Our Events in 2018

2019 ASA traveling short course: Analysis of Big Healthcare Databases

The ASA Princeton-Trenton chapter is hosting a half-day 2019 ASA traveling short course:


Analysis of Big Healthcare Databases


Instructor: Rebecca Hubbard, University of Pennsylvania


When: Friday, September 20, 2019 from 9:15 AM to 2:30 PM


West Lecture Hall, Rutgers Robert Wood Johnson Medical School

675 Hoes Lane West, 

Piscataway, NJ 08854

How to Register:

Please use the following link for registration:


Regular registration: $50; student/retiree: $25

Note that refreshments and a lunch buffet are included in each registration, and parking is complimentary and available in Lots A, B & C. Event Parking signage will be at the entrance to Lots A, B & C. Please note that all event parking at Rutgers now requires an online registration (registration is free, link to be shared).






8:45 – 9:10 am

Registration, Coffee, Breakfast, and Check In

9:10 – 9:15 am

Greetings and Introduction

9:15 – 10:30 am

Data structure and extracting data elements from the EHR

10:30 – 10:45 am


10:45 – 12:00 PM

Missing data and outcome misclassification

12:00 – 1:00 pm


1:00 – 2:30 pm

Correcting for bias due to EHR data errors


Parking Map:



This is a great opportunity for learning and networking. We look forward to seeing you at this exciting event.


Yong Lin

President, ASA Princeton-Trenton Chapter


Abstract: The widespread adoption of electronic health records (EHR) as a means of documenting medical care has created a vast resource for the study of health conditions, interventions, and outcomes in routine clinical practice. Using healthcare databases, including EHR and administrative claims data, for research facilitates the efficient creation of large research databases, execution of pragmatic clinical trials, and study of rare diseases. Despite these advantages, there are many challenges for research conducted using these data. To make valid inference, statisticians must be aware of data generation, capture, and availability issues and utilize appropriate study designs and statistical analysis methods to account for these issues. In this course, we will discuss topics related to the design and analysis of research studies using big healthcare databases. We will cover issues related to the structure and quality of the data, including data types and methods for extracting variables of interest; sources of missing data; error in covariates and outcomes extracted from EHR and claims data; and data capture considerations such as informative visit processes and medical records coding procedures. In the second half of the course, we will discuss statistical approaches to address some of the challenges and unique features of healthcare databases, including missing data and error in automated algorithm-derived covariates and outcomes. We will also discuss some cutting-edge methods developed to address the unique challenges of this context such as privacy-preserving computation for use in distributed research networks. The overarching objective of this course is to provide participants with an introduction to the structure and content of healthcare databases and equip them with a set of appropriate tools to investigate and analyze this rich data resource.

About the instructor: Rebecca Hubbard is an Associate Professor of Biostatistics in the Department of Biostatistics, Epidemiology and Informatics at the University of Pennsylvania. Her methodological research emphasizes development of statistical tools to support valid inference for EHR-based analyses, accounting for complex data availability and data quality issues, and has been applied across a variety of domains including studies of cancer epidemiology, aging and dementia, and pharmacoepidemiology. She has experience conducting analyses using data from a number of large healthcare databases including Medicare, PCORnet, Kaiser Permanente, Flatiron Health, and Optum. Results of this work have been published in over 100 peer-reviewed papers in the statistical and medical literature. She has taught short courses at ENAR, the FDA, and the Summer Institutes in Statistical Genetics and Statistics for Clinical Research at the University of Washington over the past 10 years.


May, 2017 



  • ASA Prinecton-Trenton Chapter Spring 2017 Symposium


Dear Colleagues,

 The ASA Princeton-Trenton Chapter in New Jersey is organizing a Spring Statistics Symposium from 9:00 AM to 2:30 PM EDT on Friday, May 5, 2017 in the East Lecture Hall, Rutgers Robert Wood Johnson Medical School.

The focus of the Spring Symposium is on New Advancement in Biomarkers/Precision Medicine/Enrichment Design. It aims to introduce participants to the new advancement of statistical methods and applications in precision medicine, to discuss the challenges and solutions related to the use of biomarkers in design and analyses for clinical trials and translational research, and to review present knowledge on biomarkers.

Tentative agenda:




8:30 - 9:00 AM



9:00 - 9:05 AM

Greetings and Introduction

9:05 - 9:50 AM

The Reality of Personalized Oncology – 2017


Richard Simon,Associate Director, Division of Cancer Treatment & Diagnosis, Biometric Research Program

Chief, Computational & Systems Biology Branch, National Cancer Institute

9:50 - 10:35 AM

Statistical Challenges in Precision Medicine with a Focus on Companion Diagnostics


Gregory Campbell,President, GCStat Consulting, LLC. Former Director, Division of Biostatistics, CDRH/FDA

10:35 - 10:50 AM


10:50 - 11:35 AM

Two Novel Designs for Small Populations: The Confirmatory Basket Trial and the Informational Design


Robert A. Beckman, MD,Professor of Oncology and of Biostatistics, Bioinformatics, and Biomathematics,Georgetown University Medical Center

11:35 - 12:20 PM

Developing an Immunohistochemistry Test for Programmed Cell Death Ligand 1 (PD-L1) as a Companion Diagnostic for Pembrolizumab

Deepti Auroa-Garg,Director, Biomarker and Diagnostics Leader, Merck Oncology

12:20 - 1:30 PM

Lunch Break

1:30 - 2:30 PM

Panel discussion

Richard Simon, Greg Campbell,
Robert A. Beckman,Deepti Auroa-Garg


light breakfast and lunch buffet are provided with registration. The parking lots [Parking Lot A, B, and C] are conveniently located near the event location, and will have Signs indicating event parking. For your information, a parking map is attached. 






1. Richard Simon: The Reality of Personalized Oncology– 2017

Oncology is a set of diseases thought to result in large part from the development of genomic mutations and other DNA alterations. The development of inexpensive DNA sequencing has enabled the development of new cancer classifications based on the presence or absence of genomic alterations. This classification has been strongly correlated with response to molecularly targeted treatment and this has had a profound influence on pharmaceutical drug development strategies. These strategies have led to changes in clinical development plans as the old “aspirin paradigm” is no longer appropriate as a basis for clinical trial design in many cases.


In my talk I will describe some successes in accommodating clinical trial design to developments in cancer biology. I will also describe some remaining challenges such as utilization of quantitative biomarkers, prospective-retrospective studies and designs which adaptively identify the intended use population.


2. Gregory Campbell: Statistical Challenges in Precision Medicine with a Focus on Companion Diagnostics

Since the successful sequencing of the human genome early this century, the public has begun to see practical breakthroughs through advancements in precision medicine; namely, therapeutic medical products tailored to the patient using a biomarker. With the recent genetic and genomic advancements, there has been an explosion in the amount of biomarker data. One statistical challenge is that of co-development: how to confirm that a particular therapeutic product is safer or more efficacious for an individual based on the particular result of a companion diagnostic (usually genomic) test. Various types of diagnostic tests or biomarkers are introduced and illustrated. Another statistical challenge is that of design: how can diagnostic tests and therapeutic products be co-developed, especially if the drug clinical trial precedes the development of a market ready diagnostic test to be used in concert with the drug. Statistical designs that allow for adaptation or for the use of retrospective data in a scientifically valid manner are discussed. Analysis of data from such trials is also challenging, particularly of multiplicity, selection of a cutoff for the test, and missing data. Several drug-diagnostic examples are reviewed and a number of clinical trials are discussed. The implications for the future of individualized medicine are enormous. An interdisciplinary effort involving.


3. Robert Beckman: Two Novel Designs for Small Populations: The Confirmatory Basket Trial and the Informational Design

Two novel designs will be presented, the confirmatory basket trial and the informational design.

Increasingly, tumors are defined on a molecular basis rather than only on histology, and targeted agents which address these molecular subtypes are being approved. This profusion of molecular subtypes creates “rare” diseases as subsets of common cancers, leading to difficulties in enrolling sufficiently large cohorts for confirmatory trials. However, if the molecular subtype is shared across various histologies, these may be pooled into a basket trial. To date, basket trials have been primarily for exploratory early development. We present a new confirmatory basket trial design which will provide patients in niche indications with enhanced access to novel therapies, facilitate development and full approval for niche indications, allow accelerated approval for indications within a basket based on a surrogate endpoint, reduce development cost by combining trials into one, and enhance the ability of regulatory authorities to evaluate risk and benefit in niche indications.

The informational design allows adaptation when the ability of interim endpoints to predict final definitive endpoints is uncertain. A fraction of the patients are designated as an informational cohort and their data governs an adaptation at the end of the study. These patients are also used in the final analysis with suitable alpha adjustment. Applications include alpha allocation between a full population and a subset, setting a threshold for a continuous biomarker, and a speculative application of a confirmatory trial which selects its own primary endpoint. The latter may be useful for rare diseases where the natural history is unknown and difficult to obtain. 



4. Deepti Auroa-GargDeveloping an Immunohistochemistry Test for Programmed Cell Death Ligand 1 (PD-L1) as a Companion Diagnostic for Pembrolizumab


Keytruda® (Pembrolizumab) an anti-PDL1 therapy has been approved in first and second line Non-Small Cell Lung Cancer (NSCLC). This approval is restricted to patients whose tumors express PDL1 ≥ 50% for first line and ≥ 1% for second line thereby heralding a new era in immune-oncology and personalized medicine. The talk will focus on the development of a PDL1 IHC diagnostic test for Keytruda eligibility and cover aspects of selection of scoring guidelines in training sets and their application in Phase III trials.





ASA Princeton-Trenton Chapter