Hero

Videos

Slideshows

Articles

Podcasts

Poll

Childhood: a carefree time for many. However, children today face countless sources of stress: academic pressure, schedules crammed with activities, lack of sleep, bullying, family instability, world events, social media, and more. Mental health issues among children are growing at a dangerous rate, and it is important that we use all of our innovative technologies to help them.

It is clear that far too many children are having trouble coping with the stresses of daily life. An estimated 7.7 million children

 have at least 1 mental health disorder, and half are not receiving the treatment they need. More broadly, an estimated 15 million children in the United States

 alone are in need of a psychiatrist. More than 4 million children and teenagers have been diagnosed with anxiety, and nearly 2 million have been diagnosed with depression, according to a study published in 

 This is resulting in far too many tragic outcomes. A study published in 

 reported that a fifth of deaths among children and adolescents in 2016 were suicides.

Yet with such a high demand for doctors, there are only approximately 8300 practicing child psychiatrists in this country.

 At a time when health care professionals are stretched thin, access to child psychiatry

 is at an all-time low, and the inequities of our health care system have been made clearer than ever. Can artificial intelligence (AI) help alleviate some of these problems?

AI, more broadly, is already “transforming the world,”

 and has found uses reading CT scan images, powering autonomous vehicles, assisting with national defense, and more. It is a powerful technology that will continue to be used in more areas, but how can it be applied to children’s mental health? Enter AI chatbots: programs designed to carry on realistic text- or voice-based conversations with patients and offer advice. They can be available to talk 24/7 and could be infinitely scalable, providing help to millions more people.

They are already helping in many situations. More than 40 different kinds of mental health chatbots exist, mostly focused on treating depression or autism.

 And they work: According to findings from a 2017 study by investigators from Stanford University School of Medicine and the developers of the aptly named “Woebot,” the bot was effective in reducing depression and seen as empathetic among college students after just 2 weeks of therapy modeled after cognitive behavioral therapy, a well-known method employed by many human therapists.

“The number of participants reporting that the bot felt empathic is noteworthy, and comments that referred to the bot as ‘he,’ ‘a friend,’ and a ‘fun little dude’ suggest that the perceived source of empathy was Woebot rather than the bot’s developers.”

In another study, Australian teenagers were tasked with helping develop a mental health chatbot, providing input on its gender and the type of slang it should use. (They chose a gender-neutral bot named Ash that was familiar with “brekky” for breakfast and “footy” for football.) The teenagers in that study said they were not always able to get help from existing school resources, and the investigators hoped the chatbot would be able to fill some of those gaps.

Chatbots have the potential to be a powerful tool to help millions—likely even more than the number of individuals diagnosed with anxiety or depression. As many individuals struggle with their mental well-being without a formal diagnosis of a mental health disorder, perhaps mental health chatbots could also help reduce the stigma associated with asking for help. But chatbots cannot be the only way we provide help to the most vulnerable among us. AI tools must be part of a holistic approach led by qualified and experienced doctors. Just as medication is not the only tool doctors use to treat illness or injury, chatbots must not be the only way we improve the mental health of children.

“Good mental health and well-being is about more than just a lack of illness,” the authors of the Australian study wrote. “Focusing on positive coping and building resiliency is important in establishing a holistic approach to health, addressing both physical and psychological experiences of youth. It is clear that young people being aware of mental health and well-being literacy has implications for the promotion of healthy well-being and early prevention of mental health disorders.”

We are facing a serious mental health crisis among young individuals, and we must use every tool at our disposal to help. AI is not replacing doctors—rather, it is a tool that allows for the democratization of care and increased access to combat noncomplex mental health issues. Doctors will remain a valuable asset for treating more complex disorders. However, chatbots are a powerful tool that should be adopted more widely because anything that can be used to improve the well-being of children must be considered.

is a graduate fellow in bioengineering and computer science at Stanford University School of Medicine. 

 is chief fellow in child and adolescent psychiatry at Stanford University School of Medicine.

US national and state-level prevalence of mental health disorders and disparities of mental health care use in children

The child psychiatrist will not see you now

. June 2, 2017. Accessed November 4, 2021.

3. Ghandour RM, Sherman LJ, Vladutiu CJ, et al. 

Prevalence and treatment of depression, anxiety, and conduct problems in US children

The major causes of death in children and adolescents in the United States

How artificial intelligence is transforming the world

. The Brookings Institution. April 24, 2018. Accessed November 4, 2021.

6. Abd-Alrazaq AA, Alajlani M, Alalwan AA, et al. 

An overview of the features of chatbots in mental health: a scoping review

Delivering cognitive behavior therapy to young adults with symptoms of depression and anxiety using a fully automated conversational agent (Woebot): a randomized controlled trial

Co-developing a mental health and wellbeing chatbot with and for young people

An estimated 7.7 million children have at least 1 mental health disorder. How can chatbots help?

Chatbots for Child Mental Health Care: Helpful, but Limited

First used decades ago, virtual psychiatric health care has been established as a viable alternative to in-person psychiatric services. In 2016, Hubley and colleagues performed a review of available research and concluded that a “large evidence base supports telepsychiatry as a delivery method for mental health services.”

 Luo et al established electronically delivered cognitive behavioral therapy (eCBT) as noninferior to in-person cognitive behavioral therapy (CBT) with respect to a variety of outcomes. They also found that eCBT is actually more effective than face-to-face CBT for reducing depression symptom severity.

While implementation of telepsychiatric care had been steadily increasing, barriers to widespread use remained. The 2 primary barriers were regulatory burdens and logistical difficulties associated with reimbursement, according to Keesara et al.

 Then, in spring 2020, the COVID-19 pandemic necessitated virtual care, and these barriers were rapidly removed. The pandemic prompted a shift in outpatient psychiatry services from primarily face-to-face to primarily virtual. Given the unprecedented nature of this paradigm shift, literature from the pre-COVID-19 era has lost some degree of relevance. However, a wealth of research on telepsychiatric health care has been performed since March 2020.

Chen et al described early experiences, reviewed key regulatory changes enabling widespread telepsychiatry, and provided clinical insights and suggestions.

 Yellowlees and colleagues detailed anecdotal experience at University of California, Davis Health and described generally positive experiences from both the provider and patient points of view.

 Another group of investigators published a nationwide multisite survey on patient attitudes. Of the 14,000 patients who were surveyed, approximately 3000 responded and results generally favored use of telepsychiatry.

In a retrospective cohort study of nearly 39,000 patient surveys, Ramaswamy et al found satisfaction with video visits to be significantly higher than satisfaction with in-person visits (94.9% versus 92.5%; p < 0.001).

 Additionally, Serhal and colleagues developed and validated a survey on patient satisfaction. They found that access, timeliness, and safety were predictors of patient satisfaction.

 Finally, a review on determinants of satisfaction showed the main determinants to be convenience and cost.

 It stands to reason that telepsychiatric care provides benefits associated with access, timeliness, safety, convenience, and cost, and current research generally indicates patient preference for virtual health care visits.

The psychiatric outpatient experience at Cleveland Clinic Akron General (CCAG) and its resident-run clinic has been rapidly and dramatically changed in response to the COVID-19 pandemic. Patient attitudes toward these changes (in particular, telepsychiatric care) have not been assessed. As such, our goal was to gather and report feedback—in both a quantitative and a qualitative manner—from our patients regarding their attitudes toward the conversion to virtual psychiatric care. We hope this feedback will allow us to improve delivery of care and other patient outcomes in the future.

The CCAG Institutional Research Review Board approved use of the minimum necessary protected health information for the conduct of this quality assurance initiative. Psychiatric patients at CCAG and its resident-run clinic who attended virtual visits (by video or phone conference) from April 1, 2020, to December 31, 2020, were identified by a review of electronic medical records. Email addresses for these patients were extracted. A short anonymous survey including a demographics section and a questionnaire was emailed to patients via the Research Electronic Data Capture (REDCap®) application, a secure web-based platform that automatically collects and stores responses. The questionnaire included 4 multiple-choice questions and 1 open-ended prompt and was designed to assess difficulty setting up video software/telephone, quality of visit, convenience, and overall preference (

Descriptive statistics were performed for data analysis using IBM SPSS (Version 24.0, Armonk, NY) statistical software. Frequencies with percentages, measures of central tendency, and measures of spread were presented as appropriate. Qualitative data in the form of responses to open-ended questions were also reported.

A total of 1259 surveys were administered, and 120 (9.5%) patients responded. Self-reported demographics are presented in 

One hundred one (84.2%) patients indicated setup was no challenge or only mildly inconvenient, and 18 (15.0%) indicated they had significant setup difficulties. Forty-eight (40.0%) patients felt virtual visits were of higher quality, 32 (26.7%) expressed neutrality, and 40 (33.3%) indicated in-person visits were of higher quality. Ninety-four (78.3%) patients indicated virtual visits were more convenient, 8 (6.7%) indicated neutrality, and 16 (13.3%) indicated in-person visits were more convenient. Lastly, 65 (54.2%) patients expressed an overall preference for virtual visits, 19 (15.8%) expressed no preference, and 36 (30.0%) expressed a slight or significant preference toward in-person visits (

Of the 120 respondents, 69 (57.5%) provided responses to the open-ended prompt. Of these, 32 (46.4%) expressed a preference for virtual visits, commented only on benefits of virtual visits, or commented only on drawbacks to in-person visits. Of their comments, 24 (75%) included language on convenience, 12 (37.5%) included language on comfort, 6 (18.8%) included language on safety (largely with respect to the COVID-19 pandemic), and 7 (21.9%) included language indicating virtual visits benefitted family and childcare situations. Thirteen (18.8%) respondents expressed a preference for in-person visits, commented only on benefits of in-person visits, or commented only on drawbacks to virtual visits. Their comments included language on the benefits of personal and emotional interactions, the ability to read body language, and the capacity to comfort and express care. The remaining 24 (34.8%) respondents did not express a preference, commented on benefits/drawbacks of both options, or did not provide a clear or relevant response.

Our results indicate that a great deal of benefit can come from telepsychiatric care. These findings corroborate those of other groups.

 Conceivably, increased convenience coupled with comparable quality would predict a significant preference toward virtual visits. However, the actual trend observed was modest. This may be explained by the heterogeneous nature of our patient population. For example, differences in access to technology, technological proficiency, access to transportation, travel time, availability, and personal finances are just a few variables that account for differences in preferences and opinions.

Even differences in diagnoses play a role. Within free-response answers, many patients indicated that they felt more comfortable attending virtual visits given their struggles with agoraphobia. It is worthy to note, however, that this “benefit” may not be in some patients’ best interest. For example, some providers may feel that exposure to places and situations that cause discomfort is necessary for recovery.

There were some important limitations to this initiative. Results may have been influenced by selection and non-response bias. For example, given that the survey was administered via email, it stands to reason that patients who are more comfortable using computer technology may have been more likely to respond. Also, demographic data show that 60.5% of responders reported attainment of a bachelor’s degree or higher, and 28.6% reported having attended some college or obtaining an associate’s degree. As such, results largely reflect the opinions and preferences of more highly educated patients and, therefore, do not accurately represent those of our patient population as a whole. Other types of bias that may be at play include recall, acquiescence, and question order bias.

The best way to meet diverse needs is to offer diverse options. Telepsychiatry represents an option that may be well suited to some and not as well suited to others. To test for associations between patient opinions and factors such as demographics, baseline clinical characteristics, and diagnoses, formal research is required. Telepsychiatry is a branch of health care delivery still in its infancy, and there is much to be discovered. Future research will inform on optimal ways to provide individualized psychiatric care.

is a psychiatrist at LifeStance Health, Upper Arlington, Ohio.

is a health scientist in the Department of Health Sciences at Cleveland Clinic Akron General, Akron, Ohio; she is also an assistant professor of psychiatry and a clinical assistant professor of obstetrics and gynecology at Northeast Ohio Medical University, Rootstown, Ohio.

is professor and chairman, Department of Psychiatry, Creighton University School of Medicine, Omaha, Nebraska, and chairman, Catholic Health Initiatives (CHI) Health Behavioral Health Services, Omaha, Nebraska.

1. Hubley S, Lynch SB, Schneck C, et al. 

A comparison of electronically-delivered and face to face cognitive behavioural therapies in depressive disorders: a systematic review and meta-analysis

 [published correction appears in EClinicalMedicine. 2021 Nov 02;41:101182]. 

Covid-19 and health care’s digital revolution

COVID-19 and telepsychiatry: early outpatient experiences and implications for the future

5. Yellowlees P, Nakagawa K, Pakyurek M, et al. 

Rapid conversion of an outpatient psychiatric clinic to a 100% virtual telepsychiatry clinic in response to COVID-19

Patient attitudes toward telepsychiatry during the COVID-19 pandemic: a nationwide, multisite survey

7. Ramaswamy A, Yu M, Drangsholt S, et al. 

Patient satisfaction with telemedicine during the COVID-19 pandemic: retrospective cohort study

8. Serhal E, Kirvan A, Sanches M, et al. 

Client satisfaction and experience with telepsychiatry: development and validation of a survey using clinical quality domains

A review of patient and provider satisfaction with telemedicine

COVID-19 has prompted an unprecedented shift toward virtual psychiatric health care—but how do patients and providers really feel about it?

Telepsychiatry: Diverse Needs Require Diverse Options

Avrim B. Fishkind, MD, discusses prescription digital therapeutics, highlighting one product in particular. He believes these FDA-approved hard- and software will be popping up more and more over the coming years.

has practiced emergency and community psychiatry for the last 25 years. He is a past president of the American Association for Emergency Psychiatry. In 2007, he founded JSA Health Telepsychiatry, one of the nation’s first 24/7 telepsychiatry companies. He is a frequent speaker at the American Psychiatric Association’s conferences on subjects from verbal de-escalation of agitation to digital futures for psychiatrists including telemedicine and prescription digital therapeutics. Dr Fishkind is currently Physician Advisor to Freespira and Chief Clinical Officer of Orbis Health. In his spare time, he races vintage cars and rebuilds vintage audio equipment, and enjoys exploring Houston with his wife and his college age twins.

Here's why the next step in mental health treatment is digital.

Prescription Digital Therapeutics and the Hunger for New Treatments

Alison M. Darcy, PhD, talks about how digital therapeutics for mental health can maximize work with therapists, and what a recent survey reports on the mental health statues of the United States.

is a clinical research psychologist, an adjunct faculty member in the Department of Psychiatry and Behavioral Sciences at Stanford University School of Medicine, and the founder and president of

68% of Americans reported their mental health could be improved. 

Surveying Attitudes Towards Digital Mental Health

For the past decade, I have taught 6-hour annual seminars on the history of psychiatry to psychiatric residents. This course was assigned to me, in part, because I have lived through the history of psychiatry for more than 50 years. Through this seminar, my residents and I have come to realize that each generation’s members, ourselves included, look back on earlier psychiatric concepts and practices, from the ancients right through recent times, with a mixture of bemusement, pity, and horror. Our bemusement and pity stem from previous eras’ well-intentioned but quaint or misguided ideas based on scientific ignorance, whereas our horror is evoked by seemingly inhumane practices inflicted on psychiatrically vulnerable populations.

Assuming humans survive and continue doing competent science, we can imagine psychiatrists of the year 2500, enjoying knowledge far more advanced than ours, and reacting to our ideas and practices, too, with bemusement, pity, and horror. From this viewpoint, let us try to envision the history of future psychiatry, to not only guess what might be possible in 2500, but also to imagine how 26th-century psychiatrists might perceive how we think and practice during our psychiatric moments in time.

Overall, we hope that they might view us sympathetically, as trying hard to do good work with what we have, constrained by our limited concepts; by our physics, chemistry, biology, psychological, and social science tools; and by our sociocultural-economic structures. This perspective reminds us that current theories, diagnoses, and methods of assessment, treatment, and prevention are at best temporary placeholders in the arc of psychiatric history, many destined to be replaced within a generation or 2.

Unable to search 2500’s PubMed, we cannot cite the next half-millennium’s scholarly publications. Although we are unable to foresee what are certain to be surprising discoveries, we can future-cast by extending contemporary research and technologies. We can also invoke developments envisioned by today’s keenest scientists.

 From such perspectives, what contemporary psychiatric sacred cows, practices, and professional delusions (or to be kinder, illusions) might 26th-century psychiatrists see as deserving to be thrown into the trash bin of psychiatric history?

In 2500, most of today’s aminergic, cognitive, psychoanalytic, and family process theories are likely to be viewed as 

; 26th-century psychiatry might view our biopsychosocial approach as a good start for an overarching framework, but as generally simplistic in details and understanding.

By 2500, scientists may have clarified the mechanisms and impacts of genetic, epigenetic, and huge varieties of elusive physical and psychosocial environmental forces. We may know how psychobiological processes, from molecular to transpersonal over lifespans, influence psychiatric health. Consequently, models of the mind and brain, coping and adaptation, and family and social function will evolve accordingly. As more nuanced mechanisms are delineated, old theories will drop and new ones will emerge.

In 2500, most of today’s assessment methods will be considered archaic and excessively labor intensive. Predominantly dependent on prolonged self-reported histories elicited face to face, human-to-human psychiatric assessment interviews will be regarded as inefficient, incomplete, and remarkably error-prone, faulted for being susceptible to ubiquitous and influential 

By 2500, such biases will have been (hopefully) largely eliminated with the use of artificial intelligence (AI)–assisted assessment tools. Human assessors will come into play only after initial screening assessments have been completed. Constantly refined over years, mechanisms for obtaining dynamical biopsychosocial fingerprints might produce highly nuanced and individualized 

 Starting with genomic and epigenomic analyses in utero (“You on a Chip”),

 such fingerprints will utilize massive data stores, integrating information about neurobiology, temperament, coping styles, personality, and life events derived from individual, family, school, social connection, AI usage, and other registry sources. To all this might be added information from responses to individualized, virtual reality–based scenario simulations and from high-definition, whole-brain, dynamic scans assessing connectivity among numerous interacting brain regions. Each fingerprint will be compared with those of billions of individuals going back for generations.

In 2500, most of today’s psychiatric diagnoses as compiled in DSM-5 will be viewed as primitive, superficial descriptions at best, on par with antiquated terms like rubor, calor, and dolor, or perhaps pneumonia in general medicine.

By 2500, psychiatric diagnoses might be calculated along lines of precision psychiatry, encompassing individuals’ cell-specific diagnoses, brain functional neuroanatomy and connectivity patterns, interpersonal interactions, life events, and appraisals of deviant behaviors, all contextualized within patients’ micro- and macro-cultural frameworks. What we now call multiple comorbidities and varieties of major psychiatric disorders will be cast as highly elaborated and differentiated 

 These patterns might calculate patients’ stress–vulnerability balances and psychological reserves, indicate warning signs and risks, and predict the likelihood of a specific patient erupting in an episode of suicide or violence.

By 2500, virtually all of today’s psychiatric medication, somatic, and psychotherapeutic treatments will be viewed as archaic. Psychiatrists in 2500 will shudder at the hit-or-miss application and modest effectiveness of our biological and psychotherapeutic interventions as well as their many adverse effects.

 They will find it hard to understand why clinicians in the early 21st century permitted patients with substantial psychic pain related to psychosis, depression, and cravings to suffer for days and weeks before being administered substantial relief.

 They will feel sorry that patients had to waste so much time in ineffective treatments (in addition to traveling to and from these treatments) and endure years of poor quality of life.

By the year 2500, psychiatric patients might access a host of wearable and implantable devices that both monitor ongoing status and administer experiential (eg, educational, interactive, psychotherapeutic) and direct biological interventions.

For reasons of efficiency and effectiveness, AI-based personalized robots (robotherapeutic i-BOTs

) treatments might prevail. I-BOTs could provide social support and companionship on one hand, and, on the other, train patients to develop more effective social skills with specific individuals, individuals in general, and groups. Based on moment-to-moment data from personalized smartphones and auditory-visual-physiological sensors, these devices might detect when individuals experience difficulties and initiate first-order personal coaching, which might up-regulate to personalized psychotherapies. When i-BOTS sense that their resources are insufficient to adequately manage difficulties, they might signal a need for higher-level services to the patient’s personal health concierge, formerly known as a primary care clinician, for triage to higher levels of care, including human clinicians, who can (as it were) think outside the BOT.

By means of i-SIM therapeutics and virtual reality–based games, which could use holographic avatars to reenact earlier life events and current stressful interpersonal or other toxic environments, patients might learn to overcome fears, fixations, deficient emotional regulatory capacities, and other challenges.

For interventions requiring biologicals, the psychiatrists of 2500 may be able to tailor everything from medication types and dosing schedules to delivery systems and routes of administration, basing them on patient-centered analyses using genes, epigenetic patterns, brain circuitry, microbiomes, and psychosocial factors.

Neurostimulatory interventions might feature stereotactic membrane–like skullcaps, capable of focusing deep ultrasound and other noninvasive stimuli at precise targets to promote local neogenesis.

 Optogenetic-guided retroviral interventions might enhance and sculpt specific brain nuclei and connection pathways.

 Umpteenth-generation CRISPR-like genetic interventions might reverse neuropathological processes associated with excessive pruning, dysregulated emotional and attentional centers, impulsivity, habit formation, and craving.

 Anti-neuroinflammatory agents and neuroplaque removers might inhibit and reverse degenerative and other psychiatric diseases,

 while individualized biomic brews might enhance healthier microbiomes.

Since unrelenting psychic pain in anxiety, mood, psychotic, and other disorders can activate and kindle terrifying, auto-traumatizing memory patterns (further augmenting posttraumatic stress disorder), patients in extreme psychic pain might be treated with psych-anesthetics. These agents might acutely alleviate psych-aches and induce therapeutic sleep, during which neurogenerative treatments might be applied.

With respect to treatment settings, few ambulatory patients will travel to clinicians’ offices or treatment centers, as virtually all interventions will be amenable to remote administration via teletreatment.

 Who would want to spend hours traveling to and from clinicians’ offices when 3-D telemedicine permits individual, tele-family, and tele-group therapy, even with participants in different locations, time zones, and continents?

When necessary, individuals who are highly disturbed will still be treated in specialized sanctuary-like medical treatment centers, usually for brief periods of time. Thanks to thorough assessments and deep data mining, extremely accurate assessments of needs and risks should be available for comprehensive discharge planning.

For socialization, education, training, and rehabilitation purposes, actual live human gatherings are likely to endure, as they have for many centuries.

In 2500, psychiatrists will view the early 21st century as virtually devoid of meaningful efforts to prevent psychiatric disorders. They will see that even when risk factors were well known and identified, few public health efforts were designed to minimize subsequent emergence of psychiatrist disorders.

By 2500 (with hopefully few biases), public health programs might identify and conduct more intensive assessments of high-risk neighborhoods, individuals, and families, and better provide indicated social and individual-based interventions. Thanks to immediate diagnoses of new pregnancies—and based on intrauterine genome and epigenetic scans as well as prenatal neurotemperamental assessments—precision interventions could assist with repairs based on advanced gene- and epigene-manipulation techniques, while other individually tailored preventive interventions may also be implemented.

In 2500, psychiatrists will find many of our contemporary practices for training, certifying, and recertifying psychiatrists to be laughable. They may appreciate that the quasi-apprentice model was probably the best that could be achieved under current circumstances. They will also note that, although well intended, cumbersome administrative superstructures required that practitioners, academic faculty, trainees, and staff waste huge amounts of time in mind-numbing educational training and assessment rituals of unproven worth. They will further note that many tedious rituals revolved around scheduling, electronic medical records, charting, billing, ineffective meetings, and other time- and energy-swallowing administrative activities.

By 2500, based on increasingly effective and evidence-based learning methods, psychiatric training will utilize large numbers of simulators that will challenge, train, and meticulously assess trainees’ competencies in handling nuanced scenarios. Carefully constructed games will assess clinical decision-making and practice skills, knowledge, professionalism, communication, and systems-based care, each tailored for specific environments in which the clinicians serve. Certifications for specific necessary skills will be fluidly conducted, with in-house, on-the-job, and just-in-time learning, training and certification, renewable as new techniques and situations arise. Medical record keeping and other administrative functions will largely be allocated to natural language, administratively coded robo-scribes. These AI devices might also summarize and bring clinicians’ attention to salient portions of patients’ medical information before their scheduled encounters.

In 2500, psychiatrists will find themselves sharing several characteristics with early 21st century psychiatrists. Given human nature, at least some citizens will still be prone to intermittently go off the rails; psychiatric disorders are still likely to be stigmatized; and psychiatric gurus are still likely to believe that there are too many patients, too few clinicians, too little knowledge, too few resources, and too little money for what they do. And, given human nature, in 2500 a high value will still be placed on deep, confiding, empathic interpersonal relationships, including clinician-patient relationships.

Why bother teaching the history of psychiatry by highlighting how future generations might appraise today’s psychiatric practices, and envisioning how they might practice? My intention is not to throw cold water on the sincere efforts of today’s dedicated, well-intentioned professionals. Rather, I hope these musings offer a perspective on our place in the history of psychiatry’s arc over time. We try to do the best we can with what we have, but we should not be self-satisfied about our efforts.

If any of these fanciful visions of future psychiatric practice seem worthwhile, they might provide aspirational goals, and motivation to realize these goals. And, if any of these fanciful visions seem dystopian, they might provoke us to develop better alternatives.

 is a professor in the Department of Psychiatry, University of Colorado School of Medicine, Anschutz Medical Campus.

A History of Psychiatry: From the Era of the Asylum to the Age of Prozac

The Future of the Mind: The Scientific Quest to Understand, Enhance, and Empower the Mind

3. Ghaemi SN. Paradigms of psychiatry: eclecticism and its discontents. 

4. Ghaemi SN. The rise and fall of the biopsychosocial model. 

5. Braus BR, Rummans TA, Lapid MI, et al. Clinicians and cognitive bias: a case of frontotemporal dementia misdiagnosed as conversion disorder. 

6. Shah NJ, Arrubla J, Rajkumar R, at al. Multimodal fingerprints of resting state networks as assessed by simultaneous trimodal MR-PET-EEG imaging. 

7. Herland A, Maoz BM, Das D, et al. Quantitative prediction of human pharmacokinetic responses to drugs via fluidically coupled vascularized organ chips. 

8. Rush AJ, Ibrahim HM. Speculations on the future of psychiatric diagnosis. 

9. Tarrier N, Khan S, Cater J, Picken A. The subjective consequences of suffering a first episode psychosis: trauma and suicide behaviour. 

10. Cormack F, McCue M, Taptiklis N, et al. Wearable technology for high-frequency cognitive and mood assessment in major depressive disorder: longitudinal observational study. 

11. Sequeira L, Battaglia M, Perrotta S, et al. Digital phenotyping with mobile and wearable devices: advanced symptom measurement in child and adolescent depression.

12. Fiske A, Henningsen P, Buyx A. Your robot therapist will see you now: ethical implications of embodied artificial intelligence in psychiatry, psychology, and psychotherapy.

14. Serretti A. The present and future of precision medicine in psychiatry: focus on clinical psychopharmacology of antidepressants. 

15. Bowary P, Greenberg BD. Noninvasive focused ultrasound for neuromodulation: a review. 

16. Lewis PM, Thomson RH, Rosenfeld JV, Fitzgerald PB. Brain neuromodulation techniques: a review. 

17. Barnett SC, Perry BAL, Dalrymple-Alford JC, Parr-Brownlie LC. Optogenetic stimulation: understanding memory and treating deficits. 

18. Duan J, Sanders AR, Gejman PV. From schizophrenia genetics to disease biology: harnessing new concepts and technologies. 

19. Rosenblat JD. Targeting the immune system in the treatment of bipolar disorder. 

20. Evrensel A, Ünsalver BÖ, Ceylan ME. Therapeutic potential of the microbiome in the treatment of neuropsychiatric disorders. 

21. Hulsbosch AM, Nugter MA, Tamis P, Kroon H. Videoconferencing in a mental health service in The Netherlands: a randomized controlled trial on patient satisfaction and clinical outcomes for outpatients with severe mental illness. 

Telepsychiatry and Health Technologies: A Guide for Mental Health Professional

. American Psychiatric Association Publishing; 2018

23. Arango C, Díaz-Caneja CM, McGorry PD, et al. Preventive strategies for mental health. 

24. Miller DD, Brown EW. Artificial intelligence in medical practice: the question to the answer? 

25. Zulman DM, Haverfield MC, Shaw JG, et al. Practices to foster physician presence and connection with patients in the clinical encounter. 

What will psychiatric practice look like in the future?

Psychiatry in the Year 2500: Lessons Learned From History

It can be hard to gauge medication adherence when relying only on patients’ self-reported usage. Inaccurate reports can weaken the therapeutic alliance and lead to worse treatment outcomes. Could digital mental health tools provide more objective data, and thereby improve patient care?

In this Mental Health Minute, Alexis Skoufalos, EdD, considers how digital tools might allow patients to track, document, and report outcomes and other measures of care quality. This video builds on her latest study in 

 is associate dean and director of the Doctor of Health Science in Population Health at Thomas Jefferson University.

Have you used digital therapeutics in your practice? 

Share comments with your colleagues by emailing 

. Comments may be shared online pending review and editing for style.

Patients’ reports of medication adherence are sometimes inaccurate, but digital tools offer opportunities to collect objective data.     

How Can Digital Tools Improve Medication Adherence? 

Are You Ready for the Annual Psychiatric Times World CME Conference? 

At the Annual Psychiatric Times World CME Conference, an all-star lineup of presenters will share their latest research. Check out this preview of the full agenda. 

Paranoia, or fear of deliberate harm from others, is among the most common symptoms of schizophrenia-spectrum psychoses, and it also occurs in emotional disorders and other severe mental illnesses, placing 

 on the lives of patients and their families.

 Innovative therapeutic interventions are desperately needed to overcome paranoia and improve outcomes.

Existing therapies, whether pharmacological or psychotherapeutic, have widely recognized limitations in engagement, adherence, and effectiveness. Antipsychotic medication is the mainstay of treatment for most patients. However, distressing symptoms persist in 

 of those who take them, functional outcomes are frequently poor, and many discontinue treatment periodically or permanently, concerned about adverse effects, such as weight gain and metabolic syndrome.

 Psychological therapies, such as cognitive behavioral therapy for psychosis (CBTp), have been shown to improve psychotic symptoms, including paranoia, when added to treatment as usual in large clinical trials.

 But effects need to be improved, and implementation in routine care is poor. Individuals are often not willing or able to access and make use of therapy, particularly those from minoritized groups. There is a scarcity of trained therapists, with estimates suggesting only 0.3% of mental health consumers in the United States have access to CBTp.

A transformation in psychosis care is needed; digital technology has the potential to deliver it by improving the patient experience, therapy effectiveness, and 

 Digital interfaces can be designed to support personalization and accessibility, and they can assist engagement and adherence. Web and mobile technologies are also more readily scaled up than traditional interventions and can be designed to reduce training demands.

A New Frontier: Digital Therapeutics for Paranoia

To the best of our knowledge, metacognitive training (MCT), developed by Steffen Moritz, PhD, MSc, and colleagues, was the first psychological intervention for delusional beliefs that incorporated technology. It used desktop programming and presentation software to provide illustrative games and tasks. Group versions of this training approach were shown to have small to moderate effects on delusions, although it is important to note that patients with severe delusions were excluded and the basic digital materials were not personalized, limiting its ability to address challenges to therapy accessibility and adherence. More recently, individualized delivery of metacognitive therapy (MCT+) has been developed and evidence suggests it may have better effects on delusions, though there are only a few relatively

Building on this pioneering work, recent digital therapeutics for psychosis have sought to harness innovative technology such as virtual reality, social networking, mobile apps, and web-based therapies. Whilst preliminary findings are encouraging, few of these have been tested in large clinical trials, and those 

 In 1 of the largest trials to date, Depp et al (2019)

 investigated the effects of augmenting a single session of CBTp with an app, CBT2Go, compared to a single session of self-monitoring supported by an app or treatment as usual (TAU), in individuals with severe mental illness (N = 255). They found a small effect on general symptomatology favoring the CBT2Go and self-monitoring conditions over TAU. However, the impact on paranoia was not specifically examined, and the authors also note that baseline levels of psychosis symptoms were low. The FOCUS mobile app adopts a similar therapeutic approach, although it was investigated as a standalone app to support self-management in its most rigorous evaluation. Ben-Zeev et al (2019)

 compared FOCUS to a recovery group intervention in 163 patients with severe mental illness. They found no between-group differences on clinical outcomes, nor any within-group changes in psychosis symptoms.

These trials suggest that app-based technologies may safely and acceptably augment therapy for paranoia although, but they have yet to demonstrate meaningful clinical outcomes in robust designs. Crucially, there is little evidence that digital therapeutics have been designed to address healthcare inequalities. Indeed, emerging findings indicate digital therapies may actually exacerbate exclusion, with those from privileged demographic groups more likely to access and benefit from them.

In addressing the challenge of transforming psychosis care, we have developed SlowMo, an innovative digitally supported therapy for paranoia. SlowMo is based on the principles of CBTp and leverages technology to address health care inequalities, through enhancing therapy appeal and user friendliness for a 

 SlowMo therapy targets an evidence-based mechanism, or active ingredient, that we and others have shown can fuel paranoia.

 Drawing on the term popularized by Daniel Kahneman, PhD,

, comprising of a jumping to conclusions bias (forming rapid judgements using limited information) and belief inflexibility bias (reduced capacity for reflecting on one’s own beliefs, changing them in the light of new evidence, and considering alternatives). SlowMo is designed to help patients to slow down for a moment, encouraging awareness of a tendency to jump to conclusions and find new ways of feeling safer.

We followed human-centered design principles in the hopes of improving the user experience of therapy and radically enhancing engagement and adherence.

 Our multidisciplinary collaboration of individuals with lived experience, industrial designers, software developers, clinicians, and researchers developed SlowMo over 3 years through 7 cycles of prototyping. To address health care inequalities, we codesigned with stakeholders from diverse demographic backgrounds, who could identify therapy barriers and facilitators, and optimize the inclusivity of the design. SlowMo therapy consists of a web app delivered on a touchscreen laptop to support the delivery of face-to-face individual cognitive-behavioral therapy sessions, which is synchronized with a mobile app for use in daily life (

A Journey of Visualizing Thoughts and Thinking Habits

SlowMo uses new, responsive technologies to support the visualization of thoughts—as bubbles—and their associated thinking habits. This visual concept aims to enhance cognitive therapy by providing an appealing and tangible means of communicating subjective experience. The design minimizes the information processing demands, thereby lowering a significant barrier to engagement with traditional talking therapies. Patients interact with their personalized SlowMo thought bubbles using a touchscreen, altering speed, size, and transparency to reflect related thinking habits, distress, and conviction.

The SlowMo webapp used during sessions presents therapy as a journey, with interactive features including information, animated stories, and games. As well as tailoring worry and positive thought bubbles, patients record key learning points and future plans, with all content synchronized to the mobile phone app for access outside of sessions. The mobile app allows patients to notice their fears and thinking habits as they occur in daily life and helps them to find other ways of managing distressing experiences. It consists of a redesigned CBT thought record for managing paranoia and attempts to overcome the limitations of paper versions, which can be an inaccessible means of support during peak moments of distress. SlowMo uses a native mobile app with opt-in data transfer, addressing patients’ privacy concerns, which was a salient theme during the design research. The mobile app relies on user-initiated interaction and optional push notifications, to accommodate those who wish to have control over their data and find passive monitoring intrusive.

Delivering Evidence-Based Benefits for Recovery in Psychosis

SlowMo therapy has been iteratively developed over the past 10 years, with a number of studies demonstrating the efficacy of previous versions of the 

a large-scale randomized controlled trial

 with N = 362 participants recruited from United Kingdom (UK) clinical psychosis services (

, compared to the minimum 16 sessions recommended by the UK’s National Institute of Health and Care Excellence guidelines for schizophrenia treatment.

Remarkably, despite being half the minimum recommended length, SlowMo had a moderate effect on the measures of delusion severity commonly used in trials (Scales for Assessment of Positive Symptoms, SAPS

 and Psychotic Symptom Rating Scales, PSYRATS

)—for which meta-analyses have found only small effects.

 These positive effects sustained or grew over 6 months, with benefits cascading into other areas, including self-concept, wellbeing, and quality of life. Less consistent improvements were found for referential delusions, meaning that the Green Paranoid Thoughts Scale (GPTS)

 total, which is comprised of ideas of reference and persecution subscales, was not significant at 6 months. However, a psychometrically improved version of this measure, published whilst the trial was being conducted, did reach conventional levels of 

A new digital therapeutic helps patients with psychosis to slow down and manage their fears.

Digital Psychiatry