Goal of the project:

Re-design SEER*Explorer, a research tool for visualizing cancer statistic data, to be usable to wider audiences. In specific, cancer patients, journalists, and newer cancer researchers. In addition, address both new and existing usability issues that hindered task completion for this tool.

Challenges for the project:

First, was that I did not have direct access to users. Given that NCI is a good steward for data, they do not reach any of these users directly: these would instead be handled by hospital systems. Therefore, I would need to rely on other methods for not only learning about users, but designing a system that would fit the user’s needs.

Secondly, was that I would not be coding. We had specific contractors who were dedicated towards coding what t he end product would be. Therefore, I needed to make sure that the design properly conveyed the problematic areas not only to stakeholders, but to developers as well.

Methodology

I used IDEO’s design thinking framework to approach this project. As I was somewhat new to the field of cancer, the most important step was to empathize.  While I began to do user research into the larger audience, I quickly realized that empathy was going to have to play a much larger role into the design process.

I began doing research into the chosen audiences which had been decided by key stakeholders. These were Patient Survivors, Journalists, and Newer Researchers. I relied on a number of different resources for this. These included:

• Health Information National Trends (HINTS) survey, an NCI-published survey which provided medical research into Patients, their information-seeking habits, their level of medical terminology, and other statistical information as it relates to medical communication.
• Academic research on cancer reporting, understanding where gaps of knowledge may occur and issues with communications.
• Interviews with key stakeholders within NCI, to understand both the organizational and business goals as well as to leverage their insight into understanding other resources to examine
• Other in-house NCI documentation that provided insight into specific audiences.

However, upon doing further research, I began to understand that empathizing with the user needed to be a much deeper process. These were some of the challenges I realized I would face when talking with users.

Patient survivors

Cancer Survivors were one of the primary audiences that I was designing for, and very early on I realized that I couldn’t treat this like any other user. The very first things I realized was that they might be encountering this tool in an emotional vulnerable state.

According to my user research, while healthcare providers were the most trusted source of information, internet tools such as SEER*EXPLORER might be the second most. In addition, patients had some contradictory habits with information-seeking: they often times would take the initiative in trying to find relevant information, but too much information would be overwhelming and drive them away.

While my design might not harm them physically, I could very easily ruin someone’s day if I did not present the design in an appropriate manner. (“How would you feel if you found out that you only have a 1% chance of surviving 5-years?”)

In addition, this audience might not know a lot of medical terminology, which would be problematic when using the existing design of SEER*EXPLORER: a significant amount of upfront knowledge would be required to complete many of the necessary tasks. Therefore, I needed to understand how they would approach this issue with sensitivity.

Journalists

Another audience that we would be targeting with this revision was journalists. NCI has invested in outreach opportunities to the public recently, with the major one being the Annual Report to the Nation. However, SEER*EXPLORER has the functionality to provide on-demand statistics based on the cancer registry database, an important need for this audience. It also has the option to provide trend data which can give a more nuanced image of cancer from year to year.

The main difficulty with this audience comes from health communication. While the tool has vast functionality, there is a lot of upfront knowledge that is required. Given the limited technical knowledge as well as possible time constraints, we wanted to make sure that they could not only complete the tasks that were given, but also obtain help quickly if necessary.

Newer Researchers

An interesting trend appeared when examining the current userbase of SEER*Explorer. Newer researchers would often be the main users of SEER*Explorer, while more experienced researchers would use more statistical tools such as SEER*Stat. However, both of them would mainly use the program for the visualizations. The ability to generate charts and graphs of relevant data and export either the image or the selected slice of data from the database was seen as the key advantage of this program over other tools.

Research Findings and Design

Doing research allowed me to gain some insight into how the tool might allow these user groups to utilize the tool. However, it also allowed me to understand what the design issues were currently faced. First and foremost was that much of the functionality that these users would need was actually implemented within the tool. Rather than needing to add additional functionality, my main design task would be to structure the information in way that would allow these users to complete their necessary tasks.

Therefore, I had to then consider what was the correct method for presenting this information.  To understand that, I had to figure out what was the point.  I had a couple of existing resources that I had encountered as part of my user research. These included the HINTS survey, which gave statistical surveys of patient habits and needs, as well as a previous usability report for the last re-design of this tool. 

From this user research, I was able to examine the problems that not only arose from the latest revision, but also which ones might have persisted through previous versions.  I was then able to get a good idea of the problems that I needed to address with my design.

I began my design process by paper prototyping. I find that paper prototyping provides an avenue to slow down and think about the design of elements of the page that you might otherwise assume is standard.  This paper prototype allowed me to quickly examine several different ways of presenting information.

(Paper Prototype goes here)

I used these sketches not only to figure out what might be the best design approach, but also in my meetings with stakeholders. When discussing possible options with stakeholders, they often had valuable design input based on their own experiences. By having these sketches on hand, I was able to help them visualize and gain greater insight into what they would like to see implemented.

While making these sketches, I also did a Heuristic Evaluation of the website in order to assess the current usability of the website. This allowed me to understand the issues with the current site when it came to usability issues.

This highlighted a number of key problems with this design. The number one issue with this was that while several important usability issues had been addressed in previous versions, the implementation had led to a number of problems for our new target audiences.

Therefore, I spent the majority of my time continuing to refine both the scope and size of the problems, as well as possible solutions to these issues. Given that I would not be coding the final product, I made sure to refine the suggestions to made as well as presenting these ideas to relevant stakeholders.

Lessons Learned

Given the unique challenges that I faced, it is no surprise that I was able to learn a number of unique lessons from this. It should also not be surprising to learn that a new understanding of empathizing with users to be the primary one.

This project required me to deep dive into not only the motivations that users might have, but also how to approach presenting this information and keeping in mind their emotional state as they complete these tasks. It required me to think about time constraints, terminology, and a number of other factors that might affect usage.

Terminology and adaptation of methods was another lesson, one which had parallels for both myself and stakeholders. Design terms, such as Visibility and Constraints, often had to be replaced with much simpler explanations as they meant different things to my users. Talking with one stakeholder, they thought constraints meant a loss of mobility which might affect the way that a patient would use our tool.

Likewise, I had to convince my stakeholders that terms that they knew like back of their hand, such as Incidence, Mortality, and Survival, had to be explained clearly throughout the tool, or adequate help needed to be provided to define these terms.

Lastly, though, was the importance of communication with stakeholders. As I worked with prototyping various designs and design methodology, I often had to make decisions based on my experience and skills. However, explaining not only what the problems were but also my design recommendations were was often something I needed to think deeply on. NCI is a data-driven culture, with research based on the cancer database and registry. As such, some UX methods, such as personas (which were not based on real patients) was something that I had to learn to advocate for.