Digital Interventions for Adults With Symptoms of Depression and Children and Adolescents With Symptoms of Obsessive-Compulsive Disorder

JAMA. 2021;325(18):1839-1840. doi:10.1001/jama.2021.5047

The need for accessible and effective mental health interventions remains a global health priority, with internet and smartphone technology offering a feasible means to deliver care. The global pandemic has accelerated interest in, use of, and support for digital mental health solutions. Thus, 2 articles in this issue of JAMA offer relevant and timely evidence for offering digitally enabled care and highlight emerging frontiers for the behavioral health field., Both articles present an impressive assessment of digital approaches to improve access and outcomes and use high-quality methodology (randomization, control groups, multisite) that is often lacking in studies evaluating these type of digital interventions. Both reports move beyond the simple question of accessibility and feasibility and address important questions about effectiveness and implementation.

Although the studies are seemingly different, they share important commonalities. Araya et al examined how a smartphone app could deliver behavioral activation to adults with depressive symptoms and medical comorbidities across study sites in Peru and Brazil. In 2 randomized trials involving 880 adults in Brazil and 432 in Peru, participants were randomized to receive 6 weeks of the study digital intervention or enhanced standard care. The primary outcome was a reduction of at least 50% from baseline in the Patient Health Questionnaire-9 score (range 0-27, higher scores indicate more severe depression) at 3 months. A greater percentage of participants in the digital intervention group, compared with the enhanced usual care group, experienced a reduction in the Patient Health Questionnaire-9 score of at least 50% at 3 months in Sao Paulo (40.7% vs 28.6%) and in Lima (52.7% vs 34.1%). However, at 6 months there was no statistically significant difference in the primary outcome between the 2 groups.

Aspvall et al examined how a computerized cognitive behavioral therapy (CBT) program, when used as part of a stepped-care model, could improve symptoms in children and adolescents with obsessive-compulsive disorder (OCD) in Sweden. In this randomized noninferiority clinical trial of 152 children with OCD (mean age, 13.4 years), participants were offered traditional in-person care delivered by a specialist or a stepped-care model with 16 weeks of internet-delivered CBT followed by traditional in-person specialist care if needed. At 3 months, after completion of the first treatment phase, 46% of participants in the stepped-care model were classified as nonresponders and received traditional care compared with 30% of participants in the traditional care model, who were also offered further care. The primary outcome was the clinician-rated Children’s Yale-Brown Obsessive-Compulsive Scale score (range, 0-40; higher scores indicate greater severity). At 6 months, those receiving stepped care had a numerically similar reduction in symptoms to those receiving traditional in-person care, with mean Children’s Yale-Brown Obsessive-Compulsive Scale scores of 11.57 vs 10.57 and a difference of 0.91 points (1-sided 97.5% CI, −∞ to 3.28), which met the noninferiority criterion of 4 points.

Although these 2 studies featured different countries (high-income vs low-/middle-income), different digital technology approaches (self-help app vs computerized stepped care), different diseases (depression vs OCD), and different ages (children and adolescents vs adults), they shared underlying commonalities around the universal application of digital mental health, including the importance of human support, the need to monitor and assess digital interventions, and the need for assessment in the clinical population.

The studies were conducted in different continents and across 3 countries that span the spectrum from overall high to low access to mental health services. Yet, even in Sweden, where the population is offered relatively higher access to free mental health compared with lower access in Brazil and Peru, there is a need for more comprehensive services. The findings of these 2 studies suggest that digital mental health could help fill critical gaps in specialty care (CBT for OCD for children and adolescents in Sweden, as reported by Aspvall et al) while filling the need for broader care services (behavioral activation for depression in adults in Brazil and Peru, as reported by Araya et al). The ability of new digital offerings to serve diverse needs is a notable benefit of technology-based approaches; however, this also raises concerns around serving diverse populations. Ensuring access to devices, as illustrated by the offer of free smartphones to the intervention cohort in the study by Araya et al, as well as digital health literacy and skill training are critical for equitable care and to avoid a “digital divide.”,

Although the interventions in each study were unique, they offer a window into the role of human support in facilitating engagement and outcomes with digital health. Authors of both studies noted that their digital approach offers scalability through reduced reliance on human support, yet humans still remained a critical factor “in the loop” for both. Araya et al offered a comparison in their app study because the Lima cohort received access to nursing support and the San Paulo cohort did not. Without this supportive accountability offered by nurses who would check in on those with less engagement with the app, the San Paulo cohort only had 65% adherence, while the Lima cohort, with support, had 92% adherence. The Aspvall et al study offered child and adolescent therapist contact as well as on-demand telephone support during the computerized CBT program targeting OCD symptoms. Although data were not presented examining associations between therapist contact/support and success with the computerized CBT program, there is mounting evidence that this type of human support is critical for improving outcomes. In both studies, the level of human support was clearly less than in traditional care settings, and therefore represents increased access to care. But the question today is to what degree is human support necessary? The answer will influence the related questions of cost-effectiveness and of tangibly increased access to care.

Both studies also offer behavioral treatments that are low risk and thus may not be considered a high priority for clinician monitoring. The US Food and Drug Administration does not intend to closely regulate software offering CBT or behavioral activation skills given the low risk posed to patients. However, both studies show that low risk does not mean that there is no risk or need to follow and monitor patient progress. Araya et al reported worsening of depression in 10% of patients in the behavioral action smartphone app group (compared with 12% in the control group), reflecting that any treatment will have nonresponders and those who may even experience clinical deterioration. Aspvall et al reported increased anxiety and depressive symptoms as the most common adverse events in both groups, with the rate of adverse events being similar between the groups (64% for stepped care and 67% for traditional care).

Behavior change is difficult; not every patient is expected to respond to treatment, and some will even feel worse after treatment. Offering therapies via technology does not change concerns about potential adverse effects, necessitating active monitoring and the capacity to respond to patient needs even for these lower-risk treatments. Aspvall et al offer a practical solution with their stepped care model, in that patients who were not responding to the technology-delivered care were then offered more traditional therapy and closer monitoring. Such blended approaches to mental health technology continue to gain traction given advantages for both safety and engagement.

Both studies reported successful mental health outcomes that highlight the potential of their digital approaches. However, in the study by Araya et al, even though it was noted that there were statistically significant beneficial outcomes in the intervention group, compared with the control group, at 3 months (but not 6 months), the authors discuss what outcomes and degree of clinical improvement are “worth pursuing.” The translation of improvements on symptom scales to clinically meaningful change for patients is a challenge not unique to digital mental health approaches, but one for which this field has the opportunity to bring innovation. New study end points around sleep, physical activity metrics (eg, step count), cognition measures of attention and memory, dynamic symptom trends, and functional outcomes in nonresearch settings are easily accessible via smartphones. Thus, the same devices offering digital intervention present the opportunity for longitudinal, temporally dense, and multimodal assessments at no extra burden or cost. These enhanced end points beyond simple symptom reporting may help elucidate mechanism of actions of digital treatments and identify personalized outcomes of most interest to each patient. Such reporting could help explain why effects of the smartphone-based behavioral activation in the studies by Araya et al were not sustained at 6 months and potentially could be used to offer customized targeted “booster” sessions in response to patients’ evolving needs. This type of precision psychiatry is feasible today and will likely first be realized with digital interventions such as those presented in these 2 reports.

The 2 studies reported in this issue of JAMA are timely contributions that underscore how digital interventions can help patients today. They also highlight how high-quality research can advance digital health science and raise the next generation of questions to make digital approaches even more effective and clinically important.

Article Information

Corresponding Author: John Torous, MD, Beth lsreal Deaconess Medical Center, 330 Brookline Ave, Rabb 2, Boston, MA 02215 ([email protected]).

Conflict of Interest Disclosures: None reported.

References

7.Enforcement Policy for Digital Health Devices for Treating Psychiatric Disorders During the Coronavirus Disease 2019 (COVID-19) Public Health Emergency: Guidance for Industry and Food and Drug Administration Staff. US Food and Drug Administration; 2020.

Source://jamanetwork.com/journals/jama/fullarticle/2779843?guestAccessKey=d65ef722-3b4c-4318-8689-2c06891340c5&utm_source=silverchair&utm_medium=email&utm_campaign=article_alert-jama&utm_content=etoc&utm_term=051121