Open any app store and mental-health support appears abundant: mood trackers, meditation libraries, chatbots offering cognitive behavioural techniques, sleep coaches and crisis lines, many free and all available at 3 a.m., when human services mostly are not. Given long waits for therapy in many countries, the appeal is obvious. So is the question regulators and researchers keep asking: which of these apps actually help, which merely feel helpful, and which could do harm?
The honest answer is uneven. the World Health Organization has recognised digital tools as one way to widen access to mental-health support, while warning about quality and safety gaps. Our evidence check, summarised here and tracked across our homepage page and Blog coverage, sorts the claims into what studies support, what remains unproven and what should make any user pause before downloading.
Where the Evidence Is Encouraging
The strongest support belongs to apps that deliver structured, recognised therapy — especially cognitive behavioural therapy for insomnia, anxiety and mild-to-moderate depression — and that keep a human in the loop somewhere, even lightly. Trials of such guided programmes show modest but real benefits over waiting lists, which is a low bar that still matters for people who would otherwise receive nothing. Unguided versions show smaller effects and much higher abandonment, a pattern so consistent it is practically the field’s law: people stop using mental-health apps, usually within weeks.
Meditation and mindfulness apps sit a step below: trials suggest small average improvements in stress and wellbeing for consistent users, with quality varying and publication bias flattering the results. Mood-tracking alone shows little direct benefit, though some users value the self-knowledge. The overall picture is not that apps are useless; it is that the average effect is small, the best designs borrow from real therapy, and engagement — the thing apps are worst at — is where benefits live or die.
Where Claims Collapse
Chatbots are the fastest-growing and least-proven category. Scripted companions that simulate empathy can comfort in the moment, and some users genuinely value a non-judgemental ear at night. But large language models can also affirm delusions, miss crisis signals, give confident nonsense and create emotional dependence by design. Several documented cases of chatbots responding badly to users in crisis have pushed regulators to demand clearer limits: these are not therapists, and apps implying otherwise are making a claim evidence cannot carry.
Privacy is the other collapse zone. Investigations of popular mental-health apps have repeatedly found sensitive data — mood logs, journal entries, even session content — flowing to advertisers and analytics firms under vague consent. For a category whose entire value rests on trust and candour, that finding is damning. An app that helps your anxiety while selling your diary has not solved your problem; it has monetised it.
Choosing and Using Apps Safely
Practical screening helps. Favour apps linked to health services, universities or published trials; read the privacy policy’s data-sharing section, not its promises; be wary of subscriptions that are easy to start and hard to cancel; and treat any app that diagnoses, or claims to replace treatment, as a warning sign. If you are in crisis, an app is not the right tool at all — contact local emergency services or a crisis line and a human being.
Used with open eyes, the best apps earn a modest place in mental-health care: between nothing and therapy, for mild difficulties, as structured self-help with human backup. That is less than the marketing promises and more than cynics allow. As evidence standards tighten — several countries now certify apps for prescription — the gap between the helpful minority and the hopeful majority should finally become visible before you download, not after.
