Romain S. Neugebauer, PhD

People

Kaiser Permanente Division of Research
2000 Broadway
Oakland, CA 94612

(510) 891-3234 (phone)
romain.s.neugebauer@kp.org
Kaiser Permanente Researcher Profiles

Romain S. Neugebauer, PhD, is a research scientist at the Kaiser Permanente Northern California Division of Research. His research focuses on the development, application, and dissemination of statistical methodologies to evaluate treatment effects in randomized experiments and observational studies. His particular interest is in causal methods to evaluate the effects of treatment regimens with complex longitudinal data. Prior to joining the Division of Research, Dr. Neugebauer was a researcher and lecturer in the School of Public Health at the University of California, Berkeley where he also received his PhD in biostatistics.

Current Position(s):

Research Scientist II, Division of Research, Kaiser Permanente Northern California

Section Affiliations:

Biostatistics Core

Related Website(s):

Causal Inference software

Studies

Leveraging Machine Learning to Improve Risk Prediction for Chemotherapy Induced Neuropathy

Chemotherapy-associated peripheral neuropathy (CPN) affects nearly 70% of cancer patients. An estimated 30% of cancer survivors have persistent CPN, which can have a devastating impact on their quality of life and functioning. Our objective is to develop and validate a predictive model that will identify those patients who are at high risk of developing severe CPN that can result in their receiving fewer planned chemotherapy cycles. We will focus on two subgroups our preliminary studies suggest have a higher risk for CPN: people who have diabetes or are African American/ Black.

Investigator: Adams, Alyce; Neugebauer, Romain
Funder: National Cancer Institute

Evaluation of Complex Interventions on Drug (Re)Fill Behavior Using Electronic Health Care Data

This study develops generalized per-protocol analyses based on Inverse Probability Weighting and Targeted Learning to control for or evaluate the effect of medication nonadherence in observational studies that use electronic health databases to inform the management of chronic conditions. Unlike current implementations of these analyses that require investigators to make untestable and often arbitrary assumptions regarding patients’ use of drugs from pharmacy dispensing data, the proposed causal methodologies directly evaluate the health effects of drug (re)fill regimens. To develop these analytic approaches, we seek guidance from stakeholder partners representing a broad range of constituencies to illustrate and ensure the practical relevance of the proposed methods from two prior studies with electronic health databases conducted to inform the management of type 2 diabetes.
Investigator: Neugebauer, Romain
Funder: Patient-Centered Outcomes Research Institute

Feasibility and Performance Assessments of TMLE and hdPS Methodologies to Fit MSM in a Real-World CER Study

This study aims to evaluate the practicability and practical advantages of: 1) Targeted Minimum Loss based Estimation (TMLE), and 2) a high-dimensional propensity score (hdPS) algorithm to fit dynamic Marginal Structural Models (MSM) in a longitudinal comparative effectiveness research (CER) study of adults with type 2 diabetes.
Investigator: Neugebauer, Romain
Funder: Agency for Healthcare Research and Quality

Software to Automate Implementation of Causal Inference Methods with Time-Varying Interventions

This project aims to develop SAS macros to automate the data structuring required for implementation of Causal Inference methods with electronic health records data in studies with time-varying exposures, confounding and selection bias (e.g., marginal structural modeling with inverse probability weighting). SAS macros are validated with simulated data and their applicability is evaluated using secondary data analyses in cardiovascular, diabetes, and health services research. End-user feedback is used to validate and refine the implementation and documentation of the SAS software.
Investigator: Neugebauer, Romain
Funder:

Causal Analyses of Electronic Health Record Data for Assessing the Comparative Effectiveness of Treatment Regimens

This study aims to strengthen causal inferences based on Marginal Structural Modeling (MSM) and large healthcare databases for Comparative Effectiveness Research in Type 2 Diabetes Mellitus (T2DM), specifically, it aims to: 1) Develop a consensus for the types and quantity of MSM parameters relevant to T2DM patients and other stakeholders; 2) Evaluate the feasibility and performance of MSM approaches based on Targeted Learning to estimate these MSM parameters more efficiently and with more robustness to modeling assumptions than the current Inverse Probability Weighting approaches; 3) Explore the biases that may arise from current MSM applications with data where patient monitoring is not fixed but random or where the outcome is not right-censored but interval-censored and evaluate innovative MSM methods that aim to address these biases.
Investigator: Neugebauer, Romain
Funder: Patient-Centered Outcomes Research Institute

Highlighted Publications

A Case Study of the Impact of Data-Adaptive Versus Model-Based Estimation of the Propensity Scores on Causal Inferences from Three Inverse Probability Weighting Estimators

Author(s): Neugebauer R; Schmittdiel JA; van der Laan MJ

Int J Biostat. 2016 May 1;12(1):131-55.

Current Position(s):

Section Affiliations:

Related Website(s):

Studies

Leveraging Machine Learning to Improve Risk Prediction for Chemotherapy Induced Neuropathy

Evaluation of Complex Interventions on Drug (Re)Fill Behavior Using Electronic Health Care Data

Feasibility and Performance Assessments of TMLE and hdPS Methodologies to Fit MSM in a Real-World CER Study

Software to Automate Implementation of Causal Inference Methods with Time-Varying Interventions

Causal Analyses of Electronic Health Record Data for Assessing the Comparative Effectiveness of Treatment Regimens

Highlighted Publications

A Case Study of the Impact of Data-Adaptive Versus Model-Based Estimation of the Propensity Scores on Causal Inferences from Three Inverse Probability Weighting Estimators

High-dimensional propensity score algorithm in comparative effectiveness research with time-varying interventions

Targeted learning in real-world comparative effectiveness research with time-varying interventions

Causal inference in longitudinal studies with history-restricted marginal structural models

Publications

Exploiting nonsystematic covariate monitoring to broaden the scope of evidence about the causal effects of adaptive treatment strategies

Determinants of Oral Bisphosphonate Use Beyond 5 Years

Comparison of Mortality and Major Cardiovascular Events Among Adults With Type 2 Diabetes Using Human vs Analogue Insulins

Identifying hypertension in pregnancy using electronic medical records: The importance of blood pressure values

Exercise During the First Trimester and Infant Size at Birth: Targeted Maximum Likelihood Estimation of the Causal Risk Difference

Patients with complex chronic conditions: Health care use and clinical events associated with access to a patient portal

Using Pharmacy Data and Adherence to Define Long-Term Bisphosphonate Exposure in Women

Association of immunosuppression and HIV viremia with anal cancer risk in persons living with HIV in the United States and Canada

Targeted learning with daily EHR data

Developing a Prognostic Information System for Personalized Care in Real Time

Association of immunosuppression and HIV viraemia with non-Hodgkin lymphoma risk overall and by subtype in people living with HIV in Canada and the USA: a multicentre cohort study.

Comparative effectiveness of treatment of actinic keratosis with topical fluorouracil and imiquimod in the prevention of keratinocyte carcinoma: a cohort study

Automated symptom and treatment side effect monitoring for improved quality of life among adults with diabetic peripheral neuropathy in primary care: a pragmatic, cluster, randomized, controlled trial.

simcausal R Package: Conducting Transparent and Reproducible Simulation Studies of Causal Effect Estimation with Complex Longitudinal Data

Cutaneous Melanoma Risk among People with HIV in the United States and Canada

A real-world, community-based cohort study comparing the effectiveness of topical fluoruracil vs. topical imiquimod for the treatment of actinic keratosis

simcausal R Package: Conducting Transparent and Reproducible Simulation Studies of Causal Effect Estimation with Complex Longitudinal Data

Adjuvant gemcitabine-docetaxel chemotherapy for stage I uterine leiomyosarcoma: Trends and survival outcomes

Minimally Invasive Hysterectomy and Power Morcellation Trends in a West Coast Integrated Health System

Associations of CD4+ T-cell count, HIV-1 RNA viral load, and antiretroviral therapy with Kaposi sarcoma risk among HIV-infected persons in the United States and Canada

Community-Based Practice Variations in Topical Treatment of Actinic Keratoses

The impact of telephonic wellness coaching on weight loss: A "Natural Experiments for Translation in Diabetes (NEXT-D)" study

Identification of the joint effect of a dynamic treatment intervention and a stochastic monitoring intervention under the no direct effect assumption

Analysis of Body Mass Index and Mortality in Patients With Colorectal Cancer Using Causal Diagrams

A Case Study of the Impact of Data-Adaptive Versus Model-Based Estimation of the Propensity Scores on Causal Inferences from Three Inverse Probability Weighting Estimators

The Diabetes Telephone Study; Design and challenges of a pragmatic cluster randomized trial to improve diabetic peripheral neuropathy treatment

The association of race/ethnicity and risk of atypical femur fracture among older women receiving oral bisphosphonate therapy

Occult Uterine Sarcoma and Leiomyosarcoma: Incidence of and Survival Associated With Morcellation

Wellness Coaching for People With Prediabetes: A Randomized Encouragement Trial to Evaluate Outreach Methods at Kaiser Permanente, Northern California, 2013

Telephone-Based Coaching: A Comparison of Tobacco Cessation Programs in an Integrated Health Care System

Use of abacavir and risk of cardiovascular disease among HIV-infected individuals

Employer-Based Screening for Diabetes and Prediabetes in an Integrated Health Care Delivery System: A Natural Experiment for Translation in Diabetes (NEXT-D) Study

Adherence to Metformin, Statins, and ACE/ARBs Within the Diabetes Health Plan (DHP)

High-dimensional propensity score algorithm in comparative effectiveness research with time-varying interventions

Targeted learning in real-world comparative effectiveness research with time-varying interventions

Racial/ethnic differences in hip and diaphyseal femur fractures

Exercise as a Vital Sign: A Quasi-Experimental Analysis of a Health System Intervention to Collect Patient-Reported Exercise Levels

Accuracy of hemoglobin A1c imputation using fasting plasma glucose in diabetes research using electronic health records data

Impact of Specific Glucose-Control Strategies on Microvascular and Macrovascular Outcomes in 58,000 Adults With Type 2 Diabetes

Patient Satisfaction and Perceived Success with a Telephonic Health Coaching Program: The Natural Experiments for Translation in Diabetes (NEXT-D) Study, Northern California, 2011

Super learning to hedge against incorrect inference from arbitrary parametric assumptions in marginal structural modeling

Health-Plan and Employer-Based Wellness Programs to Reduce Diabetes Risk; The Kaiser Permanente Northern California NEXT-D Study

Causal inference in epidemiological studies with strong confounding

Dynamic marginal structural modeling to evaluate the comparative effectiveness of more or less aggressive treatment intensification strategies in adults with type 2 diabetes

Predictors of patient satisfaction with mohs surgery: analysis of preoperative, intraoperative, and postoperative factors in a prospective cohort

HIV Infection, Immunodeficiency, Viral Replication, and the Risk of Cancer

Robust extraction of covariate information to improve estimation efficiency in randomized trials

Ambient ozone concentrations and cardiac mortality in Southern California 1983-2000: application of a new marginal structural model approach

Cohort Profile: The Diabetes Study of Northern California (DISTANCE)--objectives and design of a survey follow-up study of social health disparities in a managed care population

Causal inference in longitudinal studies with history-restricted marginal structural models