BiosBeat

Welcome to the BiosBeat communication tool for the Department of Biostatistics at the UNC Gillings School of Global Public Health! Here you will find a collection of the latest department news, special features, dates to save, and so much more. So, read on, enjoy, and be sure to regularly check back for updates!


In this edition:

BIOS student honored with ENAR’s Distinguished Paper Award
Biostatistics researcher awarded ASA’s Norman Breslow Prize for outstanding paper
Alumnus honored with Presidential Early Career Award for Scientists and Engineers
Loop uses heat maps to show geographic variations in cardiovascular disease risk factors
Chen, Zeng and Kosorok publish discussion paper in JASA
Save the Dates


BIOS student honored with ENAR’s Distinguished Paper Award

Fei Gao

Fei Gao

Fei Gao, a biostatistics doctoral student at the UNC Gillings School of Global Public Health, has won a Distinguished Student Paper Award, presented by the Eastern North America Region (ENAR) of the International Biometric Society. Her advisers, Donglin Zeng, PhD, professor of biostatistics, and Danyu Lin, PhD, Dennis Gillings Distinguished Professor of biostatistics, are co-authors of the paper.

In some clinical studies, such as studies of chronic disease, researchers may not be able to pinpoint the time at which a particular health event occurs. However, study participants are interviewed or examined at a number of points along a timeline, and by collecting data at intervals, researchers are better able to place the health event on the timeline.

When participants do not attend one or more examination sessions, possibly for health-related reasons, data collection is diminished.

“For example, if you are sicker, you may choose to drop out of a study prematurely,” said Gao. “I wanted to do a joint analysis for this problem in which we would model the event time of interest and the dropout time together, while using a random effect to capture their association.”

She illustrated the effectiveness of the method by analyzing data from the prospective epidemiologic Atherosclerosis Risk in Communities (ARIC) study and correcting the prediction of a diabetes event among ARIC participants by adjusting for dropout due to death.

“Fei’s work addresses an important problem in clinical and epidemiological studies,” said Lin. “Her work requires strong theoretical and computational skills and is highly relevant to medical and public health research. We’re very proud of her for winning this award.”

Gao will present her paper at the ENAR 2017 spring meeting, to be held March 12-15, in Washington, D.C.

The UNC Collaborative Studies Coordinating Center is the coordinating center for ARIC, which is sponsored by the National Heart, Lung and Blood Institute.


Biostatistics researcher awarded ASA’s Norman Breslow Prize for outstanding paper

Dr. Qingning Zhou

Dr. Qingning Zhou

Qingning Zhou, PhD, postdoctoral research associate in the UNC Gillings School of Global Public Health’s Department of Biostatistics, has received the American Statistical Association (ASA) Statistics in Epidemiology section’s Norman Breslow Prize for her paper, “Outcome-dependent Sampling with Interval-censored Failure Time Data.”

Zhou’s advisers – Haibo Zhou, PhD, biostatistics professor, and Jianwen Cai, PhD, Cary C. Boshamer Distinguished Professor and interim chair of biostatistics – are co-authors.

Researchers who conduct epidemiological studies and disease prevention trials often characterize the relationship between an exposure and a “failure time,” or time-to-disease, outcome. In many applications, the failure rate is low, and the observed data suffer from interval-censoring, which occurs when the failure time is not exactly observed but known only to fall within a certain interval.

In such cases, large cohort studies may be required to reach a reliable conclusion on the exposure-failure time relationship. Conducting such studies, however, could be prohibitive for researchers on limited budgets, especially when the exposure measurements are expensive to obtain.

To enhance study efficiency and reduce costs, Zhou proposes an outcome-dependent sampling (ODS) design and an efficient inference procedure for studies concerning interval-censored failure time outcome.

The paper also illustrates how this design and method may be applied to a dataset on incident diabetes from the Atherosclerosis Risk in Communities (ARIC) prospective epidemiologic study.

“The basic idea of ODS design is to oversample subjects believed to be more informative in terms of the exposure-outcome relationship,” said [Qingning] Zhou. “We enrich the observed sample by selectively including subjects who experience the failure at an early or late time.”

The Norman Breslow Prize, the Statistics in Epidemiology section’s top award presented to young investigators, is given to papers with both methodological contributions and substantive epidemiological applications.

Norman Breslow, PhD, a prominent biostatistician, was known for his work on case-control study design, which compares disease exposure in a case group that has undergone certain outcome to exposure in a control group without that outcome. The development of ODS design was inspired by the case-control study design.

“This work provides a much needed statistical tool for epidemiologists to conduct studies more powerfully without increasing the study budget,” said Cai. “I am excited and proud that Qingning won this prestigious award.”

Zhou will present the paper at the 2017 Joint Statistical Meetings (JSM), to be held in Baltimore from July 29 to Aug. 3. She has been awarded a $1,000 stipend to defray travel costs to JSM.

The UNC Collaborative Studies Coordinating Center, based in the biostatistics department, is the coordinating center for ARIC.


Alumnus honored with Presidential Early Career Award for Scientists and Engineers
Dr. Matthew Wheeler

Dr. Matthew Wheeler

Matthew Wheeler, PhD, alumnus of UNC Gillings School of Global Public Health and researcher at the Centers for Disease Control and Prevention, received the Presidential Early Career Award for Scientists and Engineers, the highest honor bestowed by the United States government upon science and engineering professionals in the early stages of their independent research careers.

President Barack Obama announced the award for Wheeler and 101 other American scientists on Jan. 9.

“I congratulate these outstanding scientists and engineers on their impactful work,” President Obama said. “These innovators are working to help keep the United States on the cutting edge, showing that federal investments in science lead to advancements that expand our knowledge of the world around us and contribute to our economy.”

The awards, established by President Clinton in 1996, are coordinated by the Executive Office of the President’s Office of Science and Technology Policy. Awardees are selected for their pursuit of innovative research at the frontiers of science and technology and their commitment to community service as demonstrated through scientific leadership, public education or community outreach.

Wheeler works on the development of risk assessment methods at the CDC’s National Institute for Occupational Safety and Health (NIOSH).

He recently returned from temporary assignment to the U.S. Environmental Protection Agency, where he developed software to be included in their risk assessment evaluations. He also has developed novel ways to estimate adverse response to chemical hazards of any dose size, using only the chemical structure information.

“This work, if replicated and extended,” Wheeler said, “has potential to be groundbreaking. It will allow us to estimate chemical responses across a range of doses instead of a single dose, which is the current standard, and to develop methods of chemical toxicity screening before the chemical even is developed – obviously a great benefit to occupational safety and health. As these methods are similar to those used in pharmaceutical drug development, it also has potential to be extended in this area, where it could be used to hone in on chemicals with therapeutic attributes, using computer models, before the chemical is developed and tested in the lab.”

“Matt distinguished himself by being an excellent researcher from day 1, with a great knack for identifying important scientific problems and relentlessly pursuing a solution,” said Amy Herring, ScD, Carol Remmer Angle Distinguished Professor of Children’s Environmental Health and associate chair in the Gillings School’s Department of Biostatistics. “I am delighted the President chose to honor him for his strong contributions to health science.”

Wheeler earned bachelor’s and master’s degrees from Miami University (Ohio) and a doctoral degree in biostatistics from the Gillings School in 2013.


Loop uses heat maps to show geographic variations in cardiovascular disease risk factors
Matthew_Loop_Profile_2016

Dr. Matthew Loop

Analyzing data from the longitudinal, population-based cohort Reasons for Geographic and Racial Differences in Stroke (REGARDS) study, which focuses on stroke mortality among blacks and residents of the southeast United States, Matthew Loop, PhD, clinical assistant professor of biostatistics, uses heat maps to illustrate geographic variation in cardiovascular disease risk factors including hypertension, diabetes mellitus, and smoking in the continental U.S. In his paper, recently published in the American Heart Association journal, Circulation: Cardiovascular Quality and Outcomes, Loop looks beyond the administrative county lines that often define disparities because data are only available from county-level maps. With these heat maps, Loop is able to show that the prevalences of hypertenstion, diabetes mellitus and smoking vary more finely than by state or county, and differ between blacks and whitesuseful information for future public health interventions.

Point map of REGARDS participant locations (n=28 879). Blue indicates black race and red indicates white race. REGARDS indicates Reasons for Geographic and Racial Differences in Stroke (Contributed image).

Point map of REGARDS participant locations (n=28 879). Blue indicates black race and red indicates white race. REGARDS indicates Reasons for Geographic and Racial Differences in Stroke. (Contributed image)

Maps of estimated hypertension, diabetes mellitus, and current smoking prevalence among whites and blacks, adjusted for age and sex. High prevalence is indicated by red, while low prevalence is indicated by blue. Predicted prevalences assumed a population with the same proportion of women for each race and the same age as the mean age of each race. Thus, the prevalences reflect the sex and age composition of REGARDS participants of each race. REGARDS indicates Reasons for Geographic and Racial Differences in Stroke (Contributed image).

Maps of estimated hypertension, diabetes mellitus, and current smoking prevalence among whites and blacks, adjusted for age and sex. High prevalence is indicated by red, while low prevalence is indicated by blue. Predicted prevalences assumed a population with the same proportion of women for each race and the same age as the mean age of each race. Thus, the prevalences reflect the sex and age composition of REGARDS participants of each race. REGARDS indicates Reasons for Geographic and Racial Differences in Stroke. (Contributed image)


Chen, Zeng and Kosorok publish discussion paper in JASA
Dr. Guanhua Chen

Dr. Guanhua Chen

zeng_donglin_optio2 copy

Dr. Donglin Zeng

Dr. Michael Kosorok

Dr. Michael Kosorok

Guanhua Chen, PhD, who earned master’s (2010) and doctoral (2015) degrees at the Gillings School and now is assistant professor of biostatistics in the Vanderbilt University School of Medicine; Donglin Zeng, PhD, professor of biostatistics; and Michael R. Kosorok, PhD, W.R. Kenan Jr. Distinguished Professor of biostatistics, have published their discussion paper, “Personalized Dose Finding Using Outcome Weighted Learning” in the Journal of the American Statistical Association.

Abstract:

In dose-finding clinical trials, it is becoming increasingly important to account for individual-level heterogeneity while searching for optimal doses to ensure an optimal individualized dose rule (IDR) maximizes the expected beneficial clinical outcome for each individual. In this article, we advocate a randomized trial design where candidate dose levels assigned to study subjects are randomly chosen from a continuous distribution within a safe range. To estimate the optimal IDR using such data, we propose an outcome weighted learning method based on a nonconvex loss function, which can be solved efficiently using a difference of convex functions algorithm. The consistency and convergence rate for the estimated IDR are derived, and its small-sample performance is evaluated via simulation studies. We demonstrate that the proposed method outperforms competing approaches. Finally, we illustrate this method using data from a cohort study for warfarin (an anti-thrombotic drug) dosing. Supplementary materials for this article are available online.

Read the full paper and discussion here.


Save the Dates

Thursday, February 16

New Student Recruitment
Contact Veronica Stallings for more information.

Friday, February 17

New Student Recruitment
Contact Veronica Stallings for more information.

Thursday, February 23

Doctoral Alumni Panel
1301 McGavran-Greenberg Hall
5:15-6:15 p.m.

Tuesday, February 28

Master’s Alumni Panel
1301 McGavran-Greenberg Hall
5:15-6:15 p.m.

Friday, March 10

Spring Break begins at 5 p.m.

Sunday, March 18

Spring Break ends

Thursday, March 23

Biostatistics Awards Day
BIOS 843 Seminar
Seunggeun (Shawn) Lee, PhD, Assistant Professor, 2017 James E. Grizzle Distinguished Alumni Award Recipient
The University of Michigan School of Public Health, Department of Biostatistics

Friday, March 24

Experience Gillings: A Program for Admitted Graduate Students

Thursday, March 30

BIOS 843 Seminar
Xiaoming Huo, PhD, Professor
Georgia Institute of Technology H. Milton Stewart School of Industrial & Systems Engineering

Thursday, April 6

BIOS 843 Seminar
Veerabhadran Baladandayuthapani, PhD, Associate Professor
The University of Texas MD Anderson Cancer Center, Division of Quantitative Sciences, Department of Biostatistics

Thursday, April 13

BIOS 843 Seminar
Chiung-Yu Huang, PhD, Associate Professor
Johns Hopkins University Bloomberg School of Public Health, Department of Biostatistics

Friday, April 14

Class holiday; no classes held

Thursday, April 20

BIOS 843 Seminar
Gabor Szekely, PhD, Program Director
National Science Foundation, Division of Mathematical Sciences, Statistics Program

Thursday, April 27

BIOS 843 Seminar
Wenyi Wang, PhD, Associate Professor
The University of Texas MD Anderson Cancer Center, Division of Quantitative Sciences, Department of Bioinformatics and Computational Biology

Saturday, May 13

Gillings School of Global Public Health commencement

Tuesday, May 23

2017 Atlantic Causal Inference Conference

Wednesday, May 24

2017 Atlantic Causal Inference Conference

Thursday, May 25

2017 Atlantic Causal Inference Conference

For more Gillings School events, view the school’s event calendar.

Have news or story ideas for BiosBeat? Submit them to Britt Clampitt.