Returning Individual Research Results from Digital Phenotyping in Psychiatry.

Psychiatry is rapidly adopting digital phenotyping and artificial intelligence/machine learning tools to study mental illness based on tracking participants' locations, online activity, phone and text message usage, heart rate, sleep, physical activity, and more. Existing ethical frameworks for return of individual research results (IRRs) are inadequate to guide researchers for when, if, and how to return this unprecedented number of potentially sensitive results about each participant's real-world behavior. To address this gap, we convened an interdisciplinary expert working group, supported by a National Institute of Mental Health grant. Building on established guidelines and the emerging norm of returning results in participant-centered research, we present a novel framework specific to the ethical, legal, and social implications of returning IRRs in digital phenotyping research. Our framework offers researchers, clinicians, and Institutional Review Boards (IRBs) urgently needed guidance, and the principles developed here in the context of psychiatry will be readily adaptable to other therapeutic areas.

Predicting states of elevated negative affect in adolescents from smartphone sensors: a novel personalized machine learning approach.

BACKGROUND: Adolescence is characterized by profound change, including increases in negative emotions. Approximately 84% of American adolescents own a smartphone, which can continuously and unobtrusively track variables potentially predictive of heightened negative emotions (e.g. activity levels, location, pattern of phone usage). The extent to which built-in smartphone sensors can reliably predict states of elevated negative affect in adolescents is an open question. METHODS: Adolescent participants (n = 22; ages 13-18) with low to high levels of depressive symptoms were followed for 15 weeks using a combination of ecological momentary assessments (EMAs) and continuously collected passive smartphone sensor data. EMAs probed negative emotional states (i.e. anger, sadness and anxiety) 2-3 times per day every other week throughout the study (total: 1145 EMA measurements). Smartphone accelerometer, location and device state data were collected to derive 14 discrete estimates of behavior, including activity level, percentage of time spent at home, sleep onset and duration, and phone usage. RESULTS: A personalized ensemble machine learning model derived from smartphone sensor data outperformed other statistical approaches (e.g. linear mixed model) and predicted states of elevated anger and anxiety with acceptable discrimination ability (area under the curve (AUC) = 74% and 71%, respectively), but demonstrated more modest discrimination ability for predicting states of high sadness (AUC = 66%). CONCLUSIONS: To the extent that smartphone data could provide reasonably accurate real-time predictions of states of high negative affect in teens, brief 'just-in-time' interventions could be immediately deployed via smartphone notifications or mental health apps to alleviate these states.

A comparison of methods to harmonize cortical thickness measurements across scanners and sites.

Results of neuroimaging datasets aggregated from multiple sites may be biased by site-specific profiles in participants' demographic and clinical characteristics, as well as MRI acquisition protocols and scanning platforms. We compared the impact of four different harmonization methods on results obtained from analyses of cortical thickness data: (1) linear mixed-effects model (LME) that models site-specific random intercepts (LMEINT), (2) LME that models both site-specific random intercepts and age-related random slopes (LMEINT+SLP), (3) ComBat, and (4) ComBat with a generalized additive model (ComBat-GAM). Our test case for comparing harmonization methods was cortical thickness data aggregated from 29 sites, which included 1,340 cases with posttraumatic stress disorder (PTSD) (6.2-81.8 years old) and 2,057 trauma-exposed controls without PTSD (6.3-85.2 years old). We found that, compared to the other data harmonization methods, data processed with ComBat-GAM was more sensitive to the detection of significant case-control differences (Χ2(3) = 63.704, p < 0.001) as well as case-control differences in age-related cortical thinning (Χ2(3) = 12.082, p = 0.007). Both ComBat and ComBat-GAM outperformed LME methods in detecting sex differences (Χ2(3) = 9.114, p = 0.028) in regional cortical thickness. ComBat-GAM also led to stronger estimates of age-related declines in cortical thickness (corrected p-values < 0.001), stronger estimates of case-related cortical thickness reduction (corrected p-values < 0.001), weaker estimates of age-related declines in cortical thickness in cases than controls (corrected p-values < 0.001), stronger estimates of cortical thickness reduction in females than males (corrected p-values < 0.001), and stronger estimates of cortical thickness reduction in females relative to males in cases than controls (corrected p-values < 0.001). Our results support the use of ComBat-GAM to minimize confounds and increase statistical power when harmonizing data with non-linear effects, and the use of either ComBat or ComBat-GAM for harmonizing data with linear effects.

Functional connectomics of affective and psychotic pathology.

Converging evidence indicates that groups of patients with nominally distinct psychiatric diagnoses are not separated by sharp or discontinuous neurobiological boundaries. In healthy populations, individual differences in behavior are reflected in variability across the collective set of functional brain connections (functional connectome). These data suggest that the spectra of transdiagnostic symptom profiles observed in psychiatric patients may map onto detectable patterns of network function. To examine the manner through which neurobiological variation might underlie clinical presentation, we obtained fMRI data from over 1,000 individuals, including 210 diagnosed with a primary psychotic disorder or affective psychosis (bipolar disorder with psychosis and schizophrenia or schizoaffective disorder), 192 presenting with a primary affective disorder without psychosis (unipolar depression, bipolar disorder without psychosis), and 608 demographically matched healthy comparison participants recruited through a large-scale study of brain imaging and genetics. Here, we examine variation in functional connectomes across psychiatric diagnoses, finding striking evidence for disease connectomic "fingerprints" that are commonly disrupted across distinct forms of pathology and appear to scale as a function of illness severity. The presence of affective and psychotic illnesses was associated with graded disruptions in frontoparietal network connectivity (encompassing aspects of dorsolateral prefrontal, dorsomedial prefrontal, lateral parietal, and posterior temporal cortices). Conversely, other properties of network connectivity, including default network integrity, were preferentially disrupted in patients with psychotic illness, but not patients without psychotic symptoms. This work allows us to establish key biological and clinical features of the functional connectomes of severe mental disease.

An Ethics Checklist for Digital Health Research in Psychiatry: Viewpoint.

BACKGROUND: Psychiatry has long needed a better and more scalable way to capture the dynamics of behavior and its disturbances, quantitatively across multiple data channels, at high temporal resolution in real time. By combining 24/7 data-on location, movement, email and text communications, and social media-with brain scans, genetics, genomics, neuropsychological batteries, and clinical interviews, researchers will have an unprecedented amount of objective, individual-level data. Analyzing these data with ever-evolving artificial intelligence could one day include bringing interventions to patients where they are in the real world in a convenient, efficient, effective, and timely way. Yet, the road to this innovative future is fraught with ethical dilemmas as well as ethical, legal, and social implications (ELSI). OBJECTIVE: The goal of the Ethics Checklist is to promote careful design and execution of research. It is not meant to mandate particular research designs; indeed, at this early stage and without consensus guidance, there are a range of reasonable choices researchers may make. However, the checklist is meant to make those ethical choices explicit, and to require researchers to give reasons for their decisions related to ELSI issues. The Ethics Checklist is primarily focused on procedural safeguards, such as consulting with experts outside the research group and documenting standard operating procedures for clearly actionable data (eg, expressed suicidality) within written research protocols. METHODS: We explored the ELSI of digital health research in psychiatry, with a particular focus on what we label "deep phenotyping" psychiatric research, which combines the potential for virtually boundless data collection and increasingly sophisticated techniques to analyze those data. We convened an interdisciplinary expert stakeholder workshop in May 2020, and this checklist emerges out of that dialogue. RESULTS: Consistent with recent ELSI analyses, we find that existing ethical guidance and legal regulations are not sufficient for deep phenotyping research in psychiatry. At present, there are regulatory gaps, inconsistencies across research teams in ethics protocols, and a lack of consensus among institutional review boards on when and how deep phenotyping research should proceed. We thus developed a new instrument, an Ethics Checklist for Digital Health Research in Psychiatry ("the Ethics Checklist"). The Ethics Checklist is composed of 20 key questions, subdivided into 6 interrelated domains: (1) informed consent; (2) equity, diversity, and access; (3) privacy and partnerships; (4) regulation and law; (5) return of results; and (6) duty to warn and duty to report. CONCLUSIONS: Deep phenotyping research offers a vision for vastly more effective care for people with, or at risk for, psychiatric disease. The potential perils en route to realizing this vision are significant; however, and researchers must be willing to address the questions in the Ethics Checklist before embarking on each leg of the journey.

Correction: Individual-specific functional connectivity markers track dimensional and categorical features of psychotic illness.

The original version of this article omitted the author "Roscoe O. Brady Jr." from the "Psychotic Disorders Division, McLean Hospital, Harvard Medical School, Belmont, MA, USA" and the "Department of Psychiatry, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA". This has been corrected in both the PDF and HTML versions of the article.

Individual-specific functional connectivity markers track dimensional and categorical features of psychotic illness.

Neuroimaging studies of psychotic disorders have demonstrated abnormalities in structural and functional connectivity involving widespread brain networks. However, these group-level observations have failed to yield any biomarkers that can provide confirmatory evidence of a patient's current symptoms, predict future symptoms, or predict a treatment response. Lack of precision in both neuroanatomical and clinical boundaries have likely contributed to the inability of even well-powered studies to resolve these key relationships. Here, we employed a novel approach to defining individual-specific functional connectivity in 158 patients diagnosed with schizophrenia (n = 49), schizoaffective disorder (n = 37), or bipolar disorder with psychosis (n = 72), and identified neuroimaging features that track psychotic symptoms in a dimension- or disorder-specific fashion. Using individually specified functional connectivity, we were able to estimate positive, negative, and manic symptoms that showed correlations ranging from r = 0.35 to r = 0.51 with the observed symptom scores. Comparing optimized estimation models among schizophrenia spectrum patients, positive and negative symptoms were associated with largely non-overlapping sets of cortical connections. Comparing between schizophrenia spectrum and bipolar disorder patients, the models for positive symptoms were largely non-overlapping between the two disorder classes. Finally, models derived using conventional region definition strategies performed at chance levels for most symptom domains. Individual-specific functional connectivity analyses revealed important new distinctions among cortical circuits responsible for the positive and negative symptoms, as well as key new information about how circuits underlying symptom expressions may vary depending on the underlying etiology and illness syndrome from which they manifest.

Reproducibility of Cognitive Profiles in Psychosis Using Cluster Analysis.

OBJECTIVES: Cognitive dysfunction is a core symptom dimension that cuts across the psychoses. Recent findings support classification of patients along the cognitive dimension using cluster analysis; however, data-derived groupings may be highly determined by sampling characteristics and the measures used to derive the clusters, and so their interpretability must be established. We examined cognitive clusters in a cross-diagnostic sample of patients with psychosis and associations with clinical and functional outcomes. We then compared our findings to a previous report of cognitive clusters in a separate sample using a different cognitive battery. METHODS: Participants with affective or non-affective psychosis (n=120) and healthy controls (n=31) were administered the MATRICS Consensus Cognitive Battery, and clinical and community functioning assessments. Cluster analyses were performed on cognitive variables, and clusters were compared on demographic, cognitive, and clinical measures. Results were compared to findings from our previous report. RESULTS: A four-cluster solution provided a good fit to the data; profiles included a neuropsychologically normal cluster, a globally impaired cluster, and two clusters of mixed profiles. Cognitive burden was associated with symptom severity and poorer community functioning. The patterns of cognitive performance by cluster were highly consistent with our previous findings. CONCLUSIONS: We found evidence of four cognitive subgroups of patients with psychosis, with cognitive profiles that map closely to those produced in our previous work. Clusters were associated with clinical and community variables and a measure of premorbid functioning, suggesting that they reflect meaningful groupings: replicable, and related to clinical presentation and functional outcomes. (JINS, 2018, 24, 382-390).

Nicotine Increases Activation to Anticipatory Valence Cues in Anterior Insula and Striatum.

Introduction: Smoking is associated with significant morbidity and mortality. Understanding the neurobiology of the rewarding effects of nicotine promises to aid treatment development for nicotine dependence. Through its actions on mesolimbic dopaminergic systems, nicotine engenders enhanced responses to drug-related cues signaling rewards, a mechanism hypothesized to underlie the development and maintenance of nicotine addiction. Methods: We evaluated the effects of acute nicotine on neural responses to anticipatory cues signaling (nondrug) monetary reward or loss among 11 nonsmokers who had no prior history of tobacco smoking. In a double-blind, crossover design, participants completed study procedures while wearing nicotine or placebo patches at least 1 week apart. In each drug condition, participants underwent functional magnetic resonance imaging while performing the monetary incentive delay task and performed a probabilistic monetary reward task, probing reward responsiveness as measured by response bias toward a more frequently rewarded stimulus. Results: Nicotine administration was associated with enhanced activation, compared with placebo, of right fronto-anterior insular cortex and striatal regions in response to cues predicting possible rewards or losses and to dorsal anterior cingulate for rewards. Response bias toward rewarded stimuli correlated positively with insular activation to anticipatory cues. Conclusion: Nicotinic enhancement of monetary reward-related brain activation in the insula and striatum in nonsmokers dissociated acute effects of nicotine from effects on reward processing due to chronic smoking. Reward responsiveness predicted a greater nicotinic effect on insular activation to salient stimuli. Implications: Previous research demonstrates that nicotine enhances anticipatory responses to rewards in regions targeted by midbrain dopaminergic systems. The current study provides evidence that nicotine also enhances responses to rewards and losses in the anterior insula. A previous study found enhanced insular activation to rewards and losses in smokers and ex-smokers, a finding that could be due to nicotine sensitization or factors related to current or past smoking. Our finding of enhanced anterior insula response after acute administration of nicotine in nonsmokers provides support for nicotine-induced sensitization of insular response to rewards and losses.

Testing unit root non-stationarity in the presence of missing data in univariate time series of mobile health studies.

The use of digital devices to collect data in mobile health studies introduces a novel application of time series methods, with the constraint of potential data missing at random or missing not at random (MNAR). In time-series analysis, testing for stationarity is an important preliminary step to inform appropriate subsequent analyses. The Dickey-Fuller test evaluates the null hypothesis of unit root non-stationarity, under no missing data. Beyond recommendations under data missing completely at random for complete case analysis or last observation carry forward imputation, researchers have not extended unit root non-stationarity testing to more complex missing data mechanisms. Multiple imputation with chained equations, Kalman smoothing imputation, and linear interpolation have also been used for time-series data, however such methods impose constraints on the autocorrelation structure and impact unit root testing. We propose maximum likelihood estimation and multiple imputation using state space model approaches to adapt the augmented Dickey-Fuller test to a context with missing data. We further develop sensitivity analyses to examine the impact of MNAR data. We evaluate the performance of existing and proposed methods across missing mechanisms in extensive simulations and in their application to a multi-year smartphone study of bipolar patients.

Ethological computational psychiatry: Challenges and opportunities.

Studying the intricacies of individual subjects' moods and cognitive processing over extended periods of time presents a formidable challenge in medicine. While much of systems neuroscience appropriately focuses on the link between neural circuit functions and well-constrained behaviors over short timescales (e.g., trials, hours), many mental health conditions involve complex interactions of mood and cognition that are non-stationary across behavioral contexts and evolve over extended timescales. Here, we discuss opportunities, challenges, and possible future directions in computational psychiatry to quantify non-stationary continuously monitored behaviors. We suggest that this exploratory effort may contribute to a more precision-based approach to treating mental disorders and facilitate a more robust reverse translation across animal species. We conclude with ethical considerations for any field that aims to bridge artificial intelligence and patient monitoring.

Evaluating the Acceptability and Feasibility of Collecting Passive Smartphone Data to Estimate Psychological Functioning in U.S. Service Members and Veterans: A Pilot Study.

INTRODUCTION: This study investigated the acceptability and feasibility of digital phenotyping in a military sample with a history of traumatic brain injury and co-occurring psychological and cognitive symptoms. The first aim was to evaluate the acceptability of digital phenotyping by (1a) quantifying the proportion of participants willing to download the app and rates of dropout and app discontinuation and (1b) reviewing the stated reasons for both refusing and discontinuing use of the app. The second aim was to investigate technical feasibility by (2a) characterizing the amount and frequency of transferred data and (2b) documenting technical challenges. Exploratory aim 3 sought to leverage data on phone and keyboard interactions to predict if a participant (a) is depressed and (b) has depression that improves over the course of the study. MATERIALS AND METHODS: A passive digital phenotyping app (Mindstrong Discovery) functioned in the background of the participants' smartphones and passively collected phone usage and typing kinematics data. RESULTS: Fifteen out of 16 participants (93.8%) consented to install the app on their personal smartphone devices. Four participants (26.7%) discontinued the use of the app partway through the study, primarily because of keyboard usability and technical issues. Fourteen out of 15 participants (93.3%) had at least one data transfer, and the median number of days with data was 40 out of a possible 57 days. The exploratory machine learning models predicting depression status and improvement in depression performed better than chance. CONCLUSIONS: The findings of this pilot study suggest that digital phenotyping is acceptable and feasible in a military sample and provides support for future larger investigations of this technology.

Precision Assessment of Real-World Associations Between Stress and Sleep Duration Using Actigraphy Data Collected Continuously for an Academic Year: Individual-Level Modeling Study.

BACKGROUND: Heightened stress and insufficient sleep are common in the transition to college, often co-occur, and have both been linked to negative health outcomes. A challenge concerns disentangling whether perceived stress precedes or succeeds changes in sleep. These day-to-day associations may vary across individuals, but short study periods and group-level analyses in prior research may have obscured person-specific phenotypes. OBJECTIVE: This study aims to obtain stable estimates of lead-lag associations between perceived stress and objective sleep duration in the individual, unbiased by the group, by developing an individual-level linear model that can leverage intensive longitudinal data while remaining parsimonious. METHODS: In total, 55 college students (n=6, 11% second-year students and n=49, 89% first-year students) volunteered to provide daily self-reports of perceived stress via a smartphone app and wore an actigraphy wristband for the estimation of daily sleep duration continuously throughout the academic year (median usable daily observations per participant: 178, IQR 65.5). The individual-level linear model, developed in a Bayesian framework, included the predictor and outcome of interest and a covariate for the day of the week to account for weekly patterns. We validated the model on the cohort of second-year students (n=6, used as a pilot sample) by applying it to variables expected to correlate positively within individuals: objective sleep duration and self-reported sleep quality. The model was then applied to the fully independent target sample of first-year students (n=49) for the examination of bidirectional associations between daily stress levels and sleep duration. RESULTS: Proof-of-concept analyses captured expected associations between objective sleep duration and subjective sleep quality in every pilot participant. Target analyses revealed negative associations between sleep duration and perceived stress in most of the participants (45/49, 92%), but their temporal association varied. Of the 49 participants, 19 (39%) showed a significant association (probability of direction>0.975): 8 (16%) showed elevated stress in the day associated with shorter sleep later that night, 5 (10%) showed shorter sleep associated with elevated stress the next day, and 6 (12%) showed both directions of association. Of note, when analyzed using a group-based multilevel model, individual estimates were systematically attenuated, and some even reversed sign. CONCLUSIONS: The dynamic interplay of stress and sleep in daily life is likely person specific. Paired with intensive longitudinal data, our individual-level linear model provides a precision framework for the estimation of stable real-world behavioral and psychological dynamics and may support the personalized prioritization of intervention targets for health and well-being.

The Transdiagnostic Connectome Project: a richly phenotyped open dataset for advancing the study of brain-behavior relationships in psychiatry.

An important aim in psychiatry is the establishment of valid and reliable associations linking profiles of brain functioning to clinically relevant symptoms and behaviors across patient populations. To advance progress in this area, we introduce an open dataset containing behavioral and neuroimaging data from 241 individuals aged 18 to 70, comprising 148 individuals meeting diagnostic criteria for a broad range of psychiatric illnesses and a healthy comparison group of 93 individuals. These data include high-resolution anatomical scans, multiple resting-state, and task-based functional MRI runs. Additionally, participants completed over 50 psychological and cognitive assessments. Here, we detail available behavioral data as well as raw and processed MRI derivatives. Associations between data processing and quality metrics, such as head motion, are reported. Processed data exhibit classic task activation effects and canonical functional network organization. Overall, we provide a comprehensive and analysis-ready transdiagnostic dataset, which we hope will accelerate the identification of illness-relevant features of brain functioning, enabling future discoveries in basic and clinical neuroscience.

Examining the association between posttraumatic stress disorder and disruptions in cortical networks identified using data-driven methods.

Posttraumatic stress disorder (PTSD) is associated with lower cortical thickness (CT) in prefrontal, cingulate, and insular cortices in diverse trauma-affected samples. However, some studies have failed to detect differences between PTSD patients and healthy controls or reported that PTSD is associated with greater CT. Using data-driven dimensionality reduction, we sought to conduct a well-powered study to identify vulnerable networks without regard to neuroanatomic boundaries. Moreover, this approach enabled us to avoid the excessive burden of multiple comparison correction that plagues vertex-wise methods. We derived structural covariance networks (SCNs) by applying non-negative matrix factorization (NMF) to CT data from 961 PTSD patients and 1124 trauma-exposed controls without PTSD. We used regression analyses to investigate associations between CT within SCNs and PTSD diagnosis (with and without accounting for the potential confounding effect of trauma type) and symptom severity in the full sample. We performed additional regression analyses in subsets of the data to examine associations between SCNs and comorbid depression, childhood trauma severity, and alcohol abuse. NMF identified 20 unbiased SCNs, which aligned closely with functionally defined brain networks. PTSD diagnosis was most strongly associated with diminished CT in SCNs that encompassed the bilateral superior frontal cortex, motor cortex, insular cortex, orbitofrontal cortex, medial occipital cortex, anterior cingulate cortex, and posterior cingulate cortex. CT in these networks was significantly negatively correlated with PTSD symptom severity. Collectively, these findings suggest that PTSD diagnosis is associated with widespread reductions in CT, particularly within prefrontal regulatory regions and broader emotion and sensory processing cortical regions.

Accelerating Medicines Partnership® Schizophrenia (AMP® SCZ): Rationale and Study Design of the Largest Global Prospective Cohort Study of Clinical High Risk for Psychosis.

This article describes the rationale, aims, and methodology of the Accelerating Medicines Partnership® Schizophrenia (AMP® SCZ). This is the largest international collaboration to date that will develop algorithms to predict trajectories and outcomes of individuals at clinical high risk (CHR) for psychosis and to advance the development and use of novel pharmacological interventions for CHR individuals. We present a description of the participating research networks and the data processing analysis and coordination center, their processes for data harmonization across 43 sites from 13 participating countries (recruitment across North America, Australia, Europe, Asia, and South America), data flow and quality assessment processes, data analyses, and the transfer of data to the National Institute of Mental Health (NIMH) Data Archive (NDA) for use by the research community. In an expected sample of approximately 2000 CHR individuals and 640 matched healthy controls, AMP SCZ will collect clinical, environmental, and cognitive data along with multimodal biomarkers, including neuroimaging, electrophysiology, fluid biospecimens, speech and facial expression samples, novel measures derived from digital health technologies including smartphone-based daily surveys, and passive sensing as well as actigraphy. The study will investigate a range of clinical outcomes over a 2-year period, including transition to psychosis, remission or persistence of CHR status, attenuated positive symptoms, persistent negative symptoms, mood and anxiety symptoms, and psychosocial functioning. The global reach of AMP SCZ and its harmonized innovative methods promise to catalyze the development of new treatments to address critical unmet clinical and public health needs in CHR individuals.

Topographic organization of macaque area LIP.

Despite several attempts to define retinotopic maps in the macaque lateral intraparietal area (LIP) using histological, electrophysiological, and neuroimaging methods, the degree to which this area is topographically organized remains controversial. We recorded blood oxygenation level-dependent signals with functional MRI from two macaques performing a difficult visual search task on stimuli presented at the fovea or in the periphery of the visual field. The results revealed the presence of a single topographic representation of the contralateral hemifield in the ventral subdivision of the LIP (LIPv) in both hemispheres of both monkeys. Also, a foveal representation was localized in rostral LIPv rather than in dorsal LIP (LIPd) as previous experiments had suggested. Finally, both LIPd and LIPv responded only to contralateral stimuli. In contrast, human studies have reported multiple topographic maps in intraparietal cortex and robust responses to ipsilateral stimuli. These blood oxygenation level-dependent functional MRI results provide clear evidence for the topographic organization of macaque LIP that complements the results of previous electrophysiology studies, and also reveal some unexpected characteristics of this organization that have eluded these previous studies. The results also delineate organizational differences between LIPv and LIPd, providing support for these two histologically defined areas may subserve different visuospatial functions. Finally, these findings point to potential evolutionary differences in functional organization with human posterior parietal cortex.