X Close

CBC Digi-Hub Blog



Tackling the last (and first) mile of medicine with behavioral science

By Artur Direito, on 10 August 2018

By Katie Baca-Motes, Scripps Translational Science Institute

The medical field is taking notice of advances in behavioral science –improving medication adherence, increasing vaccination ratesimproving show rates for healthcare visits, decreasing opioid prescriptions, and increasing generic drug prescriptions. Behavioral solutions such as these are sometimes referred to as a “nudge” after the popular book of the same title. And while some argue that “nudging” is overrated, approaching health-related challenges through a behavioral science framework is becoming increasingly common.

Almost all these efforts focus on using behavioral insights to address “last mile” problems – focusing on delivery of proven treatments and medicines.  Popular TED talks and book titles reference the need for behavioral science to take us “the last mile,” Mitesh Patel and co-authors urge healthcare systems to establish “nudge units because “the final common pathway for the application of nearly every advance in medicine is human behavior.” Healthcare providers are even proactively thinking about the “last mile challenge” of returning genetic information.

But, in addition to helping address the last mile challenge, behavioral science is desperately needed at the “first mile” of all medical advances. Currently, less than 3% of eligible patients participate in any clinical trial, 35% of all clinic trials fail due to inadequate enrollment, and 89% of all trials must extend the recruitment period or increase recruitment funds. And often, clinical trials lack an appropriate level of diversity in their cohorts, limiting the known efficacy of treatments to a limited population (commonly, white, middle-upper class males).  While some have highlighted the potential of using behavioral insights to increase clinical trial enrollment, published research in this space is sparse.

In Sendhil Mullainathan’s TED talk he laments, “See what’s really puzzling and frustrating about the last mile, to me, is that the first 999 miles are all about science. No one would say, “Hey, I think this medicine works, go ahead and use it.” We have testing, we go to the lab, we try it again, we have refinement. But you know what we do on the last mile? “Oh, this is a good idea. People will like this. Let’s put it out there.” The amount of resources we put in are disparate.”

It’s equally frustrating at the first mile – the design of the trial is all science, but the outreach, recruitment, and return of information to participants often fails to incorporate a similar scientific process.  Advances in behavioral science and the ease of testing digital communications and websites open opportunities to understand the behavioral aspects of participant involvement in medical research.

We recently had an opportunity to conduct a direct-to-patient site-less clinic trial where all aspects of the trial were handled remotely – outreach and recruitment, informed consent, and monitoring.  Our first attempt at enrollment resulting in a depressing .08% response rate. We were well on our way to be a trial that either failed due to inadequate enrollment or in need of extended time or funds.  But using a behavioral science approach in which we tested variations in content, design, frequency, and cadence, we were able to dramatically increase the enrollment rate to 9.4%.

In another program, we used a key tenant of behavioral science, “make it easy,” offering people the opportunity to join a medical research program through a smart phone app they frequently accessed.  By simply inviting participants to consent, and subsequently donate health and activity data through this familiar app, we amassed a wide range of data with a diverse, engaged cohort in the first 10 months.

The proliferation of digitized trials and the launch of large research programs point to a growing need to understand motivations and barriers to participation and to understand the nuances around returning data back to participants and healthcare providers.  Relying on our intuition won’t get us to the level of participation and the level of diversity we need in medical research to lead us to the breakthroughs that are possible.

Applying the behavioral science or “nudge” approach to the first mile of medicine has the potential to radically increase the number of people engaged in medical research, giving rise to new treatments and medicines that we need to then get down thelastmile in medicine.


  • Where/ how can behavioral science and experimentation be best applied to recruitment and retention in digital health/clinical trials?


Katie Baca-Motes

Katie Baca-Motes (@BacaMotes) is the Director of The Participant Center for the All of Us Research at Scripps Research. Past experience includes managing behavioral research projects with the Scripps Translational Science Institute, Walt Disney Company, Aetna, Janssen, and others.

Leave a Reply