From mood trackers to guided cognitive-behavioral therapy (CBT) and peer support, depression apps are crowding the digital marketplace. As more patients turn to this technology to ease their symptoms, physicians are faced with figuring out their potential role in clinical practice.

Unlike other psychiatric disorders, which have few digital therapeutic options, there are myriad apps aimed at depression, although only a handful are evidence-based or FDA-cleared.

Many are marketed as “wellness” interventions, bypassing regulatory oversight, leaving clinicians to navigate which have the potential to augment traditional treatment safely and effectively, how to discuss them with patients, and the ethical considerations involved.

“You don’t have to use these apps yourself, but you need to be at least aware of what your patients are using,” John Torous, MD, psychiatrist and director of Digital Psychiatry Division at Beth Israel Deaconess Medical Center (BIDMC) in Boston, told Medscape Medical News.

“I think the future of outpatient mental health is hybrid care, where you do the best you can with in-person visits but also use technology to help outside the visit,” Torous added.

The Latest Apps

In recent years, multiple studies have demonstrated the efficacy of depression apps. A 2023 meta-analysis of 13 randomized trials involving more than 1400 participants showed that mobile mental health apps — used alone or alongside conventional treatment — were effective for adults with depression.

In 2024, the FDA approved Rejoyn (Otsuka Pharmaceutical and Click Therapeutics), the first prescription digital therapeutic for adults with major depressive disorder (MDD). This app is intended to be used as an adjunct to antidepressant therapy and under the guidance of a clinician.

Rejoyn’s developers noted that it is designed to improve emotion regulation and processing through CBT-based video lessons and cognitive-emotional training exercises.

A recent meta-analysis of 45 randomized trials that focused specifically on mindfulness apps showed the technology was associated with significant reductions in symptoms of depression and anxiety compared with control interventions.

At this year’s American Psychiatric Association (APA) annual meeting, Elombe Issa Calvert, MBBS, from the Division of Digital Psychiatry at BIDMC, presented a 12-week feasibility study using his lab’s mindLAMP app and platform to develop a “digital phenotype” of depression.

In combination with a wearable device, the app collects real-time data on behavioral patterns, including geolocation, activity levels, and screen use, with the potential to guide personalized care. Calvert told Medscape Medical News he foresees a future in which it could be prescribed for joint use by patients and psychiatrists.

Recently published studies on the CBT apps MoodMission and RESiLIENT, involving roughly 4000 participants each, have added to the evidence base. MoodMission, which recommends CBT strategies they call “Missions,” first showed positive results in a 2018 randomized controlled trial.

While all three of the assessed apps (MoodMission, MoodKit, and MoodPrism) were linked to greater well-being than a control condition, only MoodMission and MoodKit were associated with reduced depression scores.

A large study published last year further showed that “non social Missions,” those that did not involve interaction with other people such as exercise or self-care practices, were tied to significant reductions in distress.

RESiLIENT, meanwhile, implements five CBT skills — behavioral activation, cognitive restructuring, problem solving, assertion training, and insomnia-focused behavior therapy.

In a study published in April, each of these skills, whether applied individually or in combination, proved superior to control conditions for improving depression, as measured by changes at 6 weeks on the Nine-Item Patient Health Questionnaire.

Popular general wellness apps include the mindfulness and meditation apps Calm and Headspace, with 180 million and 80 million downloads, respectively.

Torous noted that it’s expensive and time-consuming to take any app through the FDA process. “You could have the same app be FDA cleared or not cleared depending on how specific you are on the medical claim that it makes,” he said.

Many wellness apps could probably gain FDA clearance as medical devices with enough time and resources — but for developers, it may not be worth the effort, Torous noted.

A Role for Chatbots?

The conversation also extends to chatbots. Do they work in clinical practice?

In a randomized controlled trial published in April, a generative artificial intelligence (AI)-powered therapy chatbot called Therabot was linked to a significant reduction in MDD, along with generalized anxiety and eating disorders.

On the other hand, a human therapist outperformed an AI model (ChatGPT3.5) for feedback, collaboration, and guided discovery in the delivery of CBT in a small pilot study presented at the 2025 APA meeting.

Still, popular chatbots include Youper — a mental health manager that provides daily check-ins and CBT techniques — and Wysa — which guides users through interactive, text-based conversations centered around CBT and dialectical behavior therapy. Wysa received FDA Breakthrough Device Designation in 2022.

Howard Liu, MD, chair of the Department of Psychiatry at the University of Nebraska Medical Center in Omaha, Nebraska, noted the importance of involving clinicians when using any of these apps.

“It’s just making sure there is a human element to make sure the screening for high-risk behaviors has some sort of safety net,” he told Medscape Medical News at the APA meeting.

“The saying is ‘garbage in, garbage out.’ So whatever you’re feeding in to train the system, that’s what you’re going to get out of it. So the data need to be current and updated. I think it’s useful but only with guardrails in place,” Liu said.

Uptake and Evaluation

A recent report by the Pew Research Center noted that 91% of all Americans now own a smartphone, as do 95% of teens. In addition, up to 39 different apps are used monthly by an average smartphone owner.

However, there’s a quick and substantial plummet in engagement. For psychiatric patients, the drop-out rate for a new app is a whopping 90% within the first 10 days, Manuel Gardea Reséndez, MD, professor of psychiatry, Universidad Autónoma de Nuevo León, San Nicolás de los Garza, Mexico, reported during another APA meeting presentation.

“I think this has a lot to do with the design of the apps, with many developed by software engineers and without input from those who would benefit most from it,” Gardea Reséndez told Medscape Medical News.

Torous said he thinks engagement also falls because apps need at least a little bit of human support. “We can build apps, but we need to build ways to help people engage in the use of these apps,” he noted.

He added that although technology moves fast, it’s important for clinicians to try to stay as up-to-date as possible on the latest information. But with so many options out there, what’s the best way to determine if an app could be helpful for an individual patient?

The APA provides several helpful tools, including its App Adviser and “App Evaluation Model” for clinicians. It lists a series of questions that should be answered while determining an app’s usefulness, including which platforms/operating systems the app works on, whether it’s been updated in the past 180 days, if it has a transparent privacy policy, if it’s easy to use, and if it has a “clinical/recovery foundation relevant to your intended use.”

“The hierarchical model is useful as it highlights the presence and absence of features APA deems important for the patient and helps in making an informed choice,” the organization noted on the Model’s website.

The APA has also published a “Digital Mental Health 101” document to help guide clinicians through the digital landscape.

Integration Into Clinical Practice

The Mindapps.org website provides free and public access to a large database of information on commercially available mental health apps with descriptions and ratings. The database can be searched using a number of different filters, including privacy, price, and language.

“It’s like a menu approach you can use to go over [things] with patients,” Torous said.

It’s also important for clinicians to ask about which apps patients are already using — and then integrate parts of those apps into their treatment, if possible, Gardea Reséndez noted.

Some apps act as a form of “homework,” giving patients the chance to practice skills at home and review their progress in session. Others are designed specifically to help organize and track therapeutic assignments in one place.

Gardea Reséndez said he also asks about any wearable devices patients may be using, which can provide insights into factors such as sleep patterns. “Regardless of their own perceptions, it provides clear data,” he said.

Risks and Ethical Concerns

One key question is what ethical issues this type of technology raises. A recent study reported by Medscape Medical News highlights some troubling trends. Only 15% of mental health apps include the 988 Suicide and Crisis Lifeline, the national standard for crisis support.

Even more concerning, 14 apps with a combined 3.5 million downloads provided incorrect or nonfunctional alternative hotlines.

Paul Appelbaum, MD, director of the Division of Law, Ethics, and Psychiatry at Columbia University, New York City, told Medscape Medical News that he is also concerned about privacy issues and the use of chatbots.

Individuals need to know upfront who — or what — they’re interacting with and the level of human oversight involved, Appelbaum noted. He added that it is also important to know how quickly an inappropriate chatbot response would be identified and addressed through human intervention.

A study published in Psychiatric Services showed that the chatbots ChatGPT by OpenAI and Claude by Anthropic “provided direct responses to very low-risk queries 100% of the time,” but both (plus Gemini by Google) did not provide direct responses to any query judged to be very high risk.

“This suggests a need for further refinement to ensure that chatbots provide safe and effective mental health information, especially in high-stakes scenarios involving suicidal ideation,” lead investigator Ryan McBain, PhD, a senior policy researcher at RAND, Arlington, Virginia, said in a release.

A letter sent last year from the American Psychological Association to the Federal Trade Commission highlighted concerns over the underregulated development and deployment of generative AI, particularly chatbots represented as therapists.

In response to some of these concerns, the FDA’s Digital Health Advisory Committee on Generative AI-Enabled Digital Mental Health Medical Devices will hold a public meeting on November 6.

“Many of these medical devices pose novel risks, and as mental health devices continue to evolve in complexity, regulatory approaches ideally will also evolve to accommodate these novel challenges and opportunities to provide a reasonable assurance of their safety and effectiveness,” the Agency noted on the Federal Register website.