Section Newsletter
Spring 2005

Message from the Chair

Marcia A. Testa, MPH, PhD
The Statistics Section ends spring 2005 with a number of new plans and exciting announcements. This edition of the Newsletter looks back at past activites and ahead toward plans for the upcoming APHA 133rd Annual Meeting taking place this year in New Orleans. First and foremost, I want to thank all the Section Officers and members who have contributed their time, effort and continuing dedication and service to the Section. I would like to extend a well-deserved note of appreciation to Larry Moulton, our current Chair-Elect and 2005 Statistics Section Program Chair, who has managed to organize a great 2005 Program, to Past-Chair Andy White for getting together a great ballot and for offering his service on a number of very important APHA committees, to Stuart Gansky for his role as Secretary, and Brenda as Secretary-Elect, to Michael Stoto, Deborah Ingram, Craig Turnbull, Charity Moore, Peter Imrey, Diane Makuc, Xihong Lin, Martin Weinrich, Brenda Gillespie, Daniel Freeman, Jack Barnette, Janet Eyster, Elizabeth Zell, Mark van der Laan (just to name a few), and all those who have served as Committee Chairs, Program Organizers and active members. I offer a round of applause to all.

Reflections on the Practice of Quantitative Methods and Statistics in Public Health

For the past two years, I have become much more actively involved in applying statistical methods to the practice of public health. My activities as Head of the Evaluation Core for the Harvard Center for Public Health Preparedness, along with fellow Statistics Section members, Michael Stoto (Chair, 2003) and Elena Savoia (Student Liaison, 2005) have brought me closer to answering the question of why statistics and quantitative methods are such an integral part of ensuring the health security of our communities. In addition to working with local, state and federal agencies on evaluating and measuring public health preparedness, I recently became a member of my town's Board of Health -- which is about as close as you can get to public health "primary care" practice. These experiences have greatly reinforced the previous "Message from the Chair" columns written by Andy White and Michael Stoto -- that quantitative methods and training are an absolute necessity at all levels of public health practice and research. On that note, I wanted to take a bit of a different approach to my "Message from the Chair" column in this Newsletter by stressing the importance of the practice of statistics in public health by relaying a series of personal reflections on people, situations and experiences relating to applying our skills in public health practice to relieve human suffering. I hope you find these stories and scenarios of interest.

Remembering Professor I. Richard Savage -- Statistics Practitioner of Public Policy
More than 25 years ago, my statistics professor at Yale, Professor I. Richard Savage, informed his non-parametric statistics class that “it doesn’t make any difference how good your statistical analysis is, or whether you use ranks or not -- if your raw data isn’t any good, your analysis will be worthless.” Basically, I believe he was pointing out that applied statisticians have to be deeply knowledgeable and involved with their data in order for their analysis efforts to bear fruit. Born in 1925, Savage was an emeritus professor and former Chair of the Department of Statistics at Yale University and a world-renowned expert in the use of statistics in public affairs when he died last year on June 4, 2004. The memory of his words comes back to me every time I design a survey or attempt to interpret the results of a statistical analysis. Professor Savage was one of a few mathematical statisticians of his generation who chose to pursue the application of statistical principles and concepts to problems of public policy. He worked in AIDS diffusion, DNA fingerprinting, human rights and national defense. Early on in his career, he was asked to participate in the first National Research Council study of undercount in the U.S. census. In 1995, recalling that experience with fellow department chairs Allan Sampson and Bruce Spencer he recollected:

Savage: That was my first introduction to Washington, and my first work in public policy. I found that panel pretty interesting. It struck me almost from the beginning that nobody was looking at whether it was important whether there was an undercount or not. In the panel report, there’s no discussion of the implications of the undercount. It’s all concerned with the origins of the undercount and mainly how to get rid of it.
Spencer: So you got interested in how the statistics get used?
Savage: Right after that, I became involved with other National Research Council panels.”(1)

Professor Savage knew it was not only important, but imperative, to question how “statistics” are used in public policy. His work entitled "Ability Testing of Handicapped People: Dilemma for Government, Science and the Public" (1992) brought to the forefront of public policy issues critical to the disability community. Professor Savage recalled in 1995,

“Until 1990, the ability testing groups —- the commercial groups that do it —- had stonewalled to prevent any modifications that would tend to improve the situation for the disabled. A negative thing they do is that they allow disabled people to take tests under nonstandard conditions. But then the test scores are flagged when they are sent to institutions, where the flag says that these scores are not reliable and should not be compared with any other scores.”(1)

Professor Savage knew a great deal about the importance of measurement when interpreting statistical results concerning disability data. Afflicted with severe myopia since childhood and limited mobility from polio contracted at the age of 20 in 1945, his work on disability performance measurement pointed out not only how statistics were being used incorrectly, but also the importance of validity and reliability of the measures used in policy decisions. I reflect on his contributions every time I question whether devoting so much time to applied statistics is a smart career choice for my students. And with his passing, I realize that life is too short not to try to make an impact on that which can directly alleviate human suffering.

1. Sampson AR and Spencer B. Conversations with I. Richard Savage, Statistical Science 1999. 14:126 - 148

A Tribute to David S. Salsburg -- Creating Magic Bullets to Cure Disease
I recently attended a special symposium and reception in honor of the life-long accomplishments of another great mathematical statistician who continues to dedicate his life to applied statistics for solving problems of human suffering, David S. Salsburg. In the preface of his 1992 book entitled “The Use of Restricted Significance Tests in Clinical Trials,” he captured the dilemma of all scientists when making career choices. He wrote

“Throughout human history, there has been a clash between the view that knowledge is a superior good by itself and the view that knowledge should be used to make life better for mankind.” (2)

He attributes his final decision to dedicate his career to applied statistics to his wife, Fran, who “made me appreciate the efforts of Paul Ehrlich, who wanted to take the abstract knowledge of selective dyes and create “magic bullets” that would cure diseases, or of Benjamin Franklin, whose scientific investigations led to efficient stoves and protection against lightning”, and who made it possible “for me to face the practical consequences that result when we apply abstract mathematical ideas to the problems of human suffering.”(2) In his more recent book entitled “The Lady Tasting Tea: How Statistics Revolutionized Science in the Twentieth Century” Howard Holt and Company, New York 2001, Salsburg points out how the field of statistics can be applied to real-world problems and how they can impact science and public policy. He uses the famous smoking/lung cancer debate to describe how statistical concepts were used to help formally link smoking to lung cancer. The paperback version costs $11 -- a good investment for educating the non-statistician. If you want to know who the "Lady" was and what statistics has to do with "tea tasting," here are a few hints -- a summer afternoon in Cambridge, England, in the late 1920s and a famous statistician with a Vandyke beard. For the answer to this riddle, you have to read the book. When I reflect upon how we as statisticians can help solve problems of human suffering, I think of David Salsburg and how he is an excellent role model for our students as they try to make their career choices.

2. Salsburg DS. The Use of Restricted Significance Tests in Clinical Trials. Springer-Verlag, New York, 1992, pp. v - vi.

The Need for Greater Statistical Competency in the Field of Measurement and Communication when Guiding Public Health Policy
As 2004 ended, many news stories and events were reported concerning the health of individuals, the nation and the world. Individual health care consumers learned through an advocacy group reporting on the results of a federally sponsored survey that "two-thirds of FDA scientists lacked confidence that the FDA adequately monitors the safety of prescription drugs." State residents were warned by a non-profit organization, Trust for America, that "despite a surge in federal aid for state public health efforts, the effort to comprehensively fix the nation’s public health system is falling short with only six states achieving green status for administering and distributing vaccines and antidotes in the event of an emergency." And when a Sumatra earthquake touched off a series of tsunamis, killing more than 150,000 people and putting the health of millions, particularly 1.5 million children, in serious jeopardy, the world learned that even the most basic components of public health -- clean water, adequate nutrition and secure shelter -- should never be taken for granted.

Whether the problem is drug safety, public health preparedness or providing access to safe food and sanitation, correctly measuring where we stand and how far we must go to achieve a remedy is essential. Interpreting what the data are telling us involves understanding not only how the data were collected and analyzed, but how the measures were derived. In fact, measurement is the very first step in formulating decisions and taking corrective action.

For example, exactly what does it mean “to lack confidence that the FDA adequately monitors the safety of prescription drugs”? There are at least two unobservable constructs in this simple statement; namely “confidence” and “monitoring.” How was “confidence” measured and assessed? What are the multiple processes involved in “monitoring”? How low a score do you need to “lack” confidence? Many news media and advocacy-based reports are devoid of the basic tenets of statistical cause and effect. Rare serious adverse events will always only show up after millions of prescriptions are written -- long after drug approval, which is based typically on less than 6,000 subjects. The issue involves not only an understanding of statistical uncertainty, but also a general consensus on the degree of trade-off between risk and benefit given the information provided by clinical trial subjects. And yet, news media and advocacy groups sometimes push policy makers to believe that decisions could and should have been made earlier, even prior to the collection of data which would scientifically support their position. However, if one believes that the role of news, media and advocacy groups is to use data to incite action and not guide action, then perhaps they are doing their job when they ignore the science (or lack of science) behind the numbers. But as statistical scientists, are we doing our job as effectively as we could?

The role of the statistical scientist is to use data to help guide public health action. However, there are those who use data to incite action. For example, the non-profit group, Trust in America, recently ranked states in the United States using a 10-point public health preparedness score by summing 10 binomial indicators and then ranking the states by those sums. After reviewing the data and results in their highly publicized report, I had to conclude that their purpose was to incite, not to guide action. I came to this conclusion for a number of reasons. For example, South Carolina was penalized 1 point (binomial outcome of "increase" versus "no increase") for not increasing the rate of flu vaccinations between 2002 and 2003 (S.C. went from 69.4 percent to 69.3 percent), but its neighbor North Carolina received 1 point (it went from 68.2 percent to 68.8 percent). I'm sure that caused a bit of tension between public health preparedness workers in those two states. Did anyone constructing these scores and performing this analysis ever consider measurement error here, or whether the difference in rates of change between the two states was clinically meaningful, or whether baseline values should somehow be incorporated into the score to adjust for a ceiling effect? In 2003, South Carolina actually had a higher rate as compared to North Carolina, but received no credit for that. On another indicator, states were also penalized if they did not increase spending for public health between 2002 and 2003. It made no difference whether public health spending was initially 10 times higher in one state versus another in 2002, or by how much spending was increased (a 1 percent decrease was given no point, the same as a 30 percent decrease).

Also in the report, the aging of the public health workforce's age was cited as "dire" because the state public health workforce was 46.6 years, while the national average age of the workforce was 40.0 years. I’m still not exactly sure what this means. Should I suddenly be alarmed that the Senate is in "dire" condition because the average age of Senators is "well over" 47 years. I wonder what would be said about the average age of the Officers of the Statistics Section for that matter. What should the average age of state public health workers be? The differences among states were small to begin with, but clever gaming of the indicators, and conversions to a competitive (albeit invalid) state ranking, were used successfully to capture the attention of the press and the public media.

However, public health statisticians are not usually as vocal, as well organized or as aggressive as advocacy groups in garnishing public attention. And what happens if in the end all the public sees are the flawed numbers? What happens when it comes time to guide public policy action and all we have are flawed state rankings, with money and resources being allocated on the basis of random measurement error? I reflect on the need to strengthen our role as advocates for truth in numbers when I speak to our statistics students about focusing career plans toward public health policy and practice.

Teaching Measurement, Evaluation, Reporting and Communications as Part of Training in Public Health Statistics
Reflecting on the issues raised above, I think that while public health statisticians are typically well trained in methods for analyzing “data,” they are much less well trained in knowing exactly what their data actually measure, or how to communicate their results successfuly to the stakeholders. Public health data take on many forms, including individual indicators of health, such as clinical and laboratory measures, behavioral, health status, and quality of life as well as aggregate measures of performance and outcomes such as compliance with guidelines and rates of morbidity and mortality. And yet, biostatistics curricula seem to singularly omit the discipline of measurement which tells us about the strengths, weaknesses and limitations of the construction of public health measures, scales and indicators. Similiary, communicating our results to the important stakeholders is the probably the most important stage of public health practice analysis.

We must teach our students that by better understanding our measures, we can aid the policy makers and medical and social scientists when interpreting statistical results in order to enact the appropriate policies and programs to bring about a remedy. Measurement is closely linked to the process of determining improvement through the process of evaluation. Considering that public health deals with populations, and that interventions to improve public health occurs through programs, and evaluation is the analytical tool to determine the effectivenes of programs, where in our statistics curriculum do we teach our students about statistical and analytical methodology for evaluation of public health programs? And where to we teach our students to become good communicators of their results and analysis? Looking through course catalogues of Schools of Public Health, I came across the following definition of Biostatistics at my former alma mater, Yale University, "Biostatistics provides the essential methodological underpinning necessary for much of Public Health practice. Appropriate study design, data collection and analysis are all needed in order to be able to quantify the effect of risk factors and health intervention on individuals or populations." <> However, looking through the publications of faculty in the biostatistics departments of many Schools of Public Health, I find very few papers that deal specifically with public health research and practice .
As members of the Statistical Section of the American Public Health Association, we need to help shape public policy. We should take time to reflect upon the need to learn a lesson from our advocacy colleagues and be vocal, self-assured, and confident in our ability to use science to guide policy decisions.

Evidence Based Policy and Practice -- A Quest for Statistical Justice
You’ll see by visiting the APHA Annual Meeting Web site that the theme of the 2005 Annual Meeting is “Evidence Based Policy and Practice.” According to APHA,

“Evidence-based policy and practice are processes of systematically finding, appraising and using contemporaneous research findings as the basis for decisions. It is the best knowledge based on rigorous, comprehensive syntheses and analyses of the scientific literature on topics relevant to (clinical, social, science/behavioral, economic and other) health care organizations. The reason for evidenced based decision making is to improve performance, health outcomes and make more efficient use of resources. Another crucial role of evidence-based decision making is to translate science to policy makers and assist them to evaluate the merits of competing demands for limited resources. Evidence-based science surrounding public health leads to conclusions based on proven facts.”

This year’s Annual Meeting theme “Evidence Based Policy and Practice” offers the Statistics Section an opportunity to be more vocal. We have an unique opportunity this year to use what we know to help solve the problems of human suffering through changes in how we use quantitative methods to impact public health policy and practice. This year, I ask each member to join us in advocating against statistical injustices -- gaming of measures, distortions of data, ignoring statistical inference and uncertainty. Often the advocacy and scientific groups seem to be in conflict, but we certainly can learn from each other. Let us reflect upon ways that we can use advocacy to fight for statistical justice, so that we will be better able to promote evidence-based decisions to solve problems of human suffering.

I wish all of our Section members a great year, and I look forward to your continued active participation.

NEWS FLASH - Jay Glasser, PhD, to Give Lowell Reed Lecture in New Orleans

Jay Glasser, PhD, 2005 Lowell Reed Lecturer and Past (2003) President of APHA
The Statistics Section is extremely proud to have one of its members and former Statistics Section Chair, APHA Board Member, APHA Treasurer and 2003 APHA President, Jay H. Glasser, MSc, PhD, give the 2005 Lowell Reed Lecture at the APHA Annual Meetings in New Orleans this coming November 2005. Glasser’s message throughout his service in APHA is one that emphasizes connecting people to public health, strengthening the membership of the association, forging ahead with new strategic alliances and engaging the next generation of public health professionals.

Glasser received his BA from the University of Connecticut in Zoology in 1957, his MSc from Columbia University in Biostatistics in 1960 and his PhD from North Carolina State University in Experimental Statistics in 1967. He has been on the faculty of the University of Texas Health Science Center at Houston, School of Public Health since 1969 and a full professor of biometry since 1989. He is also co-director of the International Program on Health Technology Assessment at the University of Texas Health Science Center, adjunct professor in the Department of Ophthalmology at the University of Texas Health Science Center, and adjunct professor in the Department of Senior Citizens Health at Hokkaido University in Japan.

NEWS FLASH – Rebecca Betensky, PhD, is 2005 Spiegelman Winner

Rebecca Betensky, PhD

Spiegelman Award Committee 2005 Chair Xihong Lin, PhD, announced that Rebecca Betensky, PhD, Associate Professor of Biostatistics, Harvard School of Public Health, will receive the Spiegelman Award during the Annual APHA meeting in New Orleans this November.

The Statistics Section will honor Rebecca Betensky, PhD, Associate Professor of Biostatistics, Harvard School of Public Health, during the Statistics Section Award Ceremony, which will take place TUESDAY, NOV. 8, 2005: 2:30 p.m.-4:00 p.m. Xihong Lin (2002 Winner and Award Committee Chair for 2005) was assisted by Marie Davidian, Craig Turnbull (APHA rep), Michael Newton (2003 Winner and Award Committee Chair for 2006), Ron Brookmeyer, Dan Freeman (APHA rep), Mark van der Laan (2004 Winner and Award Committee Chair for 2007), Louise Ryan and Michael Stoto (APHA rep), who all served on the 2005 Nominations and Selection Committee.

Rebecca Betensky received her PhD in 1992 from Stanford University. Her current methodological research interests are in the areas of failure time data and correlated multivariate binary data. In the area of failure time data, she has developed computationally simple methods for proportional odds and accelerated failure time regression with interval censored data. These methods view the interval censored data as a series of correlated current status observations. She has recently developed a test for whether the required independence between failure time and visit compliance holds. In previous work, she developed local likelihood based methods for smoothing the distribution function based on interval censored data and for proportional hazards regression with interval censored data. In the area of correlated multivariate binary data, she has developed models for estimation of the aggregation and co-aggregation of diseases within families, under non-random ascertainment. She is currently extending these models to allow for more complex ascertainment and to handle ordinal and censored outcomes.

Congratulations, Rebecca!

Come to the APHA 133rd Annual Meeting

This year's theme of Evidence-Based Policy and Practice focuses directly on the Core Mission of the Statistics Section since evidence requires data and interpretation of data requires individuals skilled in quantitative methods. The APHA Annual Meeting is the premier platform to share successes and failures, discover exceptional best practices and learn from expert colleagues and the latest research in the field. At APHA in New Orleans, you will learn how to stay on top of the trends in public health. Scientific sessions, networking opportunities and events, and the largest public health exposition will equip you with the tools needed to succeed. So make sure to come explore a world of ideas and innovation with more than 13,000 peers and leaders in public health. APHA has a world of public health in store for you. Simply click on <> and register today.

Late Breaker Abstracts in Epidemiology -- July 15, 2005 Deadline

The Epidemiology Section invites the submission of late breaker abstracts. Studies completed after the general February abstract deadline and abstracts that are specific to the theme of the conference will receive high priority.

Only abstracts including data analyzed will be considered. The deadline for late breakers is July 15, 2005. All abstracts should be submitted via <>. Please follow the directions on the Web site. Decisions will be sent to abstract submitters in mid-August.

The Epidemiology Section will consider very late abstracts on Oct. 1, 2005. Only abstracts with data collected after June 2005 will be considered. Abstracts submitted for this very late abstract deadline should be submitted directly to Dr. Louise-Anne McNutt at <>.

Abstracts must be 250 words or less with the standard background, methods, results, and discussion sections. Decisions will be sent to submitters by Oct. 14, 2005.

SAVE THE DATE AND TIME -- 2005 Statistics Section Business Meeting in New Orleans


Statistics Section Business Meeting and Social Hour

Tuesday, Nov. 8, 2005: 6:30 p.m.-8:00 p.m.
This very important session is open to all APHA members interested in the activities of the Statistics Section. New members are welcome. The meeting will begin with a social hour so that members can network and learn about the activities of their colleagues and associates. This is an excellent opportunity to discover new contacts and make new friends. The social hour is followed by the annual Statistics Section Business Meeting during which Council representatives and officers will brief members about the Section's prior year's activities. Plans for the upcoming year will also be discussed. New officers will be inducted at this meeting.

After the Business Meeting, the traditional Statistics Section Dinner, organized by Secretary Elect Brenda Gillespie, ( will take place.

To see some pictures from last year's Statistics Section Dinner organized by Secretary Stuart Gansky, click on the URL at the end of this article. Please save the date and plan to come!

Related Files:

2004 Lowell Reed Lecture: Biostatistics, the FDA and Public Health

Robert O’Neil, FDA, gives the Lowell Reed Lecture in New Orleans, 2004 and receives a plaque from Marcia A. Testa, 2004 Session Organizer and Statistics Section 2004 Chair-Elect.
Robert T. O'Neill was the 2004 APHA Statistics Section Lowell Reed Lecturer. Each year the Statistics Section selects an outstanding statistician who has contributed to the field of statistics and public health through significant contributions in research, teaching and service. The award is given in memory of Lowell Reed, who is most noted for his discovery of the ED-50, a tool for toxicology, and for his work on the Reed Frost epidemic model. The awardee is selected to give the Keynote Address during the Lowell Reed Lecture, Spiegelman and Statistics Section Awards Session. In November of 2004 the Statistics Section was proud to honor O'Neill, who is the Director of the Office of Biostatistics within the U.S. Food and Drug Administration's Center for Drug Evaluation and Research.

O'Neill exemplifies the professional statistician who combines excellence in the areas of research, teaching and service. He contributes enormously to the public health of the United States through his tireless dedication to public service, his wise and insightful publications, and his commitment to educating statisticians involved in therapeutic drug research and clinical trials. His presentation provided an overview documenting the evolution of the current regulations and statistical involvement. He also highlighted the current methodological challenges facing the field of biostatistics within drug development. His talk emphasized the international role that biostatistics, regulation and public health has assumed in the new global economy and demonstrated the important link between regulation, biostatistics and public health.

Statistics Section Partners Again with ASA for Continuing Education Institute 2005

Ralph Turner, PhD,MPH, Psychologist, Outcomes Researcher and CEI Course Director for November 2005 APHA Annual Meeting
This year the Statistics Section in partnership with the American Statistical Association and the Epidemiology Section will host two of the 22 Continuing Education Institutes at the 2005 APHA Annual meeting. The number of contributed CEI's was reduced this year, and course proposal submissions were very competitive. (Go to <> for online CEI information).

Ralph Turner, Marcia Testa and Linda Marc (all APHA Statistics Section members) will give a six hour CEU course entitled Psychometric, Measurement and Analysis Issues for Health Outcomes Survey Research (CEI 1010.0) on Saturday, Nov. 5, 2005: 9:00 a.m.-5:00 p.m. The target audience is public health statisticians, epidemiologists, evaluation researchers, social workers and mental health workers. The purpose of this Institute to promote the effective use of health survey outcomes research, which is being used increasingly in public health prevention, screening and health education programs. More and more public health databases and surveys are using questionnaires to report the health of various populations for the purpose of developing and evaluating public health programs and for monitoring the health of populations. This course is designed to give public health researchers and practitioners a foundation in measurement and psychometrics for the purpose of collecting, analyzing and interpreting data obtained from questionnaires. The learning objectives are:

  • Conceptually define the meaning and purpose of health outcomes survey research.
  • Recognize the role of psychometrics in obtaining health outcomes survey research.
  • Evaluate the usefulness of health outcomes survey measures for your organization.
  • Recognize the different types of measures used in outcomes research, including clinical, mental health, health status, quality-of-life, work/role and health care utilization, and consumer satisfaction.
  • Adopt new methods for modeling survey responses.
  • Interpret the meaning of statistical measurement concepts, such as reliability, validity, responsiveness, sensitivity and power.
  • Obtain a basic appreciation of the statistical analysis appropriate for health outcomes survey research

More information on this course can be found at <>.

A second three-hour CEU course, An Introduction to Item Response Theory (IRT) for Public Health Professionals (CEI 2014), will be given by Adam C. Carle (also a Statistics Section member) on Sunday, Nov. 6, 2005: 8:00 a.m.-11:30 a.m. The purpose of this session is to introduce participants to the theory and application of item response theory (IRT) models and provide experience interpreting the outcomes of analyses and studies employing these models. The course will briefly cover basic psychometric concepts, describing the importance of establishing reliable and valid measurement in public health research, and then focus chiefly on IRT measurement models. Material will cover the Rasch, 2PL, and 3PL models and their associated measurement parameters. Polytomous models will also be similarly addressed. The course will provide examples of the models in real data and describe their interpretation. The program will address the settings and manner in which IRT models may be employed to increase the validity and reliability of public health research, and, finally, it will also describe the use of IRT models in a variety of settings, e.g. computer adaptive testing, measurement bias studies, and test equating studies. The learning objectives are:

  • Discuss the importance of establishing and employing valid measurement tools in epidemiological and public health research
  • Understand basic psychometric concepts;
  • Recognize and understand fundamental item response theory models (IRT) and related concepts; and
  • Be an effective and knowledgeable consumer of research employing IRT models.

More information on this course can be found at <>.

Please forward these URLS to your associates who might have staff or students interested in attending.

Mark van der Laan, PhD, is Recipient of 2004 Spiegelman Award

Andrew White, 2004 Statistics Section Chair, (center) and Michael Newton (right), 2003 Spiegelman Award Winner, congratulates Mark van der Laan (left)
At the November, 2004 APHA Annual Meeting in Washington, D.C., the Statistics Section was pleased to honor Mark van der Laan, PhD, Professor of Biostatistics and Statistics, UC Berkeley, as the 2004 Spiegelman Award Winner. Since 1970, the Statistics Section of APHA has presented the Mortimer Spiegelman Award to an outstanding public health statistician under age 40. The award serves three purposes: to honor the outstanding achievenments of both the recipient and Mortimer Spiegelman; to encourage further involvement in public health of the finest young statisticians; and to increase awareness of APHA and the Statistics Section in the academic statistical community. Mark’s research interests include statistical methods in genomics, survival analysis, censored data, semiparametric models and causal inference, data adaptive loss based estimation, and multiple testing.

Mark went to UC Berkeley from the Netherlands' University of Utrecht, where he studied mathematics (1985-1990) and obtained his PhD. (1993). He completed his thesis, "Efficient and inefficient estimation in semiparametric models," under the guidance of Prof. Richard D. Gill. In 1994, Mark was a Neyman Visiting Professor in the Statistics Department at UC Berkeley. After accepting a tenure track position in the Division of Biostatistics, School of Public Health at Berkeley, he became an Associate Professor in Biostatistics in July 1997 and was promoted to full Professor in Biostatistics and Statistics in 2000.

Having taught several introductory courses in Biostatistics for Public Health students, he currently teaches classes on censored data, survival analysis, causal inference, data adaptive loss-based estimation, and multiple testing. He has been Associate Editor for Biometrics (1997-2003) and Lifetime Data Models (1996-2000), and Vice-President of the Bay-Area Chapter of the American Statistical Association. He is currently Associate Editor for Journal of Statistical Planning and Inference, Statistical Applications in Genetics and Molecular Biology, and the Annals of Statistics, and is Director of the Biostatistics and Computing core of the Superfund Research Program of the Center on Genomics in Environmental Science in the School of Public Health, headed by Professor Martyn Smith.

In addition to these numerous activities, Mark is also founding editor of a new electronic International Journal in Biostatistics, which is led by Prof. Dr. Nick Jewell. Several grants have been awarded to Mark, including an NIH FIRST Award research grant for the period 1996-2001 to work on "Locally Efficient Estimation in Censored Data Models"; an NIAID grant for 1999-2002 to develop a unified methodology for censored data and causal inference; a three-year grant he received in March 2001 from the Life Science Informatics Institute and its industrial partner, biotech company Chiron, to create statistical methods for data structures involving microarray data on complete (human) genomes; an NIH grant "Statistical Analysis of Longitudinal Studies with Gene Expression Data" (2002-2006); and an NIH grant "Data Adaptive Estimation in Epidemiology and Genomics" (2004-2007).

Mark and his wife, Martine (also Dutch), have a daughter, Laura, and two sons, Lars and Robin. His hobbies are tennis, chess, running marathons, and chasing grizzly bears, among other outdoor activities.

Congratulations to Mark and his family for his great accomplishments -- from all the members of the APHA Statistics Section.

NEWS FLASH --2005 Annual Statistics Methodology Session Talks in New Orleans are Announced

As the 2004 award winner, Mark van der Laan, PhD, 2004 Spiegelman Award Recipient, has organized the Annual Statistical Methodology Session in New Orleans entitled Marginal Structural Models for Causal Inference to be given Tuesday, Nov. 8, 2005: 4:30 p.m.-6:00 p.m., with talks on Population Intervention Models in Causal Inference by Alan Hubbard, PhD, History Adjusted Marginal Structural Models: Applications in AIDS Research by Maya Petersen and Honest Confidence Intervals for Causal Effects in the Presence of Many Potential Confounders by James M. Robins, MD. Mark will continue to serve on the Spiegelman Award Selection Committee for the next three years.

Go to <> to learn more about the Statistical Methodology Session.

Message from Our Student Liason: The Ideal Situation

Elena Savoia, MD, MPH
The “ideal situation for a student” is when you develop an interest for a specific topic, and you decide to dedicate yourself to strengthening the skills that will permit you to better understand and explore the different aspects of that topic. Living in an academic environment enables you to strengthen and to keep that special interest alive. Finally, you must meet a professor who is able to stimulate your intellect, curiosity, and offer you a concrete opportunity to transform your desired initial interest into a career.

I consider myself extremely lucky because the “ideal situation” happened to me. It happened in the fall of 2003 when I enrolled for the degree of Master of Public Health at Harvard University. I was a medical doctor coming from Italy where I had just completed my residency in preventive medicine. Having a strong background in medicine, some knowledge of epidemiology and biostatistics along with an intense interest in infectious diseases and evaluative sciences, I started the MPH program focusing in the area of quantitative methods. The first semester was dedicated to acquiring technical competence in study design and data analysis.

My experience at Harvard University became more interesting during the second part of the program. It all began in the winter of 2003 during an orientation session with Professor Marcia Testa, core leader of my concentration. She offered the students the possibility to accomplish the research practicum requirements by conducting a research project in the field of emergency public health preparedness. I and four other dedicated students took on the task with great initiative and enthusiasm. This was an exciting opportunity because it was going to require a great deal of creativity and teamwork. I knew it was going to be a challenge because it was all new to me; nevertheless, I was ready to embrace whatever was necessary to finish the project.

For five months the group analyzed the data provided by the “Massachusetts Department of Public Health Emergency Preparedness Survey,” evaluating the preparedness of cities and towns across the state. The data were collected with the purpose of measuring the level of emergency preparedness of the different types of agencies (city/town administration, local public health agency, law enforcement, fire department, emergency medical services) in the municipalities throughout the counties of Massachusetts. We were asked to develop a method to quantitatively assess the different aspects of public health preparedness. Very soon we realized how challenging this project was going to be. The reason is that quantifying preparedness for terrorist attacks, by any measure, is extremely difficult and in a certain way elusive because by definition the assessment must anticipate and prioritize the events and circumstances for which preparedness is desired.

Preparedness can be fairly assessed only for a specified event of a given magnitude, and it should be based on objective measures of response and performance. However, from the beginning we followed a rigorous but also creative method and began the project by identifying the main focus areas of emergency preparedness. Only after having conceptualized the construct through a detailed literature review were we able to approach our data and create scales to describe the level of preparedness of the agencies across the state.

We were able to demonstrate that preparedness efforts vary across towns and counties, supported by our scoring system that seemed to take this variation into account. Much more research needs to be conducted to establish the critical link between performance measurement and performance evaluation. Benchmarks of high levels and low levels of emergency preparedness must relate to outcome assessment, so that financial resources can effectively target those interventions and programs that are most cost-effective.

In November, 2004 I had the honor to present our preliminary results at the Statistics Section of the 132nd APHA Annual Meeting in Washington, D.C. The event was of great magnitude, attracting about 14,000 public health professionals from all over the country. The seminar gave me the opportunity to meet with others in public health conducting similar studies, appreciate their comments and feedback while establishing a vast network with people of similar interests. I have been fortunate to have found all the necessary elements to be in the ideal situation as a student. My academic experiences have proven to be very rewarding to my professional development. I currently consult professionally with local and state departments of health as both an independent consultant and a research associate in the Division of Public Health Practice, Harvard School of Public Health.

The importance of attending a meeting like the one offered by APHA every year goes far beyond the presentation of the study results that can be easily achieved with a publication in a scientific magazine. The real benefit is in meeting people with different backgrounds and similar interests, people who are authors of unpublished papers and went through the same process of problem solving and study design hopefully coming up with very different opinions to be discussed. I truly enjoyed the experience of conducting the research and presenting my findings to a unique group of individuals who share common interests and values. I encourage all students to become involved in applying quantitative methods to public health problems and joining our Statistics Section.

The 100th Anniversary Campaign Planning Committee Kicks Off

Elizabeth Zell staffs the Statistics Section Booth at the 2004 Annual APHA meetings -- only four more years to 100


This year kicks off the campaign toward the Statistics Section's 100th Anniversary. As we approach this very important landmark event, we continue to look for new members to ensure future longevity.

In addition, we also announce the beginning of our 100th anniversary campaign to support students and new members. Larry Moulton (Chair-Elect) and Brenda Gillepsie (Secretary- Elect) look forward to receiving your input regarding potential campaign contributors and pledges in celebration of our 100th Birthday.

Remember to tell your friends who have not selected a Section to join the Statistics Section.

2004 Annual Statistics Section Awards

(L to R) Martin Weinrich, Gooloo Wunderlich, Charles Rothwell, Janet Eyster and Andrew White
Each year the Statistics Section presents three awards to Section members who have made outstanding contributions to statistics and public health. Each award winner is chosen from a different membership population representing affiliations with academia, government and industry/non-governmental organizations. Section members had the special privilege of honoring three new award winners during the November 2004 Annual Meeting in Washington, D.C. This past year the Awards Committee was chaired by Martin Weinrich and winners were as follows:

Academic: Janet Eyster PhD, Department of Epidemiology, Michigan State University
Eyster was recognized for her dedication to the Statistics Section, particularly for her exemplary assistance in placing the Spiegelman Award on a sound footing, for her philanthropic work, and for her contributions to hierarchical statistical modeling.

Government: Charles Rothwell, Associate Director for Data Processing and Services, NCHS
Rothwell was recognized for distinguished service at the state, national, and international levels, and for designing health information and planning systems, providing consultation expertise to Eastern Europe, building partnerships between universities and the National Science Foundations, and working to advance bipartisan health care reforms.

Industry: Gooloo Wunderlich, PhD, Senior Program Officer, Institutes of Medicine of the National Academies
Wunderlich was recognized for her contributions to the success of the Statistics Section, as well as her leadership at the Institute of Medicine of the National Academies on studies of the crises in rural health care, hunger and Social Security Disability.

Interested in the Awards Committee

If you have an interest in recommending or nominating future Award Winners in any of the following areas: 1) academia, 2) government and 3) industry/non-governmental organizations, please contact Awards Chair Martin C. Weinrich, PhD, Professor of Nursing & Biostatistics, Department of Physiological and Technological Nursing, Medical College of Georgia, Augusta, Ga., <>.

Special 2004 Statistics Awards -- Gary Koch and Michael Stoto

Gary Koch speaks to the Statistics Section after receiving his Special Lifetime Contributions Award from the Statistics Section during the November, 2004 Annual Meeting

Gary Koch was honored at the 2004 APHA meeting in November for his Lifetime Contributions to the Statistics Section. This award was also a tribute to Koch's generosity in supporting the Spiegelman Fund endowment with a significant contribution that helps to gaurantee the endowment's longevity.

Gary Koch, PhD, is Professor of Biostatistics, School of Public Health (1976-present) and Director, Biometric Consulting Laboratory (1987-present) at University of North Carolina at Chapel Hill. Koch's research interest include
Categorical Data Analysis and Nonparametric Methods as applied to Cardiovascular Disease, Child Development and Violence Prevention. Koch's contriubtions to the field of Biostatistics research and practice is enormous, with 378 publications (at last count).

Michael Stoto, Associate Director, Public Health and Senior Statistical Scientist, Rand Corporation and Adjunct Professor, Harvard School of Public Health (Biostatistics) and George Washington University (Epidemiology and Biostatistics), was also honored at the same meeting with a special Chair's Award for his service and dedication to the Section.

Before going to RAND in 2001, Stoto was Professor and Chair of the Department of Epidemiology and Biostatistics at the George Washington University School of Public Health and Health Services. Stoto's interests include the use of statistical data and quantitative analysis in clinical and public health policy. He has published more than 100 articles on various topics in epidemiology, biostatistics, demography, community health assistant, risk analysis and management, and the evaluation of public health interventions. His work also addresses substantive issues in public health policy and practice. Since joining RAND in 2001, he has been helping to develop research efforts on bioterrorism, focusing on public health issues.

Stoto also worked for more than 10 years as a member of the research staff at the Institute of Medicine, where he was responsible for numerous projects in public health practice, as well as a variety of projects on specific public health issues. Before joining the IOM, Stoto was a faculty member at Harvard's John F. Kennedy School of Government.

Stoto has made great contributions to the field of statistics and biostatistics with his research focusing on the application of statistical and other quantitative methods in the analysis of public health policies and practice, especially under conditions of uncertainty. His research interests include meta-analysis and research synthesis, epidemiologic surveillance research on statistical and demographic projections and estimates, measurement of community health, and the measurement and communication of uncertainty in statistical estimates. At the RAND Corporation, Stoto has led the analysis of the statistical and practical issues associated with early detection of bioterrorism events, and evaluation of regional and syndromic surveillance systems in the District of Columbia. He has also participated in a review of the adequacy of California’s public health system to protect and improve the health of local communities and case studies of SARS, West Nile virus, and monkeypox to assess public health preparedness in the United States.

For a sample of Stoto's publications go to <> and <>.

American Public Health Association Statistics Section

APHA Home Page

Remember to login to Members Only to Update your Contact Information

APHA Sections Web Page

Statistics Section Web Page


Chair: Marcia A. Testa, MPH, MPhil, PhD
Department of Biostatistics
Harvard School of Public Health
655 Huntington Ave.
Boston MA 02115
(617) 432-2818

Chair-Elect: Larry Moulton, PhD
Department of International Health
Johns Hopkins Bloomberg School of Public Health
615 N. Wolfe St.
Baltimore, MD 21205
(410) 955-6370

Immediate Past-Chair: Andrew A. White, MPH, PhD
Stat Tech, Inc.
7705 Brookville Road
Chevy Chase, MD 20815
(301) 379-7859

Secretary: Stuart Gansky, MS, DrPH
Center to Address Disparities in Children's Oral Health
Division of Oral Epidemiology & Dental Public Health
3333 California St, Suite 495
San Francisco CA 94143-1361
(couriers: 94118)
voice: (415) 502-8094
FAX: (415) 502-8447

Secretary-Elect: Brenda Gillespie, PhD
University of Michigan, CSCAR
3514 Rackham Building
Ann Arbor, MI 48109-1070
(734) 647-4609

Section Councilors:

Janet T. Eyster, PhD (2005)
Department of Epidemiology
Michigan State University
4660 S. Hagadorn Road, Ste.600
E. Lansing, MI 48823
(517) 353-8623, ext. 123

Martin C. Weinrich, MA, PhD (2005)
Professor of Nursing & Biostatistics
Department of Physiological
and Technological Nursing
Medical College of Georgia, Augusta, GA
(706) 721-6792

Deborah D. Ingram, PhD (2006)
National Center for Health Statistics
Centers for Disease Control and Prevention
3311 Toledo Rd.
Hyattsville, MD 20782
(301) 458-4733

Craig D. Turnbull, PhD (2006)
(919) 966-7259

Charity Moore, PhD, MSPH (2005-2007)
School of Public Health
University of North Carolina
725 Airport Dr
Chapel Hill , NC 27599
(919) 843-3403

Peter Imrey, PhD (2005-2007)
Department of Quantitative Health Sciences / Wb4
Cleveland Clinic Foundation
9500 Euclid Ave.
Cleveland OH 44195
(216) 444-0923

Governing Council Representatives:

Diane Makuc, DrPH (2004-2005)
National Center for Health Statistics
Centers for Disease Control and Prevention
3311 Toledo Rd.
Hyattsville, MD 20782
(301) 458-4360

Andrew White (2005-2006)

Note: Section Chair is ex-officio (without vote) member of the GC.

Intersectional Council:
Section Chair-Elect, Chair, and Past Chair are members.

Action Board Rep:
Charity Moore


APHA Annual Meeting Program Committee
Larry Moulton (Program Organizer)
Marcia Testa (Program Chair)

Spiegelman Award Committee:
Xihong Lin (Chair) (2003-2005)
Marie Davidian (2003-2005)
Craig Turnbull (APHA rep) (2005)
Michael Newton (2004-2006, Chair in 2006)
Ronald Brookmeyer (2004-2006)
Daniel Freeman, Jr. (APHA rep) (2004-2006)
Mark van der Laan (2005-2007, Chair in 2007)
Louise Ryan (2005-2007)
Michael Stoto (APHA rep) (2005-2007)

Section Awards Committee:
Martin Weinrich, Chair
Andrew White
Elizabeth Zell
Plaques: Deborah Ingram

Annual Meeting Booth Committee
Brenda Gillespie
Marcia Testa

Communications Committee
Stuart Gansky
Marcia Testa

Membership Chair and webmaster:
Larry Moulton

Brochure Committee:
Brenda Gillespie
Larry Moulton