VLVincent Liu

Vincent Liu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Vincent Liu has a total of 40 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Consensus Statement on Electronic Health Predictive Analytics: A Guiding Framework to Address Challenges

    by Ruben Amarasingham, Anne-Marie J. Audet, David W. Bates, I. Glenn Cohen, Martin Entwistle, G. J. Escobar, Vincent Liu, Lynn Etheredge, Bernard Lo, Lucila Ohno-Machado, et al.

    Context: The recent explosion in available electronic health record (EHR) data is motivating a rapid expansion of electronic health care predictive analytic (e-HPA) applications, defined as the use of electronic algorithms that forecast clinical events in real time with the intent to improve patient outcomes and reduce costs. There is an urgent need for a systematic framework to guide the development and application of e-HPA to ensure that the field develops in a scientifically sound, ethical, and efficient manner.Objectives: Building upon earlier frameworks of model development and utilization, we identify the emerging opportunities and challenges of e-HPA, propose a framework that enables us to realize these opportunities, address these challenges, and motivate e-HPA stakeholders to both adopt and continuously refine the framework as the applications of e-HPA emerge.Methods: To achieve these objectives, 17 experts with diverse expertise including methodology, ethics, legal, regulation, and health care delivery systems were assembled to identify emerging opportunities and challenges of e-HPA and to propose a framework to guide the development and application of e-HPA.Findings: The framework proposed by the panel includes three key domains where e-HPA differs qualitatively from earlier generations of models and algorithms (Data Barriers, Transparency, and Ethics) and areas where current frameworks are insufficient to address the emerging opportunities and challenges of e-HPA (Regulation and Certification; and Education and Training). The following list of recommendations summarizes the key points of the framework:1. Data Barriers: Establish mechanisms within the scientific community to support data sharing for predictive model development and testing.2. Transparency: Set standards around e-HPA validation based on principles of scientific transparency and reproducibility.3. Ethics: Develop both individual-centered and society-centered risk-benefit approaches to evaluate e-HPA.4. Regulation and Certification: Construct a self-regulation and certification framework within e-HPA.5. Education and Training: Make significant changes to medical, nursing, and paraprofessional curricula by including training for understanding, evaluating, and utilizing predictive models.


  • Open Access

    DISCOVERING PATIENT PHENOTYPES USING GENERALIZED LOW RANK MODELS

    by Alejandro Schuler, Vincent Liu, Joe Wan, Alison Callahan, Madeleine Udell, David E. Stark, Nigam H. Shah

    The practice of medicine is predicated on discovering commonalities or distinguishing characteristics among patients to inform corresponding treatment. Given a patient grouping (hereafter referred to as a phenotype), clinicians can implement a treatment pathway accounting for the underlying cause of disease in that phenotype. Traditionally, phenotypes have been discovered by intuition, experience in practice, and advancements in basic science, but these approaches are often heuristic, labor intensive, and can take decades to produce actionable knowledge. Although our understanding of disease has progressed substantially in the past century, there are still important domains in which our phenotypes are murky, such as in behavioral health or in hospital settings. To accelerate phenotype discovery, researchers have used machine learning to find patterns in electronic health records, but have often been thwarted by missing data, sparsity, and data heterogeneity. In this study, we use a flexible framework called Generalized Low Rank Modeling (GLRM) to overcome these barriers and discover phenotypes in two sources of patient data. First, we analyze data from the 2010 Healthcare Cost and Utilization Project National Inpatient Sample (NIS), which contains upwards of 8 million hospitalization records consisting of administrative codes and demographic information. Second, we analyze a small (N=1746), local dataset documenting the clinical progression of autism spectrum disorder patients using granular features from the electronic health record, including text from physician notes. We demonstrate that low rank modeling successfully captures known and putative phenotypes in these vastly different datasets.


  • Adverse outcomes associated with delayed intensive care unit transfers in an integrated healthcare system

    by Vincent Liu, Patricia Kipnis, Norman W. Rizk, Gabriel J. Escobar

    BACKGROUND: Patients with intensive care unit (ICU) transfers from hospital wards have higher mortality than those directly admitted from the emergency department.OBJECTIVE: To describe the association between the timing of unplanned ICU transfers and hospital outcomes.DESIGN, SETTING, PATIENTS: Evaluation of 6369 early (within 24 hours of hospital admission) unplanned ICU transfer cases and matched directly admitted ICU controls from an integrated healthcare system. Cohorts were matched by predicted mortality, age, gender, diagnosis, and admission characteristics. Hospital mortality of cases and controls were compared based on elapsed time and diagnosis.RESULTS: More than 5% of patients admitted through the emergency department experienced an unplanned ICU transfer; the incidence and rates of transfers were highest within the first 24 hours of hospitalization. Multivariable matching produced 5839 (92%) case‐control pairs. Median length of stay was higher among cases (5.0 days) than controls (4.1 days, P < 0.01); mortality was also higher among cases (11.6%) than controls (8.5%, P < 0.01). Patients with early unplanned transfers were at an increased risk of death (odds ratio, 1.44; 95% confidence interval, 1.26‐1.64; P < 0.01); an increased risk of death was observed even among patients transferred within 8 hours of hospitalization. Hospital mortality differed based on admitting diagnosis categories. While it was higher among cases admitted for respiratory infections and gastrointestinal bleeding, it was not different for those with acute myocardial infarction, sepsis, and stroke.CONCLUSIONS: Early unplanned ICU transfers—even within 8 hours of hospitalization—are associated with increased mortality; outcomes vary by elapsed time to transfer and admitting diagnosis. Journal of Hospital Medicine 2011;. © 2011 Society of Hospital Medicine.


  • Persistent Racial Disparities in Survival After Heart Transplantation

    by Vincent Liu, Jay Bhattacharya, David Weill, Mark A. Hlatky

    Background—: Racial and ethnic disparities are well documented in many areas of health care, but have not been comprehensively evaluated among recipients of heart transplants. Methods and Results—: We performed a retrospective cohort study of 39075 adult primary heart transplant recipients from 1987 to 2009 using national data from the United Network of Organ Sharing and compared mortality for nonwhite and white patients using the Cox proportional hazards model. During the study period, 8082 nonwhite and 30 993 white patients underwent heart transplantation. Nonwhite heart transplant recipients increased over time, comprising nearly 30% of transplantations since 2005. Nonwhite recipients had a higher clinical risk profile than white recipients at the time of transplantation, but had significantly higher posttransplantation mortality even after adjustment for baseline risk. Among the nonwhite group, only black recipients had an increased risk of death compared with white recipients after multivariable adjustment for recipient, transplant, and socioeconomic factors (hazard ratio, 1.34; 95% confidence interval, 1.21 to 1.47; P <0.001). Five-year mortality was 35.7% (95% confidence interval, 35.2 to 38.3) among black and 26.5% (95% confidence interval, 26.0 to 27.0) among white recipients. Black patients were more likely to die of graft failure or a cardiovascular cause than white patients, but less likely to die of infection or malignancy. Although mortality decreased over time for all transplant recipients, the disparity in mortality between blacks and whites remained essentially unchanged. Conclusions—: Black heart transplant recipients have had persistently higher mortality than whites recipients over the past 2 decades, perhaps because of a higher rate of graft failure.


  • Open Access

    Racial Disparities in Survival After Lung Transplantation

    by Vincent Liu, David Weill, Jay Bhattacharya

    Racial disparities in access to health care and survival have been well described for many medical conditions.1-3 Establishing equity by eliminating these disparities, a goal of the Healthy People 2010 initiative,4 has been proposed as a means of preventing tens of thousands of deaths each year.5,6 Previous data, however, suggest that some racial disparities in health care have not changed, or perhaps have even worsened, with time.2,7,8


  • Length of Stay Predictions

    by Vincent Liu, Patricia Kipnis, Michael K. Gould, Gabriel J. Escobar

    Length of stay (LOS) is a common measure of hospital resource utilization. Most methods for risk-adjusting LOS are limited by the use of only administrative data. Recent studies suggest that adding automated clinical data to these models improves performance. To evaluate the utility of adding "point of admission" automated laboratory and comorbidity measures-the Laboratory Acute Physiology Score (LAPS) and Comorbidity Point Score (COPS)-to risk adjustment models that are based on administrative data. We performed a retrospective analysis of 155,474 hospitalizations between 2002 and 2005 at 17 Northern California Kaiser Permanente hospitals. We evaluated the benefit of adding LAPS and COPS in linear regression models using full, trimmed, truncated, and log-transformed LOS, as well as in logistic and generalized linear models. Mean age was 61 +/- 19 years; females represented 55.2% of subjects. The mean LOS was 4.5 +/- 7.7 days; median LOS was 2.8 days (interquartile range, 1.3-5.1 days). Adding LAPS and COPS to the linear regression model improved R by 29% from 0.113 to 0.146. Similar improvements with the inclusion of LAPS and COPS were observed in other regression models. Together, these variables were responsible for >50% of the predictive ability of logistic regression models that identified outliers with longer LOS. The inclusion of automated laboratory and comorbidity data improved LOS predictions in all models, underscoring the need for more widespread adoption of comprehensive electronic medical records.


  • A multi-drug regimen for respiratory syncytial virus and parainfluenza virus infections in adult lung and heart-lung transplant recipients

    by V. Liu, G.S. Dhillon, D. Weill

    V. Liu, G.S. Dhillon, D. Weill. A multi-drug regimen for respiratory syncytial virus and parainfluenza virus infections in adult lung and heart–lung transplant recipients.Transpl Infect Dis 2010: 12: 38–44. All rights reserved Background. Respiratory syncytial virus (RSV) and parainfluenza virus (PIV) can cause significant morbidity and mortality in lung and heart–lung transplant recipients. We evaluated the utility of a multi-drug protocol for the treatment of RSV- and PIV-related infections. Patients and methods. RSV or PIV was identified in 25 patients with a total of 29 infectious episodes between January 2006 and December 2007. The study included 20 women and 5 men, mean age 42 ± 13 years. Fifteen patients had received bilateral lung transplant and the remainder either received single lung or heart–lung transplant. Mean time from transplant to infection was 1192 days. RSV was identified in 23 cases, PIV in 7 cases. Patients underwent treatment with inhaled ribavirin, methylprednisolone, and intravenous immunoglobulin (IVIG). RSV-positive patients were also treated with palivizumab. We retrospectively evaluated their clinical status and pulmonary function for a 1-year interval before and after the date of infection. Results. Average baseline forced expiratory volume in 1 s (FEV1) before infection was 2.14 ± 0.68 L/min. Average decline in FEV1 was 5.7% at the time of infection. Average FEV1 during post-treatment follow-up was not significantly different than baseline (2.16 ± 0.80 L/min). Among patients with bronchiolitis obliterans syndrome (BOS) stages 1, 2, or 3 at the time of infection, average FEV1 declined by 14.8% and remained lower at 9.1% during follow-up when compared with patients with BOS stages 0 or 0p. No complications resulted from treatment. One patient died during follow-up as a result of pre-existing liver failure. Conclusions. This study of lung and heart–lung transplant recipients infected with RSV and PIV shows that a multi-drug regimen including inhaled ribavirin, corticosteroids, and IVIG (with or without palivizumab) is safe and effective. Prompt diagnosis and therapy for patients with RSV or PIV infections are critical for maintaining lung function.


  • Pulmonary Toxicity Associated With Erlotinib

    by Vincent Liu, Dorothy A. White, Maureen F. Zakowski, William Travis, Mark G. Kris, Michelle S. Ginsberg, Vincent A. Miller, Christopher G. Azzoli