Medical Futures Lab

Recently I had a brief meeting/discussion with a doctor from the Texas Medical Center who conducts research in the realm of public health. We met so that I could receive feedback on some health-related software that I had developed for a class. Among the many things I learned during the conversation, there was an intriguing observation that I perceived: software engineering methodologies for creating a solution are philosophically different than the techniques the doctor was employing, and the methods I was learning about in class.

Source: Wikipedia

In software development, one of the core principles is to operate continually, using an loop of building software and getting feedback. This is absolutely necessary in the world of software where goals, technologies, and requirements change at dizzifying rate. The answer to the speed of software was to use a set of adaptive and flexible methodologies, often referred to as “Agile” practices.

http://i.stack.imgur.com/aexha.gif

The method on the other end of the spectrum is to engage in research that results in a deep understanding of users and the solution domain. Rather than prototype quickly and get feedback, the idea behind user-centered design is that users are sufficiently understood so that when it is time to design, prototype, and evaluate, time spent will be more efficient and effective because there is a solid understanding of the stakeholders and their environment. This is effective in most situations, because the assumptions are not changing even while one is prototyping.

In reality, the methods are not too far apart from each other, as they both operate in iterations that contain similar elements. However, there is an important philosophical difference; in the two figures, user centered design places the main elements in a circular loop while agile development reinforces the continual and parallel nature of the processes. This is only because of the ridiculous speed of the software industry. But in the end, each party at both ends of the spectrum could take notes from each other. A healthy balance of user-centered design combined with fast prototyping and quick iteration cycles would provide an approach that is both responsive and precise.

Source: Wikipedia

In order to tackle the problem of low clinical trial accrual, my group has decided to work on designing the general solution of a Clinical Trials Navigator (CTN). While I am satisfied with this as a good general solution, there are other specific solutions that could be beneficial. Especially being a computer science major who loves to build software, I can’t help but toy with these other solutions. So just for fun, I will change gears and write a slightly more technical blog post that describes the architecture for a solution that I would like to implement in the future (when I have time!).

The idea is a service that sends patients email digests for new trials that are relevant to them. A digest contains a simple list of potential clinical trials that the patient might qualify for, and is interested in. The flow of the application is described below.

Basic Architecture

The use case for a user is that they register online for the digest service and choose the corresponding types of trials that they would be interested in. After this simple registration, the user does not have to initiate any more contact with the service. From the user’s perspective, digests with aggregated clinical trials information will come in at a fairly steady pace. This saves the patient from the headache of sifting through trials themselves.

But how does the user receive this neat packaged digest of relevant clinical trials? The solution I have thought up is simple:

1. The listing of all clinical trials on clinicaltrials.gov is freely available and is updated every day. The end goal is to detect new trials, so what we will do is index all current trials, and then check every day to see if there are non-indexed trials (which means they are new).

2. When new trials are added, we check against our database of registered users to check whether or not the trials would be relevant. If they would be, then we make a note to include the trial in the user’s next digest.

3. Once a digest has reached a tipping point (a point where any more trials would clutter the digest) or a time of inactivity (below the tipping point, but there are trials that are getting stale), then we send the user the email digest.

4. In the email the user has the option to unsubscribe themselves from the digest.

Although this solution does not help patients who are not already engaged patients, I hope that it might help those who are not against the idea of trials, but have not the energy, time, or knowledge to sift through trials online.

Source: Wikipedia

In Medical Media Arts Lab, we have been encouraged over and over again to consolidate our research and design through the use of stories; so far we have produced storyboards for the problem as a whole and written multiple narratives from the perspective of our problem stakeholders. Before these exercises, I could not clearly see the benefits of storytelling, but I am now very aware of their effectiveness. Witnessing the power of storytelling, I began to cogitate on why they are so potent; With this blog post I’d like to explore some perspectives on the power of representing problems in the form of stories.

The IDEO Human-Centered Design Toolkit directly addresses and prescribes storytelling. Specifically, it recommends storytelling in the step right before materials/solutions are produced. One reason for the use of storytelling is that it the stories give real, human-centered ideas and solutions that are synthesized from research and thought:

“Telling stories is about transforming the stories we heard during research into data and information that we can use to inspire opportunities, ideas and solutions. Stories are framed around real people and their lives, not summaries of information.”

The other reason why they are proponents for this type of storytelling, is that they help solutions designers think in terms of specific events, rather than in general summaries:

“Stories are useful because they are accounts of specific events, not general statements. They provide us with concrete details that help us imagine solutions to particular problems.”

Another interesting perspective is that storytelling is fundamental to being human, and is necessary for human survival. Personally, I can see the motivation for this strain of thought; for me, stories grab my attention, foster empathy, and are easy to process cognitively. For instance, today I attended some presentations from electrical engineering researchers who were describing their ongoing research. It was no surprise to me that the presentations about an ongoing problem that were framed as stories were engaging and had me quite interested, while the other ones that framed the problem via a bland description of the problem had me feeling drowsy. It is the format of a story that somehow makes the problem come to life in a way that feels natural and is captivating.

Whatever the reason for storytelling, I am grateful to have had experiences this semester that have reinforced its strengths. It has helped me already in my classes, and will undoubtedly aid me in the future for problem solving and communication.

Over the last month, Team SAVE has conducted interviews with patients who have participated in or are currently participating in clinical trials. In each interview, we asked the patient about their experiences, opinions, and suggestions. While conducting these interviews, we noticed a common thread among what many of the patients said; in almost every interview that we have conducted thus far, the interviewee has said something along the lines of:

          “I am only alive today because of pure chance and luck”

In most of the situations, the interviewee had been in a late stage of cancer, and by chance had stumbled upon a clinical trial that miraculously reversed their condition. In one case it was a faint suggestion online that had caused them to seek out the trial. In another it was a lucky reference from a doctor who recommended another doctor who just so happened to be studying the exact mutation of the patient. In these cases, the patient somehow managed to find a suitable trial, when the odds of finding one were close to none.

Luck should be erased from the equation of clinical trials. In other words, the recommendation that the doctor gave to visit a specialist for some disorder or the advice given on a website should not happen by chance. In an ideal setting, the right information about trials should always naturally travel to the patient. To remedy this situation, we have been designing the role of a clinical trials navigator, who will consistently provide education about trials, help patients obtain trials information. The navigator will be the person that the patient sees soon after diagnosis, and during the meeting with the navigator, the patient will be brought up to speed about clinical trials, and relevant trials will be suggested by the navigator. We hope that this will remove the chance required for patients with specific mutations who need cutting edge medicine.

Although most of the interviewees shared a similar story, some did not. One, in particular, had a completely opposite story. After being diagnosed with cancer, they immediately searched online for a possible mutations, and came to their doctor prepared with clinical trials in mind. It would be ideal if every patient acted like this, but unfortunately not every patient is so equipped. Going forth we are looking to design our solution in a way that engages not just the patients who would seek out help by themselves, but those who would not seek out aid, and are really the ones who need help the most.

Image Source: http://farm8.staticflickr.com/7221/7175331883_80d3ebae45_b.jpg

I come from a computer science background.

Thus, I have grown accustomed to immediately think of a technical solution when I am presented with a problem:

http://doreenrainey.com/wp-content/uploads/2012/01/TheresAnAppForThatHeader.jpg

• For obtaining driving directions, I think of Google maps.
• For spreading awareness of a cause or an event, I think of social media websites like Facebook, Twitter, and Youtube.
• For improving the organization and workflows of hospitals, I think of electronic health records.
• For banking, I think of Chase’s online platform.
• For improving education, I think of Khan Academy.

In the last month and a half, however, my tendency to naturally think of technology has been disadvantageous.

When our team was confronted with the initial problem of increasing clinical trial enrollment, we gravitated towards technical solutions. Some of out initial solutions were an online game, a Yelp/eHarmony-like website, and an online clinical trials matching system with human navigators. When I was thinking of these ideas, I was operating under the assumption that clinical trial enrollment is low because the solutions we thought of don’t exist. This was partially because our project was in its preliminary stages, but it was mostly due to my orientation towards technology.

After performing more research, we found discovered that there is a wealth of technologies and platforms that exist online. Websites like Emerging Med and The American Cancer Society’s Clinical Trials Matching Service are just a few examples of the many existing implementations of the ideas that we came up with.

http://cdn.cultofandroid.com/wp-content/uploads/2012/04/shock.jpg

I was in shock. Some implementations of our ideas had already been around for over a decade. If our solutions had already been implemented, then why is the clinical trial enrollment rate still so low? My scope of thinking began to expand. As it expanded, I began to see the importance of a holistic and interdisciplinary approach to problem solving. Although it is useful that I have the knowledge and skills to create a website or make an app, it is not the most important thing; technology is only a small piece in the puzzle that is improving clinical trial enrollment.

After this realization, our team has been approaching the problem from a greater diversity of perspectives. Lately our team has been viewing the problem from a marketing and public health standpoint. Rather than focusing on providing help online, it may be valuable to focus on connecting people to existing online resources. By expanding our focus, we hope to come up with an effective solution.