Digital Dashboard as a communicating tool for improved information exchange within staff in a hospital environment

(1)

Digital Dashboard as a communicating tool for improved information exchange

within staff in a hospital environment

Simon Edman

Supervisor at Websystem: Joakim Forsgren Supervisor at UmU: Stig Byström

Examiner: Thomas Mejtoft

M

ASTER

T

HESIS IN

I

NTERACTION

T

ECHNOLOGY AND

D

^ESIGN

, 30 ECTS

UMEÅ UNIVERSITY

DEPARTMENT OF APPLIED PHYSICS AND ELECTRONICS SE-901 87 UMEA

SWEDEN

October 19, 2020

(2)

UMEÅ UNIVERSITY

Abstract

Digital Dashboard as a communicating tool for improved information exchange within staff in a hospital environment

by Simon Edman

There is a well-known expression that reads: “Communication is key”.

That applies to our everyday life, work, relationships and more. Com- munication helps to make sure everyone is informed, and by doing so, avoiding misunderstandings. Effective and valuable internal communication is also a success factor and an essential part of an organizations’

structure. This study aimed to help staff at the Intensive Care Unit (ICU) at Örnsköldsvik hospital.

This paper presents results from a study performed, with the objective to investigate how a digital dashboard should be designed to make management and responsible specialist feel confident that they reach out to all team members and also receives communicated information in a hospital environment. This was done by conducting a user-centered design process to analyze, design and implement a digital dashboard that later on was used in a final study including a test period and evaluation. For a period of 2 weeks, the implemented digital dashboard was tested at the ICU in a hospital. After that period, the dashboard was evaluated by the staff members using a questionnaire, and an interview with the responsible administrator of content.

The result suggest that the management and responsible specialist felt that the digital dashboard makes them reach out to all other team members better, compared to their current communication channels. In addition, the vast majority of team members felt that they receive communicated information better. By using a digital dashboard as an internal communication platform, information exchange within the staff at the ICU was improved.

(3)

ii

Acknowledgements

First I want to thank all the people around me that has helped me with my work on the thesis. I am extremely grateful for this help and my thanks go particularly out to Joakim Forsgren, Kristoffer Wiklund and all developers that, on behalf of Websystem AB, supported me and gave valuable feedback. Timmy Eklund and Adam Nording for the peer reviewing and the comments. Stig Byström for the feedback and supervising on behalf of Umeå University. The ICU staff at Örnsköldsviks hospital that made it possible to carry out the work even with the circumstances that arose as a result of Corona Virus. To finish off, a special thank you is sent out to my fiancee and my family for the overall support during the time writing my master’s thesis at Websystem AB.

(4)

List of Figures

3.1 UX venn diagram . . . 8

3.2 UX cyclic design process . . . 10

5.1 One of the whiteboards used at the ICU . . . 23

5.2 The pause room at the ICU . . . 24

5.3 Sketch of digital dashboard . . . 25

5.4 Lo-Fi of digital dashboard . . . 26

5.5 Figur representing icuregswe.org . . . 27

5.6 Color palette of the Digital Dashboard . . . 27

5.7 Hi-Fi of digital dashboard . . . 28

5.8 Final implementation of administration page for the digital dashboard . . . 29

5.9 Final implementation of digital dashboard . . . 30

5.10 Final implementation of notes feed. . . 31

5.11 Final implementation of SIR data visualization. . . 32

5.12 Final implementation of the routines feed. . . 33

5.13 Final implementation of the calendar. . . 34

5.14 Placement of digital dashboard during Final Study . . . 34

5.15 Result of participants answers in terms of Comparison. . . 37

5.16 Result of participants answers in terms of When they watch the DD. . . 38

5.17 Result of participants answers in terms of discussing the content with colleagues. . . 39

5.18 Result of participants answers in terms of MOST useful section. . . 40

5.19 Result of participants answers in terms of LEAST useful section. . . 41

(7)

vi

List of Abbreviations

IC Internal Communication UX User Experience

UI User Interface DD Digital Dashboard ICU Intensive Care Unit DOM Document Object Model CMS Content Management System JSON JavaScript Object Notation

(8)

Chapter 1

Introduction

There is a well known expression that reads: “Communication is key”. That applies to our everyday life, work, relationships and more. Communica- tion helps making sure everyone is informed and by doing so, avoiding misunderstandings. Effective and valuable internal communication is also a success factor and an essential part of an organizations’ structure [1]. It can create a good working relationship between staff members, which can in turn improve morale and efficiency. It can be challenging to make sure all employees at a workplace receives and acknowledges communicated information. Many organizations use some type of internal communication platform to achieve this and by that engage the employees. Bill Quirke described in his book Using Internal Communica- tion to Turn Strategy into Action [2], the value of internal communication as following:

“The value that it can add is immense – faster change, more flexibility and innovation, better quality decisions, better knowledge shar- ing, and a more motivated workforce. Although, as a whole, lead- ers believe in the power of communication, even leading companies tend to fail at harnessing that power to deliver the necessary results.

Internal communication is vital to success and when done well can provide strategic advantage through aligning employee efforts, shar- ing knowledge, and engaging their passion.”

There are a number of different ways on how to communicate with and within staff at a workplace, papers on a wall, internal social platform, sending out emails, large digital displays showing information and more.

Social networking tools such as LinkedIn¹and Facebook² are often used by companies for recruitment and branding purposes. Anshu Sharma and Jyotsna Bhatnagar published in, 2016, the research Enterprise social media at work: web-based solutions for employee engagement. The findings

1LinkedIn,https://www.linkedin.com

2Facebook,https://www.facebook.com

(9)

Chapter 1. Introduction 2

of that research was that organizations should use social media tools in- ternally for employee engagement initiatives, and not only for recruitment and branding purposes [3]. Holger Sievert and Christina Scholz performed an interview survey on about 500 German corporations regarding the growing importance of digital engagement via internal social media. The result of that survey showed that internal social media, as one amongst more, can be used to engage the employees [4]. However, in the study it is also mentioned that within the past few years, other research has proved that companies with implemented internal social media do not take advantage of the possibilities because they are simply not using it. Often not all employees are reached, and many benefits remain un- tapped [4]. So, the effectiveness and value of an internal communication system are, of course, also depending on the people using it, not only the platform itself.

The use of an internal communication tool, such as internal social media, demands that employees identify themselves on the platform. Some companies and workplaces have staff without regular access to company computers and staff needs to be reached with printed news-letters or with bulletin boards. A large display is especially useful for them [5]. Further- more, some workplaces are not intended to use an internal communication system that requires staff to sit down at a computer to access the information. Instead, they need a solution that is visible for all staff. That is why a large display is appropriate to use when communicating the same information to a broad audience. A display can either be placed in public, semi-public, or not public at all, depending on the environment it is used in and the information it communicates. This type of use of a large digital display is often referred to as digital signage; they could be non- interactive or interactive. There are different ways of interacting with such a screen; it could be with buttons, touchscreens, gestures, voice, and more.

The ongoing transformation from paper-based advertisement to advertisement using digital signage is not that surprising; signs are becoming ubiquitous. It is an effect of the availability of suitable technologies for outdoor use and the sinking prices for large display hardware [6, 7]. It has become feasible to build large video walls for airports and also eco- nomical enough to add small screens to vending machines. They have rapidly become so natural that we use them without even thinking [7].

In Sweden, invoices, receipts and more is preferred to be handled digital instead of paper-based. It is also a motivation for companies and organizations that aims to achieve a more environmentally friendly work environment. It is a well-recognized fact that excessive use of paper in the workplace has substantial effects on the environment. The transformation from paper-based information exchange to large digital displays could also apply when it comes to internal communication. So, instead

(10)

of whiteboards or bulletin boards with papers on them, a large digital display can be used.

This master thesis was written in collaboration with Websystem AB³, described in section 2.1, and aimed to help the Intensive Care Unit (ICU) at Örnsköldsvik hospital. A digital dashboard (DD) was designed and implemented to help the ICU communicate to all staff members about upcoming events, educations, consolidated data and more, i.e. an internal communication platform.

1.1 Objective

The objective of this master thesis is to analyze the current communication channels at the ICU and investigate how to design a digital dashboard that makes the staff members at the ICU feel that information exchange is improved. A literature study will be conducted to establish a theoretical framework for administration and communication with a large digital display. The study will include testing of prototypes and designs, interviews, and evaluation, which will, later on, lead to a final design proposal to be used in a final study. By performing this, the following primary and secondary questions will be answered:

Primary

• How can we design a digital dashboard that makes the staff at the ICU feel that information exchange is improved?

Secondary

• How to create an effective solution regarding presentation of the information?

• How to create an effective solution regarding the administration of the information?

1.2 Limitations

The time period of this thesis was limited to 20 weeks. During that time period, of which this thesis was written, the Corona virus⁴ (COVID-19) spread rapidly around the world. This lead to regulations and general advice from the government [8], including hospitals not letting in visi- tors and social distancing among residents, which also applied to Örn- sköldsviks hospital.

3Websystem AB,https://websystem.se

4Corona virus,http://coronavirus.com

(11)

Chapter 1. Introduction 4

1.3 Thesis outline

The structure of the chapters and its’ content in this Master Thesis is as follows:

• Chapter 1 contains an introduction to the subject and objective.

• Chapter 2 describes Websystem AB, background and problem state- ment and related work.

• Chapter 3 is the theoretical framework about User Experience, as- pects of design, quality research, communication with large displays, Digital Dashboards and content management.

• Chapter 4 describes the used methodology for this thesis.

• Chapter 5 presents the results from the design process, the final prototype implementation and final study.

• Chapter 6 contains discussions about the different states in this project.

• Chapter 7 present conclusions from the result and lifts future work about this area.

(12)

Chapter 2

Background

This chapter introduces Websystem AB, describes the stated problem and introduces related work concerning communications via large displays.

2.1 About Websystem AB

Websystem is a technology company based in Örnsköldsvik, Sweden, which uses the open source Content Management System (CMS) tool Drupal¹. They specialize in developing web services with a focus on communication and information flows. They work together with advertising agencies and web agencies where Websystem are the ones responsible for the technology. They also work with companies that aims to integrate their services to the web. Websystem are there to help them with Appli- cation Programming Interfaces (API:s) and modules in Drupal [9].

2.2 Problem statement

For the ICU at Örnsköldsvik hospital, information is mostly communicated with email, whiteboards and verbally. Members of the staff have repeatedly mentioned that important information that today is mainly communicated with email is easily missed due to the lack of time to ac- tually sit down and check their email during a stressful workday. And respectively the management, responsible specialist and administrators of the information experiences that they do not reach out. The ICU needs some sort of digital dashboard that can be used to communicate to all staff members about upcoming events, educations, statistics, and more, i.e. an internal communication platform. The dashboard is not supposed to be personalized or linked to an individual person. Instead, displayed on a large screen and mainly focuses on information dissemination and for one-to-many communication. The placement of the display is vital;

1Drupal,https://drupal.com

(13)

Chapter 2. Background 6

in a room where all staff members is present during their shift and have access to, without exposing it to the public and unauthorized people.

2.3 Related Work

More workplaces include digital signage to communicate with employees. In 2014, Weber Shandwick, in partnership with KRC Research, re- leased Employees Rising: Seizing the Opportunity in Employee Activism. It concludes the result of a conducted global online survey of 2,300 employees. One of the findings of that survey presented was that more than half of offices claim to use digital signage in their communications strategy to communicate to employees, share notices, provide data and give feedback from customers [10].

Keith Kelsen wrote in his book Unleashing the Power of Digital Signage:

Content Strategies for the 5th Screen [5] that digital signage can be an in- credibly effective tool in regards to communicate with employees. Fur- thermore, he mentions that when placed in areas where the employees congregate, for example break rooms and cafeterias, digital signs can readily become a focal point of those areas and command attention. And because they provide high-resolution images, these displays captures the employees attention more effectively than email [5].

(14)

Chapter 3

Theoretical Framework

This chapter provides a theoretical framework for the reader and give an explanation on how a large digital display can be used to display a DD as communication platform and how it can be administrated. Initially, this chapter will explain user experience in section 3.1 User Experience.

3.1 User Experience

Dr. Don Norman¹ and Dr. Jakob Nielsen² are often mentioned when talking about User Experience (UX). Norman is the writer of the book The Design of Everyday Things [11]. He is also a member of the Board of Trustees of IIT’s Institute of Design in Chicago and principal emeritus of Nielsen Norman Group. Nielsen, Ph.D., is a User Advocate and principal of the Nielsen Norman Group. The two has written an article regarding the definition of UX, a part of that definition is stated below:

“User experience encompasses all aspects of the end-user’s interaction with the company, its services, and its products. True user experience goes far beyond giving customers what they say they want, or providing checklist features.” [12]

In the same article they point out two requirements to follow to achieve an exemplary user experience:

• First requirement is to make sure to meet the exact needs of the customer.

• The second is to produce a product that are a joy to own, and a joy to use.

UX has become an established term in today’s human-computer interaction and interaction design field [13]. Figure 3.1 is a simplified Venn diagram that gives an overview of UX and the different areas it includes.

1Norman,https://www.nngroup.com/people/don-norman/

2Nielsen,https://www.nngroup.com/people/jakob-nielsen/

(15)

Chapter 3. Theoretical Framework 8

FIGURE3.1: The figure illustrates a UX venn diagram. In- spired by one found on the website UXplanet [14].

• Business analysis is very important. There has to be a vision and clear direction on what the aim is. If there is no clear vision we can not expect all parts successfully working together.

• Prototypes are the base of everything else. The functionality also has a big impact on UX. Even if the prototype is perfect the functionality has to work as well, bugs in the system can easily lose a users interest in the product.

• Visual Design must make sense for the customers, the look and feel.

It is essential that the solution has a good interaction for the user, remember that the design is for the user and not for the designer. If the design and interaction does not meet the users expectations the user would probably not use it. The designer has to remember who it is aimed for.

Business analysis, Prototype and Visual Design together form the overall UX. Where these areas overlap new areas of responsibility occur. Proto- type and Business Analysis together creates the performance. Prototype and Visual Design creates the interface. Finally, Business Analysis and Visual Design creates the interaction.

(16)

3.2 UX process lifecycle

When designing something new, it is important to understand the user needs and the purpose of the design. The user is in the center of the design process, also known as User-Centered Design (UCD). The Inter- action Design Foundation has defined UCD as:

“User-centered design is an iterative design process in which de- signers focus on the users and their needs in each phase of the design process. In UCD, design teams involve users throughout the design process via a variety of research and design techniques, to create highly usable and accessible products for them.” [15]

Whether the design process aims to design a new website, new car concept or other, the iterative design process cycle of the four activities Ana- lyze, Design, Implement, and Evaluate are often applied [16]. In the book The UX book [16] written by Rex Hartson and Pardha Pyla a UX process lifecycle is described. This cycle is specified for the context of interaction design and UX. They also mention that in a UX design the four elemen- tal activities are translated and together they build the UX lifecycle that they call the Wheel see Figure 3.2. Analyze translates to understanding the user work and needs. Design translates to creating conceptual design and determining interaction behavior and look and feel. Implementation translates to prototyping. Finally, Evaluation translates to ways to see if the design is on the right track to meet the user needs and requirements.

This is no strict cycle with demands that has to be followed, it is simply an elaborated and evaluated guide to help the designer on how to go from a idea to a finished result that meets the user needs and requirements.

(17)

FIGURE3.2: The figure illustrates the cyclic design process.

Inspired by the UX Book [16].

• Analyze - The first activity of the design process. It is an important stage because it lays the foundation for the entire design process. If the analyze is performed poorly and misleading, there is a high risk of the final design, not meeting the customers’ expectation. During this phase, user needs and requirements are determined. It is about creating an image on why, for example, a new and improved design of a current solution is needed or why the customer wants an entirely new design and a new system. Interviews, field studies or questionnaires can be conducted to help to create that image.

• Design - The design activity consists of design thinking, creating conceptual designs and also design production. That could include sketching, workshops, ideation and more. The goal is to describe what the system is, how it works and what it should do.

• Prototype - The prototype activity is about creating an early model for the interaction design that is easily changed. There are a number of different types of prototypes, and they all have their purpose, the most common prototypes are: Low-Fidelity (Lo-Fi), Medium- Fidelity (Mid-Fi) and High-Fidelity (Hi-Fi).

(18)

• Evaluate - The evaluation activity is about testing and determine things that are good and bad with a design. By evaluating each of the designs created, it is possible to refine a design as soon as possible.

3.3 Mental models and Familiarity in design

All the people on this planet have each of their own beliefs. A mental model is something that is based on those beliefs and not on facts. It is a model of what the user knows about something, perhaps a website.

Dr. Jakob Nielsen published an article about Mental Models in 2010 on NNGroup. In that article he wrote that it is essential to consider the users’

beliefs. What users believe they know about a UX and User Interface (UI) strongly impacts how they use it. He also points out that mismatched mental models are common, especially with designs that try something new [17]. If a new design includes a groundbreaking way of interaction, a user might end up not using it because they simply has no idea how to.

The same Nielsen has stated a law that is called "Jakob’s Law of Internet User Experience" and that law is:

“Users spend most of their time on other sites. This means that users prefer your site to work the same way as all the other sites they already know. Design for patterns for which users are accustomed.”

In the book Mental models: aligning design strategy with human behavior, written by Indi Young, the three C’s are introduced. They are the three main reasons used in the book describing the advantages of mental mod- els all begin with the letter C – Confidence in design, Clarity in direction and, Continuity of strategy [18].

There is a reason to why the trash can icon on most computer operating systems and other platforms look like a trash can from the real world. It is because we are familiar with what a trashcan looks like and therefore know the usage of it. By using the power of familiarity in design and things from the real world when designing, the user might more easily understand the design and how to use it properly and as expected. It can also apply to the choosing of fonts, colors and more.

3.4 Qualitative Research

Interviews are a tool commonly used when performing a qualitative research [19] as part of a user-centered design process. When conducting interviews, it is important to ensure that an appropriate interviewing style is to be used. Interviews can be structured, semi-structured and

(19)

unstructured [19]. Each of the different styles of interviews have their purpose:

• Structured - Are similar to questionnaires. The questions are read precisely as constructed to every participant. This style is appropriate to use when, for example, wanting to collect uniform data from a large sample of participants, comparing results across different groups of users or obtaining general information about de- mographics such as age, gender, education and so on.

• Semi-structured - Does not have as structured questions as structured interviews. These interviews are appropriate to use when, for example, wanting to gather facts, attitudes or opinions, understand the user goals, gather information about tasks, task flows and so on.

• Unstructured - Has only a few structured questions and can, therefore, reveal issues that the interviewer had not considered before.

Are appropriate to use when, for example, wanting to develop new insights about the user’s interaction with technology, gather data on general themes rather than specific questions or explore a new domain where you are not certain of the major issues facing users.

3.5 Digital Signage and Display Blindness

As mentioned in the introduction, paper-based advertisement is used less to give room for advertising using digital signage. However, there are also flaws when using large digital displays; one of them is something called display blindness. This phenomenon has been a significant motivation for much of the research on public displays. However, there is not that much that has been done to give an understanding of display blindness. The phenomenon describes the situation when users do not see things that are on a screen. It is a kind of selective attention [20]

that makes people overlook things outside their focus of interest. When walking in a city, many screens can be present and each of them sending out a message of some sort, it is well known that people often ignore the screens and mind their own business. Display Blindness is similar to the long-known web user behaviour that is called Banner Blindness. It is an instance of selective attention [20], in which peoples attention only is directed to a subset to the stimuli in the environment, usually, those subsets are related to their goals [21].

(20)

3.6 Digital Dashboard and Data Visualization

An easy and great way to aggregate and visualize data from multiple sources, such as databases, locally hosted files, and web services is with a digital dashboard. They can be personal or public depending on the goals. A personal DD demands that a user can identify themselves and therefore get information that is primarily meant for them. With a public DD, the goal is to present information that is valid to more people and therefore not in need of knowing who is using it, the information it offers is equal for everyone watching. However, depending on the sensitivity of the presented information on the DD, it might be useful to have some general identification to limit the access.

When designing a DD that will be shown on a display, placed in a more open space with movement, and not on your own desk, it is important that the viewers can easily acknowledge the information. The Five second test³ is valuable to take in consideration. It is a method of user research that helps measure what impression users get within the first five seconds of viewing a design and what information users take away and acknowledges.

A popular way to visualize data on the web and on DDs is by using the JavaScript library Data-Driven Documents, often referred to as d3.js⁴. D3 allows the developer to bind arbitrary data to a Document Object Model (DOM). DOM methods gives the opportunity to change the structure, style or content of a document, for example a HTML document. A JavaScript file using DOM methods can therefore change the content of a HTML document. The usage of D3 can help generating an interactive bar chart from data with smooth interaction and transitions.

3.7 Content Management

Content that is created and later on needed to be retrieved, presented or edited has to be stored somewhere. It could be on a database hosted by a server, a cloud storage solution, locally stored on a computer or other.

For someone with minimum experience in web development and programming languages in general, creating a website from scratch can seem impossible, and things like HTML, CSS and JavaScript can be confus- ing. However, today there is an easy way to develop a website and manage the web content with something called Content Management System (CMS). That is an application that allows multiple contributors to create, edit and publish websites – in other words, used to manage web content.

3Five second test,https://fivesecondtest.com

4Data-Driven Documents,https://d3js.org

(21)

The content in a CMS is often stored in a database and the content could then be displayed in a presentation layer.

There are a number of different variants of CMS, all of which serve more or less the same purpose. The most used one is Wordpress⁵ which had, when this was written a market coverage of 62.6% [22]. Another one is Drupal⁶with a market coverage of 2.9% [22], that is also the CMS Web- system AB is using when developing.

Drupal is an open-source CMS and is made by a their community. Any- one can use it, and it is free to use. Something that sets Drupal apart from other CMS:s is its flexibility; modularity is one of its core princi- ples [23]. That modularity is possible due to all the different modules that can be used when developing a website. One of them is JSON:API⁷, which makes it possible to generate an API server. This server can then be used to create JavaScript Object Notation (JSON) requests to the website from a different website and then retrieve JSON responses. JSON is a lightweight data-interchange format. That format is easy for humans to read and write, and for machines, it is easy to parse and generate [24].

5Drupal,https://wordpress.org

6Drupal,https://www.drupal.com

7JSON:API,https://www.drupal.org/project/jsonapi

(22)

Chapter 4

Methodology

The methodology used to determine and develop a DD concept for the ICU will be described in this section. The method is divided in three parts, Literature Study, Design Process and, Final Study.

4.1 Literature Study

To gain an understanding of the research field, an extensive literature study was conducted. This literature study aimed to gain knowledge in the area of internal communication, DD, CMS, Drupal, how to design, test and how to present data and other information in a way that is easy to understand. The result of this research can be found in Chapter 2. The decisions and design choices taken based on the literature study can be found in section 5.1.

A lot of information was found from books, articles, short videos, jour- nals and other material. All of the information was analyzed critically.

Reliable sources were used such as Google Scholar¹, ResearchGate², NNgroup³, ScienceDirect⁴, The Interaction Design Foundation⁵and references in other literature.

4.2 Design Process

During this thesis, the design process that was used is based on the UX process life cycle from The UX book [16] described in section 3.2, although it was modified. The design process contained five iterations. First an individual analyze iteration, including a field study to understand the user needs, requirements, and the current way of communication. After that

1Google Scholar,https://scholar.google.com

2ResearchGate,https://www.researchgate.net

3NNgruop,https://www.nngroup.com

4ScienceDirect,https://www.sciencedirect.com

5The Interaction Design Foundation,https://www.interaction-design.org

(23)

Chapter 4. Methodology 16

four design iterations, Sketches and Ideation, Lo-Fi prototype, Hi-Fi prototype and, Implementation. The activities design and prototype was combined into one activity. That activity and the evaluation was repeated for each of the different iterations. Each of the different sketches and prototypes were evaluated before moving on to the next step in the process. Evalua- tion meetings were held with the unit manager and people not involved in the project and ongoing discussions with the experienced web developers at Websystem.

4.2.1 Analyze

As mentioned in the problem statement, section 2.2, the current problem is clearly stated. Members of the ICU staff has repeatedly indicated that valuable information, mainly communicated with email, is not acknowl- edged by all concerned staff due to the lack of time to sit down and check their email during a stressful workday. A plausible solution to this and by that improve information exchange within staff is with a DD. Although the problem was stated, further knowledge about what information is disclosed, what statistics to visualize, and more needed to be sorted out before any work with sketches and prototypes start. To achieve this, a field study was conducted to understand the user needs, requirements, and the current way of communication.

To achieve a qualitative research, interviews were held with management and responsible specialist at Örnsköldsviks hospital. The method for the interviews was unstructured interviews; there was some written questions to each of the interviews that was used as a discussion starter. This was decided after evaluating the different techniques described in section 3.4. Unstructured interviews felt most appropriate to use to achieve an open interview with room for discussions of different thoughts.

A startup interview was held with the operations developer at the Region Västernorrland. The purpose of this interview was to get a bigger picture of what the new solution should be and why it was needed. Also, what information and data the new solution should be able to communicate.

Finally, who are the receivers of the information and how it should be communicated.

Thereafter an interview with the unit manager was held. The unit manager is responsible for a lot of the communication to the staff, a content administrator. That interview aimed to specify more about what the actual content should be and how it is communicated today.

To complete the field study, a tour was conducted in the open areas of the ICU:s department at the hospital to get a closer look at the current solution in action. Also, to get an idea of where the placement of the DD display could be.

(24)

4.2.2 Sketches and Ideation

Design and Prototype

Based on the literature study and the field study, sketches were drawn on a whiteboard to get a draft overview of the potential layout of the dashboard. Smaller workshops were held with developers at Websys- tem with the aims of discussing different ideas and thoughts regarding design. Things like, were should a specific section be placed, what size should it be and more. This stage did not include an evaluation phase since the whole stage contained a constant evaluation of different ideas.

4.2.3 Lo-Fi Prototype

Creating a Lo-Fi prototype is a great approach when in need of translat- ing a high-level concept of design into something more tangible. The role of a Lo-Fi is foremost to decide and evaluate functionality, visual appear- ance such as colors, font or more are not important.

Based on the sketches and ideation, Lo-Fi prototypes were drawn with pen and papers. This was done to conclude the different sketches and to get a more specific idea about the design. The aim of the Lo-Fi was to get an overview of the design. No interaction ideas or thoughts were discussed; this was still an early stage.

Evaluation

When the first draft of the prototypes was finished, an evaluation phase took place. This was to get an early opportunity to change obvious flaws in the design. The prototypes were discussed with the unit manager.

Also, 5 people not involved in the project were asked to give there opinion and thoughts; this was done in person with an open discussion.

4.2.4 Hi-Fi Prototype

A Hi-Fi prototype is created as similar as possible to the actual final solution. Colors, more accurate content, functionality and more is added.

They are often created when there is a solid understanding of what to build and can be used to either test it with real users or get final-design approval.

When the Lo-Fi prototypes had been evaluated and analyzed a Hi-Fi prototype was designed. This prototype was created to be able to get a more

(25)

realistic view of how the implementation could look like. The Hi-fi prototype was created in Adobe XD⁶ which is a platform for creating designs for websites, mobile apps, and more. With Adobe XD, a prototype can look very similar to an actual website and also include interactions.

Adobe XD was chosen because of previous knowledge of the tool and the large amount of documentation and tutorials. No functional interactions were included in the designed Hi-Fi for this project.

Evaluation

The Hi-Fi prototype was evaluated with the aim to look more critically of the actual presentation, position and size of the content in each of the sections, the size of the different sections and the functionality of the SIR section. The design of the prototype was evaluated in the same way as the Lo-Fi, with the unit manager and 5 people not involved in the project.

Also, with an open discussion with developers at Websystem,

To evaluate the interaction and usability of the SIR section the thinking aloud method of user testing was performed. That method was chosen because it is, to quote what Dr. Jakob Nielson wrote in his book Usability Engineering [25] published as early as 1996 and still stand by [26]:

“Thinking aloud may be the single most valuable usability engineering method.”

It is a usability test method in which the tester asks the test participants to use the system while continuously thinking out loud and by that ver- balising their thoughts [26]. By doing so the designer can piece together much of the users mental model. 5 participants took part in the test, one of which was the unit manager. The given task was to change Unit and Report of the SIR section. As mentioned, the Hi-Fi did not include interaction since the interaction itself was on such a small scale. Instead the participants used their finger to point on the Hi-Fi and the tester described what happened depending on where they pointed.

4.2.5 Implementation

The final implementation of the new solution contains two separate websites, a Drupal site as the administration website and a regular website, without any CMS, the actual DD website itself. The administration website was created to handle all the entries created by the administrators of content and will not be used by any other than them. The design of the administration site is not taken in consideration and was not evaluated.

6Adobe XD,https://www.adobe.com/se/products/xd.html

(26)

Administration website

With help from the developers at Websystem, a back-end website was created using Drupal and a temporary database was established. The website had the responsibility to administrate the content that the DD should retrieve and present, and the database to store said content. On this website, management and responsible specialists could create, edit and delete content. Three different content-types were created. Notes, Routines and Calendar. Each of these different content-types contains fields which are adapted to each section.

Digital Dashboard website

The DD was created using HTML, CSS and JavaScript. This was chosen due to previous knowledge and the large amount of tutorials and documentation available. The DD was responsible of retrieving and presenting the information stored on the Administrator website server and data from SIR.

Demo and Evaluation

Before the final implementation was tested in the hospital environment, a demo and evaluation meeting was held with the unit manager, the medi- cally responsible doctor for surgery and intensive care and the unit manager for surgery. The purpose of this was both to demo the DD for the staff that was going to use it but foremost this is the management and responsible specialists that is reporting data to SIR and communicates information to other staff members. Another thing that had to be decided was what the time span should be, to indicate that a post is new. The interaction of the SIR was once again evaluated, but this time with actual functionality implemented. It was possible to change Unit and Report, and by doing so, the corresponding bar chart with data was visualized.

4.3 Final Study

The final study included a user test of the design developed during the implementation process. The purpose of this test was to get a better understanding of how the staff felt after using the DD for a period of time.

This was done to answer the primary research question and determine if this design makes the staff members at the ICU feel that information exchange is improved?

The implemented DD was tested during a period of 2 weeks at the ICU at Örnsköldsviks hospital. The DD was used on a laptop that was placed on a table in the pause room, a laptop with a mouse and an external screen

(27)

was used. The administrators of the content was informed on how to create, edit and delete each of the different entries.

After the testing period, an interview was held with the administrator of the content, the unit manager, and a questionnaire created with Google form was sent out. The chosen interview method was semi-structured since the aim was to gather attitudes and opinions. The questionnaire was directed to the staff, the receivers of the content. The questions of the interviews and the questionnaire were written and asked in Swedish.

The interview questions were chosen in cooperation with the supervisor at Websystem; the questions can be seen in Appendix A, translated to English. The questions for the questionnaire were chosen in cooperation with the supervisor at Websystem and also the operations developer at the ICU; the questions can be seen in Appendix B.

(28)

Chapter 5

Result

In this chapter, the results from the literature study, the field study with interviews and tours, the design and evaluation state, and the resulting implementation and final study is stated.

5.1 Result Literature Study

After the conducted literature study it was determined to use Drupal to create a back-end website with the responsibility to act as a database to store the different entries and its information that later on could be requested with the API from a different website. This is an effective solution of administrating the information. The DD itself is developed using the languages HTML, CSS, and JavaScript, all of which are often used when developing for the web [27]. The data retrieved from SIR is not stored in the Drupal database; it is requested from the SIR website directly and had to be visualized in a way that is easy for the user to com- prehend. To achieve this, the JavaScript library d3.js can be used to create individual bar charts representing each report and its values. The data could, therefore, be visualized in a way that resembles the way SIR vi- sualizes the data. This in regard to Familiarity in design, mentioned in section 3.3. This together creates an effective solution regarding presentation of the information.

5.2 Result Analyze

5.2.1 Interview with the operations developer

The result from the interview with the operations developer was that is was determined that the DD should be used as a compliment to their current communication via email. A lot of the staff works in shifts, the day and night are divided into three parts, day, evening and night. It is important that the information exceeds the limits of the shifts, so that all

(29)

Chapter 5. Result 22

staff are reached no matter what shift they are on. The operation developer pointed out that it is not possible to completely replace their email communication. Instead, the staff should be able to read a summary that points out the meaning of a sent out email on the DD and be advised to read the whole email to get further information. Another thing was that it should be possible to interact with the DD, preferably by touch, but not required. Most parts of the DD should work as a feed and not be required to interact with. Those sections could be some notifications feed, a bullet list that could work as a checklist or inform the staff about new routines.

Both those sections would need to have some sort of indication if a post is new, in regards to a not yet decided time. The section that would need to be interacted with is the part regarding data visualized, retrieved from SIR (Svenska Intensivvårdsregistret). SIR is a national quality register for intensive care in Sweden. It is a significant amount of data and different categories, the only way to visualize them in a way that is usable for the users is to be able to choose between those different units and reports. It was also discussed if there should be a fourth section, but nothing that resulted in a possible idea.

5.2.2 Interview with the unit manager

The result from the interview with the unit manager was more specific on what content the DD should contain. A total of four separate sections should be included in the DD and those were:

• A feed for smaller notes.

• A bullet list for routines.

• Visualized data from SIR.

• An event calendar.

The notes should be smaller informational posts, resembling the tradi- tional post-it on a board. A note could, for example, announce and shortly describe an upcoming event, education or meeting. The bullet list is meant to be used as an informational list representing a change in current routines and also new ones. The routines do not change that often and could be presented in a way resembling a paper. Regarding the SIR section, the unit manager wanted the ability to visualize the SIR data portal

"Koll på nuläget", translated to English: Check on the current situation.

It concludes a number of, by SIR, chosen reports as bar charts see Figure 5.5. It gives the user the option to chose between all ICU:s in Sweden and get the corresponding reports. For the DD the unit manager would like to have the further option to choose which ICU and also which specific report to view from SIR. The calendar was to replace the current one they use, a printed paper with written events on the corresponding date of occurrence.

(30)

5.2.3 Tour at the ICU

Apart from the regular communication between staff verbally and information sent by email, a variety of whiteboards and other boards are used.

Each of the whiteboards had papers attached to them and also information written on them. On the whiteboard regarding staff-information, there was the previously mentioned calendar with written events see Fig- ure 5.1.

FIGURE 5.1: The figure demonstrates one of the whiteboards used at the ICU, this specific one is Staff- Information. The content itself of the papers is censored

due to confidentiality.

A good placement for the display would be in the staffs pause room, that is a room where the staff can go to take a quick break, drink some water and more.

(31)

FIGURE5.2: The figure demonstrates the pause room at the ICU, this is where the DD might be placed.

5.2.4 Summarize of field study

The field study gave a lot of significant knowledge about what content sections the DD should contain, the four different sections: Notes feed, Routines feed, SIR data, and a Calendar. It was determined that only the SIR section has to be interacted with to be able to chose between the different units and reports. Also, what each of the different sections could contain, finally, were the DD could be placed, the pause room. This knowledge, in combination with the research, was a great foundation for the next step in the process, Sketches and Ideation.

5.3 Result Sketches and Ideation

The sketches that were created during the workshops resulted in two different ways of separating the content sections see Figure 5.3. The sketches were made to be usable with or without the need for interaction. This because in a stressful hospital environment, it is not possible to demand someone to interact with the DD to use it.

The upper one had a single page design of the four sections and if a section is pressed it would fill the whole screen with its content. This design made it possible to view a summary of each section at once. That single page overview layout would stay in that mode until someone interacted with it and chose to view a specific section in full-screen mode.

(32)

The bottom one had a fixed navigation menu to the left, and the content associated with each page was supposed to be to the right. The prototype with a menu would without interaction switch between the different sites based on a timed interval and with interaction, the user could choose by themselves which site to view.

FIGURE5.3: The figure illustrates the sketches of the dashboard, created during workshops and ideation.

5.4 Result Lo-Fi prototype

By this stage, it was not yet determined exactly how the data from SIR was to be visualized. Therefore the “SIR?” expression in that section see Figure 5.4. After further evaluation with the unit manager, developers at Websystem and others, it was decided that the best solution would be a combination of both previous sketches. It contained both the content overview site and the menu with navigation. The menu is fixed to the left and the overview to the right containing a section for each of the different content types.

(33)

FIGURE5.4: The figure illustrates the Lo-Fi prototype of the digital dashboard.

5.5 Result Hi-Fi prototype

The Hi-Fi prototype is very much the same as the Lo-Fi, but with colors and a more detailed layout, see Figure 5.7. The design of the SIR section was decided to resemble the way SIR visualize the data. This was determined after further discussions and evaluation of the Lo-Fi with the ICU unit manager and developers at Webystem. By placing the menu on one side instead of above or below the sections, each section can use more height and therefore include more content. That solution turned out to be the best option in regards to both being interactive but not entirely in need to be interacted with. The overview makes it possible for someone, with limited time, to get a quick overview of the information communicated by the notes feed, routines feed, and calendar. The different menu icons represent each of the different sections, the overview, and settings.

Only the overview page will be implemented during this thesis. The others are, for now, just placeholders and not intended to be implemented during this thesis. They present how a solution might look like in future versions, if the administration site and DD were to be merged into one.

The user could then administrate and create entries directly on the DD.

The colors were chosen to resemble the colors used on the SIR data portal website¹, this to take advantage of mental models and familiarity in design mentioned in section 3.3 see Figures 5.5, 5.6.

1SIR data portal,https://portal.icuregswe.org/utdata/sv/home

(34)

FIGURE 5.5: The figure demonstrates a screenshot of the current design of Koll på nuläget on icuregswe.org which is

SIR. [28]

FIGURE 5.6: The figure illustrates the used color palette when designing the digital dashboard.

During the evaluation of the Hi-Fi the overall feedback was positive. The central part that was mentioned and questioned was that they felt that there should be a more significant number of notes and routines possible to view. The way the Hi-Fi was designed, there could at most be three notes visible and depending on the length of the content of a routine post at most ten. Another thing that was mentioned was that the routines feed should look more like the notes feed, with a more prominent separation of each routine and with a header and content. Both notes and routines need to include the date and time when created. All of the participants of the think aloud test felt it easy to understand how to change both the unit

(35)

and report. Regarding the calendar section, planned events should be presented in that section. Also, an indication in the calendar in regards to dates that events occur on and perhaps a summary of all planned events.

FIGURE5.7: The figure illustrates the Hi-Fi prototype of the dashboard, created after the evaluation of the Lo-Fi.

5.6 Result Implementation

5.6.1 Administration website

The implemented Administration website contains three different content types, Notes, Routines and Calendars, see Figure 5.8. That content is stored on a server provided by Websystem. Notes contain a label field to represent its title and a text field to represent the information. Routines are built the same way as Notes. It includes a label field to represent its title and a text field to represent the information. The Calendar, on the other hand, contains a label field to represent an event and a date field to represent the date when the event occurs. All of the content types created include the time and date of when they are created; this is by default of the Drupal site and not implemented specifically. The website has a homepage presenting all of the created entries as a list, chronologically sorted with the latest entry at the top. There is a menu in the upper right corner where the administrator can choose either of the different content sites to create, edit or delete entries. Also, administrate their account and log out. This website does not depend nor relies on the DD; it works independently.

(36)

FIGURE5.8: The figure illustrates the final implementation of the administration website for the digital dashboard.

5.6.2 Digital Dashboard website

The final implemented design proposal that was developed in the last iteration of the design process is a little bit different compared to the Hi- Fi prototype, see Figure 5.9. The notes feed, and routines feed are placed next to each other instead of on top of each other. This was determined based on the feedback from the evaluation of the Hi-Fi. The notes feed resembles a typical feed on today’s social media platforms and news sites, separated boxes for each note. The text, both the notes and routines contain, were not supposed to be that long and therefore, the priority was to be able to show more notes and routines in the overview. If the amount of any of those feeds overflows the height of the display, the user has the ability to individually scroll down through the feeds to watch earlier posted information. The calendar now included a section on the right side of the calendar that lists all planned events, and in the calendar the date of which an event occurs are underlined. The overview and each of the different sections are shown in Figures 5.9, 5.10, 5.11, 5.12 and, 5.13 and further described.

One error that needed to be handled was regarding data retrieved from SIR. In some cases there was no data at all, it turned out that the ICU:s in Sweden were very different in relation to each other in terms of reporting data. In some cases there were no data to retrieve for a specific Unit and Report. To cope with that, the expected bar chart was replaced with the message “Oops, avdelningen har tyvärr inte rapporterat data för vald rapport...”, which in English translates to “Oops, unfortunately the unit did not report data for the selected report...”. The DD is automatically refreshed every hour, this is something that is easily changed if wanted.

(37)

5.6.3 Demo and Evaluation

The feedback from the demo and evaluation was very positive. They felt that the DD had features that they in there current work environment were missing. One thing that was pointed out was that the SIR section should have a default unit and report when the website is loaded or refreshed, so that there always is a bar chart showing. The version that was used during the demo had no such default, instead it was only the white background with the option above to chose unit. When the unit was chosen another option appeared were a report could be chosen. The unit and report that was requested to be set as the default was the unit Örnsköldsvik and report Ventilatorbehandling, invasiv. This because it was, according to their opinion, the most interesting and probably most watched one. The time span in regards to if a post is classified as new was determined to be 24 hours. They felt that there will not be that many entries created each day and a smaller time span would not be necessary.

This time span is easy to change if wanted.

Overview

Below is the final implemented design proposal illustrated. All text in each of the different content types, not including SIR, are just made up for demo purposes, due to confidentiality and not the actual text entered by the ICU.

FIGURE5.9: The figure illustrates the final implementation of the dashboard, created after the evaluation of the Hi-Fi.

(38)

Notes

The notes feed shows all created notes in order of the newest on top.

If the amount of notes would exceed the height of the screen, the user can scroll down the notes feed individually, without scrolling down the website itself. A note that is created within the last 24 hours are separated in its design compared to the others. On the left side of the note, a blue border is added, the size of the fonts are increased and also the size of the note itself. Also, the text “Skapad inom 24h”, which in English translates to “Created within 24h”, is added to the right of the time and date.

FIGURE 5.10: The figure illustrates the final implementation of the notes feed, and shows the indication of a new

note.

SIR

Regarding the SIR section, the unit Örnsköldsvik and report Ventilator- behandling, invasiv are chosen as default. This was determined after the demo and evaluation of the final implementation. To view a different report or ICU, the user just has to chose another option in the drop-down menus. The name of the chosen report is placed above the bar chart, in large font. Below that, the dates of which period the data correspond to. Finally, the bar chart itself, each of the axis have their corresponding name and units. All that info is retrieved from SIR. By hovering the mouse over a specific bar, the exact value is presented next to the mouse, resembling the SIR website.

(39)

FIGURE 5.11: The figure illustrates the final implementation of the SIR data visualization.

Routine

The routines feed shows all created routines in order of the newest on top.

If the amount of routines would exceed the height of the screen, the user can scroll down the routines feed individually, without scrolling down the website itself. A routine that is created within the last 24 hours are separated from the others, the text “Skapad inom 24h”, which in English translates to “Created within 24h”, is added to the right of the time and date.

(40)

FIGURE 5.12: The figure illustrates the final implementation of the routines feed, and shows the indication of a new

routine.

Calendar

The calendar shows the current year, month, and days of that month.

The current date is highlighted with a blue background. If an event is occurring on a date, that date is underlined. All the events planned are presented in the right of the calendar. If an event has already occurred, it is removed from the list. If an event is occurring on the current day, that event has a black font color. If an event is occurring in the future the font color is grey, and the font size is a little bit smaller compared to one occurring on the current day, this to emphasis the events, if there are any, that occur on the current day.

(41)

FIGURE 5.13: The figure illustrates the final implementation of the calendar.

5.7 Result Final Study

The final study was divided into two parts, a testing part were the DD was used at the ICU for a period of two weeks and after that a evaluation part were the use of the DD was compared to their current solution. The evaluation was conducted using a questionnaire and interview with the content administrator. The placement of the setup that was used during the test period can be seen in Figure 5.14 below.

FIGURE 5.14: The figure illustrates the setup and placement of digital dashboard during Final Study.

(42)

5.7.1 Interview with content administrator

When the test period started, the staff was informed about it via mail, daily management meeting the day the test period started, a note on the dashboard and also occasionally mentioned during the first few days.

The unit manager felt that the digital dashboard helps in terms of reach- ing out to the staff.

It was pointed out that in a stressful environment, it is difficult to reach out to all staff because the workload looks so different between each day.

It can take several days of which the staff has no time to open their mail.

In addition, the communication of information is made more difficult by the fact that they work in shifts. Something that might be discussed verbally during the day is not always passed on to the staff on the other shifts. Relay staff who only work on certain days can easily miss paper- based notes and routines. However, with the digital dashboard, such information is easily accessible to all staff regardless of time of day and workload.

Further mentioned was that it takes time for the staff to adjust and include the dashboard within such a relatively short period. Also, a para- dox exists when the underlying problem is to reach out to all staff; therefore, it is also challenging to reach out with information that a new information channel is tested. On that note, the unit manager partly trust that all the staff receives the communicated information. A more extended period is required to trust the system entirely. It is about the staff having to get used to a new routine. It is challenging to be able to trust it in two weeks. However, there is an incredible potential for the future.

Noticed during the test period was that staff discussed the content communicated on the dashboard, and perceives a positive attitude towards it. It was experienced that the staff felt a greater sense of security not to miss anything as all information is gathered in the same place. New content were on average posted 1-2 times a day, mostly Notes. In the future, more content administrators are needed, not only the unit manager.

During the test period, the unit manager felt that some things could be further improved. The Calendar needed added functionality, including the ability to switch between months to view all planned events. Also, Notes should, after a specific time or amount, for example, after a month, be removed to reduce the number of items. A new Routine should be emphasised more, like Notes.

Regarding the administration website, it should be more adapted for use on smaller screens, e.g. smartphones. Some buttons are placed in- correctly; for example, the button to save an entry is placed above the field where the information is written. Since the phone is always carried around and it is necessary to be able to create entries on the go, the page could have been adapted even more.

(43)

As further comments, the unit manager mentioned that there is great potential for those with specialist functions within the department to communicate information about their areas – staff such as pharmaceu- tical manager, donor manager, medical technology and more. Over time, there is great potential that this can become an included routine for how information should be communicated within the staff. Staff that works at night receives the same information as the staff who work during the day, without meeting each other. It was experienced that staff felt more involved when not required to log in to a computer and look for information themselves, which is not always possible due to lack of time. Giving staff information on new things and essential information that otherwise disappear among lots of other papers or mail. The staff, with the responsibility of reporting data to SIR, felt more motivated doing so as an effect of the usage of the dashboard. It became evident if no data was reported, since it was not presented in SIR for the given period.

Summary of Interview

The unit manager felt that the digital dashboard helps in terms of reach- ing out to staff members. With the digital dashboard, information is easily accessible to all staff regardless of time of day and workload. It is challenging to, during a period two weeks, be convinced that all team members receives the communicated information. However, there is an incredible potential for the future.

5.7.2 Questionnaire

The amount of participants taking part in answering the questionnaire was 16, all of which are staff members of the ICU at Örnsköldsviks hospital. The questionnaire was divided into three parts, Comparison, Habits and, Evaluation. The options shown in the figures is in Swedish since that was the language the questions and answers were written in.

Comparison

The result of the first question, How do you feel that you are reached by important information when using the digital dashboard compared to before?, is stated below.

Digital Dashboard as a communicating tool for improved information exchange within staff in a hospital environment