Can Smartphone CBT Prevent Major Depression? What the RESiLIENT Trial Adds
Why Prevention Is The Right Frame
Digital mental health tools are often judged as if they were trying to replace clinicians. That is the wrong starting point for the RESiLIENT trial. The more useful question is not whether a smartphone app can treat severe major depression as well as psychotherapy, medication, or combined care. The better question is whether structured cognitive and behavioural skills, delivered through a phone, can reduce the risk that subthreshold depression becomes major depression.
Subthreshold depression is clinically important precisely because it sits between wellness and disorder. People in this group have depressive symptoms, but they do not meet full criteria for a major depressive episode. Some will improve without formal treatment. Others will remain symptomatic, lose functioning, or develop major depression over time. In public-health terms, this is a high-yield prevention group.
That makes smartphone CBT interesting in a practical, less glamorous way. It is portable, low-intensity, and potentially scalable. It can reach people who would not seek therapy, could not afford regular sessions, or would not yet be referred to specialty care. Its promise is not dramatic transformation. Its promise is earlier skill-building before symptoms harden into a more disabling episode.
RESiLIENT should therefore be read as a prevention study, not as another app-hype story.
What the RESiLIENT Trial Actually Tested
RESiLIENT was a 50-week follow-up analysis of smartphone CBT randomized trials in adults from the general population with subthreshold depression. The study used a master protocol with four 2 × 2 factorial trials and randomized 3,280 participants to one of nine intervention arms or a self-check control group. Follow-up at week 50 was high, at 89%.
The interventions were not vague wellness prompts. They were specific CBT skills delivered through a smartphone app: behavioural activation, cognitive restructuring, problem solving, assertion training, and behavioural therapy for insomnia. Some arms used a single skill; others used combinations. This component-based design is one of the trial’s strengths. CBT is often discussed as one package, but it contains different active ingredients. Behavioural activation encourages people to re-engage with rewarding or meaningful activity. Cognitive restructuring targets negative automatic thoughts and distorted interpretations. Problem solving teaches a structured way to handle stressors. Assertion training focuses on communication and interpersonal agency. Behavioural therapy for insomnia addresses sleep patterns, a major risk factor and maintaining factor for depression.
The primary outcome was time to onset of a major depressive episode by week 50. It was assessed using a smartphone-based version of the depression section of the WHO Composite International Diagnostic Interview. Secondary outcomes included total burden of depression across the follow-up period and changes in PHQ-9 scores.
This design matters because prevention cannot be captured only by symptom severity at one time point. A person who avoids major depression for most of a year, or spends fewer weeks with depressive symptoms, may experience a meaningful benefit even if their final PHQ-9 score is only modestly different. RESiLIENT measured both categorical onset and depressive burden over time.
What the Results Add
The primary results favored smartphone CBT. Hazard ratios for the interventions ranged from 0.52 to 0.63 compared with self-check control. The strongest preventive effect was seen with behavioural activation plus assertion training, with a number needed to treat of 23.3. Behavioural therapy for insomnia, behavioural activation plus behavioural therapy for insomnia, and cognitive restructuring also showed strong preventive signals.
When all active intervention arms were compared with the self-check control group, the hazard ratio was 0.66. In plain terms, smartphone CBT skills were associated with a lower risk of developing major depression over 50 weeks.
The secondary outcomes also support the prevention frame. All interventions reduced total burden of depression compared with control. Behavioural activation plus cognitive restructuring showed the largest reduction in depressive burden and the greatest improvement in PHQ-9 change at week 50. Effect sizes at week 50 ranged from small to modest, approximately −0.34 to −0.07, depending on the intervention.
Those effect sizes should not be oversold. This is not a story about an app producing a dramatic clinical remission effect in severe depression. The gain is more incremental. But incremental effects can be meaningful in prevention if the intervention is safe, low-cost, scalable, and delivered to people at elevated risk. The total-burden measure is especially useful. Major depression is a categorical diagnosis, but depressive suffering is lived across days and weeks. Two people may both avoid a formal major depressive episode, but one may spend much of the year with persistent symptoms while the other experiences only brief periods of low mood. A prevention trial that tracks cumulative symptom burden is closer to what patients actually experience.
Safety findings were reassuring. No serious adverse events were reported, although suicidality alerts were triggered in a minority of participants. That detail is important: even low-intensity digital interventions need monitoring pathways when depressive symptoms or suicidal thoughts emerge.
Why Smartphone CBT Could Be Useful At Scale
Depression prevention requires reach. Traditional psychotherapy is effective, but it is limited by workforce shortages, cost, geography, stigma, scheduling, and patient readiness. Many people with subthreshold depression will not enter formal care, especially if they view their symptoms as “not serious enough” or assume they should manage alone.
Smartphone CBT fits a stepped-care model. It can offer structured skills early, before the patient needs more intensive treatment. It can also help normalize psychological self-management: scheduling activity, challenging depressive thinking, improving sleep routines, solving problems in smaller steps, and communicating more effectively.
The trial’s modular findings are clinically interesting. Behavioural activation plus assertion training performed strongly for prevention of major depression. Behavioural therapy for insomnia also stood out, both alone and in combination. Behavioural activation plus cognitive restructuring was strongest for reducing total depressive burden. This suggests that not all CBT skills operate the same way. Sleep-focused behavioural therapy may reduce risk by stabilizing a biological and behavioural vulnerability. Behavioural activation may interrupt withdrawal and inactivity. Cognitive restructuring may reduce depressive rumination and hopeless interpretations. Assertion training may improve interpersonal stress handling. The practical value is not only that smartphone CBT can work, but that different skills may serve different preventive functions.
This does not mean an app should replace clinicians. It means digital CBT could become one layer in a broader prevention system: self-guided first-line support for people with subthreshold symptoms, with escalation to therapy, medication, or urgent care when symptoms worsen.
What the Trial Does Not Prove
RESiLIENT does not prove that any commercial mental health app can prevent depression. The intervention was structured, research-tested, and built around defined CBT components. Most apps on the market have not been evaluated with comparable rigor.
The study also does not apply equally to all clinical populations. Participants were adults from the general population with subthreshold depression. The findings should not be generalized to people with severe major depression, bipolar disorder, psychosis, high suicide risk, complex trauma, substance use disorder, or severe functional impairment without additional evidence. Those patients need clinical assessment, not only self-guided digital support.
Several implementation questions remain. Engagement can decline over time. Digital access is unequal. Privacy and data governance matter. Some users may need human support to use CBT skills effectively. Others may require monitoring when suicidal thoughts appear. The durability of benefit beyond 50 weeks also remains uncertain.
Still, RESiLIENT adds something practical to the depression-prevention conversation. It shows that structured smartphone CBT skills can reduce the risk of major depression and lower cumulative depressive burden over nearly a year in adults with subthreshold symptoms.
The conclusion should be modest but positive. Smartphone CBT is not a replacement for treatment when major depression is already present. It is not a universal solution to the depression burden. But as a preventive tool, it may be useful precisely because it is brief, portable, and scalable. Prevention rarely looks dramatic. If it keeps some people from crossing the threshold into major depression, that is already a meaningful clinical and public-health gain.
References
- Akechi, T., Noma, H., Tajika, A., Toyomoto, R., Sakata, M., Luo, Y., Horikoshi, M., Kawakami, N., Nakayama, T., Kondo, N., Fukuma, S., & Furukawa, T. A. (2026). Cognitive and behavioural skills to prevent major depression among adults with subthreshold depression: 50-week follow-up analysis of smartphone CBT randomised trials (RESiLIENT trial). The British Journal of Psychiatry. Advance online publication. https://doi.org/10.1192/bjp.2026.10630
- The Effectiveness and Limitations of Mental Health Apps