February 20th, 2019 Meeting

Past Meeting

Download Ben’s Presentation
Download Arjun and Mohan’s Presentation

If you are interested in a receiving a copy of Greg’s presentation, please contact him directly at gmckinne@pacbell.net.

Wednesday, February 20th, 2019

12:30 PM (Registration)
1:30 – 4:30 PM (Meeting)

Jazz Pharmaceutical
3170 Porter Drive
Palo Alto, CA 94304

3180 Porter Drive
Palo Alto, CA 94304

Facility Host:
Diane Androvich

Event Host:
Thomas Leung (415) 956-3611

Speaker 1

Ben Wang, Lead Medical Affairs Statistical Programmer, Biomarin

“Improving Efficiency by making a Data Driven Macro”


This paper presents an example of building a data driven macro for demographics to reduce effort and likelihood of errors. A demographics table is fairly standard across industries and is an often used tool to gain insight into the population of interest. By using metadata, SAS macro language, and programming management techniques programmers can edit code in one section and allow the effects to cascade through the rest of the code.


A demographics table often includes two types of variable attributes: continuous and categorical. Each type of variable has its own type of analysis. For categorical variables, the analysis may include the counts and percent for each level. For continuous, the analysis is often the mean, standard deviation, median, range, and Inter Quartile Range.

To build a demographics table we first examine desired output. Then we examine the dataset that houses subject level information. Each subject will have its own unique id and associated gender, age, etc. It is possible to obtain the metadata for this dataset and allow this to drive subsequent analyses.


From this presentation, we can take away the idea to build a company standard for demographic tables. This has many benefits such as saving on cost of programming validation time. An established standard also has the benefit of simplifying programming efforts if well maintained.

About the Speaker:

My name is Ben Wang. I am working at Biomarin as the lead Medical Affairs statistical programmer. I graduated from Carnegie Mellon with a BS in statistics and Rutgers Medical School with a Masters in Epidemiology. Prior to working at Biomarin, I worked as a statistician at Roche Molecular Diagnostics. I have eight plus years of pharmaceutical experience.

Speaker 2

Mohan Madhanagopal & Arjun Malipeddi, Jazz Pharmaceuticals

“Learnings from Statistical Programming supporting audits for regulatory agencies (FDA and EMA)”


It is a common practice for regulatory agencies to use onsite inspections to ensure the clinical investigators and sponsors comply with GCP regulations while developing investigational drugs. Food and Drug Administration (FDA) established Bioresearch Monitoring (BIMO) Program, to develop cross-center guidelines for inspections of clinical investigators and sponsors, which is in turn, managed by Office of Scientific Investigations (OSI).

Similarly, for applications submitted to the European Medicines Agency (EMA), the Committee for Medicinal Products for Human Use (CHMP) often requests the Good Clinical Practice (GCP) inspections.

Both FDA and EMA have established standard guidance, that sponsor is expected to adhere to while presenting site level information. In this presentation, we look to explore those requirements of presenting that site level information and the statistical programming strategy that we implemented.

About the Speakers:

Mohan is a Senior Manager-Statistical Programming at Jazz Pharmaceuticals in Palo Alto, CA. He has over 12 years of SAS programming experience working at Clinical Research Organizations, at different capacities, with Statistical Programming teams leading to regulatory submissions. Mohan has an educational background in computer science with a Masters from Texas State University-San Marcos and Bachelors from University of Madras-Chennai.

Arjun is currently a Senior Statistical Programmer at Jazz Pharmaceuticals with around 7 years of experience in Pharmaceutical and Medical Device industries. I have supported various in-house programming activities leading to regulatory submissions. I have a Master’s Degree in Analytical Sciences from Texas A&M University, Commerce and Bachelors in Pharmacy from JNTU-Hyderabad.

Speaker 3

Gregory McKinney, Independent

“Markov Chains as a Predictive Analytics Technique Using SAS/IML”


As a predictive analytics approach, Markov Chains provide a powerful framework for modeling complex multi-state dynamic systems. Because of this power, Markov Chain models have been used to address many real-world problems across many industries and functions. In this paper, we look at a disease progression model using a Markov Chain. SAS/IML provides a potent tool for implementing Markov chain models in your organization. Sample code for the use of SAS/IML for Markov chain models is included.

About the Speaker:

A SAS user since 1979, Greg has worked in the Data Analytics, Data Warehouse, and Machine Learning fields within the Healthcare, Financial Services, Energy, and High-Tech industries. During his professional career, he has used SAS on IBM Mainframe, Unix, and Windows platforms. Greg has developed analytic solutions using Base SAS, SAS STAT, SAS/IML, SAS EG SAS Graph, SAS Info Maps, and SAS OLAP Cube Studio, among other SAS tools.

Currently an independent consultant, Greg helps organizations achieve strategic success through the innovative application of Data Analytics and Machine Learning to create actionable business knowledge from volumes of raw data.

He received his Bachelor of Science degree from the University of Illinois in Chicago with a dual major in Information & Decision Sciences, and Economics. He also received a Master of Science degree in Management Sciences in from Northwestern University. He is passionate about fine-art photography and sports. Greg is active as a volunteer coach (swimming and softball) for Special Olympics of Northern California.