The study, led by Michael J. Mason of the University of Tennessee’s Center for Behavioral Health Research, recruited 102 young adults between the ages of 18 and 25 through social media. This age band matters epidemiologically: according to federal survey data cited by the team, more U.S. young adults report moderate to severe symptoms of generalized anxiety disorder and major depressive episodes than any other age group. The comorbidity figures are striking. Almost half of people with major depressive disorder will also experience generalized anxiety disorder, and roughly two-thirds of those with generalized anxiety disorder will experience major depression. Any treatment that can reach both conditions with a single, low-cost intervention could reshape how this population is served.

The trial design was a three-month randomized clinical trial comparing the automated texting program against a wait-list control condition. Participants in the treatment arm received CBT-txt-A, a program that had already demonstrated efficacy in treating generalized anxiety disorder in the team’s earlier work. Depression symptoms were measured with the PHQ-8, an eight-item instrument widely used to assess current depression in general populations. Using repeated measures analysis of variance, the researchers found significant differences between the treatment and wait-list conditions on depressive symptoms over time, producing a medium treatment effect of Cohen’s d equal to 0.61. In practical terms, a medium effect means the difference between groups was large enough to be clinically meaningful, not merely a statistical artifact.

Clinical significance testing, which goes beyond statistical significance to ask whether individual participants moved into healthier diagnostic ranges, reinforced the result. At the three-month follow-up, 27 percent of participants in the treatment group had no or minimal depressive symptoms, and an additional 33 percent had only mild symptoms. For an intervention that requires no therapist, no clinic visits, and no appointments, that proportion of participants reaching minimal or mild symptom levels is notable. The team frames the program as a potential low-burden support option for young adults in settings where access to conventional treatment is limited, whether because of cost, distance, workforce shortages, or stigma.

What elevates the study beyond a simple secondary-outcomes report is its analysis of mechanism. The researchers did not merely document that depression scores fell; they tested how the treatment produced those reductions. Drawing on the tradition of mediation analysis in psychotherapy research, they hypothesized three clinical pathways through which an anxiety-focused treatment could indirectly ease depression: increased behavioral activation, reduced perseverative thinking, and reduced cognitive distortion. All three hypotheses were supported. Participants in the treatment condition showed greater gains on each of these mechanisms relative to controls, and changes in each mechanism were associated with statistically significant reductions in depressive symptoms.

Each mechanism has a well-established theoretical pedigree. Behavioral activation, the process of re-engaging with rewarding and meaningful activities, is a core component of depression treatment, and the team measured it with the short form of the Behavioral Activation for Depression Scale. Perseverative thinking, the repetitive, content-independent negative thought loop that fuels both worry and rumination, was assessed with the Perseverative Thinking Questionnaire. Cognitive distortion, the systematic errors in thinking described in Aaron Beck’s foundational cognitive theory of depression, was measured with the Cognitive Distortions Scale. The finding that an anxiety treatment moved all three levers simultaneously lends empirical weight to the generic cognitive model, which holds that cognitive and behavioral change processes cut across emotional disorders rather than belonging to one diagnosis at a time.

This transdiagnostic interpretation is where the study connects to a broader shift in clinical psychology. Increasingly, researchers argue that anxiety and depression share common underlying processes, including repetitive negative thinking, avoidance, and maladaptive cognitions, and that treatments targeting those shared processes should produce crossover effects. The current results provide precisely that kind of evidence in a digital format. Prior trials by the same group found that CBT-txt-A reduced anxiety symptoms and, in a companion analysis, that a digital depression treatment produced secondary benefits for anxiety, which the authors describe as empirical support for the generic cognitive model. The new study completes the loop by demonstrating the reverse crossover: treating anxiety secondarily benefits depression.

The delivery mechanism itself is part of the scientific story. Text-message-delivered interventions sit within a growing family of digital mental health tools, including app-supported smartphone programs and internet-delivered psychotherapy, that meta-analyses have found effective across a range of common mental health problems. But CBT-txt-A differs from app-based approaches in an important way: it is automated and arrives in the messaging channel people already check constantly throughout the day. There is nothing to download, no login to remember, and no therapist waitlist to join. For young adults, a demographic that notoriously underuses traditional mental health services, embedding evidence-based therapeutic content into an everyday communication medium may lower the barriers that keep people from seeking help.

Recruitment through social media also deserves attention as a methodological feature. Rather than waiting for distressed young adults to find a clinic, the team met them on the platforms where they already spend time. This approach raises questions about sample representativeness that the researchers will need to address in future work, and the authors note that data will be made available upon request once the research team has completed all analyses. The trial was approved by the institutional review board of the University of Tennessee and funded by the Betsey R. Bush endowment for Children and Families at Risk and Behavioral Health Research at the university’s Center for Behavioral Health Research.

The mediation findings, in particular, point toward future refinements. If reductions in perseverative thinking, cognitive distortion, and increases in behavioral activation statistically account for the improvements in depression, then digital programs could be deliberately engineered to strengthen those specific ingredients rather than simply transplanting face-to-face protocols into text format. Mediation analysis of this kind also informs dosage questions: understanding which mechanisms activate earliest and most strongly could guide how program content is sequenced. The research team’s prior work testing mechanisms of change for text-message-delivered cognitive behavioral therapy in young adult depression laid the groundwork for the current mediation models, and the convergence of results across trials strengthens confidence that these are genuine change processes rather than statistical noise.

Limitations remain. The comparison condition was a wait-list rather than an active control, which means some of the observed effect could reflect nonspecific factors such as expectation or attention, although the presence of specific, hypothesized mediation pathways argues against a purely placebo interpretation. The three-month window captures short-term benefit, and longer follow-up will be needed to determine whether gains persist. The sample of 102 participants is adequate for the mediation models used but modest in absolute terms, and replication in larger and more diverse samples will be essential. Still, the combination of a medium effect size, a meaningful proportion of participants reaching minimal symptoms, and three empirically supported mechanisms of change makes a compelling case that automated texting programs can do more than their designers intended.

The practical implications are considerable. A program that requires no clinician time can be scaled at near-zero marginal cost, and if it simultaneously treats anxiety and reduces depression, it functions as a force multiplier for stretched mental health systems. For the millions of young adults experiencing comorbid anxiety and depression who never make it to a therapist’s office, a stream of carefully constructed text messages may be the only evidence-based intervention they ever receive. The study suggests that this unlikely delivery channel, powered by the same cognitive and behavioral mechanisms that drive traditional therapy, can produce clinically meaningful change, and that the boundary between treating one disorder and treating its most common companion may be far more permeable than diagnostic manuals imply.

Subject of Research: Secondary effects and clinical mechanisms of a text-message-delivered cognitive behavioral therapy for anxiety on depressive symptoms in young adults

Article Title: Secondary Effects and Mechanisms of Digital Anxiety Treatment on Depressive Symptoms: A Randomized Clinical Trial with Young Adults

Article References: Mason, M. J., Coatsworth, J. D., Zaharakis, N., Brown, A., Russell, M., & Mennis, J. (2026). Secondary Effects and Mechanisms of Digital Anxiety Treatment on Depressive Symptoms: A Randomized Clinical Trial with Young Adults. International Journal of Mental Health and Addiction. https://doi.org/10.1007/s11469-026-01733-0

Image Credits: AI Generated

DOI: 10.1007/s11469-026-01733-0

Keywords: digital mental health, text message therapy, cognitive behavioral therapy, generalized anxiety disorder, major depressive disorder, young adults, randomized clinical trial, behavioral activation, perseverative thinking, cognitive distortion, comorbidity, CBT-txt-A