Document text
Principal Investigator: Christopher James Kennedy
Organization: MASSACHUSETTS GENERAL HOSPITAL
Fiscal Year: 2024
Award: $181,346
Funding agency: National Institute of Mental Health
Background: The United States is experiencing a crisis in mental health care, in which long-term shifts
towards the emergency department (ED) for psychiatric treatment have been exacerbated by the COVID-19
pandemic – reduced inpatient hospitalization capacity, delayed outpatient appointments, extended ED
boarding, and staffing shortages, combined with substantial growth in demand for mental health services (due
in part to pandemic-related increases in stress, depression, social isolation, and anxiety, and with differential
impacts on vulnerable populations). Data-driven approaches have not been widely applied to understand
mechanisms and predictors of ED utilization for psychiatric patients, or to optimize clinical decision-making to
improve patient outcomes. Research: In this study, I will conduct secondary data analysis of 141,431 ED visits
for psychiatric care in a large regional health system from 2017-2023, extracting and preparing over 5,000
patient characteristics from the electronic health record (diagnoses, labs, medications, procedures, visits, and
notes). Aim 1: I will create a transdiagnostic subtyping system, using latent class analysis, hierarchical density-
based clustering, and neural network autoencoders, to reliably categorize and track major trends in patient
populations receiving emergency psychiatric care. Aim 2: I will develop a suite of ensemble machine learning
models to predict ED length of stay, ED boarding duration, short-term ED re-admission, transfer to inpatient
hospitalization, and post-discharge adverse events (suicide attempt, overdose, or psychotic episode). Aim 3: I
will estimate a precision treatment rule, using targeted causal inference and ensemble machine learning, to
capture heterogeneous treatment effects for primary ED disposition decisions (inpatient hospitalization, partial
hospitalization / intensive outpatient program, or outpatient monitoring) on adverse psychiatric events.
Candidate's Career Development, Goals, and Environment: This proposal's research aims and the
candidate's career development will be supported by the extensive resources available at Massachusetts
General Hospital (MGH) and Harvard Medical School, as well as formal training, coursework, and mentorship
in (T1) mental disorders and psychopathology, (T2) precision treatment optimization, (T3) hybrid pragmatic
effectiveness trials, and (T4) professional development in preparation for a future R01 submission. The
mentorship team includes Primary Mentor Dr. Jordan Smoller, a leading expert in precision psychiatry; Co-
Mentors Dr. Matthew Nock, a leader in mental disorders and suicide prevention; Dr. Susan Murphy, a leader in
precision treatment optimization; and Dr. Stephen Bartels, a leader in implementation science and hybrid
pragmatic trials; and Consultants Dr. Suzanne Bird, expert in emergency psychiatry and director of MGH Acute
Psychiatry Services; Dr. Edwin Boudreaux, expert in hybrid pragmatic trials in emergency psychiatry; and Dr.
Soroush Saghafian, expert in reinforcement learning to optimize hospital decision-making. This award will
enable the candidate to develop an independent, rigorous research program in computational mental health.
Terms: <Accident and Emergency department><Acute><Address><Admission><Admission activity><Adverse Experience><Adverse event><Agitation><Ambulatory Care><Ambulatory Monitoring><Anxiety><Appointment><Area><Automated Clinical Decision Support><Aves><Avian><Award><Beds><Big Data><BigData><Birds><COVID crisis><COVID epidemic><COVID pandemic><COVID-19 crisis><COVID-19 epidemic><COVID-19 era><COVID-19 global health crisis><COVID-19 global pandemic><COVID-19 health crisis><COVID-19 pandemic><COVID-19 pandemic affected><COVID-19 pandemic consequence><COVID-19 pandemic effects><COVID-19 pandemic impact><COVID-19 pandemic impacted><COVID-19 period><COVID-19 public health crisis><COVID-19 years><Calibration><Caring><Categories><Characteristics><Chest Pain><Clinical><Cognitive Discrimination><Complex><Computing Methodologies><Continuity of Care><Continuity of Patient Care><Continuum of Care><Coupled><Crowding><Data><Data Analyses><Data Analysis><Data Bases><Data Science><Databases><Day Care><Decision Making><Development><Diagnosis><Diagnostic><Discrimination><Drugs><ED care><ED visit><ER care><ER visit><Electronic Health Record><Emergencies><Emergency Care><Emergency Department><Emergency Department care><Emergency Room care><Emergency Situation><Emergency care visit><Emergency department visit><Emergency health care><Emergency healthcare><Emergency hospital visit><Emergency medical care><Emergency room><Emergency room visit><Environment><Ethnic Origin><Ethnicity><Event><Fatigue><Future><General Hospitals><General Taxonomy><Generalized Growth><Goals><Grant><Grouping><Growth><Health><Health Care Systems><Health Status><Health system><Healthcare Systems><Hospital Admission><Hospitalization><Hospitals><Hybrids><Iatrogenesis><Inpatients><Intervention><Intervention Strategies><Jordan><Label><Lack of Energy><Leadership><Learning><Length of Stay><Level of Health><Machine Learning><Massachusetts><Medication><Mental Depression><Mental Health><Mental Health Services><Mental Hygiene><Mental Hygiene Services><Mental disorders><Mental health disorders><Mentors><Mentorship><Methods><Modeling><NIMH><National Institute of Mental Health><Nature><New England><Northeastern United States><Number of Days in Hospital><Out-patients><Outcome><Outpatient Care><Outpatient Monitoring><Outpatients><Overdose><Partial Hospitalization><Patient Care><Patient Care Delivery><Patient outcome><Patient-Centered Outcomes><Patient-Focused Outcomes><Patients><Pattern><Pharmaceutical Preparations><Precision therapeutics><Preparation><Procedures><Provider><Psychiatric Disease><Psychiatric Disorder><Psychiatric therapeutic procedure><Psychiatry><Psychological Health><Psychological reinforcement><Psychometrics><Psychomotor Agitation><Psychomotor Excitement><Psychomotor Hyperactivity><Psychomotor Restlessness><Psychopathology><Race><Races><Reinforcement><Research><Research Proposals><Research Resources><Resources><Restlessness><Risk><Risk Factors><SAMHSA><SARS-CoV-2 epidemic><SARS-CoV-2 global health crisis><SARS-CoV-2 global pandemic><SARS-CoV-2 pandemic><SARS-coronavirus-2 epidemic><SARS-coronavirus-2 pandemic><Series><Services><Severe Acute Respiratory Syndrome CoV 2 epidemic><Severe Acute Respiratory Syndrome CoV 2 pandemic><Severe acute respiratory syndrome coronavirus 2 epidemic><Severe acute respiratory syndrome coronavirus 2 pandemic><Social isolation><Statistical Methods><Stress><Structure><Subgroup><Substance Abuse and Mental Health Services Administration><Suicide attempt><Suicide precaution><Suicide prevention><System><Taxonomy><Testing><Therapeutic><Time><Tissue Growth><Training><Treatment Day Care><Uninsured><United States><United States Substance Abuse and Mental Health Services Administration><Visit><Vulnerable Populations><Work><Writing><Youth><Youth 10-21><abnormal psychology><adverse event risk><autoencoder><autoencoding neural network><automated decision support><autonomous decision support><care for patients><care of patients><career development><caring for patients><clinical care><clinical decision-making><computational methodology><computational methods><computer based method><computer based prediction><computer methods><computing method><coronavirus disease 2019 crisis><coronavirus disease 2019 epidemic><coronavirus disease 2019 global health crisis><coronavirus disease 2019 global pandemic><coronavirus disease 2019 health crisis><coronavirus disease 2019 pandemic><coronavirus disease 2019 pandemic consequence><coronavirus disease 2019 pandemic impact><coronavirus disease 2019 public health crisis><coronavirus disease crisis><coronavirus disease epidemic><coronavirus disease pandemic><coronavirus disease-19 global pandemic><coronavirus disease-19 pandemic><data base><data interpretation><deep learning><deep learning method><deep learning strategy><density><depression><developmental><drug/agent><effects following the COVID-19 pandemic><electronic health care record><electronic health medical record><electronic health plan record><electronic health registry><electronic medical health record><emergency settings><experience><group of color><groupings><health care quality><health equity><health level><healthcare quality><hospital days><hospital length of stay><hospital re-admission><hospital readmission><hospital stay><iatrogenic><iatrogenically><iatrogenicity><impact of the SARS-CoV-2 pandemic><implementation science><improved><indexing><individual of color><insight><interventional strategy><machine based learning><machine learning based model><machine learning model><malleable risk><medical college><medical schools><mental disorder prevention><mental health care><mental healthcare><mental illness><modifiable risk><neural network><non fatal attempt><nonfatal attempt><novel><ontogeny><outpatient programs><outpatient services><outpatient treatment><pandemic><pandemic disease><pandemic effect><pandemic impact><pandemic outcome><pandemic repercussions><patient oriented outcomes><patient population><patient stratification><people of color><person of color><personalization of treatment><personalized medicine><personalized therapy><personalized treatment><population of color><pragmatic effectiveness trial><pragmatic trial><precision therapies><precision treatment><predictive modeling><preparations><prevent suicidality><prevent suicide><programs><psychiatric care><psychiatric emergency><psychiatric illness><psychiatric therapy><psychiatric treatment><psychological disorder><psychotic><racial><racial background><racial origin><re-admission><re-hospitalization><readmission><rehospitalization><resilience factor><resiliency factor><response to therapy><response to treatment><school of medicine><severe acute respiratory syndrome coronavirus 2 global health crisis><severe acute respiratory syndrome coronavirus 2 global pandemic><statistic methods><statistics><stratified patient><suicidal attempt><suicidality prevention><suicide intervention><support tools><therapeutic response><therapy optimization><therapy response><tool><treatment effect><treatment optimization><treatment planning><treatment response><treatment responsiveness><treatment strategy><trend><unsupervised clustering><vulnerable group><vulnerable individual><vulnerable people>