Mapping Out Mental Health Tech: 70+ Startups Changing Lives

As we continue to fight COVID-19, the nation’s mental health crisis continues to intensify.

In a recent study by the U.S. Centers for Disease Control and Prevention, 1/3 of Americans reported symptoms of anxiety and depression, compared to 1/9 in Q1 of last year. SAMSHA, the National Helpline for mental health crises, has seen an 891% increase in call volume from this time last year. Employers are taking note and showing greater interest in finding solutions for their employees. As shown below, the number of earnings calls mentioning the term “mental health” skyrocketed in Q2 of this year.

Number of mentions of “mental health” in earnings calls (Source: Crunchbase)

Even before COVID-19, the world was already in a mental health crisis.

According to the World Health Organization, 1 in 4 people in the world will experience mental or neurological disorders at some point in their lives.[1]

In the U.S., 90% of suicides are related to mental illness, and more than 47,000 people take their own lives each year.

Numbers, of course, only tell part of the story. They help illustrate the scale and severity of our nation’s mental health crisis, but even more importantly, each of these numbers represents a valuable human life. In the last few years, I have thought deeply about mental health and have seen close friends struggle with depression and anxiety.

Mental health is an issue I care deeply about, and I am hopeful that mental health technologies can positively impact the many around the world.

Below, I outline 1) key challenges and opportunities in mental healthcare in the U.S., 2) the current market map of startups tackling mental health globally, and 3) investable trends.

Key Takeaways:

1. Mental health is a critical issue, affecting 450 million people worldwide and 1 in 4 adults in the U.S.

2. Four of the biggest challenges in mental healthcare are lack of access, inconsistent quality, regulation at times constraining innovation, and discrimination.

3. There are many ways to segment the market. I’ve mapped six categories: self-care, chatbots, P2P, telehealth, digital therapeutics, and hardware.

4. Six areas I predict continued momentum are: 1) solutions for mild and moderate cases of mental illness, 2) solutions focused on prevention and upstream intervention through seamless monitoring and diagnoses, 3) the unbundling of mental health services, 4) solutions targeting youth, 5) senior-focused solutions, and 6) provider tools that monitor and improve patient outcomes.

Want to read this story later? Save it in Journal.

Challenges (& Opportunities) in the U.S.

Access:

1. There is a huge supply shortage. More than half of all U.S. counties have no psychiatrists.[2]
2. There are barriers to payment and affordability. A 2019 report found that a behavioral health office visit is over five times more likely to be out of network than a primary care appointment.[3] Furthermore, 50–60% of private practice clinicians no longer accept insurance.[4]
3. Mental and physical healthcare remain siloed. In the last few years, there has been movement towards integrated and collaborative care models, where mental health clinicians are embedded in existing primary care systems. However, most physical and mental health systems still operate independently, with separate insurance, payment policies, and patient records.
4. Access is not equal. Racial and ethnic minorities, the LGBTQ+ community, and other minority groups experience lower levels of access and quality. Many studies support this,[5] and disparities are continuing to increase.[6]

Quality:

1. We lack effective outcome metrics, as improvements in mental health can be difficult to quantify. Furthermore, the few assessments we do have are used infrequently. Only 18% of psychiatrists and 11% of psychologists in the U.S. regularly use symptom rating scales — such as the GAD-7 for anxiety and the PH-Q9 for depression — to monitor treatment effectiveness and patient improvement.[7]
2. Not all treatment is evidence-based,[8] resulting in less effective outcomes. According to Dr. Schueller, the creator of mental health app guide Psyberguide, only 3% of mental health apps in the app store are evidence-based.

Regulation:

COVID-19 has led to regulatory changes that have enabled the remote delivery of mental healthcare. Although it remains uncertain which regulatory changes will persist, four key ones are:

1. Under an exemption of the Ryan Haight Act, controlled substances can now be prescribed via telemedicine without a pre-existing relationship (i.e. a prior in-person exam).
2. The federal government’s Centers for Medicare & Medicaid Services (CMS) has temporarily waived state licensure requirements, allowing physicians and other healthcare professionals[9] participating in federal healthcare programs to receive payment for telemedicine services in states where they do not hold a license.[10]
3. The Health Insurance Portability and Accountability Act (HIPAA) has been relaxed to allow the use of FaceTime, Zoom, Skype, and Google Hangouts.
4. CMS has allowed reimbursement of in-home care for Medicare patients. Previously, patients could only receive reimbursement for in-home care if they met certain conditions, such as being homebound, needing care on an intermittent basis, or receiving care from a Medicare-certified home health agency (HHA).

Stigma and discrimination:

While conversations surrounding mental health have increased, we are far from reaching complete de-stigmatization.

1. People still feel uncomfortable discussing mental health with the people and communities closest to them. In a survey from 2016 with over 2,000 respondents from the U.S., 56% of respondents said they’d be uncomfortable talking to friends and family about mental health, and 84% said they’d be uncomfortable talking to their employer.[11]
2. Dr. Insel, the former Director of the National Institute of Mental Health, points out that the issue is not simply stigma, but discrimination. This is evident from the high rate of mental illness in incarcerated populations — ten times as many people suffer from serious mental illnesses in state prisons and county jails as those receiving care in state mental hospitals.[12] Yet only one in three prison inmates receive any form of mental health treatment.[13]

Mapping the Market

The Mental Health Tech Landscape

*This spreadsheet includes websites of the above startups and a couple more.

The following is based on my research of 70+ startups from articles, interviews, and first-hand conversations with founders. For each of the six categories, I offer a definition, relevant insights, and key questions for analysis.

Several startups fall within multiple categories. As there are more than a thousand mental health startups,[14] this does not encompass all startups in the space. The startups I’ve chosen represent category leaders as well as startups I find uniquely innovative.

Self-Care:

Definition: solutions focused on mental well-being and resilience and not intended to treat severe mental illnesses. Current solutions center around meditation and mindfulness, journaling (with some tools loosely based on cognitive behavioral therapy — CBT), and mood or behavior tracking.

Additional insights, based on my testing of 20+ solutions in this space:

Maintaining user engagement is a key challenge. How can solutions create a feedback loop where users feel like they are deriving significant value? Four user engagement strategies I’ve observed are:

1. Establishing a clear (and rewarding) daily task. This helps the user form a daily routine, ensuring they open the app at least once a day.
2. Providing a structured program with clear next steps and an end goal. This mitigates the problem of choice paralysis when deciding what to do once inside the app.
3. Offering insights based on usage. This incentivizes users to engage with the app. Actionable insights, especially ones that connect back to features/usage of the app, are particularly engaging.
4. Building a sense of community, where users engage in facilitated dialogue.

Given how crowded the space is, three potential moats are:

1. Community: the unique community and interactions of the user base.
2. Content: unique content, such as guided meditations and journaling prompts.
3. Data: the unique application(s) of user data, which cannot be transferred to other platforms.

By nature of being a consumer app, the user experience (how appealing and intuitive the design is) is critical — a necessary but insufficient factor for success.

Key questions:

1. Differentiation: This space is crowded — what differentiates the product from other solutions? In regard to potential moats — what is the value of the original content? How does the product utilize user data to generate value?
2. User engagement: A study conducted by the University of Haifa’s Dr. Baumel found that the median percentage of daily active users for mental health apps was only 4.0%. How does the product create and sustain user engagement?
3. Business model: Since many apps operate under a freemium model — what is the free-to-paid user conversion rate? As with any consumer-facing app, what is the customer LTV and power user curve?[15]

Chatbots:

Definition: solutions applying AI to simulate human conversation. I’ve focused here on D2C solutions.

Additional insights:

Advantages of chatbots:

1. Lack of bias and ability to remain objective, assuming the underlying algorithms are trained in the right way.
2. Anonymity and a sense of privacy from the user’s standpoint (assuming transparent and robust data protection).
3. Scalability.

Areas for growth, based on my own user testing:

1. Conversations feel un-human — somewhat slow and unnatural, with a limited sense of continuity between conversations.
2. Interactions feel simplistic — current solutions are largely based on CBT principles. Hence, users with basic familiarity with CBT may be able to predict what the bot might say or what exercise it will suggest, diminishing their need for the product.

Key questions:

1. User engagement: how can chatbots create human rapport and deliver emotional support and other benefits of human interaction?
2. Clinical effectiveness: what does long-term patient engagement look like? What kind of health outcomes are achieved?
3. Business model: given these solutions are mostly D2C and many are currently free or freemium, how can they scale and earn revenue?

Peer-to-Peer:

Definition: solutions centered around peer support. I’ve seen two main models: 1) one-to-one peer interactions where an individual seeking help is matched with an individual offering help, and 2) group interactions where peers facing similar challenges engage in discussion.

Additional insights:

Advantages of P2P:

1. Lower costs for both the business and users, as there are usually no or few professional providers involved.
2. Ability to serve as a source of empathy and emotional support. Hence, P2P in its current form is best suited to general mental health as opposed to the treatment of more severe illnesses.
3. Potential to be anonymous and hence more appealing to typically underserved populations — such as men — that may be more hesitant to seek help or treatment.

Interestingly, Dr. Baumel’s study suggests that peer support apps are better at user engagement than apps focused on tracking, breathing exercises, or psychoeducation, with 17% daily active users as opposed to the 4% median for all mental health apps.

Key questions:

1. User safety: how do solutions ensure user safety, prevent online trolls, and build trust?
2. Clinical effectiveness: For one-to-one interaction models — how are helpers trained and monitored? For group interaction models — how do solutions form the most relevant and effective peer groups?

Telehealth:

Definition: solutions that facilitate the patient-client relationship and enable remote treatment. I’ve observed two main types: 1) platforms and managed marketplaces that offer a specific service (e.g. Talkspace), and 2) full-stack solutions that function as a “digital clinic” and offer comprehensive mental healthcare (e.g. Ginger).

Additional insights:

The value proposition for patients — a more patient-centered experience with greater agency in one’s own care. Many solutions offer some combination of the following:

1. Higher quality care, with vetted providers and a focus on evidence-based care.
2. Tools to measure and track treatment progress.
3. A platform to communicate directly with the provider(s).
4. Greater education regarding treatment methods and processes.
5. Increased transparency in and assistance with payment/reimbursement processes.
6. Easier access and shorter wait times.

The value proposition for providers (mental health professionals) — greater agency in one’s own practice. Many solutions offer some combination of the following:

1. More flexibility in scheduling.
2. The opportunity to specialize and focus on treating a specific group of patients (e.g. adolescents).
3. Less time spent on administrative work such as billing and paperwork.
4. Tools to monitor patient progress.
5. A network and community of fellow professionals to work alongside with.

Key questions:

1. Care delivery: If teletherapy is involved — how does the company ensure strong therapeutic alliances, given that the patient-therapist relationship is one of the strongest predictors of successful outcomes? While some platforms seek to build prediction algorithms for patient-therapist matching, there is not yet a standard measure or dataset for modeling therapeutic alliance.
2. Business model: How are services integrated with the current payer system (i.e. with insurance companies and Medicaid/Medicare)? If B2B — how compelling are the outcomes to employers?
3. Go-to-market: For full-stack care models, distribution will become increasingly important as the number of players increases. What are the company’s go-to-market channels?
4. Regulatory risks: What federal or state regulations could impact operations?

Digital Therapeutics (DTx):

Definition: evidence-based therapeutic interventions delivered via software. As defined by the Digital Therapeutics Alliance, DTx products may be used independently or in tandem with clinician-delivered therapy to optimize patient outcomes.

Additional insights:

1. Given that the DTx space is relatively young, the regulatory and reimbursement processes are still forming. The FDA is in the early stages of developing an approval process for therapeutic-focused software, and Medicare and most state Medicaid plans do not cover DTx.
2. Big pharma is interested in DTx, as it can diversify their revenue streams and drastically reduce their drug development costs and timelines.[17]
3. Benefits to patients: increased access to evidence-based interventions, with greater personalization and flexibility.
4. Benefits to healthcare providers and systems: increased access to treatment for patients, valuable data regarding patient engagement and outcomes.
5. Benefits to payers: improved outcomes, greater access, and lower overall costs. [16]

Key questions:

1. Clinical effectiveness: what is the clinical efficacy? What is the scope and scale of the product’s clinical trials? Does the product need to be FDA cleared or approved?
2. Business model: how can the product be reimbursed by public and private payers?
3. Go-to-market and adoption: how can companies build successful partnerships with pharmaceutical companies? How can companies gain buy-in from psychotherapists?

Hardware:

Definition: any connected hardware, including VR headsets, wearable devices, and telepresence robotics.

Additional insights:

1. While the FDA, CMS, and health plans provide a clear path for getting medical devices approved and paid for by payers, the path to payer reimbursement for hardware-software hybrid solutions is still forming.
2. Apple has been building its health capabilities since at least 2014. The Apple Watch now features panic attack detection, sleep tracking, and mobility measurements, collecting data that can inform mental and behavioral health diagnoses and treatment.[18] It will be interesting to see how Apple affects wearables in this space (as a potential competitor, strategic partner, or even acquirer) and what startups (continue to) build software for Apple’s products.

Key questions:

1. Clinical effectiveness: What FDA approval is required to gain reimbursement for the device? How are outcomes measured, and how do they compare to existing solutions? For wearables, how can biofeedback be utilized to obtain better outcomes?
2. Differentiation: What is the value of the associated software? Is the software or the hardware the primary moat?

Investment Theses

Finally, six trends I predict and hope to see:

1. Continued adoption of solutions addressing mild and moderate cases of mental illness, as more people prioritize their mental health. This includes tools that strengthen mental resilience (i.e. “self-care”: meditation, journaling, CBT exercises, etc.) as well as solutions for mild and moderate cases of depression and anxiety (i.e. “digital therapeutics”).

2. Additional solutions focused on prevention and upstream intervention. This involves utilizing biometrics and smartphone usage to conduct frequent and/or seamless diagnoses. For example, several startups are utilizing smartphones as a tool for passive data collection (e.g. Ginger[19] and Mindstrong).

3. Further unbundling of mental health services. As more attention shifts to mental healthcare, there is more demand and opportunity for specialist care as opposed to generalist care. This wonderful graphic from Jessica DaMassa and Matthew Holt illustrates this.

The unbundling of different mental health provisions

4. More solutions targeting youth.[20] Given that 75% of mental illnesses begin before the age of 25, and early identification is key to successful treatment, I am excited to see more youth solutions. These could include AI chatbots, video games (such as Akili’s EndeavorRx, the first video game to be FDA-cleared as a prescription treatment for ADHD), and digital solutions providing full-stack care (such as Brightline, Daybreak, and Mantra, which target young children, adolescents, and college students respectively).

5. Increased number of senior-focused solutions. As the U.S. population continues to age,[21] the need for remote monitoring and care will continue to heighten. Additionally, these older generations are tech-literate and increasingly avid technology adopters,[22] creating a great opportunity for digital tools.

6. Better solutions to monitor and improve patient outcomes. Mental health treatment lacks quality metrics. Hence, there is an opportunity for tools to help clinicians monitor patient satisfaction and outcomes, especially as the larger healthcare system continues to move towards value-based care.

Conclusion

We must pay attention to mental health. COVID-19 has exacerbated existing mental health concerns, and the U.N. has warned of a global mental health crisis.

I hope this provides a valuable perspective on the growing space of mental health tech. I welcome any questions or thoughts and am excited for continued innovation in the space.

If you are a founder tackling mental health, feel free to reach out, I’d love to learn more!

Addendum (8.18.2021)

A year later, I remain keen on innovation within behavioral and mental health. Here are a few areas I am especially excited about:

1. Solutions for SMI & other moderate to high acuity patient populations.
2. Solutions targeting specific demographics, including youth and seniors as noted above as well as racial and ethnic minorities and other underserved communities.
3. Solutions targeting specific behavioral health conditions such as eating disorders, trauma-related disorders, and severe depression.
4. Solutions utilizing one to many models (e.g. peer and group support models — these require less providers to scale, and in some cases offer a more palatable / less intimidating experience to patients).

Side note: my foray into mental health has naturally led me to explore women’s health — if you’re building solutions for women’s hormonal, maternal, or menopausal health, feel free to reach out.

Sources:

See below for sources and additional resources.

Shout-out to Solome Tibebu for organizing the Behavioral Health Tech Conference, which introduced me to many leaders in this space, and to the founders who took the time to speak with me.

Special thanks to Jessica DaMassa for providing invaluable feedback, and to my friends Ayushi, Carol, Hannah, and Laura for being my first readers.

[1] World Health Organization (WHO) report

[2] Estimating the Distribution of the US Psychiatric Subspecialist Workforce

[3] Health Insurers Still Don’t Adequately Cover Mental Health Treatment

[4] From Roger Dowdy’s talk during Behavioral Health Tech Conference

[5] Mental Health Disparities: Diverse Populations

[6] Trends in Racial-Ethnic Disparities in Access to Mental Health Care, 2004–2012

[7] Fixing Behavioral Health Care in America

[8] Primer on evidence-based treatment practices

[9] Great primer on the different types of mental health professionals

[10] Note that this federal waiver does not automatically apply to private health care programs. Individual states are responsible for authorizing providers to provide professional services without state licensure.

[11] Stigma of Mental Illness

[12] US jails hold 10 times more mentally ill people than state hospitals, report finds

[13] The Many Forms of Mental Illness Discrimination

[14] Approaching 1,000 Mental Health Startups in 2020. I would also be remiss if I didn’t include Ed Gaussen of Mantra Health’s piece on the mental health startup landscape from 2018.

[15] The Power User Curve: The Best Way to Understand Your Most Engaged Users

[16] DTA industry report

[17] DTx is a huge and growing space. For an in-depth look of prescription DTx, see Omers Ventures’ PDTx Investor’s Guide

[18] For more perspectives on Apple’s healthcare play, see Patricia Mou, Rex Woodbury, Nathan Baschez and Rob Litterst

[19] This video from Ginger co-founder Anmol explains how passive data can be used to detect potential mental illnesses. As an example, he notes that lower movement and communication, which can be measured by phone usage, can serve as an indicator of early depression.

[20] As a side note, I believe it’s important to integrate mental health education in our education system. Dr. Tom Insel’s TEDx talk offers a compelling case for the importance of early education, identification, and prevention.

[21] The U.S. Population is Aging

[22] 2019 and 2017 Pew Research reports on technology use among seniors.

📝 Save this story in Journal.

👩‍💻 Wake up every Sunday morning to the week’s most noteworthy stories in Tech waiting in your inbox. Read the Noteworthy in Tech newsletter.