CueMed Mobile Health

Overview

The Project

This project was a 3-week user experience consulting engagement for a startup company in the health technology sector. I was responsible for user research and worked in collaboration with interaction designer Erika Spoden and visual designer Brooke Stopherd. I delivered a research report to the client, made feature recommendations to my design team, designed usability testing protocols, and conducted usability testing.

The Client

CueMed is a smart appliance startup that is seeking to break in to the competitive marketplace of pill reminder apps for mobile devices. The founders hope to develop a product with provable efficacy that can be sold to doctors and insurers as well as end users. On the advice of a previous UX consulting team, they pivoted to focus specifically on treatment of hypertension and hired a UX designer to create a prototype. CueMed asked my team to pick up where the previous team left off, and to evaluate and test that prototype.

The Problem

Poor treatment adherence is a national problem responsible for $100 billion annual healthcare spending and involved in over 33% of medication-related hospitalizations. According to the World Health Organization, "lack of information and skills in self-management, difficulty with motivation and self-efficacy, and lack of support for behavioral change are all impediments to patient adherence."

The solution

CueMed will use mobile technology to enable and reinforce healthy behaviors; project support and positivity through the app’s written content and graphic elements to promote healthy moods; and provide useful information about medications and other educational elements to empower users to take charge of their health.

Digital Prototype

The digital prototype includes interaction design by Erika Spoden and visual design by Brooke Stopherd.

Research & Discovery

Methodology

I began research with an exploratory literature review via Google Scholar and PubMed using the search terms "technology" and "treatment adherence," as well as read consumer product reviews of competing reminder apps.

I interviewed a practicing physician subject matter expert about challenges of treatment adherence and managing hypertension.

I conducted diagnostic evaluation and user testing of CueMed's existing prototype, as well as testing of the team's redesign prototype.

I generated recommendations for features to include to address the health needs of users with hypertension.

Research Findings

I submitted a research summary to CueMed encapsulating my findings and recommendations.

Some key findings:

I learned that prescription medication adherence is particularly important for hypertension patients. Doctors treating hypertension calculate the doses of medication they prescribe based on patient blood pressure readings, and if a patient's blood pressure stays high because they've been skipping doses, a doctor who assumes they've been fully compliant is likely to respond by raising the prescribed dose. Too much blood pressure medication can cause unsafe drops in blood pressure leading to fainting, comas, or even death.

I summarized consumer product reviews of 47 different pill reminder apps to determine which features and design solutions were considered the industry gold standard.

Recommendations

Based on my research I recommended that CueMed add 3 new features to their mobile app.

• Blood Pressure Log
  • Routine blood pressure monitoring is recommended for hypertension patients and it will provide a pathway for continued engagement with the app over a long period of time, which is one of CueMed's business goals. 
  • Home blood pressure readings are considered more accurate than doctor's office readings because of white coat hypertension, a phenomenon in which people who are nervous around doctors experience unusually high blood pressure during their physical examinations.
  • More accurate blood pressure readings means more accurate dosing of prescription medications, which means fewer and less severe side effects.
• Meditation/Relaxation
  • Many doctors recommend relaxation exercises for patients with hypertension
  • Meditation and relaxation are easily enabled by mobile technology. Pair a soothing image display with white noise, the user's favorite music, or other relaxing audio.
  • American Heart Association recommends 5 minutes of relaxation prior to a home blood pressure reading so this feature synergizes with the blood pressure log feature.
• Reinforcement
  • One of the most effective ways to help people develop new habits is to reward them for performing the desired behaviors.
  • Several competitors have successfully implemented reward features such as the Care4Today app which donates money to the user's favorite charity when the user gets perfect adherence.
  • Reward could be as simple as a victory animation, such as the ones used in Candy Crush. 

Prototyping & Testing

Usability Testing of Existing Prototype

The existing prototype of the CueMed app had serious usability problems. While all users were able to find the schedule of  upcoming medications and the medication detail page, only half were able to edit the dose of their scheduled medications. None of the test users could figure out how to log the fact that they had taken or skipped a given dose of medication once the scheduled time arrived - a critical flaw, as this log feature is one of CueMed's most central functions.

Evaluation of Existing Prototype

I took a closer look at the screens in the prototype that users were having trouble with.

Usability Testing of Redesigned Prototype

I reported my recommendations for new features to the team interaction designer, Erika Spoden, who created a new prototype to test out the viability of adding blood pressure log and daily meditation to the mobile app.

Preliminary results were largely positive, but there were some problems with the new design's prescription onboarding process that would need to be addressed in the next iteration.

Reflections

Plan ahead for attrition and no-shows

One of the big challenges for this project was attrition and no-shows from test users recruited from Craigslist. Because we focused on users with sensitive health conditions, we ended up getting a surprising number of testers who had to cancel because of medical emergencies. It probably didn't help that we were running tests in the ramp-up to the holiday season, when everyone had extra stress and extra obligations on their plates. While we were ultimately able to complete testing with a convenience sample, I wish that we had been able to get more people with hypertension to validate that my ideas for new features were things they would actually want to use. Going forward I will be sure to over-recruit to give myself a margin of error for no-shows.

Get buy-ins from all stakeholders

Our main point of contact at CueMed was one of two co-founders, and I made recommendations based largely on what she described as CueMed's business needs and priorities. While she responded quite positively, her co-founder was less enthusiastic. It turned out he had very different ideas about what kind of product he wanted to create. In the future I need to make more of an effort to get face time with every important stakeholder early on in a project to be sure we're all on the same page.

Next Steps

Build out reinforcement feature

We were able to build out prototypes for the recommended blood pressure log and meditation features that tested well, but did not have time to build out any of the reinforcement features. In many ways reinforcement is the most critical of my recommendations; being able to successfully log blood pressure won't help anyone unless they're motivated to really do it.

prioritize to stay on schedule

Going forward, CueMed will need to assess which of their planned features is the highest priority. The existing version of their prototype includes a community section, coaching section, and health-related informational content that will require constant updates to keep users engaged and represents a significant effort to maintain on the back end. Realistically, some of these features will need to be eliminated to meet their current proposed timetable.