First-episode substance-induced psychosis (SIP) presents a clinical challenge in which treatment decisions cannot rely on evidence-based guidelines and long-term outcomes are not well understood. Preliminary findings from our retrospective study of psychiatry inpatients at Harborview Medical Center (HMC) suggest that patients with first-episode SIP had similar rehospitalization rates to those with first-episode psychosis (FEP) but were less likely to receive long-acting injectable antipsychotics (LAIs) even though LAIs may reduce the risk of 30- and 180-day rehospitalization. Our study has also revealed inconsistent assessment of key risk factors for rehospitalization—such as family history of psychosis and patterns of cannabis use—that may be limiting informed decision-making, including appropriate LAI use. This project seeks to improve the risk stratification of first-episode SIP by addressing these gaps. Aim 1 will use qualitative interviews with inpatient attending psychiatrists, psychiatry residents, and patients to explore factors influencing the management of first-episode SIP. Aim 2 will evaluate the acceptability and feasibility of implementing a standardized assessment of cannabis use and family history of psychosis on HMC inpatient psychiatry units. This project will lay the groundwork for future clinical interventions that optimize treatment decisions and improve patient outcomes in psychiatric inpatient settings.
Targeted Condition: Serious Mental Illness
Unraveling the genetics of schizophrenia and bipolar disorder with large-language models
Schizophrenia (SCZ) and bipolar disorder (BPD) are among the most heritable psychiatric conditions, yet their genetic foundations remain poorly understood. Historically, unraveling these disorders’ genetic architectures was limited by inadequate technology. However, breakthroughs in next-generation sequencing have recently produced expansive genomic datasets—including those from the Psychiatric Genomics Consortium (PGC)—for SCZ and BD. Simultaneously, genomic foundation models—advanced large-language models (LLMs) trained on biological data rather than corpuses of text—have emerged as next-generation artificial intelligence platforms that offer unparalleled abilities to predict the functional effects of genetic variants, many of which were previously unclassified. This proposal harnesses these models to analyze publicly available PGC genomic datasets, aiming to annotate both common and rare genetic variants associated with SCZ and BD, pinpoint disease-associated genes, and map the biological pathways they influence. By bridging these recent advances in artificial intelligence with robust genomic data, this proposal seeks to illuminate the genetic underpinnings of SCZ and BPD. The anticipated insights promise to deepen our understanding of heritable psychiatric conditions, laying the groundwork for enhanced diagnostics and novel therapies aimed at biology, rather than nosology
Strengthening financial literacy for people living with serious mental illness
Improved financial literacy among people living with serious mental illness (SMI) is associated with a higher quality of life, fewer hospitalizations, and better treatment adherence. Yet people living with SMI frequently express how their lack of financial knowledge has negative personal consequences and that they don’t know where to turn for assistance. This project will gather qualitative and quantitative data from people admitted to the Center for Behavioral Health and Learning, a psychiatric hospital, to understand the need and desire for a financial skills intervention and its role in discharge planning. The assessment will also seek input from family members/caregivers, representative payees/fiduciaries and experts in the community. Ultimately, we hope to create a replicable, standardized intervention that can be evaluated and implemented in inpatient settings and modified as necessary for outpatient settings.
Using Natural Language Processing to risk-stratify inpatient psychiatry conflict and violence
This QI project aims to expand from general medical wards to inpatient psychiatry the use of predictive risk-modeling for violence or restraint, using Natural Language Processing of clinical notes. We will also assess whether NLP paired with generative AI can accurately summarize a wider range of clinical notes relevant to behavioral emergencies
Optimizing telemental health with live artificial intelligence clinical scaffolding and feedback
This project aims to develop a clinical scaffolding system to enhance telemental health care by providing real-time coaching and actionable suggestions during video-based sessions. Modeled after live supervision methodologies, it supports clinicians by identifying intervention targets and offering text-based coaching prompts to guide care. Unlike automated chatbots, this approach enables clinicians to adapt suggestions to patient needs, balancing automation with oversight for safer AI-supported mental healthcare. The proposed in-session support will facilitate efficient implementation of strategies and clinician skill development. This project seeks to enhance data privacy by processing all data on-device and avoiding external data transfers.
Partnering with patients to re-envision psychiatric hospitalization and discharge
We will analyze people’s stories about psychiatric hospitalization, interview people with experiences surrounding psychiatric hospitalization, and co-design with them to identify alternative approaches that would help people care for themselves as they transition out of the hospital. We will build upon our prior work on understanding patients’ challenges and co-designing new systems that help patients transition from psychiatric hospitalizationto self-management. In particular, we will focus on how we could redesign psychiatric hospital systems with the people who have experienced them, identifying patient insights on the knowledge, resources, and self-efficacy they need to help them return to the community.
Partnering with community pharmacies to enhance access to long-acting injectable antipsychotics in Washington State
Medication nonadherence is common among patients with serious mental illness, including schizophrenia. The use of long-acting injectable antipsychotics (LAIAs) for schizophrenia is an evidence-based practice that improves medication adherence, decreases symptomatic recurrence and reduces hospitalizations. However, patients and clinicians often face several challenges in access and coordination resulting in the underutilization of LAIAs in care.
Administering LAIAs at community pharmacies could potentially increase accessibility, reduce barriers for treatment and improve patient outcomes. This project aims to assess the fit or compatibility of LAIA administration in community pharmacies. We will survey community pharmacy staff and behavioral healthcare providers in Washington State to assess the acceptability, appropriateness and feasibility of LAIA administration in community pharmacies. If LAIA administration at community pharmacies is found to be a good fit, the next steps will be to develop strategies to support implementation. A scalable and adoptable model for administering LAIAs at community pharmacies could have substantial impacts on public health through increasing access to treatment and expanding behavioral health services at the community level and in rural areas.
Training psychiatry residents in complex communication skills for working with clients and their supports
Individuals with serious persistent mental illness (SPMI) and their families and communities face significant challenges during psychiatric hospitalization. Persons with SPMI and their supporters express a need for enhanced communication from their behavioral healthcare teams during these pivotal periods of time where symptoms are new or intense. Yet, a substantial number of mental health providers have limited training in communicating complex topics such as diagnosis and prognosis. This can lead to providers avoiding essential conversations; individuals with SPMI can be unheard or excluded from participating in treatment planning.
This educational initiative seeks to craft an innovative curriculum for psychiatry residents focused on person-centered communication skills. Drawing from proven communication training frameworks within palliative care, the training will equip residents with strategies such as: utilizing person-centered language; conducting family meetings; delivering diagnostic and prognostic information. The curriculum will be developed with guidance from individuals with lived experienced of SPMI and their supporters. Moreover, the project will deliver a dedicated online portal featuring educational materials, recorded presentations, role-play scripts, and communication guides. Tools, such as self-assessment and evaluation rubrics, will be created to evaluate efficacy.
Cognitive-Behavioral Therapy for psychosis workforce development
Cognitive Behavioral Therapy for psychosis (CBTp) is a time-limited, structured form of talk therapy that is indicated for individuals who experience distress related to psychotic symptoms. Although evidence demonstrates effectiveness in enhancing care and outcomes for clients with psychosis, CBTp is not widely available in the United States. The UW SPIRIT Lab in the Department of Psychiatry & Behavioral Sciences (PI: Sarah Kopelovich, PhD) applies evidence-based implementation and dissemination strategies such as blended learning, train and trainer, Project ECHO, longitudinal consultation to agencies, supervisors, and practitioners, fidelity assessment and monitoring, and sustaining the first CBTp Provider Network in the United States. The CBTp workforce development project aims to sustain and expand access to CBTp across publicly-funded behavioral health settings in Washington State.
Developing a digital training resource for clinicians learning CBT for psychosis (CBTpro)
The Cognitive Behavioral Therapy Training Study will rigorously test CBTpro — a novel tool that uses spoken language technologies and conversational Artificial Intelligence to train behavioral health practitioners in Cognitive Behavioral Therapy. We conducted a 2-week field trial, followed by a Randomized Clinical Trail in community mental health agencies to evaluate both learner and client outcomes. The study aims to expand global access to CBT training to students and practitioners, support quality psychological treatments for clients with a range of behavioral health disorders (including Serious Mental Illness), and support ongoing clinical quality assurance in routine care settings.
