Mental health apps are stuck in the ’90s

The internet has transformed the ways we access mental health support. Today, anyone with a computer or smartphone can use digital mental health interventions (DMHIs) like Calm for insomnia, PTSD Coach for post-traumatic stress, and Sesame Street’s Breathe, Think, Do with Sesame for anxious kids. Given that most people facing mental illness don’t access professional help through traditional sources like therapists or psychiatrists, DMHIs’ promise to provide effective and trustworthy support globally and equitably is a big deal.

But before consumer DMHIs can transform access to effective support, they must overcome an urgent problem: Most people don’t want to use them. Our best estimate is that 96% of people who download a mental health app will have entirely stopped using it just 15 days later. The field of digital mental health has been trying to tackle this profound engagement problem for years, with little progress. As a result, the wave of pandemic-era excitement and funding for digital mental health is drying up. To advance DMHIs toward their promise of global impact, we need a revolution in these tools’ design.

I believe DMHIs struggle to engage users because of a shortage of creative innovation in their design. Many of today’s most popular DMHIs still bear a striking visual and functional resemblance to the 1990s-era psychotherapy self-help handbooks from which they originated. DMHI developers’ eagerness to adhere to these traditional intervention strategies has narrowed the horizon of interventions we see as valid. The field of DMHIs has decided too early in its short history, and with insufficient evidence, that current designs are the best we can do.

DMHI design today represents just a sliver of what might be possible. DMHIs’ only mandate is to produce healthy psychological change, something creative ideas might accomplish in unexpected ways. Instead of iterating on existing designs, we should dedicate more attention and resources to exploring innovative ideas that might produce genuine leaps in design.

To begin our search for new DMHI designs, we should cast a wide net, exploring radically different visions of how technology can support well-being. Without discarding what we have learned about DMHIs so far, we should strive for openness to approaches that challenge assumptions and broaden the scope of our design thinking.

For one, we can seek design leadership in teams with real-world expertise in creating compelling behavior change interventions, ranging from product advertisers to video game designers. Rather than simply creating DMHIs and hoping people use them, these efforts put appeal and implementation first. Some applied researchers are beginning to take such approaches. Amanda Yarnell of Harvard and Alexandra Psihogios of Northwestern both lead teams collaborating with health-related social media content creators. They help ensure content is accurate and useful, while the creators take the lead with their presentation skills and expertise on what makes their audiences tick.

In addition to working with outside collaborators and building on popular experiences, we might search for entirely new ideas. For example, my teammates and I at the Lab for Scalable Mental Health are running a crowdsourced megastudy of brief DMHIs for depression. Like other megastudies, ours aims to find a range of novel ideas and rigorously evaluate their potential for development and dissemination. Following our open-ended call for any intervention that might reduce depression in under 10 minutes, we received 63 submissions from global teams. Submitters included high school students, physicians, and popular YouTubers. The submissions were equally diverse; one was a spiritual movement practice, another used artificial intelligence to enhance expressive writing, and a third featured an animated meditative stroll through a meadow led by cute animals. Soon, we’ll rigorously evaluate which interventions have the most promise for further development and scaling.

Pursuing innovative DMHI designs won’t be a straight path to success. Some promising ideas will fail, but others may be surprisingly fruitful. To learn from these successes and failures, we need field-wide norms of rigorous data collection and open sharing. As a field, we should strive for an adventurous stance in exploring diverse new ideas, complemented by robust evidence to integrate learnings and ensure novel DMHIs’ reliability. My team’s crowdsourced megastudy models this process on a smaller scale: first searching broadly for creative new designs, then moving to rigorous and rapid evaluation, and finally disseminating the DMHIs and study insights into the real world.

We also need evaluation to make sure that new DMHIs are not only effective but safe before they reach public audiences. New kinds of DMHIs may carry a variety of risks, including unintended harm to mental health and privacy issues.

Both industry and academic teams have roles to play in shifting the field’s focus from iteration to innovation. Industry teams can build exciting, user-friendly products, while academic teams have the freedom to challenge current design norms without intense pressure to profit. Funders and institutions, too, can align their goals with discovery, incentivizing DMHI developers and researchers to try new ideas rather than sticking to well-trodden terrain.

My enthusiasm for exploring novel interventions runs contrary to much of the prevailing wisdom on DMHIs. Many academics, in particular, believe that rather than trying new approaches, we should focus on optimizing and better implementing current solutions. To them, moving from iteration to exploration may feel like an unscientific step backward into uncertainty. I understand their skepticism but disagree. While major leaps in traditional face-to-face therapy may be unlikely, I believe that DMHIs are different. Their technological affordances — accessibility, privacy, and flexibility — make truly innovative and rapidly impactful interventions possible. Pursuing innovation in design is our best bet for advancing DMHIs toward breakthroughs in design and a deeper understanding of what works and, perhaps more importantly, what doesn’t. One thing is clear: What we’re doing today isn’t working.

Benjamin Kaveladze is a National Institute of Mental Health T32 postdoctoral fellow in the Center for Behavioral Intervention Technologies and the Lab for Scalable Mental Health at Northwestern University.