PillPay

MIT Hacking Medicine Grandhack & Beyond

Leveraging behavior science and design, PillPay helps patients better adhere to their prescribed treatment. Medication nonadherence is a substantial issue which often causes harm to not only the patient but also creates rippling effects through the entire healthcare ecosystem. The idea was born at MIT's Hacking Medicine Grandhack and a few (including myself) continued to further develop the idea post-hackathon.

Utilizing the “endowment effect” and loss aversion as defined by Kahneman and Tverky, the system provides a framework for motivation and triggers along the end-to-end patient journey to help remain invested in their well-being. We aim to foster behavior change by minimizing the friction between the existing lifestyle and change added due to treatment through ease of use and self-motivation.

Role

• User Research
• Problem Definition
• Product Strategy
• Rapid Iteration
• Product Design
• Validation

Team

• Behavioral Scientist
• Pharmacist
• Clinical Researcher
• 3 Engineers

Medication nonadherence is widely recognized as a common and costly problem. Approximately 30% to 50% of US adults are not adherent to long-term medications.

The Challenge

Many patients do not follow their doctors' orders. Treatment of conditions relies on the expectation that medication will be taken as prescribed. Medication nonadherence is a substantial issue which causes harm for not only the individual but also causes rippling effects through the entire healthcare ecosystem. There are several factors involved in nonadherence which can include duration, frequency, complexity of the treatment, side effects, characteristics and severity of the disease. Furthermore, many existing frameworks to help patients adhere fall short by feeling disjointed from their lives and lack of motivation.

Kahneman and Tversky in their study on Prospect Theory discovered that losses loom larger than wins. As part of this aversion to loss, they identified the “endowment effect” which explains the greater value we put on things we own. Most reward systems become forgettable very quickly because they often lack significant value. PillPay introduces a PayBack system which leverages the endowment effect and loss aversion to help patients remain invested in their own well-being. At the start of a treatment, a patient may set aside any amount to be paid back as they succeed in adhering to their treatment.

With a deep understanding of the patients journey for treatment, we can target more impactful motivation to help them remain on track. It is important to accompany the patient along this journey rather than causing friction with erroneous processes.

An expansion of this concept includes a sponsorship model that allows family and friends support their loves ones in adhering to their treatment plan.

Introducing PillPay

A system to help patients with chronic illness better adhere to their treatment. We inspire behavior change through careful onboarding, appropriate triggers and motivation accelerated with tangible cash rewards.

Education

Taking on a treatment for a chronic condition is a huge step. As such, it's really important that the user is educated on the purpose and functionality of PillPay. This is a required step before using the app. When a physician prescribes the therapy to a patient, they can download the app and go through details of the therapy together. This initial investment sets up the reminder to pick up the prescription and continue forward with the therapy.

Onboarding

Upon picking up the prescription, it’s important that the user is able to quickly setup and learn how to use the system. To provide a cohesive and personalized experience and dealing with health-related matters, a good amount of information is required from the user. However, we understand that a long on boarding would be painful and significantly decrease the motivation to proceed onward. For this we leverage a conversational input pattern to easily collect the necessary information. We also begin by asking only for the users name to personalize the experiences going forward and a question to understand at a high-level their motivation for starting this treatment to help with future motivation. Upon completion, we take the user directly to a sandbox environment where they can learn to use the system while setting up the remaining sections.

Triggers & Payback

Leveraging the information collected throughout the onboarding process, a reminder (trigger) is sent at the appropriate time when the user has the medication available (capability) to ensure completion of the task.

The app enable the proper conditions for the behavior (taking medication) to happen. With users having "bought-in" for self-motivated reward, the loss-aversion reward mechanism facilitates a vested motivation to complete the task and continue forward.

How We Got Here

We decided to narrow in on a specific case of nonadherence to better study the challenges and explore options of its resolution. Smoking is a chronic condition, one that many patients attempt to treat with remedies that help with smoking cessation. The users journey may be triggered initially with a doctor suggesting and prescribing the treatment. Then a patient is expected to fill the prescription, understand what it entails and begin the regiment for the coming weeks. We discovered very quickly the exponential drop-off in adherence along each of the highlighted steps in the users journey. The further removed from the initial trigger, the less likely a patient would continue with the intended behavior. We knew that to continue with the intended behavior we would need provide ways to carry forward from the initial trigger and motivate for action.

Start With People

To better understand the basic motivations, goals and barriers of patients we developed high level personas. Though there may be many, we discovered two significant triggers as reflected in our personas. The trigger for most patients may be when the doctor prescribes the treatment. However, it was important to also consider the trigger of a friend or family of a patient that is made aware of the treatment that suggests it to their loved one.

What's Out There

We researched the landscape of what is available for patients to address this challenge today. We noticed a large gap in the market for motivating behavior change through incentive rewards. It also seemed as though the rewards offered currently as points for free items did not seem to create enough vested interest to promote change.

Vision Setting

To establish a framework of helping the user accomplish a desired behavior we worked with a deep understanding of their existing journey and target various behavior change techniques at specific instances along the users full journey for a supportive experience.

Designing for Behavior Change

To develop a system for behavior change we utilized strategies from Susan Michie and BJ Fogg's behavior change models. Following a process of identifying the target behaviors, understanding that behaviors occur at the intersection of three necessary conditions: capability - the ability to act, motivation - mechanisms that activate behavior, and opportunity - the environment that enables behavior. To incubate the above conditions we considered possible intervention functions and specific behavior change techniques.

Guerilla Testing the Concept

For some basic conceptual testing (post-hackathon), I developed a quick study to test the concept of a loss aversion reward mechanism. Though this sample case exhibits inherent biases of close participants and mock environments, it provided the ability for quick learnings.

The study asked participants to take vitamins daily. I sent a daily reminder, asking them to text a picture of themselves taking the vitamin which I followed up with a reward for successful completion of the task. For the loss aversion path, participants had a $10 “buy-in” which I rewarded back to them daily in $1 increments for taking vitamins. The other path, participants received points towards a gift card of similar value.

The results showed the loss aversion reward mechanism provided more incentive for participants to follow through for a longer period of time when motivation typically tapers off. However, one main learning is the threshold for “buy-in” may vary for individuals, and if not high enough for example, may not provide the necessary motivation to continue.

Towards an MVP

We defined an end-to-end user journey and supportive path towards behavior change through system architecture. I developed wireframes to begin laying out the app functionality and flow to better understand the detailed implementation of the concept. I also wireframed a web page for the sponsor user, though this was not in the MVP scope. After further refining the flow and behaviors of the system, I developed detailed design mockups and prototype including all visual and interaction design elements. This became the basis for design specs shared with the team for development.