Kaiser Permanente: Internal Website Redesign

MHCID Capstone Project: Kaiser Permanente- Southern CA Imaging & Strategy Department

Tools: Figma, Adobe Illustrator, Adobe XD, Adobe Photoshop

Team: 5 members total

My Role: Project Manager, Designer, & Supporting Researcher

Over 30,000,000 people are admitted to the hospital each year in the United States alone. Ensuring that the proper hospital equipment is safe to use and in the right place at the right time is critical. We partnered with Kaiser Permanente’s Southern California Medical Imaging Technology and Informatics department to help create a more efficient inventory management tool for hospital equipment across the entire Southern California region.

Problem:

  • The internal website is fragmented and outdated, requiring several steps to complete simple tasks.
  • Form procedures have not been put in place, resulting in the input of incorrect information and a convoluted inventory system.
  • Employees are not notified when an action is required to be taken on their end. This results in prolonged time for procedures to be completed.

4 screens of the current out-dated kaiser website

Solution:

  • Streamline the website architecture so it is easier and more intuitive for users to complete their task.
  • Redesign the process of forms with error-prevention in mind. Offer a “push-back feature” that would immediately notify a person upstream of any mistakes that needed to be corrected.
  • Alert employees through their worklist page that a task needs to be completed.

Process

We followed the Design Thinking methodogy and started the process with research to gain an understanding of our user’s needs. Through academic research, stakeholder, and user interviews, we were able to define our problems and opportunities. As we moved into our ideating phase, we explored numerous design solutions, and then prototyped and tested our solutions through each step. The iterative process allowed us to test, prototype, and refine our prototype.

diagram of design process that shows arrows going back and forward in each step to show iterative process

Research:

  • Academic Research: The UCI team gathered relevant scholarly research articles on a variety of topics on the industry standard for the digital implementation of workflow tools in the medical field and data management in medical systems. This domain knowledge would help inform our game plan and strategy in the early stages of the project.
  • Stakeholder Interviews: Six stakeholder interviews were conducted, each approximately an hour in length. These sessions helped the team develop an understanding of the target user groups.
  • Tree Testing: Using the research platform, Optimal Sort, the research team drafted a tree-test study and it was completed by eighteen Kaiser employees. The purpose of this study was to validate the client team’s optimal depiction of the site’s information architecture.
  • Opportunity Analysis: Based on the information that was gathered in previous research, the team was able to identify many different points of actionable change. In order to pinpoint critical and/or priority items, the team examined “competitors” that are implementing the features and functions that the myMITI website needs to consider in the future.

Stakeholder Interviews:

We interviewed a variety of people in different job roles via Microsoft Teams. The insights gathered from these interviews were used to inform our personas, storyboards and scenarios.

  • Key Finding 1: Communication between functional groups is via email, phone, and in-person. Currently, there is no notification system within myMITI for new work related tasks and items.
  • Key Finding 2: Users would like to be able to see their assigned tasks in a centralized location: “Worklist”.
  • Key Finding 3: There is no quality control or checks and balances to ensure that information is input properly into the myMITI database.

Description and image of regional capital PM and local PM
Description and image of local IAA and local physicist
Description and image of the service engineer and warehouse manager
Tree Testing:

A sample of Kaiser employees participated in the tree-test study. This study aims to verify the expectations of our clients and stakeholders regarding the optimal organization of the website.

  • Key Finding 1: For the 18 total participants, the average success rate across all tasks was an 89%.
  • Key Finding 2: Users had difficulty with inventory actions. For both decommissioning and relocating systems, there was some sort of disconnect from the optimized IA and realized success in the tree test. This led to deeper discussions with the client and stakeholders to better understand where these functions and actions ought to be located.

Opportunity Analysis:

A light competitive analysis-style activity was used in order to validate current design decisions and verify priority items. The team examined relevant applications and websites that currently implement workflow and data management functions. We then mapped out the curret site with opportunities for improvement.

Flow Chart:

Once we had a clear understanding of the website functionalities, we made a flow chart of a single task . This helped our team have a high-level overview of the processes on the myMiti page.

Design

With research and analysis as our foundation, we were finally ready to begin the highly iterative design process. We explored as many different solutions as possible in a short amount of time.

Goals

  • Alert employees when an action was needed.
  • Provide error-prevention when inputing information into the database.
  • Show users a worklist where users could easily see their assigned tasks.
  • Simplify the information architecture so that tasks are easier to complete.

Methods

  • We created mood boards, and then a stye guide to set the tone for consistency.
  • We started ideating quick sketches by hand and digitally, exploring various concepts.
  • Once we solidified a direction, we continued to refine until we felt confident starting the pixel-pushing process in Adobe XD.

Outcomes

Our final deliverable was a desktop prototype that focused on four main flows;

  • Searching for a system.
  • Creating a system number.
  • Decommissioning a system.
  • Relocating a system.
As a part of these flows, we added features such as the worklist and pushback mode to meet the needs of our users.

Mood Boards
a collage of different images/ screenshots of websites and web components
Style Guides
styleguide showing fonts, colors, and other assets that are used in the new website.
Lo-fi Wireframes 1.0
a mix of hand drawn and digital wireframes.
Refined Wireframes 2.0
8 black and white wireframes that are annotated with opportunities and comments
Refined Wireframes 2.1
8 black and white wireframes that show a more complete flow and are annotated with opportunities and comments
8 black and white wireframes that show a more complete flow and are annotated with opportunities and comments

Prototyping & Testing

Formative Usability Evaluation

The study was intended to evaluate and identify potential issues with the new features and designs. A sample of 9 Kaiser employees participated in this remote study, and were given a set of four core tasks.

  • Navigating the Worklist
  • 2. Using Pushback Mode on a Form Error
  • 3. Looking up System information using a Search Filter
  • 4. Initiating Action on a specific system

These tasks were identified over the course of the project as key workflows that end users would be completing on the myMITI website. We used the platform Validately.com to conduct the test.

image of desktop prototypes with opportunities
Validation Survey

Based on the results gathered in the formative usability evaluation, there were some key insights that needed to be verified. The survey questionnaire was distributed through google forms to 15 Kaiser employees. Based on the results, the team was able to steer the designs in the right direction. The followup survey was deployed to dive deeper into the following topics:

  • Work List Icons: The team determined that using standard icons would mitigate interpretation issues. Four standard icons were tested, and it was found that a “flag” icon best represented “Important Items.”
  • Pushback Mode: Toggle Location The toggle used to initiate Pushback Mode had potential to be lost in the page fold. Three button locations were evaluated, and it was found that the original location of the button was preferred by users. We resolved the possibility of losing the toggle by making the top area of the forms fixed when scrolling.
  • System Lookup: Search Bars Majority of users gravitated towards the top navigation search bar when searching for a system by system number. Users also noted that the duplicate search bars served the same function.
Hi-Fidelity Wireframes 3.0
Summative Usability Evaluation

We implemented a final usability evaluation through validately.com to evaluate the adjusted designs and implementations. At this stage, the research goals were to evaluate the perceived usefulness, clarity and general usability. Using a combination of stakeholder feedback and the results gathered in the formative evaluation and followup survey, a finalized prototype was completed. This asset was used for the final usability evaluation. Users were given a set of four core tasks:

  • Navigating the Worklist
  • 2. Using Pushback Mode on a Form Error
  • 3. Looking up System information using a Search Filter
  • 4. Initiating Action on a specific system
The Outcome

Our goal was to take Kaiser’s internal website and make it functional, intuitive, and simple to use. Our final usability test showed positive feedback: Our data showed that users had a 100% task success and completion rate for all four core tasks, with perceived usefulness. However, we found that there was some opportunities for improvement in helping users with comprehension of the new features.

  • Opportunity 1: Onboarding Some users noted that some new features were not easy to notice. An opportunity for better visibility and learnability would be to add an onboarding process, so users can learn the new features quickly and maximize the website’s new features. We added onboarding screens of the key new functions to so users would instantly understand how to use them.
  • Opportunity 2: Notifications We added a notification feature to give instant, visible feedback regarding worklist tasks. The myMITI website would alert employees via email, but we felt that allowing the user to also receive the communication through their current myMITI page would help. The new notifications feature that would keep users informed of any new worklist or status update.

Conclusion

Our final prototype with Kaiser resulted in positive feedback. Their developers have started to work on putting our designs into production. The new site will be used and implemented across all of the Kaiser Permanente Locations in Southern California.

tinahongdesigns © 2021