Sophia Vinogradov, MD

Telepsychiatry and digital technologies have been touted for over a decade as effective methods for providing care to individuals with serious mental illness, particularly those in rural areas and for whom brick-and-mortar mental health services are not accessible.

 Up until the (very) recent past, health systems and providers were slow to embrace broader use of digital health tools due to legal and regulatory concerns over privacy, confidentiality, and logistical barriers.

 has forced health systems to transition rapidly to 

, while psychiatric research has ground to a halt. The use of technology that supports remote methods of communication, assessment, and mental health care has become more timely and important for researchers to adopt, refine, and implement.

Virtual and mobile device-based methodologies are particularly suited to early psychosis research due to the age range of participants, namely adolescence and early adulthood. As “digital natives,” they are oriented to and well-versed in these modes of communication. Many of our clinical and research programs have been developing, using, and testing digital tools for some time. Here, we review currently available methods, based on our experience, that can be used to adapt and jumpstart early psychosis research as the field prepares for the new landscape that we will face, likely through 2021 and beyond.


	The move to telehealth platforms like Zoom, Doxy.me, and MyChart video visits has enabled remote clinical assessment in new and exciting ways. Research participants can be recruited widely via Facebook or other online platforms, complete online screenings, and participate in clinical assessments remotely. Individuals who might not participate otherwise may prefer remote options, such as those without access to transportation or those who live in rural areas as well as patients with significant negative symptoms or anxiety.

Telehealth allows research participation at more convenient times and locations for the individual. Although some nonverbal cues may be missed, appropriate camera placement can help raters assess for movement abnormalities or other nuanced issues. Remote assessment requires appropriate equipment and a stable internet connection for both research staff and the participant, as well as steps to ensure privacy and confidentiality such as a private space for the interview. With specific training in conducting remote assessments, the benefits can outweigh the challenges.

Digital technologies can to collect data actively (eg, a person answers questions pushed to their smartphone) or passively (eg, number of text messages sent in a 24-hour period). These methods enable the creation of an individual-level dynamic trajectory for research participants that can be modeled on their own or in conjunction with other clinical, cognitive, and biomarker data.

Digital technology in clinical research is a rapidly evolving field with many competing user platforms, sensors, and data types introduced regularly. Therefore, platform selection should continually consider and assess flexibility, robustness, data integrity, and data security.


	We reviewed a wide variety of platforms for our research and identified two as examples that seem to be particularly geared towards clinical research. Pear Therapeutics has developed the only prescription digital therapeutic and digital diagnostic platform, to date, that has received FDA Market Authorization to treat neuropsychiatric disease and deliver regular patient-facing assessments, such as 

, which is FDA-authorized for the treatment of chronic insomnia. The platform is developed under Good Manufacturing Processes (GMP), is HIPAA compliant and is supported by a call center for patients and clinicians. Moreover, it offers double-deidentified encryption of PII and PHI.

Mobi by Safari Health, Inc, allows active HIPAA-compliant data collection from user smartphones on Android and Apple devices.

 Mobi provides individual assessment with flexible timeframes (eg, daily, weekly) using brief (≤10 questions) surveys that are pushed to the individual’s smartphone and are optimized for patient engagement and clinical validity. Mobi allows user customization of survey questions at the individual or group level for flexibility. Data are visualized on an encrypted dashboard, designed for clinical or research monitoring, and data files can be easily downloaded for analysis. Flags indicate critical responses warranting immediate follow-up (eg, suicidal ideation). Such digital platforms can easily implement ecological momentary assessments that avoid memory bias and provide rich, in-depth data as well as passive data such as call and text utilization parameters, accelerometry, geolocation, heart rate variability or sleep architecture.

The measurement of cognition is another area for which digital technologies are becoming available. Although cognitive deficits are core features of psychotic disorders, accounting for a substantial portion of disability, there are no approved treatments to date despite over two decades of active efforts to develop them.

 Therefore, measuring cognition is a core aspect of clinical research phenotyping that can provide new insights into the underpinnings of 

In the era of big data analytics, major platforms for large scale cognitive testing have been developed (eg, Amazon mechanical TURK) that are essentially crowdsourcing applications that enable the collection of a wide range of data including behavioral, cognitive, and individual difference measures. These platforms are optimized for scale rather than precision and may not be well-suited for targeted phenotypic measurement in smaller, well-characterized clinical samples. However, these approaches provide a paradigm that can be further refined and adapted for clinical research.

More recent development includes tools such as Eprime Go, a version of a widely used software package for constructing and administering cognitive tasks for research.

 This version converts software running on a laboratory computer and shares it with individual participants to run on their home computer.

TestMyBrain.org provides the opportunity to collect large samples of cognitive data via online and mobile device testing. TMB has the ability to collect high quality data, particularly in areas of cognition and perception, that is comparable to what is collected in typical laboratory settings.

 These approaches offer a new tactic for conducting early psychosis research remotely.


	Rapidly advancing technology allows us to envision a not-so-distant future in which translational psychiatry researchers can use a suite of remotely collected biomarkers in their research. Currently available tests for pregnancy and urine toxicology can easily be implemented at home. Major firms such as Apple and Google have invested in health applications of remote technology for broad medical applications, including EKG tracing, pulse oximetry, glucometry, and others. Some of these can be informative for psychiatry as well, such as heart rate variability, which is abnormal in many psychiatric disorders.

Others that are not included in general medical panels can be equally useful in our field, eg, saliva cortisol testing, skin conductance responses, pupil diameter, and drug use tests. Finally, improvements in actigraphy research may lead to detection of relevant behavioral signatures such as smoking a cigarette, brushing one’s teeth, and others. We expect that the current situation will further stimulate research and development into remote biomarker collection, potentially providing important information not currently available even in laboratory-based research.


	Although brain-based assessments are less mobile than the domains described above, there have been recent advances. Mobile EEG devices are already in commercial use for biofeedback and other indications. Some of these devices use a band worn on the forehead while others use a bonnet worn over the head (

). Although data from these devices have not been of comparable quality with those from laboratory-based assessments until now, we expect that this will rapidly change because of the demands of the current environment and may become a critical tool for translational psychiatry research in the coming period.

Mobile MRI scanners are currently available and applied in special circumstances.

 With proper sanitization and other precautions, these may be a potential alternative to hospital- or university-based stationary MRI scanners that may be perceived by research participants as incurring a greater risk of COVID-19 infection.


	Online methods are already well suited to support efficient and consistent cross-site collaboration in training, study implementation, and fidelity maintenance, particularly for international collaborations where travel is challenging-a reality in the “new normal.” Online methods for disseminating evidence-based practices can be similarly used to scale up to a large number of research sites. Online training modules are useful in promoting knowledge and coupling such tools with multi-dimensional, multi-site training and supervision using telehealth methods that can deliver enhanced adherence and sustained fidelity.

 is Associate Professor, UCSF Department of Psychiatry; 

 is Chief, Center of Excellence in Psychotic Disorders, McLean Hospital and the William P. and Henry B. Test Professor of Psychiatry at Harvard Medical School; 

 is Associate Professor, Department of Psychiatry and Behavioral Sciences, UC Davis Health; 

 is Assistant Professor, Department of Psychiatry, University of Minnesota Medical School; 

 is Professor and Head, Department of Psychiatry, University of Minnesota Medical School; 

 is Associate Professor, Clinical Psychiatry, Columbia University College of Physicians and Surgeons and New York State Psychiatric Institute. 

 report no conflicts on interest concerning the subject matter of this article; 

 reports that she is a co-founder and shareholder of Mobi; 

 reports that she is an independent consultant for NAVIGATE training for first episode psychosis.

-The authors would like to thank Dr Jeffrey Leiberman for his very helpful comments on an earlier version of the manuscript for this article.

1. O’Reilly R, Bishop J, Maddox K, et al. Is telepsychiatry equivalent to face-to-face psychiatry? Results from a randomized controlled equivalence trial. 

. 2007;58:836-843.
	2. Lal S, Abdel-Baki A, Sujanani S, et al. Perspectives of young adults on receiving telepsychiatry services in an urban early intervention program for first-episode psychosis: a cross-sectional, descriptive survey study. 

. 2020;11:117.
	3. Kumar D, Tully LM, Iosif AM, et al. A mobile health platform for clinical monitoring in early psychosis: implementation in community-based outpatient early psychosis care. 

. 2018;5:e15.
	4. Niendam TA, Tully LM, Iosif A-M, et al. Enhancing early psychosis treatment using smartphone technology: a longitudinal feasibility and validity study. 

. 2018;96:239-246.
	5. Lesh T, Niendam T, Minzenberg M, Carter C. Cognitive control deficits in schizophrenia: mechanisms and meaning. 

Remote Data Collection With E-Prime 3.0: Introducing E-Prime Go for Remote Data Collection

. May 4, 2020.
	7. Fortenbaugh FC, DeGutis J, Germine L, et al. Sustained attention across the life span in a sample of 10,000: dissociating ability and strategy. 

. 2015;26:1497-1510.
	8. Hartshorne JK, Germine LT. When does cognitive functioning peak? The asynchronous rise and fall of different cognitive abilities across the life span. 

. 2015;26:433?443.
	9. Wilmer J, Germine L, Chabris C, et al. Capturing specific abilities as a window into human individuality: the example of face recognition. 

. 2012;29:360-392.
	10. Germine L, Nakayama K, Duchaine BC, et al. Is the web as good as the lab? Comparable performance from web and lab in cognitive/perceptual experiments. 

. 2012;19:847-857.
	11. Halberda J, Ly R, Wilmer JB, et al. Number sense across the lifespan as revealed by a massive Internet-based sample. 

. 2012;109:11116-11120.
	12. Vincent GM, Cope LM, King J, et al. Callous-unemotional traits modulate brain drug craving response in high-risk young offenders. 

. 2018;46:993-1009.
	13. Dixon L, Patel S. The application of implementation science to community mental health. 

. (In press).
	14. Covell N, Margolies P, Myers R, et al. State mental health policy: scaling up evidence-based behavioral health care practices in New York state. 

Articles

The use of technology that supports remote methods of communication, assessment, and mental health care has become more timely, and important for researchers to adopt, refine, and implement.

Research in the Time of COVID-19: Virtual and Digital Technology in Studies of Early Psychosis

Slideshows

Although the clinical symptoms of psychotic disorders are dramatic and are what most clinicians focus on as their treatment targets, impairments in a wide range of cognitive function are observed in these illnesses.

Common Neural Abnormalities in Psychotic Disorders

Over the last decade or so, our field has experienced a radical shift in our understanding of schizophrenia and other serious psychotic disorders, such as schizoaffective disorder and bipolar disorder with psychosis. We now understand that these are neurocognitive disorders (ie, how neural systems in the brain represent and process information). We also understand that they are neurodevelopmental disorders with genetic components and antecedents during gestation. The developmental course unfolds with increasing signs, symptoms, and cognitive dysfunction, until the onset of the first episode of psychosis during adolescence or early adulthood. Cognitive deficits are more significant determinants of functional outcome than are symptoms, although most current psychiatric treatments focus only (or mostly) on symptom management.

Accumulating evidence indicates that psychotic disorders constitute syndromes rather than diseases per se. Groups of patients can show a common clinical phenotype with multiple different etio-pathogenic factors that contribute to illness onset and expression. Patients with different clinical diagnostic phenotypes (such as schizophrenia rather than bipolar disorder with psychosis) can show similar underlying patterns of cognitive dysfunction and neurobiological abnormalities. New insights into the pathogenesis of psychotic disorders has spurred research focused on key domains of neural system dysfunction and on how brain plasticity mechanisms can be harnessed to drive healthier neural system functioning, improve cognition, and support functional recovery.

Cognitive and neural system dysfunction in psychotic disorders

Although the clinical symptoms of psychotic disorders are dramatic and are what most clinicians focus on as their treatment targets, impairments in a wide range of cognitive function are observed in these illnesses, ranging from the earliest stages of information processing in the brain, to higher-level abilities to abstract, to read social cues, to self-reflect, and to engage in meta-cognition. These various areas of cognitive dysfunction are accompanied by abnormalities in their neural system correlates. Moreover, patients exhibit a fair amount of cognitive and neural system heterogeneity, both within and across traditional diagnostic groupings-different patients will often have different patterns of impairment. Nonetheless, the following general categories of findings are commonly observed:

Sensory and perceptual processing abnormalities: A range of subtle deficiencies in representing and processing auditory and visual information, including socially relevant information, that can show functional consequences.

Learning and memory deficits: Impairments in how the brain encodes, learns, and makes decisions based on new information, including socially relevant information; responds to changes in rewarding or punishing contingencies in the environment; encodes and remembers autobiographical events (episodic memory); and learns associations and meanings (semantic memory).

Executive dysfunction (ie, impaired cognitive control): Deficits in attention and working memory; problems with inhibitory control; impaired abstraction, reasoning, planning, sequencing.

In psychotic disorders, the emergent picture is one of a brain that has undergone aberrant patterns of neurodevelopment, with reduced functional connectivity among key neural systems and reduced efficiency in its cognitive and socio-affective operations.

 This suggests that to improve cognition and functioning, it is essential to begin treatment as early as possible in the course of illness. Interventions designed to correct or compensate for abnormal neural system functioning must be started before irreversible maladaptive changes in the cortical and subcortical representational systems have taken root. Cognitive remediation and cognitive training approaches are used to improve cognition and are increasingly being studied in individuals who are in the early phases of illness.

Cognitive remediation vs cognitive training

Findings from meta-analyses indicate that a large variety of cognitive remediation protocols result in modest gains in overall cognitive functioning (effect sizes of about 0.4).

 No specific cognitive remediation method has been proved to be most effective.

The term “cognitive remediation” is a good descriptor for an intervention that is often multi-modal, using several approaches simultaneously (

). It can be conceptualized as a set of rehabilitative psychological treatments aimed at improving aspects of cognitive performance and behavior, which can of course also have effects on brain function and structure.

In contrast, the term “cognitive training” describes an explicitly neuroscience-informed approach to driving neural system plasticity in a well-defined neurocognitive operation or set of operations that may then generalize to other cognitive, behavioral, and functional domains. One of the best-studied examples of this kind of cognitive training utilizes approximately 10 hours of computerized visual speed-of-processing training in older adults. It has been associated with significantly improved cognition, lower rates of depression, and lower medical expenditures up to 10 years after the intervention.

In our own research, we often go one step further and use the term “targeted cognitive training” when describing cognitive training exercises that are defined in terms of specific neurocognitive target(s). Examples of specific neurocognitive targets include auditory processing abnormalities or social cognitive impairments (eg, eye gaze detection, facial emotion recognition, vocal prosody recognition). This approach is best conceptualized as a neurological treatment aimed at driving brain changes in neural systems that will generalize to improvements in cognitive performance and behavior (

). It can be delivered alone or as part of a larger treatment package.

 studied cognitive enhancement therapy (CET) vs enriched supportive therapy (EST) in 58 young individuals with prodromal schizophrenia; their average age was 26 years. CET combined 60 hours of computerized cognitive exercises using CogRehab software over 2 years in a specialized outpatient setting in conjunction with 45 sessions (90 minutes each) of social skills group exercises.

After 1 year of treatment, improvements on cognitive measures were not observed in CET participants compared with EST. After 2 years, moderate cognitive improvement was observable. Gains on measures of social cognition, cognitive style, social adjustment and symptoms were evident. At 1-year follow-up, gains on social and symptom measures were maintained and cognitive gains made during active training were significantly associated with improvement in functional outcome.

It is interesting to note that cortical surface area and gray matter volume at baseline, especially in temporal cortex, moderated the effects of CET on social cognitive outcomes. Compared with the EST group, the CET group had greater gray matter preservation in the left hippocampus, the parahippocampal gyrus, and the fusiform gyrus as well as gray matter increases in the left amygdala. These findings were related to improved cognition.

 The EST group also had improved resting-state and task-related functional connectivity in the prefrontal cortical regions; these findings were also associated with improved cognition.

The effects of mobile targeted cognitive training of auditory processing (TCT) versus control condition computer games (CG) were studied in a double-blind randomized trial.

 The study comprised  147 young adults with recent-onset schizophrenia (average age of 21 years) treated in two university-based first-episode clinics. Participants received laptop computers and instructions to complete 40 hours of training or the control condition games over approximately 8 weeks, performed on their own schedule at home. Check-ins and support were provided by the research team, but no other specific rehabilitation was provided beyond that which occurred as part of coordinated specialty care in the clinics.

Global cognition improvement was seen in the TCT group at immediate post-training assessments. Participants who completed at least 20 hours of TCT showed significant improvement in positive symptoms at 6-month follow-up compared with CG completers. Modeling of outcomes revealed that lower baseline global cognition and a higher level of education predicted greater improvement in global cognition after TCT.

In a subgroup of participants who agreed to undergo serial imaging, Ramsay and colleagues

 saw a significant positive correlation between left thalamic volume and increased global cognition in the TCT group (n = 22) and a negative trend in the computer games CG group (n = 22). Lower baseline symptoms were related to both left thalamic volume preservation and improvements in global cognition following the training.

This intervention was also studied in adults with persistent illness (average age, 40 years) and demonstrated improved cognition in the TCT group and higher serum brain-derived neurotrophic factor (BDNF) levels. Moreover, magneto-encephalographic and fMRI findings were consistent with enhanced cortical activation patterns during early sensory processing and prefrontal executive functions. Patterns of association were seen between enhanced cognition and/or cortical functioning and longer-term functional gains.

Are we ready to prescribe cognitive training to patients? It is one thing to have a highly promising evidence base emerging from clinical trials; it is another to know how to translate those findings into real-world treatment settings where models of implementation and reimbursement for cognitive remediation and/or cognitive training approaches are few and far between. Tools and education for clinicians that can allow them to assess and understand patterns of cognitive dysfunction in their patients are needed.

Cognitive training is a highly scalable and low-cost intervention, provided knowledgeable clinicians and staff can offer the necessary degree of support and guidance to patients. Ideally, cognitive training can be offered either as stand-alone cognitive treatment or as part of a broader enriched “remediation” program, depending on the needs of the individual; some individuals will be highly reliant on the structure and support of a formal multi-modal program and require group support, skills-building, etc, while others will be able to engage remotely, perhaps assisted only by an app that provides remote coaching (currently under study in our clinical trial NCT02782442).

Interested clinicians and consumers can consult 

, a non-profit organization whose mission is to help consumers choose effective and accessible mobile health technologies, including cognitive training programs (

 offers ratings on credibility, user experience, and the clarity of an app’s privacy policy). As we continue to refine our scientific understanding of cognitive training interventions for people with psychotic disorders, it may behoove us to begin to develop the programmatic infrastructure that allows us to assess cognition as a key treatment target for our patients.

Dr Vinogradov is Professor and Department Head, Donald W. Hastings Endowed Chair, Department of Psychiatry, University of Minnesota Medical School, Minneapolis, MN. 

Dr Vinogradov reports that she is on the Speakers’ Bureau for Alkermes and Mindstrong.

 Sheffield JM, Kandala S, Tamminga CA, et al. Transdiagnostic associations between functional brain network integrity and cognition. 

 McGurk SR, Twamley EW, Sitzer DI, et al. A meta-analysis of cognitive remediation in schizophrenia. 

 Wykes T, Huddy V, Cellard C, et al. A meta-analysis of cognitive remediation for schizophrenia: methodology and effect sizes. 

 Anaya C, Martinez Aran A, Ayuso-Mateos JL, et al. A systematic review of cognitive remediation for schizo-affective and affective disorders. 

 Edwards JD, Fausto BA, Tetlow AM, et al. Systematic review and meta-analyses of useful field of view cognitive training. 

 Eack SM, Greenwald DP, Hogarty SS, et al. Cognitive enhancement therapy for early-course schizophrenia: effects of a two-year randomized controlled trial. 

 Eack SM, Greenwald DP, Hogarty SS, Keshavan MS. One-year durability of the effects of cognitive enhancement therapy on functional outcome in early schizophrenia. 

 Eack SM, Hogarty GE, Cho RY, et al. Neuroprotective effects of cognitive enhancement therapy against gray matter loss in early schizophrenia: results from a 2-year randomized controlled trial. 

 Keshavan MS, Eack SM, Prasad KM, et al. Longitudinal functional brain imaging study in early course schizophrenia before and after cognitive enhancement therapy. 

 Eack SM, Newhill CE, Keshavan MS. Cognitive enhancement therapy improves resting-state functional connectivity in early course schizophrenia. 

 Fisher M, Loewy R, Carter C, et al. Neuroplasticity-based auditory training via laptop computer improves cognition in young individuals with recent onset schizophrenia. 

 Ramsay IS, Ma S, Fisher M, et al. Model selection and prediction of outcomes in recent onset schizophrenia patients who undergo cognitive training. 

 Ramsay IS, Fryer S, Boos A, et al. Response to targeted cognitive training correlates with change in thalamic volume in a randomized trial for early schizophrenia. 

 Fisher M, Holland C, Merzenich MM, Vinogradov S. Using neuroplasticity-based auditory training to improve verbal memory in schizophrenia. 

 Subramaniam K, Luks TL, Fisher M, et al. Computerized cognitive training restores neural activity within the reality monitoring network in schizophrenia. 

 Subramaniam K, Luks TL, Garrett C, et al. Intensive cognitive training in schizophrenia enhances working memory and associated prefrontal cortical efficiency in a manner that drives long-term functional gains. 

 Dale CL, Brown EG, Fisher M, et al. Auditory cortical plasticity drives training-induced cognitive changes in schizophrenia. 

 Fisher M, Mellon SH, Wolkowitz O, Vinogradov S. Neuroscience-informed auditory training in schizophrenia: a final report of the effects on cognition and serum brain-derived neurotrophic factor. 

Author | Sophia Vinogradov, MD

Articles

Research in the Time of COVID-19: Virtual and Digital Technology in Studies of Early Psychosis

Common Neural Abnormalities in Psychotic Disorders

Cognitive Training for Neural System Dysfunction in Psychotic Disorders