Decreasing the Public Health Burden of Eating Disorders: Commentary

Abstract

Eating disorders (EDs) are common, disabling, and costly; yet, less than 20% of those with EDs receive treatment. EDs have also skyrocketed in the COVID-19 pandemic, with access to care worse than ever, further solidifying the need to not only make EDs a priority but also embrace new approaches to address this major public health problem. Schleider and colleagues (2023) argue for the single-session intervention (SSI) as one such option and outline an agenda that would aid in building the evidence base and realizing the promise of SSIs for EDs. This commentary details three additional key issues that need to be addressed in order to realize the full potential of SSIs and related approaches and ultimately decrease the public health burden of EDs. These include conducting work to optimize interventions for greatest effectiveness, recognizing the value and working to massively increase reach of interventions like SSIs that can scale and meet diverse needs, and engaging in the work needed to address structural barriers to widespread dissemination of these approaches. Through this agenda, we will do more than embrace a single-session “mindset” and will catalyze the work needed to disseminate SSIs and related approaches at massive scale and maximize their impact.

Eating disorders (EDs) are common, disabling, and costly; yet, less than 20% of those with EDs receive treatment. EDs have also skyrocketed in the COVID-19 pandemic, with access to care worse than ever, further solidifying the need to not only make EDs a priority but also embrace new approaches to address this major public health problem. Schleider, Smith, and Ahuvia¹ argue for the single-session intervention (SSI) as one such option, which they define as a “one-at-a-time” approach, whereby any individual session yields potential for meaningful change. SSIs may be delivered in multiple ways, such as by trained providers or as digital self-help, for various purposes, including facilitating access to treatment or teaching an intervention strategy. There is preliminary indication that SSIs may hold promise for reducing some ED symptoms and risk factors, including restrictive eating and negative body image, and Schleider et al.¹ outline an agenda that would aid in building this evidence base. This includes testing whether SSIs for other mental health concerns also reduce ED symptoms; developing SSIs that directly target EDs; studying diverse implementation pathways; capitalizing on SSIs’ transdiagnostic utility to broaden funding opportunities; and educating EDs researchers and clinicians about SSIs. In this commentary, I would like to extend this agenda and detail three other key issues that need to be tackled in order to realize the full potential of SSIs and related approaches and ultimately decrease the public health burden of EDs.

1. Optimizing SSIs and other EDs interventions for greatest effectiveness

As acknowledged by Schleider et al.,¹ among those who begin psychotherapy, the most common number of sessions is just one. Even engagement in digital mental health, which circumvents some barriers individuals encounter in engaging with psychotherapy like distance and sessions being offered at inconvenient times, is typically very brief. For these reasons, I agree with Schleider et al.¹ that interventions, including SSIs, should be optimized for effectiveness by targeting the most critical mechanisms. Indeed, our team has always “front-loaded” our digital EDs programs with content to address what is believed to be the most important targets.

Interventions can and should be optimized based on data though, but there is currently a dearth of research on intervention mediators in the EDs field to inform these decisions. Even for cognitive-behavioral therapy (CBT), which is recommended by treatment guidelines across the globe, there is little information on which strategies are active ingredients, and of those, which have the biggest impact. To further complicate matters, the answers to these questions may differ for specific subpopulations. For example, it is possible that the most critical targets may depend on demographic characteristics, length of illness, or other individual or contextual factors.

As such, I argue there is work to be done to determine the foci of SSIs that would have the biggest impact. While this information should be considered especially important for informing development of brief interventions, this work is also critical for the broader field and for informing the development and refinement of more traditional interventions. Indeed, better understanding the most effective strategies vs those that may be inert or even unhelpful will allow for the creation of more robust and efficient treatments for EDs, which is critical given that even our best treatments are in need of improvement.

2. Increasing reach of interventions that can both massively scale and meet user needs

Skeptics of SSIs and other brief interventions for EDs may argue that it seems unlikely that a single session is capable of achieving sustainable benefits for an individual with a serious mental illness. Schleider et al.¹ are clear that they are not proposing SSIs are equally capable of ameliorating ED symptoms vs longer approaches across all diagnoses, severity levels, and contexts. While individual benefits are of course of the utmost importance, I argue that, at the same time, we must keep the reach and public health impact of our interventions top of mind. When considering an intervention from that angle, incremental improvements in effect size have less impact at a population level than increasing reach. For example, as demonstrated by Moessner and Bauer, an increase of 10% in intervention utilization (from 10%) would decrease the number of ED cases in a population by an additional 5%, whereas an improvement in treatment efficacy of 10% (from 10%) would only reduce the number of cases by 2%.² This work thus clearly shows that it is of the utmost importance to increase reach of interventions to address EDs, even when those interventions (e.g., digital interventions, SSIs) are less effective on the individual level than gold-standard psychotherapy approaches, as their potential for public health impact is so great. The scale of, for example, a digital SSI without human support is nearly limitless, assuming availability of funding for hosting/maintenance. In contrast, given the shortage of mental health professionals, let alone those who specialize in EDs, we could unfortunately never serve all those with EDs with traditional, one-on-one psychotherapy, even if they all needed and wanted this type of treatment.

Indeed, in considering this issue, we must acknowledge the heterogeneity of EDs, as well as the heterogeneity in the type of care for these problems that is needed, wanted, or can even be accessed. EDs span the continuum from an individual experiencing early warning signs to an individual with severe and enduring anorexia nervosa who has been in numerous types and levels of treatment over the course of years. Correspondingly, we should have different types and levels of care that can be offered to these different populations. At the same time, there is variability in what kind of care individuals want, based on such factors as motivation and individual preferences for how to address mental health concerns, as well as the type of care they can access, based on factors like insurance status, means to pay for care, and availability or proximity of a provider. These realities cannot be ignored, and even when “we” believe it would be “best” for a patient to receive gold-standard, individual psychotherapy for their ED, this may not be the type of care they want or even could receive.

Thus, SSIs and other brief interventions have the ability to scale in a way that could never be achieved by individual psychotherapy and also represent an important care offering that may meet an individual’s unique needs. We need to acknowledge these strengths and work to massively increase reach of such approaches, as well as reach of other interventions that can scale (e.g., those that can be delivered by paraprofessionals). This is in no way an either-or situation though, and we as a field should work to increase access to the full range of services for EDs, from prevention to treatment to relapse prevention, and that vary in intensity, from self-help SSIs to guided self-help delivered by a paraprofessional to multidisciplinary treatment team-based care and higher levels of care. It is this broad spectrum of services that is needed to meaningfully address the problem of EDs from a public health perspective, meet the needs of consumers, and ultimately decrease their incidence and prevalence in a population.

3. Overcoming barriers to increasing availability of SSIs and other scalable approaches

Even if the field adopts a “single-session mindset,” there are major barriers that exist in dissemination and implementation. Many healthcare systems are currently most accustomed to rollout and payment for more traditional mental health services. It may also be difficult to obtain payment for these types of brief interventions when they are for prevention or designed to be an intermediary step toward accessing another type of care—roles they may be particularly well suited for. Furthermore, in the ideal scenario, the range of intervention options for EDs would be rolled out as an organized system of care, whereby individuals could be easily offered various options, monitored for progress, and then given recommendations to step up to more intensive services as needed. This is challenging in fragmented healthcare systems, as exists in the U.S., but may be more feasible in countries with more cohesive systems.

To date, there has been a gap in the research-to-practice translation of evidence-based digital interventions for EDs, in part because of the aforementioned barriers. In order to make SSIs and other scalable approaches for EDs widely available, structural barriers will need to be addressed, and we in the field need to be advocates for change. For example, advocating for insurance reimbursement for digital interventions as well as demonstrating the potential return on investment for organizations when they invest in preventive SSIs or those that help individuals connect with care. At the same time, partnerships across sectors, such as with industry and non-profit organizations, will also need to be harnessed to fully realize the potential of these interventions and make them available at scale. It is notable that SSIs and other scalable approaches may be particularly well suited to addressing other structural barriers to mental health care as well, including interventions not being available in the language or literacy level of the consumer.

4. A case example

Since 2017, our team has worked with the National EDs Association (NEDA), the largest non-profit organization dedicated to EDs in the U.S., to disseminate our evidence-based online screen, which garners ~200,000 respondents annually and thus represents an extraordinary opportunity to help hundreds of thousands of individuals each year. While most respondents (86%) screen positive for an ED, only 14% have previously received and only 3% are currently receiving treatment.³ Findings further suggest that <16% initiate care following screening, despite being provided with a wealth of information on how to find treatment.⁴ These data made clear additional tools were needed to increase service utilization, but what was not clear was what strategies might be most effective.

In collaboration with NEDA and Cass (previously called X2AI), a mental health chatbot company, we developed a chatbot named Alex designed primarily as an SSI, for use by those identified by the NEDA screen as having an ED but reporting not being in treatment.⁵ Alex was designed to deliver four theoretically-informed components to target service use: psychoeducation, motivational interviewing, and personalized services recommendations based on preferences—all intended to be delivered as an SSI immediately following screening—along with repeated check-ins to ask about service seeking and offer support and troubleshooting in the weeks following screening. Alex underwent an iterative, user-centered design process, in order to ensure the chatbot would meet the needs of its diverse end-users, and overall, participants reflected very positively on the bot.⁵ We wanted to know, though, which components of the chatbot may be uniquely important in increasing services use, and so we conducted an optimization trial, randomizing participants using a factorial design to a chatbot with some combination of those elements (i.e., randomized to have each turned on or off). Results are forthcoming and will enable us to ultimately work with NEDA and Cass to disseminate a version of the chatbot that only includes effective components, ultimately making it as efficient as possible and ensuring no time is “wasted.” This will increase the likelihood individuals complete the intervention and engage with care, which has huge potential for impact when considered over the hundreds of thousands of NEDA screen completers who could use this intervention each year. In the future, the chatbot could also be easily adapted for delivery in multiple languages.

5. Conclusions

As detailed here, there are three key issues, beyond those detailed by Schleider et al., that need to be addressed in order to realize the full potential of SSIs and related approaches and ultimately decrease the public health burden of EDs. These include conducting work to optimize interventions for greatest effectiveness, recognizing the value and working to massively increase the reach of interventions like SSIs that can scale and meet diverse needs, and engaging in the work needed to address structural barriers to widespread dissemination of these approaches. Through this agenda, we will go beyond a single-session “mindset” and will catalyze the work needed to disseminate SSIs and related approaches at massive scale and maximize their impact.

Data Availability Statement:

Data sharing is not applicable to this article as no new data were created or analyzed in this study.