Closing the Adherence Gap
Solving medication adherence with technology and design
Medication is one of the most critical levers the health system has to improve outcomes, but its impact is limited by as much as patients are willing to use it. Medication non-adherence is one of the most challenging issues providers and payers deal with, and solving it would unlock better outcomes, savings, and efficiency across the healthcare system. Managing medication effectively will require building technology that fits seamlessly into patient’s lives.
Across the health system, medication adherence impacts organizations in different ways. For pharmaceutical companies, patient adherence is directly tied to commercial success. For health technology companies, medication is one of the biggest levers for improving outcomes in the populations they serve. Medical device manufacturers have been trying to navigate this growing demand with portfolios of connected products, from smart auto injectors to inhalers, that have yet to find large scale adoption.
Especially in the context of AI in health, the value of real time, longitudinal medication use data continues to grow. Health AI can better predict risk, monitor progress and intervene earlier when informed by accurate data sources. There is a growing ecosystem of unique devices and wearables that generate longitudinal health data, but among those medication use data is unique. Currently, since it cannot be easily monitored with existing devices, the opportunity for a new device is clear.
While solving for adherence means potentially building a device, a device alone won’t solve adherence.
One size does not fit all
Not all non-adherence is the same. From forgetfulness to side effects to cost, reasons for non-adherence vary wildly across conditions and people. Designing for adherence means deeply understanding who you’re building for. The more focused a problem can be (i.e. solving for a specific population’s need for payment assistance) the more targeted, and successful, your solution can be.
Formative studies with specific user groups can uncover insights that drive feature development and alignment within a team. By talking directly with patients, a product or service can be designed with their specific issues in mind. Smart has a long history of running user research to identify product opportunities, and then working with teams to commercialize these concepts.
How to track medication use
Once you have patient-centric insights informing an adherence program and a potential device, teams will have to make a decision on how medication use data is captured. Tracking medication means interfacing with the physical world, because of this, there is a wide range of approaches to monitoring it. Each approach has tradeoffs, and the right one will depend on the specific use case and population you’re building for.
- Integrated tracking: Build tracking into how medication is already taken by balancing system architecture against how precisely you need to confirm a dose was taken. Understanding the data fidelity needed for your use case, likeneeding to know exactly how much medication was removed from a container, will inform your system architecture which will then inform design decisions.
- Biomarker linking: Develop a biomarker that acts as a proxy for medication use and is measured via wearables, PPG sensing, or an at-home test. Depending on what is tracked, this can be achieved via an algorithm on a third party wearable, or a custom sensor package.
- Pharmacy Integration: Pull data around refills directly from a patient’s pharmacy. This data is less accurate and timely than from daily use, but does not require a device at all.
- Automated dispensing: Create a device that manages dosing for patients directly, where medication can be tracked as part of the system. Due to the added cost and complexity of this type of device, this may be most appropriate for use cases with a large addressable market like senior care or chronic condition management.
\
A deep graveyard of failed products
From smart pill bottles to reminder apps, many medication adherence solutions put patient forgetfulness front-and-center. Many of these products rely on audio/visual cues or app alerts to notify users to take medication. These signals only go so far. They’re limited in how they address many of the root causes of non adherence: unpredictable side effects, managing costs, understanding dosing schedules, and medication hesitancy.
No single tracking device can solve all of the above and the ever growing number of attempts isn’t for a lack of trying. Solutions that work take a human-centered approach to understanding and solving for barriers to adherence.
Medication adherence is about outcomes
Many adherence products center on tracking whether or not a dose was taken, and can lose sight of the broader outcome patients and providers are aiming for. Whether it’s managing high cholesterol with statins, depression with SSRIs, or multiple conditions at once, helping patients understand and reach their goals is key for any adherence program.
Especially as digital health further integrates AI within telehealth, patient medication data will become a key lever for improving these outcomes. In this context, medication tracking can allow programs to intervene faster and with higher confidence. There is a lot of opportunity to design and integrate novel devices into these kinds of programs.
How human centered design can move the needle
An iterative design process that links formative user research with design iterations can highly effective at developing solutions that stick. Different types of studies can be used to generate insights that drive meaningful design direction. Starting with general needs finding within a user population can help point your team in a direction around what to solve. A design team can then take those pointers to generate concepts, that are then fed back into usability focused research sessions. Running these loops to refine the design will sharpen the product you plan to build. Because the patient is at the center of this process, it can give your team high confidence that you’re solving the right problems.
\