing this assignment, and Erik Olierook, my mentor at ChipSoft, for providing feedback and helping me during this research project.

I also would like to thank everyone who participated in my usability tests, as without them I would not have been able to complete this research project. In addition, big thanks to all my fellow students for the great time.

Furthermore, I would like to thank my family and friends for supporting me throughout all years of various studies. A special thanks to my parents for the endless support, feedback, and encouragement to go beyond the expected. In addition, a very special thanks to my boyfriend who gave me daily support and who helped me to bring everything in perspective.

(4)

4

Summary

Health care providers interact daily with a Hospital Information System (HIS). Designing user interfaces that support different health care users in optimizing their workflow is a great challenge in software development for health care environments. Due to technological innovations, more detailed patient information is available in a HIS. Consequently, better user interfaces are needed to organize and arrange this information to provide better insight and overview. A multi-screen user interface offers a possible solution to this challenge. The aim of this thesis is to research which design guidelines can support user interface designers in creating suitable multi-screen user interfaces in the context of HIS.

This research was performed in an iterative approach by incrementally developing design guidelines in a user-centered way. The first step was to develop initial design guidelines through literature review, interviews, and observations. In scientific literature, little is known about designing multi-screen user interfaces in general, and even less in a health care environment. Nevertheless, some studies have been performed on users interacting with multiple displays or with one large display in a non-health care application. Studies have revealed that users often prefer more available space where one can divide their tasks among the displays. Some studies have examined the way in which health care providers review patient information. These studies helped to develop the initial design guidelines. The initial guidelines were tested with users by using paper and interactive prototyping.

Testing the prototypes resulted in nine final design guidelines. These design guidelines provide the basis for the design and prototype implementation of a multi- screen user interface for the HiX system (HIS of ChipSoft). Test users often experienced the novel user interface as pleasant and efficient. Furthermore, more screen space for information and seeing two information sources at once is seen as useful and effective. The interaction flow was evaluated positively by the test users.

This thesis is used to determine what design guidelines are necessary for a multi-screen user interface for HIS.

Keywords

Design Guidelines, Multi-Screen User Interface, Hospital Information Systems, Paper Prototyping, Interactive Prototyping, Usability

(8)

8

Samenvatting

Zorgverleners raadplegen dagelijks patiëntgegevens door middel van een zorginformatiesysteem (ZIS). Het ontwerpen van een gebruikersinterface die door de hele keten gebruikt kan worden vormt een uitdaging in de ontwikkeling van software voor de gezondheidszorg. Door de nieuwe technologie en de beschikbaarheid van steeds meer complexe informatie ontstaat de behoefte aan een gebruikersinterface die meer gedetailleerde, overzichtelijke informatie toont. Een multi-screen-gebruikersinterface kan een mogelijke oplossing voor deze behoefte zijn. Het doel van deze thesis is om te onderzoeken welke ontwerprichtlijnen een user interface designer kunnen ondersteunen bij het ontwerpen van een multi-screen-gebruikersinterface in het kader van een ZIS.

Dit onderzoek is uitgevoerd door middel van een iteratieve, gebruikersgerichte aanpak van stapsgewijs ontwikkelen van ontwerprichtlijnen. De eerste stap is het ontwikkelen van ontwerprichtlijnen door middel van literatuuronderzoek, interviews en observaties. In de wetenschappelijke literatuur is er niet veel bekend over het gebruik van een multi-screen-gebruikersinterface in het algemeen, en nog minder in de medische sector. Echter, sommige studies hebben onderzocht hoe gebruikers de interactie met meerdere beeldschermen of met een groot scherm in een niet-medische omgeving ervaren. Uit deze studies blijkt dat men meer werkruimte prefereert door middel van grotere of meerdere beeldschermen. Hierdoor kan men taken over twee beeldschermen verdelen. Enkele studies hebben onderzocht welke patronen zorgverleners toepassen bij het beoordelen van patiëntgegevens. Deze studies hebben geholpen met het ontwikkelen van de initiële ontwerprichtlijnen. De initiële ontwerprichtlijnen zijn door middel van

paper en interactive prototyping getest.

Het testen heeft geresulteerd in negen definitieve ontwerprichtlijnen die voor een gebruikersinterface nodig zijn. Deze ontwerprichtlijnen fungeren als basis voor het ontwerp en als prototype voor een multi-screen-gebruikersinterface van HiX (een zorginformatiesysteem van ChipSoft). De test-gebruikers ervaren de prototype als prettig en efficiënt. Meer schermruimte voor meer informatie en het zien van twee informatiebronnen tegelijk wordt gezien als nuttig en effectief. De interactie stroom wordt door test-gebruikers positief beoordeeld.

Deze thesis is gericht op het ontwerpen van ontwerprichtlijnen voor een multi-screen-gebruikersinterface voor een ZIS.

(9)

9

1. Introduction

Health care providers interact daily with a Hospital Information System (HIS). The system’s main function is to store, process, visualize, and share patient information. Patient information is used for a variety of care processes in health care environments. For example, health care providers use a HIS during medical consultations to read patients’ history, to write diagnoses, and to view X-rays. Other stakeholders might also use it for secondary processes such as finance or planning. Due to continuing technological progress, more and increasingly detailed information is becoming available about the patient, which is collected in a HIS and processed in complex data driven workflow by health care providers. There is a general need to organize and arrange information for better insight and overview of patient information. To tackle this problem, user interfaces are one of the most evident targets to realize the required innovation (1). In this thesis, the use of multi-screen user interfaces is examined to achieve this goal. Multi-screen user interfaces offer a possible solution to provide better insight and overview of patient information by combining multiple screens and thus increasing the available workspace. This allows the users to use the primary display for the main task while a secondary display can be used for additional activities or easy access to resources (2). Nowadays, more and more people are using multiple displays in their workplace, as they are relatively inexpensive. Nevertheless, the concept of multi-screen user interfaces is not commonly implemented in software products (3). As a result, multi-screen user interfaces lack support of software manufactures.

Up to now, little is known about applying multi-screen user interfaces for HIS. Therefore, the focus of this research is to apply the concepts of multi-screen user interfaces on HIS, to evaluate its usefulness, and to offer designers good-practice guidelines for developing HIS user interfaces. For this, we use concepts from literature as proposed in the domains of human-computer interaction and usability to develop design guidelines.

The main focus of human-computer interaction is to understand the process involving the creation of more effective and useful interactive computer systems (1). Furthermore, when designing user interfaces, usability is a key factor. Throughout this paper, the following definition of usability as described by ISO 13407 is used: “the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use” (4).

(10)

10 When developing a HIS, it is desirable to display detailed data in a well-organized way. Thus, patient data should be presented accurately and in the best possible way to assist health care providers in their care process. For example, a health care provider needs to enter data in a patient’s record and needs to examine an X-ray for that particular input data. For this reason, multi-screen user interface could contribute to these desires. As the user workflow is vital we need the end user to be involved in the design process. Hence, we apply user-centered design (UCD) as a framework of processes where user needs are the center of attention. Furthermore, through UCD, users are involved in the development processes of a product.

1.1 Problem Description: Multi-screen User Interface Design for a HIS

Health care providers are seeking an effective and efficient manner to deliver care. Patient data can be overwhelming, complex, and originating from different data sources. In addition, more information is becoming available; thus, it is important to arrange information in such a way health care providers can fully benefit from this information. Displaying better organized data can aid health care providers with processing and handling patient data.

A multi-screen user interface can assist a health care provider in his or her work. For example, a physician needs to review an X-ray while filling data in a patient record. The physician has only one computer screen available, but he needs to view an X-ray full screen for certain details. Consequently, he needs to minimize the X-ray when he needs to fill in data. In contrast, data can be edited with a multi-screen user interface while existing data is still available in full screen.

Furthermore, patient data need to be handled with care. In addition, due to the wide range of available information confusion and disorientation should be avoided (5)(6). Therefore, it is important to discern which information is needed and where it should be displayed.

Thus, a multi-screen user interface should be designed in accordance to certain guidelines. The purpose of this thesis is to establish design guideline what can be applied by user interface designers during the development process. In addition, these design guidelines can be used by developers as reference work.

(11)

11

1.2 Goals: UCD Guidelines for a Multi-screen User Interface for a HIS

The purpose of this study is to examine and test the design aspects required for a multi-screen user interface, especially for a HIS. This results in the following research question:

1. What design guidelines can steer to the design of a multi-screen user interface for HIS?

To answer the main research question the following sub questions are addressed:

1.1 What scientific literature is known about multi-screen design and related usability, and can this be applied to designing a multi-screen user interface for a HIS?

1.2 What are the needs of end users regarding a multi-screen user interface design?

In addition, a usability study is conducted to test the design guidelines. The following research questions are examined during this usability study:

2. Which design guidelines need to be adhered to when developing a multi-screen user interface?

3. How do users evaluate interfaces that adhere to the design guidelines?

By using the results from the above mentioned questions, we can answer the following questions concerning an existing user interface:

4. To what extent can HiX (HIS of ChipSoft) be adapted to a multi-screen user interface design?

1.3 Chapter Organization

The second chapter explains the methods used during the conduct of this study. This chapter contains the explanation of the overall study design, the collection of used methods, and the usability test methods. Chapter 3 discusses the progress towards the initial design guidelines. Furthermore, a state of the art is discussed, as well as the initial creation of the design guidelines. Chapter 4 elucidates the progress towards the final design guidelines. In this section, the evaluation of the usability tests is described leading to the final design guidelines. Chapter 5 and 6 handles the findings and discussions of this study, which are discussed in terms of limitations, strength, and weaknesses. Finally, a conclusion of this study is given.

(12)

12

2. Methods

2.1 Overall Study Design

For this research, an exploratory study that adopted an iterative approach was used. This approach allowed the design to adjust in the progress and to build the design to fit the user’s needs and expectations (7). The study was conducted in three phases as demonstrated in Figure 1. The analysis phase consisted of the literature review and conducting interviews. This is used to establish the initial guidelines. The Requirements Design Test & Review phase consisted of designing a multi-screen user interface with the use of

the initial guidelines. Consequently, usability testing with end users and evaluate the feedback to redesign the multi-screen user interface. The Final Design phase of the study involved applying the design guidelines for a final multi-screen user interface design in the form of a recommendation. The methods used during the different phases are listed in Table 1. These will be further explained in the next sections.

Phases Methods

Analysis Literature review

Observations Affinity Diagramming Usability Problem Taxonomy Requirements Design Test & Review – Iterative Observations

Hierarchical Task Analysis Paper Prototyping Interactive Prototyping Think Aloud

Table 1: Methods Used Per Phase

(13)

13

2.2 Collection of Used Methods

2.2.1 Interviews

To obtain information on how end users think and feel about the system interviews were conducted. These interviews were carried out among software consultants because of the limited availability of health care providers. Software consultants are mediators between users and developers; this means that these consultants have contact with end users and are often in collaboration with developers to reach a solution to fulfill the desires of the end user. Moreover, the software consultants were appointed by ChipSoft.

The interview questions were based on the ten heuristics of user interface design by Nielsen (8). These questions were used to identify the needs and requirements concerning multiple display use in a health care environment.

The interviews were unstructured with open questions. The questions were used to guide the conversation in a flexible manner. As such, the respondent and interviewer felt free to answer and question in their own manner, which resulted in a comfortable setting to gain insight into the interaction of the end users. In Appendix A - Interview Questions the interview questions can be found.

2.2.2 Observations

To gather information on how users interact with a HIS, end users were observed. To fully benefit from these observations, health care providers, who are in the process of a consult or processing patient data, were monitored in their own work environment. This was done to examine how users interact and perform their tasks under work pressure. This offers a more realistic in-depth representation of how health care providers interact with a HIS.

2.2.3 Affinity Diagramming

Affinity diagramming is a method to group large amounts of related information. In this research, the groups were not pre-defined, but arose in an interactive session. These sessions were carried out by the researcher and the user interface designers, and parts of relevant information were captured with notes. Each note was examined and grouped under a certain theme (6) (9) (10). At the end of the session, the literature was grouped under a number of themes which was used to establish the initial design guidelines. These sessions took place twice.

(14)

14 2.2.4 Usability Problem Taxonomy (UPT)

UPT is a classification scheme used to classify usability problems and findings encountered during usability studies. The structure of UPT is illustrated in Figure 2. As can be observed, the structure contains two components, which are artifact and task. The artifact component focuses on difficulties when the user interacts with the user interface. The task component concentrates on problems as the users move through a task. These components are divided into

several categories (11). In this study, UPT was used to categorize and classify the initial design guidelines. Moreover, UPT was used to validate the initial design guidelines upon a widely used method in usability engineering. All the initial design guidelines were coded in the following manner: each category in the UPT has red codes, as marked in Figure 2. For instance, when a user has a

problem with the location of a certain button, this should be coded as follows: A (Artifact component), V (Visualness), and OA (Object Appearance), which becomes A.V.OA.

2.2.5 Hierarchical Task Analysis (HTA)

HTA was used to guide test users through the multi-screen user interface. The tasks were created to solve a realistic problem which medical specialists experience daily. In addition, the tasks were also designed in a manner that each medical specialist can relate to and perform these tasks. At the beginning of this study, it was not clear which kind of specialist would participate in the usability test. Therefore, it was decided to adapt the HTA approach where tasks are explained by means of steps and subtasks (5).

(15)

15

2.3 Usability Test Methods

Usability test methods are used to determine which design guidelines are needed for a multi-screen user interface. Consequently, product and software development often uses iterative processes. Using iterative processes in a design study allows incorporating users at an early stage. In fact, UCD ensures that the end product fits the user’s needs. Moreover, designing a user interface or an application, calls for understanding user interaction. Therefore, interaction design is an iterative process (5). In this research, each session involved progressing feedback and adjusting the design guidelines in a more in-depth manner. These small iterations allowed the multi-screen user interface design to be refined based on feedback (5). This involved starting with a low-fidelity method to a high-fidelity method where user involvement, testing, and validation were the center of attention (5). Furthermore, user involvement ensured that the conceptual multi-screen user interface is developed in the way it is intended and supports users tasks (5). This method is therefore particular useful to validate the design guidelines (6).

2.3.1 Prototyping

To test the developing design guidelines against real users, prototyping was used. Prototyping is an imperative technique in product development and is widely used in software development; this technique is fundamental for the design phase. First, sketches and prototypes are developed in an iterative user involvement phase. Each iteration involves testing the product with the end user in an early phase of the development will provide a validation about the end product (12).

Paper prototyping is a low-fidelity technique where users are shown a paper version of the product and are asked questions about it (7). In this study, paper prototyping was involved in the usability testing where the initial guidelines were translated into mock-ups. Mock-ups are usually a full-sized scale model of a product which is not yet fully functional. This is a low-fidelity testing method because mock-ups do not take a long time to build and are easy to adjust (5). Therefore, it was decided to implement this approach in this study.

In contrast, interactive prototyping is a high-fidelity technique where the version of the product is better working and functional (6). The mock-ups were interactive and some behaviors were attached (6). End users were able to interact with the product as a real application. This method was used to validate the applied design guidelines.

(16)

16 2.3.2 Think Aloud

Think aloud is a form of observation used during the paper and interactive prototyping (6). The test users were asked to verbally express their thoughts when executing the tasks. The think-aloud method reveals information on cognitive and intuitive behaviors. Furthermore, it provides useful insight into user interface problems and how users use the interface. Additionally, the simplicity of this method is an advantage, as everybody can think aloud (6). Think aloud gives insight into the user expectations of the system. This method is well known in usability engineering (7).

2.4 Usability Test

The usability tests are part of the iterative process. According to Nielsen (13), at least two iterations are needed to improve usability. In both iterations, five test users participated in usability tests. Nielsen (14) states that conducting a usability test among no more than five users results in the best outcome. In this study two usability tests were conducted with each five test users.

The first usability test consisted of paper prototyping where mock-ups formed the paper version of the multi-screen user interface. Prior to the usability test, task scenarios were created. These task scenarios are based upon certain workflows within a HIS. The usability tests were recorded with the aid of a mobile phone. The recordings consisted of audio and visual material. This material was used to extract usability findings and problems. After the test, participants were asked to complete questionnaires which were also used to extract usability problems and findings. The test environments were in a training room and at the tester’s location.

Prior to the second usability test, the design guidelines were adjusted with the feedback of the first usability test. This resulted into a new set of mock-ups and task scenarios. For the second usability test, an interactive prototype was developed. The interactive prototype required a computer system with two monitors with a resolution of 1920 x 1080 each. Again, these tests took place at ChipSoft or at the location of the test user and five test users participated in the second test. The usability tests were recorded using usability testing software called Morea. Morea is comprehensive software which captures the audio, video, on-screen activity, keyboard, and mouse input. For the purpose, of this study only the on-screen activity, audio, keyboard, and mouse input were captured. By using Morea annotations and marks were created with the associated recorded test;

(17)

17 this provides more in-depth feedback on how users interacted with an interactive prototype. Furthermore, the software enabled the extraction of usability problems and findings. After the test, the participants were asked again to fill in questionnaires.

2.5 Questionnaires

As previously mentioned, before and after the usability tests, participants were asked to complete questionnaires. Before each test, the participants filled in an informed consent and a characteristic form, which can be found in Appendix B - Informed Consent and Appendix C - Characteristics Test Users Form. The informed consent was used to ask permission for recording the usability test sessions. The data from the characteristic form was used to make a summary of the participants’ characteristics.

After each test, the participants were required to fill in two usability questionnaires, which were called: Usefulness, Satisfaction and Ease of Use (USE) and Perceived Usefulness and Ease of Use (PUEUE) see Appendix D – USE Questionnaire: usefulness, satisfaction and ease of use and Appendix E – PUEUE perceived usefulness and ease of use. These offer more insight into how users felt and perceived the multi-screen user interface.

(18)

18

3. Towards Initial Design Guidelines

In this chapter the development process of the initial design guidelines is described. This chapter provides an overview of the scientific literature regarding a multi-screen user interface. Finally, a summary of the creation of the initial design guidelines.

3.1 Literature Review

In order to discover what is known about multi-screen user interfaces with regard to a HIS, an extensive search was performed through several databases. The main topic was a multi-screen user interface for HIS with a combination of multiple keywords. The included keywords are listed in Table 2 and in Appendix F – Search Strategy Literature Review shows the full search strategy. Furthermore, several searches were conducted with a combination of the keywords. Considering the comprehensive subject the decision was made to explore the following databases: ACM, IEEE Xplore, PubMed, Lista and Google Scholar. Moreover, articles were selected by reviewing the title and abstract. After critical appraisal, a set of 13 articles was used during this study. The full list of article used in this literature review can be found in Appendix G - Full List of Reviewed Articles.

Table 2: Search Queries Keywords

3.2 State of the Art

3.2.1 Multi-screen User Interface

The multi-screen user interface is a user interface which can be used on dual monitors. The term multi-screen sometimes refers to the same interface available on multiple devices such as laptops, mobile phones, and tablets (15). In this context, we use the following definition: a user interface which can be used on multiple screens which are controlled by one application, as seen in Figure 3.

Before elaborating further on multi-screen user interfaces, we first discuss the user interface itself. A user interface is the part where the user interacts with a software

Search Queries Synonyms

Multi-screen Multi-monitor Multi-display

Multi-display user interface (MDUIs) Health information systems Clinical information system

Usability

Human-computer interaction Interaction design

(19)

19 application. Generally, the user interface exist of diverse design elements by which a user can interact and can control a software application.

A multi-screen user interface is a type of user interface developed specifically for two or more monitors which extend the visible space for interaction. There is little knowledge about multi-screen user interfaces; in fact, during this research, it became clear that there are only a few software applications that take a second screen into consideration.

Figure 3: A Multi-Screen User Interface 3.2.2 Usability

Nowadays, it is important to consider the ease of use when designing a user interface. Often designers try to create a self-explanatory user interface in which the learning curve is low. In other words, usability is imperative in the design of a user interface. The definition of usability as described by ISO 13407 is “the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use”. The usability of a product can be measured in various ways, such as user performance and satisfaction. It can also be measured in terms of effectiveness and efficiency. Here, effectiveness means the extent in which the intended goal is achieved and efficiency concerns the resource used to achieve the intended goal. In addition, satisfaction describes the extent to which the users find the use of the product acceptable. Therefore, usability is a vital factor in designing a user interface.

3.2.3 Pattern of Reviewing Information

Patterns of how medical specialists review information and what effect design aspects have on usability have been examined in three studies (16) (17) (18). Thyvalikakath et al.

(20)

20 (16) have established three patterns of how medical specialist review information. From these patterns, two patterns are of interest, namely rapidly switching between information sources and reviewing multiple information sources simultaneously. Rapidly switching is mostly used when additional information is required, while reviewing information simultaneously was often done to confirm findings or to conclude diagnosis. Zhang et al. (17) have demonstrated how information should be structured upon workflow to decrease cognitive load in medical environments, which are called the five Ws. The five Ws stand for: who, what, where, and who/what. According to Zhang et al. (17), the structure follows the medical model that has already been used in consults of a medical specialist. Moreover, Khajouei and Jaspers have suggested that the structure should present the user’s workflow. Therefore, information should be physically grouped together based on the medical workflow (18).

3.2.4 Visualization of Information

As discussed before, Zhang et al. (17) have described the five Ws that should be used in a medical profession. This section addresses visualization techniques which are employed for large or dual monitor users.

Using a second screen for any objective creates a large amount of space which can lead to usability problems such as information overload and change blindness. Change blindness is the inability to visually detect when there is a disruption in continuity such as a shift of focus (19) (20). In literature, there are many studies on visualization techniques that can be applied in multiple screens environments.

Because of the lack of knowledge about multi-screen user interfaces, the available information concerning the usage of multiple screens in a normal setting is used in this study. Many studies have investigated the impact of large or dual monitor use in a traditional computer environment. Zhang et al. (21) have analyzed the necessity of large amounts of information in a multi-screen display set-up. The methods proposed in this study can improve the human-computer interaction and reduce cognitive load when the following two rules are applied:

1. “The object of two screens always keep consistent, while each screen displays different information”

2. “The information of two screens always keeps consistent, while each of which displays different objects” (21).

(21)

21 Many of the studies have discussed clustering and grouping information together. Clustering information in distinct groups will help medical specialist to perceive and interpret more easily. Therefore, related information should be grouped together.

Multi-screen user interfaces shows more information at once. Nonetheless, human vision can only focus on a limited area (2) (19) (22). Thus, spreading different information across multiple displays can affect visual attention that involves cognitive focus, gaze, and head or body displacement. Taking this into consideration, some significant statements were made during the interviews. According to one interviewee: “a second monitor would only cause confusion because users need to look from left to right, because there is more on-screen information to see”. Moreover, another interviewee said: “sometimes users do not know where to look”. Note that these statements refer to the current HIS with one monitor which causes the assumption that multiple screens will only result in confusion. Moreover, Rashid, Nacenta and Quigley (2) have also discussed that the effects of visual attention depend on the task as well as the design while Zhang et al. (21) have explored techniques where monitors are divided into focal and peripheral regions. Participants in this study referred to their displays as focal where their main work took place and peripheral when resource information is used as a point of reference. Bi and Balakrishnan (23) have investigated user’s behaviors when they switch from a standard computing environment (single-or dual monitor) to use large, high-resolution displays, which also referred to as focal and peripheral regions in this study. Furthermore, they have suggested that it should be possible for users to quickly switch applications from the peripheral region into the focal region and vice versa. To sum up, it is essential to present information in a way that helps users with their goal by using a multi-screen user interface instead of overwhelming them with information.

3.2.5 Grouping

An important factor of a multi-screen user interface is the position of information. Therefore, it is an important factor to take into consideration when designing a multi-screen user interface. When using an additional monitor users tend to separate content belonging to different tasks (2). Furthermore, an increase in amount of data by means of more pixels resulted in an increase in task performance and no significant decrease in accuracy (24). Bi and Balakrishnan (23) have investigated how users perform on large displays; the users were familiar with single or dual monitor use. In addition, 81% of the

(22)

22 mouse events took place in the middle of the large display; this region was appointed as focal since the main tasks occurred here. The remaining space of the display was used as peripheral where secondary tasks were performed or as information source. In this region the other 19% of mouse events occurred.

Interviewees stated that in the current HIS, when they open multiple screens within their available workspace (one monitor), screens are displayed as layers on top of each other; as a result, users lose track of their screens. This will bring confusion and unnecessary search for information (18).

In addition, splitting objects across screens has been discussed in several studies. Extending the visual display to a second screen is only useful when users gain more insight into their information (23). However, when not used properly, splitting can lead to an increase of completion time and information overload (2) (19). Moreover, not allowing users to have a simple overview of their information leads to excess scrolling or forcing them to rely on memory results in usability problems (18). Therefore, the information displayed in a multi-screen user interface should present the same information in a different form.

Moreover, overview is important in a user interface. Not only having an organized overview is imperative but also using the available space (22). Cluttered designs and scattered information in the user interface make it difficult for medical specialist to have a quick overview of their patients’ status (2) (18) (21) (22). Therefore, it is essential to adhere to certain design guidelines for displaying information. Displaying more information containing diverse data items simultaneously can reveal the interrelationships between them (22). As two interviewees stated, there is a need to display more information, but, on the other hand, one of the interviewees stated that users find that too much information is already being displayed (in the current HIS). Another interviewee said that a second screen will certainly be useful for the readability and retaining the overview. To use the multiple monitors efficiently, it is imperative to use an additional monitor to separate the content belonging to the task (2).

A multi-screen user interface provides more available space where excess navigation structures can be avoided. According to two interviewees, the information is occasionally hard to find (in the current HIS), as one needs to click more than four times to find particular information or to perform a certain action. Khajouei and Jaspers (18) have also suggested that deep navigational structures should be avoided especially when medical

(23)

23 specialist need to observe relationships between different information sources. This is when a second monitor and a multi-screen user interface becomes handy. Both interviewees stated that a second monitor can provide useful action and insights into information, certainly if users can divide their own region with respect towards a multi-screen user interface. Users can, moreover, divide monitors into focal and peripheral regions, where they can decide which information needs to be displayed. In addition, Rashid, Nacenta and Quigley (2) have also suggested that there should be design guidelines for multiple views of different information source.

3.2.6 Visual Representation

In two studies, design aspects have been examined on how they influence the decision- making of health care providers (18) (25). Both studies have claimed that visual representation of information improves the ability of time and accuracy. Using knowledge-enhanced graphical symbols (KEGS) compared to traditional information may require less time for interpretation by a medical specialist. Furthermore, Khajouei and Jaspers (18) have argued that using visual presentation as visual cues helps users with the interpretation of the different types of information. They have also stated that the consistency in clear use of visual presentation needs to be taken into consideration. Zhang et al. (17) have suggested using radial display when using body outline in a patient overview. This allows medical specialist to quickly obtain an overview and focus on the patient’s history. Furthermore, extraneous details, such as repeated menus and buttons, should be avoided to aid users to recognize relevant information (18).

3.3 Initial Design Guidelines

In this section, the development process of the initial design guidelines is described. The initial design guidelines were formulated by using affinity diagramming as presented in Appendix H - Affinity Diagramming Sessions and

Appendix I – Initial Design Guidelines. Affinity diagramming was used to group the information from the literature review and interviews together. Furthermore, combining the relevant and related parts of the articles and interviews provided a basis for the initial design guidelines. Moreover, grouping and merging this information has led to nine groups from which design guidelines were established. Furthermore, the initial design guidelines were labeled generic or multi-screen user interface specific as presented in

(24)

24 Table 3. In this section, a summary of the initial design guidelines is provided. The in-depth creation of the initial design guidelines can be found in

Appendix I – Initial Design Guidelines. During the literature review and interviews, information emerged which does not specifically concern a multi-screen user interface; nevertheless, it is relevant to incorporate this in the user interface design. Below, a summary is given in line with the design guidelines numbering in Table 3.

1. Relationships and linking involves the correlation between different information sources especially relevant for a multi-screen user interface.

2. Interaction concerns the way in which a user communicates with a user interface and suggests how information should be displayed.

3. Appearance relates to the use of workspace concerning the use of icons, symbols, and menus.

4. Cognitive, task, performance load, and routine describes how certain visual presentation, such as KEGS, and splitting of information on a multi-screen user interface affect the cognitive and performance load in clinical routine.

5. Visual representation concerns certain forms of visualization such as KEGS, radial displays, or color to code information.

6. Visual attention illustrates how users divide their attention among multiple screens. Furthermore, it explains how division of information depends on their executing task and how it affects their visual attention by taking into consideration that human vision has limited focus on a multi-screen user interface.

7. Navigation concerns the manner in which users navigate through the user interface. Moreover, it shows how medical specialist review different information sources when working on patient information and how these findings may help designing a multi-screen user interface. It also suggests that current users of HiX sometimes need to perform deep navigation to find their goal and this could be avoided by using a multi-screen user interface.

8. Workflow and switching tasks indicates where certain tasks were performed (on the screen) and also how it reflects the user’s workflow. Furthermore, literature has revealed that users mentally divided their screens into focal and peripheral regions. Therefore, this is taken into account in the set of design guidelines.

(25)

25 9. Readability prescribes that information should always be legible and interpretable for users. In addition, some functions, such as magnifying pictures upon opening can also provide readability.

10. Displaying information is divided into the following four subheadings: prediction, grouping, position, and overview.

10.1. Prediction indicates where and what type information is needed.

10.2. Grouping deals with how the different yet related categories of information should be grouped to find and interpret the relationships better.

10.3. Position literally concerns the position where the information is displayed and how to make use of the available information specifically for a multi-screen user interface.

10.4. Overview relates to having an overview of the given information. Furthermore, concerning patient information, having an overview at a single glance without being overwhelmed by cluttered screens. Thus, this allows to detect relationships between complex patient information.

1. Relationships/ Linking Generic or Specific

1.1 Keep the information of the two displays always consistent; thus, each display contains the same information presented in a different form. Update information simultaneously, when using information across a multi-display user interface

Specific

1.2 Be consistent in displaying information, as humans recognize information better than recall Generic 1.3 Some buttons are not self-explanatory such as move to the second screen Generic

2. Interaction

2.1 To facilitate peripheral regions, it should be more simplified by avoiding excess distractions such as repeated menus and icons

Specific 2.2 Simple tools should provide easy access to what medical providers use to give more insight

into the information such as focus, zoom, rotation, and translation

Generic 2.3 Use radial layout to visually display hierarchical structures. In addition, use elongated boxes

in the radial layout to show additional information

Generic

3. Appearance

3.1 Using multiple displays provide a large amount of space. To efficiently use the available space, only display important and relevant information

Specific 3.2 Do not fill space with unnecessary information, icons, or symbols Generic 3.3 Display more heterogeneous information when available to help uncover relationships

between complex data

Generic

4. Cognitive / task / performance load /routine

4.1 When possible use visual presentations instead of text, such as KEGS, to decrease cognitive load in clinical routine

Generic

(26)

26

5.1 Use colors to indicate the type of information and severity of the information. For example, full red for highest severity, yellow for warnings, and white for low severity

Generic 5.2 Use hidden text labels when one wants to see more information Generic

6. Visual attention

6.1 Human vision can only focus on a limited area; therefore, the presented information should depend on the task

Specific 6.2 Provide switching for user interface elements so that users can arrange their displays to their

own preference

Specific

7. Navigation

7.1 Medical specialists review at least two information sources at once. Thus, it should be made possible to view more information and to adjust the reviewing information. In addition, make it easy to arrange the multi-screen user interface to his or her own preference and perform rapid switching between different information sources

Specific

7.2 Navigational structure deeper than four clicks should be prevented; therefore, displaying more information at once should highlight different aspects of the same information with the possible actions

Generic

8. Workflow/switching task

8.1 It is not possible to use multiple displays in every situation; this depends on the physical space or task a user is performing. Hence, it should always be a choice to work on multiple displays and it should be easy to switch between different states

Specific

8.2 Divide the available space on content depending on the task. Avoid excess switching between content

Generic 8.3 The content should be presented in a way that it maps the user’s workflow Specific

9. Readability

9.1 To ensure readability, use symbols and if not possible, use simplified text Generic 9.2 To enhance readability, show where needed pictures, graphs, etc. full screen Generic

10. Displaying information

10.1 Overview

10.1.1 Data elements should be updated simultaneously on multi-screen user interface Specific 10.1.2 To avoid excess scrolling and relying on memory, make use of available space for (quick)

overview of the relationship between data entities

Specific 10.1.3 Each part should display the same information presented in another form. Do not scatter tasks

or workflows around the multiple displays

Specific

10.2 Position

10.2.1 Try to predict where users need to see information and separate the different content depending on tasks. For example, showing X-rays on the right screen in full screen

Specific 10.2.2 Make use of the available space (pixels) to present information Specific 10.2.3 Standardize the multiple display user interface in the focal and peripheral region, but also

provide the option for users to divide their own regions

Specific 10.2.4 At any time, let users know where they are in the system and on which patient, they are

working

Generic

(27)

27

10.3.1 Predict locations where information should be presented. For example, a physician opens a patient record on the left screen. In addition, an X-ray needs to be examined, when opening the X-ray, it is automatically presented on the additional screen

Specific

10.4 Grouping

10.4.1 A multi-screen user interface has the ability to present many categories of information and to keep it simple, information should be clustered or grouped. Each aspect should contain the same information presented in another form

Specific

10.4.2 Presenting information on a multi-display user interface calls for information to be clustered and grouped depending on task-specific information which helps users to perceive, find and interpret information easily

Generic

(28)

28

4. Towards Final Design Guidelines

This chapter describes the evaluation of the initial design guidelines. The main objective is to test the initial design guidelines with end users and to adjust the guidelines accordingly. Subsequently, the final design guidelines are presented.

4.1 Usability Test and Test Users

In total, ten usability tests were conducted to assess the initial design guidelines. Five tests were conducted using the paper prototype, while the other five tests employed the interactive prototype. In Table 4, the characteristics of the test users participated in both the usability tests can be observed. When applying the interactive prototype, three medical specialists were used and two non-medical specialists due to time constraint and the availability of medical specialists.

Paper Prototype Medical specialists n = 5 Interactive Prototype Medical specialists n = 3 Other Fields n = 2 Gender Gender Male 4 Male 4 Female 1 Female 1 Age Age 40-50 1 40-50 1 50-60 3 50-60 3 60-70 1 60-70 1

Experience with HIS? Yes = 5 Experience with HIS? Yes = 5 Experience with two displays? Yes = 4 No = 1 Experience with two displays? Yes = 4 No = 1

Table 4: Characteristics Paper Prototype and Interactive Prototype

4.2 Mock-ups and Tasks

In order to conduct usability tests, tasks and mock-ups were created. For the paper prototyping three sets of tasks were created, adding an allergy to the patient record,

opening media content and adding an ECG procedure and opening media content and adding a procedure. The mock-ups were created during three sessions in collaboration

with a user interface designer. To clarify, throughout this section the mock-ups are referred to as main and second screen as seen in Figure 4. The in-depth creation of these mock-ups can be found in Appendix J – Mock-ups Creation Paper Prototypeand the full description of the tasks is presented in Appendix K – Tasks Paper Prototyping. Moreover, the design guidelines were presented previously in Chapter 3.3 Initial Design Guidelines. The interactive prototype was developed based on the received feedback and refined design guidelines. In this case, three sets of tasks were also created, adding an allergy and

(29)

29

referral to the patient record, opening media content and adding procedure (ECG) at the same time, and comparing media content and updating patient's history. Subsequently, the

mock-ups were adjusted and changed with the received feedback. These mock-ups were used to create the interactive prototype. The interactive prototype is presented in Appendix M - Mock-ups Creation Interactive Prototype. The related tasks are described in Appendix N – Tasks Interactive Prototype.

Figure 4: Example Main and Second Screen

4.3 Paper Prototyping

The design guidelines were tested by means of paper prototyping. Usability findings encountered during the paper prototyping are presented per task seen in Table 5 through Table 7. In each table, the usability findings are categorized by usability problems, usability findings, and comments after the task finished along with the associated design guidelines. The associated design guidelines were assessed for positive or negative use and how many times they were encountered. The completion time and percentages divided by each task is demonstrated in Table 8. In Appendix L – In-Depth Paper Prototype Usability Test, the in-depth evaluation of the paper prototyping per tasks is described and each usability finding is described in detail. In Figure 5 through Figure 7 examples of the multi-screen user interface during this usability test are presented.

(30)

30

Figure 5: Task adding an allergy to the patient’s record

Usability

Findings/Problems Description Design Guidelines *

Interpreted Number Of Encounters Positive Negative

Usability Problems “Akkoord (Agree)” is better than “Opslaan (Save”)

1.3 X 1

Not clear what entry data needed to be filled in

5.2 X 1

Not clear that the patient record was already opened

10.2.4 X 1

Not clear where patient’s name is listed

10.2.4 X 1

Usability Findings Expects a pull-down menu where allergies can be chosen

- X 1

Testers expect to fill in the data on the second screen

1.2, 3.1, 6.1, 6.2, 10.1.2, 10.2.1, 10.2.2, 10.2.3, 10.2.4, 10.3.1, 10.4.1, 10.4.2 X 2 Checking if it is

the same patient 3.1, 6.1, 6.2, 10.2.4 X 2 After saving

attention shifts to the main screen

1.2, 3.1, 6.1, 6.2, 10.1.1, 10.2.1, 10.2.3, 10.2.4, 10.3.1, 10.4.1, 10.4.2 X 3 Present more information 3.3 X 1 Clearly main screen overview second screen details 7.1, 8.2, 10.2.3, 10.4.1, 10.4.2 X 1

Comments – After Scenario Working on two screens is pleasant: having an overview on the main screen and details on the second screen 1.1, 1.2, 2.1, 3.1, 6.1, 7.1, 8.1, 10.1.1, 10.1.2, 10.1.3, 10.2.1, 10.2.2, 10.2.3, 10.3.1, 10.4.1, 10.4.2 X 5

(31)

31

Table 5: Usability findings task adding an allergy to the patient’s record * Based on the initial design guidelines presented in Appendix I

Figure 6: Task Opening a media content and adding ECG procedure

Usability

Findings/Problems Description Design Guidelines * Positive Negative Interpreted Number Of Encounters

Usability Problems Multimedia radiology was expected to open media content 10.4.2 X 1 Grouping multimedia not logical as the media content belongs to radiology 10.4.2 X 1

Not clear which button is needed to rotate the screen

1.3 X 3

Not clear that the procedure is already saved

1.3 8.3 10.2.4 X 1

Usability Findings Showing media content: assume the most recent picture is shown first

- 1

Pleasant having an overview of the patient data and full screen media content presented on the second screen 3.1 10.1.1 10.1.2 10.2.4 10.3.1 10.4.1 10.4.2 X 2 Comments – After Scenario Second screen remains open, switching of the screen was not necessary: if there is space, small things can be done on the main screen 8.3 X 1 It is useful that the media content and overview of the 3.1 10.1.1 10.1.2 10.2.4 10.3.1 10.4.1 10.4.2 X 4

(32)

32

patients remains visible

Saving was

unclear 8.3 X 1

Table 6: Usability findings task opening media content and adding an ECG procedure * Based on the initial design guidelines presented in Appendix I

Figure 7: Opening media content and adding a procedure

Usability

Findings/Problems Description Design Guidelines * Positive Negative Interpreted Number Of Encounters

Usability Problems There is no radiology category, thus I would choose cardiology

10.4.2 X 1

Unclear that the checkmark means saving instead of saved 1.3, 4.1, 5.2, 9.1 X 2 Approve again if what entered is correct which is not insignificant - X 1

Usability Findings Immediately clear which icon should be used to rotate the screen

1.3 X 5 More convenient to display multimedia directly next to the media content 8.2, 8.3 X 1

“Then I look back at my left screen” From right (media content) to the left to enter a treatment limitation 8.2, 10.2.3, 10.4.1, 10.4.2 X 1

Comments – After Scenario “I am really focusing on the main screen and not switching between two screens”

- X 1

Useful that there

(33)

33

screen and a data

entry screen. 10.1.2, 10.1.3, 8.1, 10.1.1, 10.2.1, 10.2.2, 10.2.3, 10.3.1, 10.4.1, 10.4.2

Table 7: Usability findings task opening media content and adding a procedure * Based on the initial design guidelines presented in Appendix I

Task Completion % Mean Completion Time

Adding and allergy to the patient record

100% 3:19 min

Opening a media content and adding an ECG procedure

100% 4:14 min

Opening a media content and adding a procedure

100% 3:04 min

Table 8: Mean completion time and completion percentage per task

The main results of the usability tests were that test users expected to fill in patient data or see media content on the second screen. In some cases, such saving or closing the second screen clearly shows that test user’s attention shifts to the main screen. One finding was that test users experienced working on a multi-screen user interface on two screens as pleasant. According to the test users, using the main screen as an overview of patient data and using the second screen for more detail, data entry, or full screen media content was useful and pleasant to work with. Nevertheless, one test user felt that not in all cases it is necessary to use a multi-screen user interface because the focus is on one screen at a time.

The results of the usability tests indicate that some design guidelines were interpreted positively or negatively presented (see Table 9). Table 10 offers an overview of the adjusted design guidelines. Below, the adjusted guidelines are elaborated.

Table 9: Positively and negatively interpreted design guidelines

Positive design guidelines * Negative design guidelines *

1.2 Consistency in displaying

information 10.1.2 Avoid excess distractions 1.3 Clear buttons 1.3 Clear buttons 10.2.1 Predict the position of the

information 3.3 Display heterogeneous information 3.1 Efficient use of space 10.2.2 Make use of available space 4.1 Use visual presentations 6.1 Limited area of human vision 10.2.3 Standardize regions 5.2 Use hidden text

6.2 Provide own preference 10.2.4 Let users know 8.3 Content present workflow 7.1 Provide viewing more

information 10.3.1 Predict locations 9.1 Readability 8.2 Divide on task 10.4.1 Group information together 10.2.4 Let users know 8.3 Content present workflow 10.4.2 Group information depending on

the task 10.4.2 Group information depending on the task 10.1.1 Update simultaneously

(34)

34 Table 101.3 Certain buttons are not self-explanatory such as move to the second

screen. In addition, some buttons were not intuitive. Therefore, it was decided to extend

the design guideline with the use of simplified text on a button.

3.3 Display more heterogeneous information when available to help uncover relationships between complex data. One of the test users indicated that he wanted to see

more information in order to establish relationships between the patient data. This is according to design guideline 3.3 which is not proper adherence in this scenario, as seen in Figure 8. For this reason, it was decided to add information for the next usability test with the interactive prototype.

Figure 8: Before and after the adjusting the design guideline

4.1 When possible use visual presentations instead of text such as KEGS to decrease cognitive load in clinical routine.

This guideline was also not well applied since not all visual presentation were understood as intended. Here, as well as guideline 1.3, the guideline is extended with use simplified text.

5.2 Use hidden text labels when one wants to see more information. This guideline

was not applied, as it was not possible with paper prototyping to include this guideline. Nevertheless, it is still an important guideline to mention because it is a vital aid for users.

8.3 The content should be presented in a way that it maps the workflow of the user.

Because different kinds of end users participated in this study, it could not be determined beforehand. Yet, this design guideline is noteworthy.

9.1 To ensure readability, use symbols and if this is not possible use simplified text.

This guideline was not applied correctly, as not all symbols were accompanied by simplifying the text.

(35)

35

10.2.4 At any time, let users know where they are in the system and on which patient they are working. Let them know where actions/screens are opened. This guideline does not

need any explanation, yet it was not implemented correctly for all screens. Some screens were experienced as confusing; this was taken into account when designing the interactive prototype by clearly indicating at the top of the screens where a user is exactly located.

10.4.2 Presenting information on a multi-display user interface calls for information to be clustered and grouped depending on task-specific information which helps users to perceive, find, and interpret information easily. In the obtained feedback, two test users

indicated that the category of the media content was not logical.

Table 10: Interpretation of design guidelines positive or negative * Based on the initial design guidelines presented in Appendix I

Table 11: Adjusted design guidelines

* Based on the initial design guidelines presented in Appendix I

Positive design guidelines * Negative design guidelines *

1.2 Consistency in displaying

information 10.1.2 Avoid excess distractions 1.3 Clear buttons 1.3 Clear buttons 10.2.1 Predict the position of the

information 3.3 Display heterogeneous information 3.1 Efficient use of space 10.2.2 Make use of available space 4.1 Use visual presentations 6.1 Limited area of human vision 10.2.3 Standardize regions 5.2 Use hidden text

6.2 Provide own preference 10.2.4 Let users know 8.3 Content present workflow 7.1 Provide viewing more

information 10.3.1 Predict locations 9.1 Readability 8.2 Divide on task 10.4.1 Group information together 10.2.4 Let users know 8.3 Content present workflow 10.4.2 Group information depending on

the task 10.4.2 Group information depending on the task 10.1.1 Update simultaneously

Current Design guideline * Adjusted Design guidelines

1.3 Certain buttons do not speak for themselves Extended with therefore it was decided to extend the design guideline with the use of simplified text on a button

3.3 Display more heterogeneous information when available to

help uncover relationships between complex data Not adjusted - need to be properly implemented 4.1 When possible use visual presentations instead of text, such as

KEGS, to decrease cognitive load in clinical routine Extended with use simplified text

5.2 Use hidden text labels when one wants to see more information Not adjusted – because of the testing method interactive and paper prototype

8.3 The content should be presented in a way that it maps the

user’s workflow Not adjusted – need to be properly implemented 9.1 To ensure readability, use symbols and colors Not adjusted - need to be properly implemented 10.2.4 At any time, let users know where they are in the system and

on which patient they are working. Let them know which actions/screens are opened

Not adjusted – need to be properly implemented 10.4.2 Presenting information on a multi-screen user interface calls

for information to be clustered and grouped depending on task-specific information which helps users to perceive, find, and interpret information easily

(36)

36

4.4 Interactive Prototype

By means of using an interactive prototype, we were able to validate the adjusted design guidelines, and it became clear which design guidelines still needed some adjustment. Again, the following three tasks were used throughout the usability test: adding an allergy

and referral to the patient record, opening media content and adding procedure (ECG) at the same time and comparing media content and updating the patient’s history. The

usability findings are presented in Table 12 through Table 14. In each table, the usability findings, problems, and comments made during the test are listed along with the associated design guideline. In Table 15, the mean completion time and overall completion percentage can be seen. In Appendix O – In-Depth Interactive Prototype Usability Test,the in-depth evaluation of the interactive prototype per task is described. Furthermore, in Figure 9 through Figure 12, examples of the interactive prototype are presented.

Figure 9: Adding an allergy and referral to the patient’s record example 1

(37)

37 Usability Findings/Problems Description Design

Guidelines * Positive Negative Interpreted Number Of Encounters

Usability Problems Copy, Ctrl-v, functionality was not working - X 1 Unclear whether patients selections are done by clicking on the patient's name - X 1 Enter is not working properly - X 1

Not clear how to search for a patient

3.2, 4.1, 9.1,

10.2.4 X 4

Unclear to test the user that the patient record is already opened

10.2.4 X 1

Unclear how the allergy can be saved to the patient record 1.3, 4.1, 5.2, 9.1 X 1 The wrong button is used to add the referral letter 1.3, 4.1, 5.2, 9.1 X 1

Usability Findings The main screen was used as a reference to fill in a second screen 1.1, 3.1, 3.1, 3.3, 6.1, 7.1, 8.2, 8.3, 10.1.1, 10.1.2, 10.1.3, 10.2.2, 10.2.3, 10.4.1, 10.4.2 X 1 The tester indicates that the steps were intuitive and the screens naturally cooperate 1.1, 1.2, 3.2, 3.3, 4.1, 6.1, 8.3, 9.1, 10.1.1, 10.1.2, 10.1.3, 10.2.4 X 2 After opening the allergy widget, a second screen will be opened 1.1, 1.3, 3.1, 3.3, 6.1, 7.1, 8.2, 8.3, 10.1.1, 10.1.2, 10.1.3, 10.2.2, 10.2.3, 10.4.1, 10.4.2 X 2 It is useful to switch between a main and second screen 1.1, 1.3, 3.1, 3.3, 6.1, 7.1, 8.2, 8.3, 10.1.1, 10.1.2, 10.1.3, 10.2.2, 10.2.3, 10.4.1, 10.4.2 X 1

Table 12: Usability findings task adding an allergy and referral to the patient’s record * Based on the initial design guidelines presented in Appendix I

Design guidelines for a multi-screen user interface: Redesigning a Hospital Information System

Redesigning a Hospital Information System

Nawishca Clemence Mook

University of Amsterdam

Design guidelines for a multi-screen user interface

Redesigning a Hospital Information System

Preface

Table of Contents

Summary

Samenvatting

1. Introduction

1.1 Problem Description: Multi-screen User Interface Design for a HIS

1.2 Goals: UCD Guidelines for a Multi-screen User Interface for a HIS

1.3 Chapter Organization

2. Methods

2.1 Overall Study Design

2.2 Collection of Used Methods

2.3 Usability Test Methods

2.4 Usability Test

2.5 Questionnaires

3. Towards Initial Design Guidelines

3.1 Literature Review

3.2 State of the Art

3.3 Initial Design Guidelines

4. Towards Final Design Guidelines

4.1 Usability Test and Test Users

4.2 Mock-ups and Tasks

4.3 Paper Prototyping

4.4 Interactive Prototype