A new study has found that a therapist-guided smartphone app designed to treat depression contains all 16 core techniques associated with cognitive behavioural therapy, offering fresh evidence that digital mental health tools can deliver structured, clinically grounded care to people who might otherwise struggle to access it.
Depression affects more than 300 million people worldwide and remains one of the leading causes of disability. Despite the availability of effective treatments, barriers including long NHS waiting lists, clinician shortages, and the cost of private therapy mean that many people go without support. Smartphone-based interventions have emerged as a promising alternative, yet questions have persisted about whether these apps genuinely reflect the evidence base or simply borrow the language of CBT without the substance.
The study, published in the Journal of Medical Internet Research, analysed every page of Mindset, an app developed by Koa Health and delivered alongside weekly 15 to 20 minute video sessions with a licensed clinical psychologist. Researchers from Massachusetts General Hospital and Harvard Medical School used a dual-coding approach they termed “mechanism mapping”, pairing established CBT technique definitions with a validated taxonomy of 93 behaviour change techniques to assess not just what the app contained but how each strategy was put into practice.
Across its eight modules, the app deployed CBT techniques a total of 528 times. Psychoeducation was the most frequently occurring element, appearing on more than half of all app pages. Skill building and cognitive restructuring were also prominent, particularly in the first module, while activity scheduling was woven consistently across the later stages of the programme in support of behavioural activation, a cornerstone of modern depression treatment.
The researchers also found 37 distinct behaviour change techniques across 13 categories, coded 878 times in total. The most commonly used categories were shaping knowledge, repetition and substitution, and feedback and monitoring. These techniques were not merely decorative features; they were directly tied to therapeutic skill enactment, meaning that self-monitoring prompts and instructional content functioned as genuine vehicles for clinical change rather than engagement gimmicks.
The co-occurrence analysis revealed meaningful variation in how deeply different CBT strategies were behaviourally reinforced. High-frequency techniques such as psychoeducation and activity scheduling were supported by multiple behaviour change mechanisms layered across the intervention. By contrast, mindfulness and guided imagery appeared infrequently and with minimal scaffolding, suggesting these elements were introduced but not systematically practised.
The authors argue that most existing mental health apps function as “black boxes”, making it impossible for clinicians or patients to assess what therapeutic content is actually being delivered. In a previous open trial of Mindset involving 28 participants, clinician-rated depression severity fell significantly from baseline to post-treatment, with gains maintained at three-month follow-up.
The findings highlight a persistent gap in the digital mental health landscape, where many commercially available apps carry the label “CBT-based” despite offering only a handful of evidence-based elements. The researchers call for mechanism mapping to become a standard method for evaluating digital mental health tools, enabling clinicians to make recommendations grounded in documented content rather than marketing claims.