The retention challenge in remote therapy and learning seen through the lens of the COVID‐19 pandemic

Abstract

What does the popularity of social media "unfriending", "blocking" and "ghosting" communicate about the success potential for online psychological treatment and online education? This question has been brought to the fore by the COVID-19 pandemic and the resultant transition to remote delivery for much of clinical care and teaching. Online psychotherapy and education platforms represent rapidly adaptable, convenient substitutes and are playing an important role in helping stressed communities traverse the trauma. As we increasingly rely on these remote alternatives, however, it is crucial to anticipate and mitigate against a recurrent problem suggested by pre-pandemic scholarship: very poor retention. Although this challenge has been borne out in studies of both online therapy1 and online education2, these fields have been mutually insular and have not benefited from each other's experience in addressing this common foe. This issue has been recently highlighted in the specialized education literature3. Here, we explore it for a mental health audience, since, besides learners and educators, countless online patients and therapists also stand to gain from retention-enhancing design. Telemental health services vary greatly, including by specific technology used, intervention type, degree of provider involvement, and target population and diagnosis. During the pandemic, video-mediated consultations have become particularly common. Up until that point, the best studied telemental health intervention had been digitally-enabled self-help, typically inspired by cognitive behavioral therapy and incorporating little or no therapist involvement. The poor retention associated with the latter has been widely documented, including in one early4 and one more recent5 landmark studies that showed disappointing completion rates of 0.5% and 18%, respectively. The same limitation is borne out in studies of remote learning. When massive open online courses (MOOCs) first appeared nearly 10 years ago, they were heralded as the long awaited antidote to education disparities. Through low-cost courses delivered online by renowned educators to a worldwide audience, they promised to democratize high-quality education like never before and challenged the very premise of location-bound learning, regardless of topic or discipline. Universities would become obsolete, went the optimistic prediction6. This echoed the older promise that therapist-optional digitally-enabled self-help would dramatically increase access to care by correcting provider shortages, especially in underserved areas and communities. The euphoria – 2012 was dubbed "the year of the MOOC"6 – was short-lived, in no small measure due to a stubborn retention problem that has been revealed in several studies. Among them, a landmark analysis of 565 MOOCs delivered by the Massachusetts Institute of Technology and Harvard University to 5.63 million learners showed completion rates that ranged from 3.13% to 5.91% across academic years7. Also disappointing was the finding that MOOC completers tended to be socio-economically advantaged, not the in-need learners at the margins of global education that MOOCs hoped to reach6. Already by 2013, the world was declared "MOOC'd out"6. Retention, of course, is not the only metric by which to mea­sure the success of online therapy and education; even if retention is poor, a massively popular intervention or course still means that many users can benefit3. Also, today's pandemic-dictated platforms are typically much smaller, less impersonal, more interactive and better coached than the typical self-paced online therapy or MOOC of yore, suggesting that retention may be a less relevant problem with current offerings. Still, there is reason to be concerned about user engagement on today's platforms, due to characteristics that seem inherent to broader online psychol­ogy. Online, regardless of the specific activity, inattention and distractibility seem like perennial obstacles and ever present personality features. Already in 2008, a British Library investigation of scholars' online reading behavior described it as "promiscuous", "horizontal", "volatile" and "squirrelling"8. Given today's obsession with such analytics as "visitor conversion", "page views", "bounce rate" and "scroll depth", it would be safe to assume that this problem has worsened as Internet-related technologies have grown more sophisticated and distractions have multiplied3. A weak attachment to content has parallels in the weak bonds that characterize many online relationships, further suggesting a medium-wide commitment shortage present across online platforms and pursuits. In that sense, online information-seeking may not be fundamentally different from online befriending. From "blocking" to "unfollowing", "unfriending" and "ghosting", the abundance and popularity of online relationship-terminating functions and behaviors speak to this phenomenon. Relatedly, attention-defici