How Mental Health Apps Are, and Are Not, Regulated

Most mental health apps are never reviewed by the FDA. The agency treats meditation, mood-tracking, and stress-management tools as general wellness products and applies enforcement discretion, which means it does not examine whether they work before they reach an app store. Only apps that claim to diagnose or treat a specific condition cross into regulated-device territory. That gap leaves the job of judging quality to you and your clinician.

The line the FDA draws

Regulators sort digital products by what they claim to do. Under the FDA's guidance "General Wellness: Policy for Low Risk Devices," first issued in 2016, updated in 2019, and revised again in January 2026, a product qualifies as a general wellness tool if its intended use falls into one of two buckets: maintaining or encouraging a general state of health or a healthy activity, or associating a healthy lifestyle with a reduced risk of, or better living with, certain chronic conditions. It also has to be low risk, meaning non-invasive, non-implanted, and free of any technology that could endanger a user. A meditation timer, a mood journal, a breathing coach, and most stress-management apps fit this description comfortably.

For products that meet those criteria, the FDA says it does not intend to examine them against the requirements that apply to medical devices. That posture is called enforcement discretion, and it is the reason most of the apps in your phone's health category never passed through any federal review.

What enforcement discretion actually means

Enforcement discretion is a decision about where the FDA spends its attention, not a judgment that a product works. When an app sits inside the general wellness lane, no one at the agency has checked whether it reduces anxiety, lifts mood, or improves sleep. There is no premarket submission, no independent look at the underlying science, and no requirement that the developer hold evidence to support the language on the marketing page. A claim like "clinically proven to lower stress" can appear next to a product the FDA has never evaluated, because the claim itself is what the developer chose to make, and the wellness framing keeps it outside the device rules.

It helps to separate this from a common misreading. Enforcement discretion is not a loophole that developers exploit; it is an explicit, published policy that lets the FDA concentrate its finite review capacity on higher-risk technology. The trade-off is that evidence quality in the consumer wellness space is uneven by design, and the checking has to happen somewhere else.

When an app crosses into regulation

The picture changes the moment an app claims to diagnose, treat, cure, mitigate, or prevent a specific disease. At that point the software meets the statutory definition of a medical device, and it generally needs FDA clearance or authorization before it can make those claims, most often through the 510(k) or De Novo pathways. A small number of prescription digital therapeutics have gone through this process for conditions such as substance use disorder and chronic insomnia. Naming that category here is only to mark the regulatory boundary: these products carry claims that required evidence and review, which is precisely what a general wellness app does not.

The January 2026 revision to the policy mostly addresses a different frontier, the non-invasive wearables that estimate physiologic signals such as heart rate variability, blood oxygen, or blood pressure. It widened the wellness lane for some of these sensors while adding guardrails: outputs that mimic clinical measurements need validation, and the device cannot flag a reading as abnormal or point to a diagnostic threshold. For the mental-health-adjacent wearables that infer stress from heart rate variability, this matters, but it does not change the basic fact that the mood, meditation, and journaling apps most people use still sit under enforcement discretion.

The evidence gap you inherit

Because the regulator steps back, the work of appraisal moves to clinicians and users. Researchers who have applied structured evaluation to popular mental health apps have repeatedly found claims running ahead of evidence, thin or absent citations, and privacy practices that vary widely from one product to the next. A stakeholder-engaged update to the American Psychiatric Association's evaluation framework, published in Psychiatric Services, was built partly in response to this problem: the market moves faster than any static "best apps" list, so the field needed a durable way to ask the right questions rather than a ranking that ages within months.

A framework for judging an app

The APA does not rate or endorse individual apps. Instead it offers an App Evaluation Model, structured as a hierarchy of five levels that a person works through in order: access and background information, privacy and security, clinical foundation and evidence, engagement and ease of use, and interoperability with the rest of someone's care. The order is deliberate. If an app fails an early level, such as having no readable privacy policy or quietly selling data, there is little reason to spend time weighing how pleasant it is to use.

In practice, a few questions carry most of the weight. Does the app have a privacy policy, and does it say whether your data is sold or shared? Is there peer-reviewed evidence for this specific app, not for the general idea of, say, cognitive behavioral techniques? Who built it, and do they disclose any clinical involvement? Can it connect to the rest of your care if you want it to? None of these questions require a regulator's stamp, which is fortunate, because for most wellness apps there will never be one.

This article is educational and is not medical advice; decisions about mental health care should be made with a qualified clinician.