Overview

Anxiety disorders are the most common mental illness in the US. They are also highly treatable, but only 37% of people who suffer from them receive treatment due to factors including cultural stigma and high cost. Welp (short for “worry help”) is a mobile iOS app developed to combat this problem by offering remote, inexpensive anxiety treatment based on Cognitive Behavioral Therapy (CBT).

The project was originally launched in November 2019, and we began the app development and beta testing stage in March 2020.

Target Users

Undergraduate students that struggle with moderate to severe academic anxiety that are actively looking for treatment options.

My Role

I am the lead mobile designer working with cross-functional teams of psychologists, software engineers, business developers, and university professors.

The Problem

Our team of business developers participated in the I-Corps program at the University of Pennsylvania and conducted market research and psychological literature reviews to better understand potential target users. 

40+ interviews were also conducted with students about their struggles with anxiety, allowing us to craft user personas and determine how to create an app that improves upon the status quo.

Some of our most significant findings:

• CBT apps significantly outperformed control conditions in treating generalized anxiety disorders.

• Student counseling centers typically provide talk therapy rather than CBT to students.

• The average estimated total medical cost for treating anxiety is $6,475.

Project Goal

Given our user and market research, we determined that limitations of anxiety treatment options available to students are accessibility and lack of long-term effectiveness. We decided to build an app that is free for students and helps them treat their anxiety using CBT techniques.

Ideation

Low-Fidelity User Flows

We first brainstormed and created a list of the features we wanted in our initial prototype:

• Initial Screening Assessment

• Weekly Assessments

• Guided Journals

• Psychoeducation

From there, we created user flows to visualize a user’s journey through the application.

After that, I started drawing out wireframes before transferring them to Figma.

Medium-Fidelity Flow Exploration

A major UX challenge was deciding how to best organize journal entries. The difficulty was that we wanted to be able to group journals together based on worry. For example, if the user made several journal entries about one upcoming Physics Midterm, we wanted to be able to group those journals together so that we could track the changes in the user’s anxiety levels related to this particular event.

Option 1 is the flow that we ended up using.  I chose this design because of its simplicity - most of the available user actions are displayed on the screen, and it logically makes sense for the user to create their own titles for their worries.

Option 2 was meant to provide a shortcut for users who were journaling about the same worry but did not want to retype their entry. We found that the nuanced language we used (“old worry” vs “new worry”) was a bit confusing and not worth the shorter amount of time needed to create a new journal.

Option 3 was an idea to group journal entries based on date (i.e. date that a worry might happen, such as a Physics Midterm) rather than title. We decided against this option because of the possibility of users having hypothetical worries with no set date in mind (e.g. worrying about being unemployed).

Slider Design Exploration

Selecting our slider design was a bit of a headache due to the limited screen space and different types of questions in our assessment.

Option 1 is the design I chose because its horizontal orientation was conventional and intuitive to users.

Option 2 was similar to Option 1, but was more suitable for answer choices that were on a continuous (instead of discrete) scale. Also, the slider number would cover the slider text descriptions.

Option 3 was our most unconventional option but the vertical direction was a bit unintuitive.

Option 4 attempted to replace the slider with radio buttons, but we realized that due to limited screen space, we probably wouldn’t be able to display more than 5 buttons at a time without shrinking them significantly.

High-Fidelity User Journeys

A new user creates an account and completes an initial screening assessment that takes about 10 minutes. Once the user receives their screening results, the user watches a video about the purpose of the app and scrolls through a carousel describing each of the app’s features.

The user goes to the Journal tab and starts a new journal. They choose whether or not they have journaled about their worry before and then answer 8 questions about their worry. Once the entry is complete, the user creates a title for the journal if the entry is about a new worry.

Before a user starts their first journal entry, they must go to the Learn tab to complete a training module that introduces them to important therapy concepts. It consists of 3 sections: concretization, practice questions, and cost anchoring. Once the user completes all 3 sections, they set a daily notification time and start their first entry.

The user can view their anxiety levels over time in detail by either going to the Data tab or clicking on the graphs on the homescreen. From there, the user can scroll through different datasets and click on them to view the data over different time intervals (i.e. monthly, weekly, etc.).

The Product

Moving Forward

This design is in the process of implementation as an iOS app. We are currently conducting user testing of the application and are collecting feedback from students on the app’s design. We are reviewing user feedback on a weekly basis and will create further iterations of the app design to better suit user needs. The next version of the application will include additional CBT features including exposure therapy and problem solving.

After reflecting upon our current development process, I would like to further improve and refine our user testing process by creating more rigorous guidelines and metrics to objectively determine how users feel when using Welp. Though we currently face limitations due to Covid-19, there are definitely more user research methods we could incorporate in the future, including focus groups, contextual studies, and A/B testing. 

In addition, we should conduct more market research into the resources and limitations of student counseling centers to further develop our app’s points of differentiation. We will be interviewing counseling center directors and employees to better understand the quality and content of their anxiety treatment for students.