Do Mental Health Apps Work? What the Research Says and How to Choose One

Mental health apps are smartphone or tablet-based software tools designed to support emotional well-being, deliver therapeutic content, or facilitate access to licensed mental health professionals. They range from FDA-cleared digital therapeutics (prescription-grade software that has undergone clinical trials and regulatory review) to entirely unregulated wellness tools that make no clinical claims but are marketed to people experiencing depression, anxiety, and stress.

More than 10,000 mental health-related apps were available in the major app stores as of 2024, and they collectively represent a meaningful, though carefully qualified, contribution to mental health care access at a time when demand for services outpaces supply. 

The most rigorous meta-analysis published to date, covering 176 randomized controlled trials (RCTs), (the gold standard study design for testing treatments) and 33,567 participants, found that mental health apps produced small but statistically relevant effects on depression (Hedges’ g = 0.28) and generalized anxiety (g = 0.26) compared to control conditions (Linardon et al., 2024).

A more recent 2025 meta-analysis of 92 RCTs and 16,728 participants found a moderate overall clinical improvement effect of g = 0.43, with apps incorporating personalization and engagement features performing better (Linardon et al., 2025).

KEY HIGHLIGHTS

• The evidence base is real but modest: The 2024 World Psychiatry meta-analysis of 176 RCTs found small but consistent effects of mental health apps on depression and anxiety, with a number needed to treat (NNT — the number of people who must use an intervention for one person to benefit) of approximately 11-12. Apps are most effective when evidence-based, structured, and combined with human support (Linardon et al., 2024).
• Over 90% of mental health apps are not FDA-regulated: Most mental health apps are classified as “wellness” or “lifestyle” tools, which means they do not require FDA clearance and are not required to demonstrate clinical efficacy before reaching consumers. Only apps that make explicit medical device claims or are prescribed as digital therapeutics (DTx) are subject to regulatory review (Scholle, 2025).
• Privacy protection is not guaranteed under HIPAA: HIPAA (the Health Insurance Portability and Accountability Act, the primary U.S. health data privacy law) does not apply to most mental health apps because they are not operated by covered healthcare entities. In 2024, the FTC amended its Health Breach Notification Rule to explicitly require mental health apps to notify users of data breaches, closing one major gap, but data selling to advertisers remains a documented concern (FTC, 2024; Scholle, 2025).
• Engagement and dropout are the biggest real-world barriers: Clinical trial effects of apps consistently exceed real-world effects because trial participants are monitored and supported in a way typical app users are not. In practice, many users stop using mental health apps within two weeks of downloading them. A 2025 meta-analysis found that only 76% of studies even reported engagement data, and definitions of “engagement” varied widely across 25 different metrics, making it difficult to know how many people actually complete app-based programs (Linardon et al., 2025).
• Adverse events are understudied: A 2024 systematic review of adverse events in clinical trials of mental health apps found that only 55 of 171 identified trials reported any adverse events at all. Among those that did, a meta-analytic deterioration rate of 6.7% was identified, meaning a meaningful minority of app users experienced worsening symptoms rather than improvement (Linardon et al., 2024b).
• Apps work best as a complement to professional care: The 2025 PLOS ONE systematic review of 38 studies covering 35 unique CBT-based apps found that over 70% of people globally who need mental health services cannot access them, and that apps represent a meaningful access-bridging tool, particularly while on waitlists or in underserved areas. They are most clinically effective as adjuncts to, not replacements for, licensed professional care (Shahsavar & Choudhury, 2025).
• Digital therapeutics are a distinct, regulated category: FDA-cleared prescription digital therapeutics (PDTs), such as Freespira for PTSD and panic disorder, and Rejoyn as an adjunct treatment for major depressive disorder, are categorically different from consumer wellness apps: they have undergone clinical trials, received FDA clearance, and are prescribed by clinicians as part of a treatment plan. They are not the same product as a general mindfulness or mood-tracking app (FDA, 2024).

What Are Mental Health Apps?

Mental health apps are software applications for smartphones and tablets that deliver content, tools, or human connections intended to support psychological well-being, reduce mental health symptoms, or facilitate access to clinical services. The category spans an enormous range of product types — from FDA-cleared medical software requiring a clinician’s prescription to free consumer apps offering guided breathing exercises — and treating them as a single homogeneous category is one of the most common errors in how they are discussed and evaluated.

The most clinically meaningful distinction within the category is between digital therapeutics (DTx) and wellness apps.

Digital therapeutics are software-based interventions that have undergone clinical investigation and, in many cases, received FDA clearance or authorization as medical devices. They make specific claims about treating, managing, or preventing a clinical condition and are prescribed by a healthcare provider. Prescription digital therapeutics (PDTs) — a subset of DTx — require a clinician’s prescription to access, just as a pharmaceutical medication does.

Rejoyn, the first FDA-authorized PDT for major depressive disorder, delivers a 10-week CBT-based program as an adjunct to antidepressant treatment for adults; its authorization followed a clinical trial program evaluating its specific therapeutic effects (FDA, 2024).

Wellness apps, which constitute the vast majority of the market, make no clinical claims, require no FDA clearance, and are not required to demonstrate efficacy before consumers download and pay for them.

Many wellness apps are marketed to people experiencing genuine mental health symptoms, but their design, content quality, and data privacy practices are not regulated in the way medical devices or clinical services are. The distinction matters clinically: a person using a wellness app believing they are receiving validated clinical intervention probably is not receiving anything equivalent.

Do Mental Health Apps Actually Work? What the Research Shows

Mental health apps produce real but modest effects on depression and anxiety symptoms compared to inactive controls, and the evidence base for this conclusion is substantially stronger in 2024 and 2025 than it was even five years earlier.

The definitive current evidence comes from a meta-analysis published in January 2024 in World Psychiatry — the field’s highest-impact journal — analyzing 176 RCTs with a combined 33,567 participants across conditions of depression and anxiety (Linardon et al., 2024). The key findings: apps produced a small but statistically pertinent decline in depression symptoms (Hedges’ g = 0.28, meaning a small standardized effect; number needed to treat, NNT = 11.5) and in generalized anxiety (g = 0.26, NNT = 12.4) relative to control conditions, which included waitlists, placebos, and care as usual. These are meaningful but modest effects, comparable in magnitude to some pharmacological treatments for mild depression, but considerably smaller than those observed in trials of therapist-delivered CBT.

A 2025 meta-analysis published in npj Digital Medicine, covering 92 RCTs and 16,728 participants, found a somewhat larger overall effect (g = 0.43) across diverse clinical outcomes, with apps incorporating personalization features and engagement reminders outperforming simpler, static-content apps (Linardon et al., 2025). CBT-based approaches were present in 66.3% of the apps studied, and 78.5% of trials included some form of human contact — a finding that critically qualifies how “app effectiveness” needs to be interpreted: many apps studied in trials are supported by human clinicians or coaches, which is not how most people use them in everyday life.

A March 2025 systematic review by Shahsavar and Choudhury in PLOS ONE, examining 38 studies across 35 CBT-based apps, similarly confirmed positive effects on depression, anxiety, and, notably, suicidal behavior in some populations, while identifying that contextual engagement (personalized, timely prompts) and cognitive change strategies were the most effective active components (Shahsavar & Choudhury, 2025).

The important qualification across all of this evidence: clinical trial results systematically overestimate real-world effectiveness. Trial participants are recruited, screened, monitored, and supported in ways typical app users are not. In routine use, completion rates are low, dropout happens early, and the structured engagement that drives trial effects is largely absent.

What Are the Different Types of Mental Health Apps?

The main types of mental health apps are digital therapeutics, CBT-based self-help apps, meditation and mindfulness apps, mood tracking apps, crisis support apps, and online therapy platforms. These types differ substantially in their evidence base, regulatory status, appropriate use cases, and clinical limitations, making the choice between them more complex than app store ratings suggest. The table below summarizes the key distinctions:

App Type	What It Does	Evidence Strength	Best For	Regulatory Status
Digital Therapeutics (DTx)	FDA-cleared software delivering evidence-based treatment (e.g., Freespira for PTSD/panic disorder, Rejoyn for MDD as adjunct to antidepressants). Prescribed by a clinician.	High: Required to demonstrate clinical efficacy through RCTs before FDA clearance	Diagnosed conditions where a clinician has prescribed a specific DTx as part of a treatment plan	FDA-cleared (De Novo or 510(k)); treated as regulated medical devices
CBT-Based Self-Help Apps	Deliver structured cognitive behavioral therapy exercises, thought records, behavioral activation schedules, and psychoeducation in self-guided format	Moderate: Supported by the largest body of RCT evidence; Linardon et al. (2024) meta-analysis of 176 RCTs confirms small but significant effects (g=0.26–0.28)	Mild-to-moderate depression and anxiety in adults; supplementing existing therapy; access support while on waitlists	Unregulated as medical devices if classified as “wellness” tools (>90% of apps); FTC breach notification rules apply
Meditation and Mindfulness Apps	Provide guided meditation sessions, breathing exercises, body scans, sleep meditations, and mindfulness-based stress reduction practices	Low-to-moderate: Studies show benefits for stress and sleep quality; effects on clinical depression and anxiety are smaller and less consistent than CBT apps	Stress management, sleep improvement, general well-being maintenance; useful adjunct for mild symptoms	Unregulated; classified as wellness/lifestyle tools; no FDA oversight required
Mood Tracking Apps	Allow users to log emotional states, sleep, activity, and triggers over time; generate pattern visualizations; some include clinician-sharing features	Low as standalone: Mood tracking alone does not produce therapeutic change; most useful as data source for clinical care	Self-monitoring alongside professional care; identifying triggers; sharing symptom trends with a therapist or psychiatrist	Unregulated unless used as a medical device with diagnostic claims; data privacy varies widely by app
Crisis Support Apps	Provide immediate access to safety planning tools, grounding exercises, emergency contacts, and crisis hotline links during acute emotional distress	Limited standalone RCT evidence; clinician-designed tools (e.g., PTSD Coach, developed by VA/DoD) have stronger backing than commercial alternatives	Safety planning between therapy sessions; not a replacement for 988 or emergency services; supplemental only	Variable; VA-developed apps (PTSD Coach, COVID Coach) meet federal safety standards; commercial apps unregulated
Online Therapy Platforms	Connect users with licensed human therapists via text, audio, or video; sessions delivered through a proprietary app interface (e.g., Talkspace, BetterHelp)	Moderate-to-high for the therapist component; equivalent to in-person therapy for mild-to-moderate conditions when therapist is fully licensed	People seeking human therapeutic relationships with flexibility of remote access; NOT equivalent to the standalone app content on the same platforms	Therapists must be licensed in the state where the client resides; platform itself subject to FTC and state consumer protection laws

Are Mental Health Apps Safe? Risks, Adverse Events, and the Engagement Problem

Mental health apps carry real risks that are underreported and rarely discussed in app store descriptions or company marketing materials. Understanding these risks is not a reason to avoid apps entirely — but it is essential context for informed use.

• Adverse events and symptom deterioration: A systematic review and meta-analysis of adverse events in clinical trials of mental health apps, published in npj Digital Medicine in December 2024, searched four databases (Medline, PsycINFO, Web of Science, ProQuest) and identified 171 clinical trials. Only 55 of those 171 trials — less than a third — reported any adverse event data at all. Among the 13 trials providing sufficient data for meta-analysis, a deterioration rate of 6.7% was identified, meaning approximately 1 in 15 app users in these trials experienced worsening symptoms. Critically, deterioration rates in app conditions did not differ from those in control conditions, but the systematic absence of adverse event reporting means the true harm profile of most mental health apps remains poorly characterized (Linardon et al., 2024b).
• Engagement and dropout: High dropout from mental health apps is one of the field’s most consistent and important findings, yet it is largely absent from competitor content on this topic. Many users disengage within days of downloading an app. The 2025 npj Digital Medicine meta-analysis found that 24% of trials did not report engagement data at all, and of those that did, 25 different engagement metrics were used, making cross-study comparisons nearly impossible. This means that the evidence base largely tells us what happens when people use apps as directed, which does have little relationship to what happens (Linardon et al., 2025).
• Substitution risk: People using mental health apps for serious or complex conditions — severe depression, suicidal ideation, psychotic disorders, bipolar disorder, complex trauma — delay or forgo evidence-based professional care that produces better outcomes. The Lancet Digital Health 2025 standalone app meta-analysis explicitly noted that app recommendations are most defensible “if no evidence-based first-line intervention is available” — positioning apps as a fallback, not a first choice (Linardon et al., 2025b). People experiencing severe symptoms receive professional evaluation before relying on any self-directed app.

Are Mental Health Apps Private? Understanding Data Security and Legal Protections

Mental health apps lack automatic privacy because the Health Insurance Portability and Accountability Act (HIPAA) applies only to covered healthcare entities, leaving most consumer wellness apps free to legally share or sell sensitive symptom and medication data for advertising. While the Federal Trade Commission (FTC) strengthened protections in 2024 by requiring non-HIPAA apps to report data breaches within 60 days, and the U.S. Department of Health and Human Services (HHS) proposed 2025 updates to cybersecurity rules, the burden of security remains on the user. To mitigate risks, consumers must prioritize apps with transparent data-deletion policies and encryption, or utilize non-commercial tools like PTSD Coach (VA/DoD, 2024), which lack profit-driven data interests.

How Do You Choose a Mental Health App? A Research-Based Checklist

Choosing a mental health app requires evaluating it across five dimensions: evidence base, clinical appropriateness, data privacy, engagement design, and cost and access. The following checklist translates research findings into actionable decision criteria.

• Look for evidence-based therapeutic content: Apps grounded in CBT (cognitive behavioral therapy — the most extensively researched psychological treatment for depression and anxiety), DBT (dialectical behavior therapy, particularly for emotion regulation and self-harm), or ACT (acceptance and commitment therapy, for psychological flexibility) have the strongest empirical foundation. Check whether the app’s website or app store description cites peer-reviewed research, names its clinical advisory board, or links to clinical studies of that specific app, not just general claims that CBT is effective.
• Match the app type to your actual situation: A crisis support app is not appropriate as a primary treatment for severe depression. A general mindfulness app is not an adequate substitute for trauma-focused PTSD treatment. Use the app type comparison table in this article to match the tool to the right role: supplemental support, waitlist bridging, self-monitoring, or stress management.
• Check the privacy policy specifically for data selling: Search for the words “sell,” “share,” “third party,” “advertising,” and “marketing” in the app’s privacy policy before entering any personal mental health information. If the privacy policy is absent, inaccessible, or uses ambiguous language around data sharing, treat the app as if your data will not be protected.
• Consider government-developed or clinically validated tools first: Apps developed by academic medical centers, government health agencies, or nonprofit organizations with publicly disclosed clinical trial data represent a higher baseline of credibility than commercially developed apps. Examples comprise PTSD Coach (VA/DoD), MindShift CBT (Anxiety Canada), and Intellicare (developed at Northwestern University, with published RCT data).
• Assess whether the app supports professional care or attempts to replace it: The most appropriate role for a mental health app is as a supplement to, not a replacement for, professional evaluation and treatment. If you are experiencing symptoms that are affecting daily functioning, work, or relationships, a clinical evaluation from a licensed mental health professional precedes or accompanies any app-based support.
• Ask your therapist or prescriber: Many mental health professionals actively integrate apps into their clinical work, recommending specific tools based on a patient’s diagnosis, treatment goals, and digital literacy. If you are already in treatment, asking your clinician what apps, if any, they will recommend for your specific situation is the most personalized and clinically appropriate path to choosing one.

When Should You Use an App, and When Should You Seek Professional Help Instead?

You should use a mental health app for mild-to-moderate symptoms, for maintaining skills learned in professional therapy, for supplementing an existing treatment plan, and for bridging access while waiting for a professional appointment. They are not appropriate as the primary or sole intervention for the following situations:

• Severe depression or suicidal thoughts: If you are experiencing persistent hopelessness, inability to function in daily life, or thoughts of ending your life, a mental health app is not adequate care. Call or text 988 — the Suicide and Crisis Lifeline — immediately. It is free, confidential, and available 24 hours a day, 7 days a week.
• Psychotic symptoms: Symptoms including hallucinations (hearing or seeing things others do not), delusions (fixed false beliefs), or disorganization of thought necessitate urgent evaluation by a psychiatrist, not a self-directed app. A 2024 adverse events review found that apps delivering symptom monitoring technology to people with schizophrenia spectrum diagnoses were associated with higher adverse event reporting rates (Linardon et al., 2024b).
• Eating disorders: Apps for eating disorder symptom management have an uneven evidence base and a specific risk: calorie-tracking and food-logging apps, even when not marketed as eating disorder tools, reinforce restriction behaviors in people with anorexia nervosa. Anyone with or at risk of an eating disorder must use an eating disorder-specific app reviewed by a clinician.
• When symptoms have persisted or worsened for two or more weeks: Persistent symptoms that are not responding to self-directed app-based support after several weeks prompt professional evaluation. Apps do not produce the same effects for everyone, and continuing with a tool that is not working while a treatable condition worsens is a meaningful risk.

If you are ready to seek professional support, your primary care physician carries out an initial mental health screening and provides referrals to appropriate specialists. The SAMHSA Behavioral Health Treatment Services Locator at findtreatment.gov and the NAMI HelpLine at 1-800-950-6264 both provide free, nationally available guidance on finding licensed mental health professionals regardless of ability to pay. If you are in crisis, call or text 988.

WHAT TO DO NEXT

Mental health apps provide a valuable bridge to care, reinforcing therapy skills and offering low-barrier support for mild symptoms; however, their effectiveness is modest and accompanied by data privacy risks and regulatory gaps. These digital tools cannot replace clinical evaluation for severe symptoms, which requires a licensed professional via resources like findtreatment.gov, the National Alliance on Mental Illness (NAMI) HelpLine (1-800-950-6264), or a primary care provider. If you are in crisis, immediate, confidential support is available 24/7 by calling or texting 988 for the Suicide & Crisis Lifeline.

REFERENCES

Federal Register. (2025, January 6). HIPAA Security Rule to strengthen the cybersecurity of electronic protected health information. Department of Health and Human Services. https://www.federalregister.gov/documents/2025/01/06/2024-30983/hipaa-security-rule-to-strengthen-the-cybersecurity-of-electronic-protected-health-information

Federal Trade Commission. (2024). Health breach notification rule: Final amendments. Federal Trade Commission. https://www.ftc.gov/legal-library/browse/rules/health-breach-notification-rule

Linardon, J., Cuijpers, P., Carlbring, P., Messer, M., & Fuller-Tyszkiewicz, M. (2024). Current evidence on the efficacy of mental health smartphone apps for symptoms of depression and anxiety: A meta-analysis of 176 randomized controlled trials. World Psychiatry, 23(1), 139–149. https://doi.org/10.1002/wps.21183

Linardon, J., Messer, M., Battista, R., Greenwood, C., Rodgers, R. F., Sharpe, L., & Mulgrew, K. (2024b). Systematic review and meta-analysis of adverse events in clinical trials of mental health apps. npj Digital Medicine, 7(1), 392. https://doi.org/10.1038/s41746-024-01388-y

Linardon, J., Rodgers, R. F., Messer, M., Fuller-Tyszkiewicz, M., Sharpe, L., & Mulgrew, K. E. (2025). A meta-analysis of persuasive design, engagement, and efficacy in 92 RCTs of mental health apps. npj Digital Medicine, 8(1), 249. https://doi.org/10.1038/s41746-025-01567-5

Linardon, J., Messer, M., Egbert, A. H., Rodgers, R. F., Sharpe, L., & Mulgrew, K. E. (2025b). Efficacy of standalone smartphone apps for mental health: An updated systematic review and meta-analysis. The Lancet Digital Health. https://doi.org/10.1016/S2589-7500(25)00105-0

Scholle, S. (2025, February 27). Increasing access to safe, effective, and secure digital mental health technology should be a federal priority. Commonwealth Fund. https://doi.org/10.26099/4t02-2787

Shahsavar, Y., & Choudhury, A. (2025). Effectiveness of evidence-based mental health apps on user health outcome: A systematic literature review. PLOS ONE, 20(3), e0319983. https://doi.org/10.1371/journal.pone.0319983

U.S. Department of Veterans Affairs & U.S. Department of Defense. (2024). PTSD Coach. National Center for PTSD. https://www.ptsd.va.gov/appvid/mobile/ptsdcoach_app.asp

U.S. Food and Drug Administration. (2024). Digital health center of excellence: Digital therapeutics. https://www.fda.gov/medical-devices/digital-health-center-excellence