Articles

The mental health app market is large and growing. Estimates consistently put the number of available apps in the thousands, and downloads run into the hundreds of millions globally. That scale is either reassuring or alarming depending on what you think the evidence actually supports.

This article tries to answer that question honestly. The short version: there is genuine evidence that certain app-delivered techniques can reduce symptoms of anxiety and depression — but the effects are modest, the research has real problems, and the difference between an app built on an evidence-based technique and an app that is itself evidence-based is a distinction almost no app makes clearly. It matters, and you deserve to understand it.

What the meta-analyses actually show

Several large reviews have pooled the results of randomised controlled trials testing smartphone-delivered psychological interventions. A 2017 meta-analysis published in World Psychiatry by Linardon and colleagues, and a widely-cited 2018 review in npj Digital Medicine by Firth and colleagues, both found small-to-moderate effect sizes for reducing symptoms of depression and anxiety compared with control conditions. [1] [2]

A Cochrane-style reading of that finding would note: small-to-moderate is not nothing. It is roughly in the range of what you see for many low-intensity psychological interventions delivered face-to-face. But it is also not the same as structured therapy with a trained clinician, and the reviews are careful to say so.

One consistent finding across reviews is that guided interventions outperform unguided ones. When a human — a therapist, a coach, even a trained peer supporter — checks in on your progress, outcomes improve meaningfully. Fully self-directed apps tend to show smaller effects. [1]

The problems the headlines leave out

The meta-analytic picture is more complicated than “apps work” or “apps don’t work.” Three specific problems are worth understanding.

Small samples and publication bias

Many of the trials feeding into these reviews are small. Small trials are more likely to produce inflated effect sizes by chance, and positive results are more likely to be published than null results. The Firth et al. review explicitly flagged high risk of bias in a significant proportion of included studies. [2] This does not invalidate the findings, but it does mean the true effect is probably smaller than the headline number suggests.

The engagement cliff

Research on digital health engagement consistently finds that most users stop using mental health apps within the first few weeks. A 2019 study in JMIR Mental Health found that average retention for mental health apps drops sharply after the first two weeks of use. [3] This matters enormously for interpreting trial results: participants in a study are not typical users. They were recruited, often screened, and frequently prompted to continue. Real-world effectiveness is almost certainly lower than trial effectiveness.

Heterogeneity

The category “mental health app” contains everything from structured, session-based cognitive behavioural therapy programmes to ambient soundscapes to chatbots with no theoretical grounding. Pooling them into one meta-analysis and concluding “apps work” is like pooling aspirin trials with homeopathy trials and concluding “tablets work.” The technique matters, not just the delivery channel.

Evidence-based technique vs. evidence-based product

This is the distinction that almost no app marketing makes clearly, and it is the most important one for a potential user to understand.

Cognitive behavioural therapy (CBT) has an extensive evidence base accumulated over decades of clinical research. [4] Dialectical behaviour therapy (DBT), developed by Marsha Linehan, has strong evidence for distress tolerance and emotional regulation. [5] Paced diaphragmatic breathing has documented physiological effects that are measurable in controlled conditions. [6] Affect labelling — putting emotions into words — has been shown in neuroimaging research to modulate emotional responses. [7]

An app that implements these techniques faithfully can legitimately say it is built on evidence-based techniques. That is a real and meaningful claim.

It is not the same as saying the app itself has been through a randomised controlled trial, demonstrated efficacy at a statistically significant level, and had those results replicated by an independent research group. Very few apps have done that. Most have not. And the ones that claim otherwise are usually conflating the two.

What to look for when evaluating an app

You do not need to read primary literature to make a reasonable judgement about an app. The following questions are practical and answerable.

• Does it name its technique? An app that cannot tell you whether it is using CBT, DBT, mindfulness-based stress reduction, or something else has no theoretical grounding you can evaluate.
• Does it cite anything? Not a marketing claim — an actual reference to a research tradition or specific body of work. Absence of this is a significant red flag.
• Does it signpost crisis support? No self-help app is appropriate for someone in acute crisis. An app that does not acknowledge this and point you elsewhere is either negligent or dishonest about its scope.
• What does it do with your data? Mental health data is among the most sensitive data you can share. A credible app will tell you clearly whether data leaves your device, who can access it, and under what legal framework it operates.
• Does its business model reward your leaving or your staying? Apps optimised for engagement — streaks designed to create anxiety about breaking them, notifications engineered to pull you back — may not have your wellbeing as their primary design goal. This is worth thinking about carefully.

What the evidence does support, plainly stated

Structured self-help based on CBT principles reduces symptoms of mild-to-moderate anxiety and depression in adults, with effects that are modest but consistent across multiple reviews. [1] [2] Paced breathing techniques have documented effects on autonomic arousal. [6] Grounding techniques derived from DBT are used in clinical practice for distress tolerance, though app-delivered versions have not been independently trialled at scale. [5] Tracking mood over time — ecological momentary assessment — has research support as a self-monitoring tool. [8]

What the evidence does not support is the idea that any app is a substitute for professional care when professional care is what someone needs. The NHS and NICE are explicit on this point: digital tools may complement treatment but are not a replacement for it. [9]

Where MoodFire fits — honestly

MoodFire is a self-help app built on techniques drawn from CBT and DBT. Its Reframe tool implements structured cognitive restructuring based on Beck’s CBT model, including cognitive distortion detection — the same theoretical framework that underlies evidence-based therapy. [4] Its Ground tool uses the 5-4-3-2-1 sensory grounding exercise derived from DBT distress tolerance practice. [5] Its Breathe tool includes a coherent 5.5-second paced breathing rhythm, which corresponds to the resonance frequency breathing studied in the heart rate variability literature. [6] The Check In feature is grounded in affect-labelling research. [7]

MoodFire has not been through a clinical trial. Its efficacy as a product has not been independently validated. That is the honest statement, and it applies to the vast majority of apps on the market. The difference is that this article is saying it plainly rather than hoping you won’t ask.

What MoodFire does tell you: which techniques it uses, where those techniques come from, that it is not therapy, and that your health data is processed under explicit consent in line with GDPR. The app is designed around behavioural change principles rather than engagement optimisation — its streak feature includes a 36-hour grace period specifically to avoid the anxiety-inducing effect that rigid streaks can create. The Insights view lets you see your own mood patterns over time, including an optional correlation view with biometric data from Apple HealthKit or Android Health Connect if you choose to connect them. What those patterns mean is for you — and, if relevant, a clinician — to interpret.

If you want to explore the CBT and DBT techniques the app is built on before deciding whether to use it, the articles below go into more depth on the underlying evidence. If you are experiencing significant symptoms of depression or anxiety, please speak to a GP or mental health professional rather than relying on any self-help tool as a primary intervention.