Vincent X. Liu, MD, MS

People

Kaiser Permanente Division of Research
2000 Broadway
Oakland, CA 94612

(510) 891-5933 (phone)
vincent.x.liu@kp.org
Kaiser Permanente Researcher Profiles

Vincent Liu, MD, MS, is a senior research scientist at the Kaiser Permanente Northern California Division of Research. He also directs the Systems Research Initiative, a multidisciplinary team at the Division of Research supported by The Permanente Medical Group (TPMG). His work focuses on the intersection of sepsis, acute severe illness, informatics, and health care delivery with a goal of building towards a learning hospital system. He is a widely recognized expert in applied informatics.

Dr. Liu is board certified in Internal Medicine, Pulmonary Disease, Critical Care Medicine, and Clinical Informatics. He practices pulmonary and critical care medicine with TPMG.

Dr. Liu is the Regional Director of Hospital Advanced Analytics in Kaiser Permanente Northern California. He has led and served on committees for the American Thoracic Society, the Society of Critical Care Medicine, the National Quality Forum, and the National Institute for General Medical Services. He has served on multiple NIH grant review committees and the National Commission on Quality Assurance's Utilization Measure Advisory Panel. He is an editorial board member for Critical Care Medicine.

He is also an active mentor and leads the TPMG Informatics Fellowship Program.

Dr. Liu received his undergraduate degree from the University of Pennsylvania and his medical degree from New York University. He completed his residency training at New York University Hospitals (Bellevue) and a chief residency at Memorial Sloan-Kettering Cancer Center. He also completed a pulmonary and critical care medicine fellowship and has master's degrees in Health Services Research and Biomedical Informatics, both from Stanford University.

Current Position(s):

Research Scientist III, Division of Research, Kaiser Permanente Northern California
Intensivist, Critical Care, Kaiser Permanente Medical Center, Santa Clara, California
Clinical Assistant Professor of Medicine (Affiliated), Stanford University, Stanford, California
Regional Director, Hospital Advanced Analytics, Kaiser Permanente Northern California

Section Affiliations:

Health Care Delivery and Policy

Primary Research Interests:

Critical Care and Inpatient Health Services Research
Sepsis
Data Science
Delivery and Implementation Science
Electronic Decision Support
Health Informatics

Studies

Machine Learning Applied to EHR Data as a Means of Improving Biosurveillance and Diagnostic Excellence: A Pilot Project

We aim to develop and extend methods of probabilistic modeling of temporal electronic health record (EHR) data, using non-stationary Gaussian processes, hidden Markov models, and composite mixture models, to identify opportunities to conduct clinical biosurveillance and to achieve diagnostic excellence. Building off collaboration between KP DOR and the Lawrence Livermore National Laboratory, this proposal will use population-level EHR data, AI/ML methods and advanced computing to evaluate associations between health system diagnostic testing/treatment and prevalent/incident health conditions and to identify areas of potential diagnostic uncertainty.
Investigator: Liu, Vincent
Funder:

Turning back the clock on sepsis: Defining the epidemiology and opportunities of pre-sepsis

We intend to determine the epidemiology of pre-sepsis, including ambulatory and prehospital care settings, to identify modifiable care delays, opportunities for primary sepsis prevention, and to improve early sepsis care for vulnerable populations. The objectives include determining the epidemiology of pre-sepsis care across ambulatory and prehospital settings using linked EHR data; measuring delays in pre-sepsis care; and understanding the patient-level variation in pre-sepsis care and delays in vulnerable populations. The project is a collaboration between KPNC and University of Pittsburgh Medical Center.
Investigator: Liu, Vincent
Funder: Gordon and Betty Moore Foundation

Evaluation of Risk-Based Predictive Models to Inform the Efficiency of Perioperative Systems

We previously developed an automated electronic health record-based risk stratification tool, called the Comorbidity Assessment and Surgical Triage (CAST) score, which we implemented in KP HealthConnect to improve triage of elective surgical patients to perioperative medicine clinic visit types in real-time. This project seeks to improve the efficiency of operative workflows by evaluating the impact of CAST implementation on pre-operative workflows and surgical outcomes; characterizing patients with discordant risk assessments between CAST and the clinician; and evaluating the association between CAST and operative resources, including surgical case duration.
Investigator: Liu, Vincent
Funder: TPMG Delivery Science Projects Program

Effectiveness and safety of treatments for hospitalized patients with COVID-19

Examine the safety and effectiveness of current treatments for hospitalized patients with COVID-19 and identify patient characteristics and time periods during the disease course that are associated with more favorable treatment response.
Investigator: Liu, Vincent
Funder: Garfield Memorial Fund

Sepsis on FHIR

Overall goal is to define and refine the current epidemiology, clinical characteristics, and treatment practices in respiratory failure in late and moderately preterm infants. Specific aims are to determine that incidence and epidemiology of respiratory failure in neonates born 32-36 weeks of gestation between 2012 and 2017 in Kaiser Permanente Northern California facilities; describe disease severity and respiratory support by diagnostic category; describe use of surfactant replacement therapy; and assess the impact of surfactant replacement therapy on length of respiratory support, length of hospitalization, and respiratory outcomes in the first year of life.
Investigator: Liu, Vincent
Funder: Moore Foundation

Understanding and Informing Early Hospital Antibiotic Prescribing for Potential Infection

The goals of this study are to (1) examine temporal trends in early hospital antibiotic prescribing patterns over the past 5 years, to assess for unintended consequences associated with earlier antibiotics in severe sepsis/septic shock; (2) quantify the patient-level harm associated with receipt of unnecessary antibiotics; and (3) identify subsets of patients, discernable in real-time with high certainty, who benefit from immediate antibiotics, versus the subsets of patients for whom it is safe to delay antibiotics by 3 hours (thereby allowing time for rapid diagnostics, such as procalcitonin).
Investigator: Liu, Vincent
Funder: Agency for Healthcare Research and Quality

Targeting Cardiovascular Events to Improve Patient Outcomes After Sepsis

Sepsis is the most common illness leading to hospitalization in the United States, affecting more than 1 million Americans each year. While short-term mortality has improved over prior decades, sepsis survivors face numerous post-hospital sequelae that negatively impact long-term morbidity and mortality. Cardiovascular complications are among the most common reasons for re-hospitalization and death after sepsis, with approximately 1 in 3 severe sepsis patients experiencing a cardiovascular complication in the year after sepsis hospitalization. This project seeks to identify risk factors for post-sepsis cardiovascular risk based on granular inpatient data collected during hospitalization, and identify potential opportunities to reduce post-sepsis cardiovascular risk.
Investigator: Liu, Vincent
Funder: National Heart, Lung, and Blood Institute

Using EHR Audit Logs to Capture Diagnostic Pathways and Time to Diagnosis

Rapid expansions in the use of electronic health record (EHR) systems also contribute to large expansions in the availability of metadata. These highly granular metadata may prove particularly useful for understanding how frontline clinicians interact with clinical information when making diagnostic decisions. This project aims to characterize EHR audit log data and advance nascent understanding of the potential to leverage these to provide novel insights into the diagnostic process, with a focus on time to diagnosis, and ultimately inform the achievement of diagnostic excellence.
Investigator: Liu, Vincent
Funder: Gordon and Betty Moore Foundation

Identifying Pre-Sepsis Opportunities for Early, Targeted Intervention

Early evidence suggests there will be ongoing waves of COVID-19 outbreak that can be mitigated through effective community-level prediction models and intervention. This project will Identify pre-sepsis opportunities for early, targeted intervention to inform community-level mitigation strategies of upcoming COVID19 surges.
Investigator: Liu, Vincent
Funder: National Institute of General Medical Sciences

Benchmarking Hospital Quality - Template Matching

In this study, Kaiser Permanente Northern California researchers will serve as consultants for U.S. Department of Veterans Affairs (VA) researchers to evaluate the utility of template matching techniques, compared with standard regression techniques, for comparing hospital quality.
Investigator: Liu, Vincent
Funder: U.S. Department of Veterans Affairs

Benefits of Intensive Care Unit Admission for Patients with Acute Respiratory Failure or Sepsis

Ideal and optimally efficient use of intensive care unit (ICU) resources would create benefits across the health care system. Ideal use of ICU resources requires a needs assessment to determine which categories of patients commonly admitted to ICUs derive true net benefit on average from ICU admission versus admission to a regular hospital floor. This project aims to conduct such a needs assessment, which has never before been done.
Investigator: Liu, Vincent
Funder: National Heart, Lung, and Blood Institute

Data-Driven Precision Medicine in Sepsis

Sepsis is a common and costly disease which contributes to nearly 1 in 2 hospital deaths in our healthcare system. While aggressive early treatment with quantitative resuscitation has demonstrated benefit, many other interventions have not demonstrated efficacy likely resulting from an incomplete understanding of sepsis disease mechanisms. This study seeks to address three questions through innovative approaches to using observational data: (1) what factors can improve our characterization and phenotyping of sepsis patients; (2) what are the potential benefits of using quantitative resuscitation in less severe sepsis subgroups; and (3) how does acute organ dysfunction impact long-term mortality and utilization among sepsis survivors
Investigator: Liu, Vincent
Funder: National Institute of General Medical Sciences

Highlighted Publications

No results found matching criteria

Publications

Use of Sepsis Clinical Prediction Models to Improve Patient Care

Author(s): Mogri, Murtaza; Grant, Richard W; Liu, Vincent X

JAMA Intern Med. 2023 Apr 03.

Current Position(s):

Section Affiliations:

Primary Research Interests:

Studies

Machine Learning Applied to EHR Data as a Means of Improving Biosurveillance and Diagnostic Excellence: A Pilot Project

Turning back the clock on sepsis: Defining the epidemiology and opportunities of pre-sepsis

Evaluation of Risk-Based Predictive Models to Inform the Efficiency of Perioperative Systems

Effectiveness and safety of treatments for hospitalized patients with COVID-19

Sepsis on FHIR

Understanding and Informing Early Hospital Antibiotic Prescribing for Potential Infection

Targeting Cardiovascular Events to Improve Patient Outcomes After Sepsis

Using EHR Audit Logs to Capture Diagnostic Pathways and Time to Diagnosis

Identifying Pre-Sepsis Opportunities for Early, Targeted Intervention

Benchmarking Hospital Quality - Template Matching

Benefits of Intensive Care Unit Admission for Patients with Acute Respiratory Failure or Sepsis

Data-Driven Precision Medicine in Sepsis

Highlighted Publications

Publications

Use of Sepsis Clinical Prediction Models to Improve Patient Care

Performance of a Prediction Model of Suicide Attempts Across Race and Ethnicity

Development and Validation of an Electronic Health Record-based Score for Triage to Perioperative Medicine

Among-Hospital Variation in ICU Admission Practices and Associated Outcomes for Patients with Acute Respiratory Failure

Practice Patterns and Outcomes Associated With Anticoagulation Use Following Sepsis Hospitalizations With New-Onset Atrial Fibrillation

Risk of venous thromboembolism in non-respiratory and respiratory presentations of COVID-19 in critically ill patients

Hospital Strain and Variation in Sepsis ICU Admission Practices and Associated Outcomes

COVID-19 bacteremic co-infection is a major risk factor for mortality, ICU admission, and mechanical ventilation

Risk of posthospital venous thromboembolism in patients with COVID-19 varies by SARS-CoV-2 period and vaccination status

Trends in Outpatient Visits and Hospital and Intensive Care Unit Admissions of Adults With COVID-19 in an Integrated US Health Care System, March 2020 to January 2022

Predicting Post-Sepsis Cardiovascular Events with Death as a Competing Risk

Association between anaemia and hospital readmissions in patients undergoing major surgery requiring postoperative intensive care

The intensity of anticoagulant dosing in hospitalized patients with COVID-19: An observational, comparative effectiveness study

Anakinra or high-dose corticosteroids in COVID-19 pneumonia patients who deteriorate on low-dose dexamethasone: An observational study of comparative effectiveness

Risk of hospitalization in a sample of COVID-19 patients with and without chronic obstructive pulmonary disease

Team is brain: leveraging EHR audit log data for new insights into acute care processes

Persistent Steroid Exposure Before Coronavirus Disease 2019 Diagnosis and Risk of Hospitalization in Patients With Chronic Obstructive Pulmonary Disease

Characterizing Equity of Intensive Care Unit Admissions for Sepsis and Acute Respiratory Failure

High-dose corticosteroids in patients hospitalized for COVID-19 pneumonia: an observational study of comparative effectiveness

Association Between Preoperative Hemodialysis Timing and Postoperative Mortality in Patients With End-stage Kidney Disease

Covid-19 Vaccination and the Timing of Surgery Following Covid-19 Infection

Quantifying Frailty Requires a Conceptual Model Before a Statistical Model-Reply

Trends and Risk Factors for Venous Thromboembolism Among Hospitalized Medical Patients

Quantifying the breadth of antibiotic exposure in sepsis and suspected infection using spectrum scores

Unsupervised probabilistic models for sequential Electronic Health Records

Comparing the effectiveness of a brief intervention to reduce unhealthy alcohol use among adult primary care patients with and without depression: A machine learning approach with augmented inverse probability weighting

Risk of severe COVID-19 infection among adults with prior exposure to children

Adults hospitalized with breakthrough COVID-19 have lower mortality than matched unvaccinated adults

Temporal Trends in Antimicrobial Prescribing During Hospitalization for Potential Infection and Sepsis

The Kaiser Permanente Northern California Advance Alert Monitor Program: An Automated Early Warning System for Adults at Risk for In-Hospital Clinical Deterioration

Validation of Respiratory Rate-Oxygenation Index in Patients With COVID-19-Related Respiratory Failure

Red Blood Cell Transfusion at a Hemoglobin Threshold of Seven g/dL in Critically Ill Patients: A Regression Discontinuity Study

New complementary perspectives for inpatient physical function assessment: matched clinician-report and patient-report short form measures from the PROMIS adult physical function item bank

Exploratory analysis of novel electronic health record variables for quantification of healthcare delivery strain, prediction of mortality, and prediction of imminent discharge

Evaluation of Incident 7-Day Infection and Sepsis Hospitalizations in an Integrated Health System

Comparison of Electronic Frailty Metrics for Prediction of Adverse Outcomes of Abdominal Surgery

Prospective evaluation of social risks, physical function, and cognitive function in prediction of non-elective rehospitalization and post-discharge mortality

Developing a shared sepsis data infrastructure: a systematic review and concept map to FHIR

Evaluation of Outcomes Following Hospital-Wide Implementation of a Subcutaneous Insulin Protocol for Diabetic Ketoacidosis

Prognostic Accuracy of Presepsis and Intrasepsis Characteristics for Prediction of Cardiovascular Events After a Sepsis Hospitalization

Association of Intensive Care Unit Admission and Outcomes in Sepsis and Acute Respiratory Failure

The COVID-19 Pandemic Strikes Again and Again and Again

Connecting Research and Practice: Implementation of Suicide Prevention Strategies in Learning Health Care Systems

Outcomes After Implementation of a Benzodiazepine-Sparing Alcohol Withdrawal Order Set in an Integrated Health Care System

Performance of Predictive Models for 30-day Hospitalization and Mortality after COVID-19 Infection

Assess COVID-19 prognosis … but be aware of your instrument's accuracy!

Hospital-specific Template Matching for Benchmarking Performance in a Diverse Multihospital System

Algorithmic prognostication in critical care: a promising but unproven technology for supporting difficult decisions

Predictive Models for In-Hospital Deterioration in Ward Patients

Validation of Postpartum Hemorrhage Admission Risk Factor Stratification in a Large Obstetrics Population

Reducing Readmissions for Chronic Obstructive Pulmonary Disease in Response to the Hospital Readmissions Reduction Program

Association Between Troponin I Levels During Sepsis and Post-Sepsis Cardiovascular Complications

Temporal Trends and Hospital Variation in Time-to-Antibiotics Among Veterans Hospitalized With Sepsis

Evaluation of an intervention targeted with predictive analytics to prevent readmissions in an integrated health system: observational study

The Coming of Age of Implementation Science and Research in Critical Care Medicine

Evaluation of Vaccination Strategies to Compare Efficient and Equitable Vaccine Allocation by Race and Ethnicity Across Time

Development of a healthcare system COVID Hotspotting Score in California: an observational study with prospective validation

Equitably Allocating Resources During Crises: Racial Differences in Mortality Prediction Models

Incidence of 30-Day Venous Thromboembolism in Adults Tested for SARS-CoV-2 Infection in an Integrated Health Care System in Northern California

Retrospective Cohort Study Comparing Infliximab-dyyb and Infliximab in Biologic-Naive Patients With Inflammatory Bowel Disease in the United States

Racial Disparities in COVID-19 Testing and Outcomes : Retrospective Cohort Study in an Integrated Health System

Association of Positive Fluid Balance at Discharge After Sepsis Management With 30-Day Readmission