Harvard School of Public Health
Telephone: (617) 432-1056
Department of Biostatistics
FAX: (617) 432-5619
Statistics plays a critical and important role in many areas of public health research and practice. For example, epidemiologic studies and monitoring of infectious diseases such as malaria, tuberculosis, Lyme disease, flu and AIDS draw heavily upon the efforts of biostatisticians. However, a new role for the profession emerged in the wake of September 11, 2001 that involves protecting the public from the threat of human health disasters arising from bioterrorist activities by using biostatistical methods to intensify disease surveillance.
Nearly four years after the attacks of 9/11, the general public's concern and anxiety seemed to be subsiding, however, the London bombings of July 7, 2005 was a grim reminder that being prepared for terrorist attacks requires a heightened level of alertness. As of mid July 2005, the death toll from the bombings in London stood at 55. The authorities had officially identified 41 of the dead. Over 700 were wounded by the 4 bomb blasts throughout London's mass transit system. The outcomes of terrorist attacks are often reported by stark mortality, morbidity and injury statistics. But one line of defense against a particular form of terrorism, bioterrorism, could be strengthened by new statistical techniques and applications being developed by members of the Department of Biostatistics, Harvard School of Public Health. This line of defense, referred to as syndromic surveillance, requires applying statistical methods to the process of disease surveillance.
Syndromic surveillance is a major area of research and practice for Dr., Michael Stoto, Adjunct Professor of Biostatistics, Harvard School of Public Health and Senior Statistical Scientist and Associate Director for Public Health, Center for Domestic and International Health Security, at the RAND Corporation in Arlington Virginia. Dr. Stoto explains that "unlike bombings, bioterrorism can be invisible, unapparent until people become ill, spreading silently as infected people interact with others." He states that "the flying bombs of September 11, 2001, the subsequent anthrax letters, and credible threats of future attacks" were primarily responsible for heightening the need for a bioterrorism early warning system. He recommends that the "sooner public health officials know about a bioterrorist attack, the more effective their response can be by accelerating the rates of quarantines, vaccinations, and treatments."
In his position at RAND, Dr. 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. His methods for syndromic surveillance and measuring public health preparedness extends beyond bioterrorism as well. Dr. Stoto 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. In addition, Dr. Stoto is part of the Evaluation Core of the CDC-funded Harvard Center for Public Health Preparedness (CPHP). The Evaluation Core, headed by Marcia A. Testa, Senior Lecturer on Biostatistics, is responsible along with other academic centers throughout the United States for developing methodologies for evaluating the level of public health preparedness of federal, state and local health agencies.
According to Dr. Stoto, "Since the terrorist attacks on September 11, 2001, many state and local health departments around the United States have started to develop syndromic surveillance systems." He stresses that syndromic surveillance is a relatively new concept in epidemiology involving the statistical analyses of data on individuals seeking care in emergency rooms or other health care settings with pre-identified sets of symptoms thought to be related to the precursors of diseases. He recommends that by "Making use of existing health care or other data, often already in electronic form, these systems are intended to give early warnings of bioterrorist attacks or other emerging health conditions." And also "by focusing on symptoms rather than confirmed diagnoses, syndromic surveillance aims to detect bioevents earlier than would be possible with traditional surveillance systems."
Because potential bioterrorist agents such as anthrax, plague, smallpox and viral hemorrhagic fevers initially exhibit symptoms of a flu-like illness, data suggesting a sudden increase of individuals with fever, headache, muscle pain, and malaise might be the first indication of a bioterrorist attack or natural disease outbreak. With concerns about potential outbreaks of West Nile Virus, Severe Acute Respiratory Syndromic (SARS) and avian flu virus migrating to humans, syndromic surveillance is also thought to be useful for early detection of natural disease outbreaks.
As part of his work in syndromic surveillance, Dr. Stoto has been studying how hospital admissions data can be used to spot abnormal trends in the patterns of diseases typically associated with the symptoms of potential bioterrorism agents. A example of a sample set of raw data used in such an analysis is shown in Figure 1. By applying smoothing techniques to these raw data it is possible to compare the numbers of emergency room visits in which gastrointestinal problems were listed as the chief complaint across time and location. Since day-to-day variation in the counts makes it difficult to identify any patterns or departures, syndromic surveillance methods generally employ one or more statistical detection algorithms to analyze data on a daily basis and raise an "alarm" when the count is significantly greater than expected, suggesting a possible outbreak or attack. Simulation studies, however, show that there is a relatively narrow window between what is obvious (e.g. 25 extra people coming into one emergency room on a Spring day with flu symptoms) and what is barely detectable with syndromic surveillance.
Another group of Harvard biostatistics researchers is also developing methods for syndromic surveillance. Professor Marcello Pagano, Al Ozonoff, Adjunct Assistant Professor and Assistant Professor at Boston University School of Public Health, Laura Forsberg, doctoral degree candidate in Biostatistics, and Caroline Jeffery, Visiting Scientist in Biostatistics, have been collaborating on research involving spatio-temporal analysis of syndromic data for biosurveillance. Dr. Pagano explains that the word surveillance is a combination of the French root 'sur' combined with the Latin 'vigilare', to yield the meaning: to watch over. He warns that "it is easier to terrorize the ignorant than the well-informed, so vigilance, and in particular active surveillance, is an important component of counterterrorism. "
The work of Pagano and colleagues shows how public health practitioners can apply particular statistical methods to the task of biosurveillance in order to improve a population's ability to withstand bioterrorism. They position the detection and reporting of outbreaks in the context of classical statistical hypothesis testing framework. In their conceptual model, the null hypothesis indicates normal levels of disease activity, and the alternative hypothesis, the presence of an outbreak. According to these researchers there are two types of possible errors involved in any early disease detection system -- not reporting an outbreak on the one hand, and raising a false alarm on the other. They note that the typical surveillance approach models the disease incidence under the null hypothesis, and then this model is used to perform a comparison between the daily prediction and what is actually observed. They warn that traditional regression approaches, such as those used by the Centers for Disease Control and Prevention for influenza surveillance, might be adequate for describing seasonal variation, however, such methods are not suitable for making future predictions because they do not take into account the auto-correlation of the individual observations. For example, they point out that by ignoring the pronounced spikes of influenza in the winter months, these methods are prone to sounding a bioterrorism alarm at the beginning of every flu season.
Syndromic surveillance systems are built to raise an alarm so that epidemiological investigation can commence. As with all detection systems there is a trade-off between sensitivity (signaling a threat when an attack or threat is real) and specificity (no signal when the there is no attack or threat). And while waiting to collect more data can increase the accuracy of the alarm system, waiting too long could render preventative measures obsolete. The solutions to these problems might be more readily solved with the application of statistical methods and analytical approaches to decision theory that can be applied to empirical data. Indeed, early detection methods are at the core of most public health disease prevention programs.
For the past thirty years, public health has been promoting disease prevention by screening individuals for cancer, diabetes, heart disease and a host of diseases and conditions so that they can seek early treatment. However, the development of instruments and tools necessary for monitoring entire populations, as opposed to screening individuals, is just now gaining momentum. The most obvious weakness of syndromic surveillance is the lack of timely and accurate population data. Because of the delay in reporting, reliance on hospital claims databases is inadequate, and even with the most sophisticated statistical methods, no surveillance system will work without access to timely and accurate databases. According to Dr. Stoto, "Beyond purely statistical issues, the value of syndromic surveillance depends on how well it is integrated into public health systems." He further emphasizes that "Syndromic surveillance is intended as an alert system, and to be effective it must be linked to the public health agencies that our nation relies on for protection against all disease outbreaks, whatever their origin."
Syndromic surveillance offers the potential for early warning of bioterrorist attacks and natural disease outbreaks, but the effectiveness of these systems will have to be weighed against the costs. However, it is apparent that efforts by leading statisticians for developing reliable surveillance databases in addition to accurate and efficient statistical methodologies moves syndromic surveillance closer to becoming a first line of defense against the harm that could result from attacks from both nature and man.
By Stephen W. Lagakos
This has been another busy year in the Department. Some of these activities are described in this issue of the newsletter, and many others are detailed on our web page (www.biostat.harvard.edu), which I encourage you to browse.
During the past year we have implemented a number of changes to our doctoral program resulting in part from our adoption of the PhD degree in 2003. Beginning this August, we are introducing a summer program for incoming students that will take place during the 3 weeks preceding the start of the fall semester. We will offer non-credit courses in probability, statistical methods, computing, and operational mathematics that are aimed at "homogenizing" the backgrounds of our diverse incoming students. The courses will be taught by Brian Healy (Operational Math and Computing), Jennifer Schumi (Methods), and Tyler VanderWeele (Probability), who are currently advanced doctoral students in our program. In conjunction with the introduction of the summer program, we have revamped our offerings in statistical methods and probability, and beginning this fall we will introduce integrated 2-semester sequences (Methods I and II and Probability I and II) in both. Another important change for first-year doctoral students is that, in accordance with GSAS guidelines, there will be no teaching (TA) or research assistantship (RA) responsibilities during the first year. This will allow our incoming students to 'sample' one or two courses in areas (such as statistical genetics, computational biology, and health decision analysis) that could not, in the past, have been squeezed into the first-year program. The breadth and support of our training programs has been enhanced by receipt of an additional NIH Training Grant neurostatistics, which is being led by Rebecca Betensky.
The department has continued its valuable partnerships with Frontier Science, Kitasato University, Lilly, Pfizer, Schering-Plough, and Wyeth. However, we were greatly saddened by the unexpected loss of Sandy Heft, Vice President for Biostatistics at Schering-Plough, over Memorial Day weekend, just prior to this year's Harvard Schering-Plough Workshop. Sandy was a great supporter of the department and a close friend to many of us, and we will miss him dearly.
Looming on the distant horizon (perhaps 7-8 years in the future) is the relocation of HSPH to Harvard's new campus being built along the Charles River in Allston. Thus full implications of the move, as well as what we ought to be doing in preparation, are still being discussed. There will be tremendous opportunities to forge new interactions with other parts of the University, especially the life-sciences, and perhaps also to develop undergraduate courses in biostatistics. However, the move will inevitably create challenges to some current interactions, especially because some of our faculty associated with Harvard's teaching hospitals will likely remain in the Longwood Medical Area.
I am delighted to report 3 new additions to our faculty: Xihong Lin (University of Michigan) and John Quackenbush (Institute for Genomic Research) have joined the department as tenured professors, and will focus their energies in environmental statistics and computational biology, respectively. Chris Paciorek, who has interests in environmental statistics and ecology, will be joining us as an Assistant Professor in July. We are happy and fortunate to have these additions to our faculty. Two faculty--Wing Wong and Marco Bonetti--have left our department for positions in California and Italy, respectively. We appreciate all that both have done for the department and wish them the best. We shall also miss Carolyn Dueck--our long-time Director of Administration--who will be moving up (literally) to the Dean's Office as of July 1. Finally, a record number of students (n=24) graduated since the June 2003 commencement and were recognized in this year's commencement. We wish all the best as they begin or resume their professional careers.
With best wishes, Steve Lagakos
Dr. Wayne A. Fuller, Emeritus Distinguished Professor in Liberal Arts and Sciences, Iowa State University was this year's winner of the Marvin Zelen Leadership Award in Statistical Science. This annual award, supported by colleagues, friends and family, was established to honor Dr. Marvin Zelen's long and distinguished career as a statistician and his major role in shaping the field of biostatistics.
Professor Fuller is internationally renowned for his more than forty years
of outstanding leadership in statistical research including survey sampling,
time series and measurement error analysis. He has nurtured many generations
of outstanding statisticians at Iowa State University. The award ceremony was
concluded by a well-received lecture by Professor Fuller on "Analytic Studies
with Complex Survey Data" at the Harvard School of Public Health on May
LJ Wei, Wayne Fuller and Louise Ryan
The 2005 Marvin Zelen Leadership Award was given to Professor Ross Prentice
on June 23, 2005. Professor Prentice was recognized for his outstanding contributions
to the discipline of Biostatistics, including a long record of seminal methodologic
research in areas such as survival analysis and epidemiologic methods and the
mentoring of doctoral students and junior faculty, and to public health, especially
through his leadership role in the Women's Health Initiative. Under his leadership,
the Division of Public Health Sciences at the Fred Hutchinson Center has grown
into one of the largest units in the world conducting research on the causes
and prevention of cancer. Professor Prentice's work has been recognized in awards
from the American Cancer Society, the American Statistical Association, the
American Association of Public Health, and the Council of Presidents of the
Statistical Societies. Professor Prentice delivered a lecture titled "Chronic
Disease Prevention: Research Strategies and Needs." His talk highlighted the
need for randomized trials in disease prevention, especially with interventions
that involve life-style changes.
Ross Prentice and Marvin Zelen
The following is a short summary of my recently NIAID-funded R01 entitled "Methods for HIV Genotype-based Antiviral Drug Selection".
Because antiretroviral drug resistance can frequently arise, thru both the direct effect of treatment, and the transmission of resistant virus, it is vital that routine patient management assess both the sensitivity of a patient to their current regimen, as well as the potential benefit of other regimens. The potential benefit of a regimen can be measured in terms of the development of future resistance, future immune response, duration of virological control and delay in onset of clinical outcome. The overall aim of the grant is to develop statistical methods that will help in understanding the relationship between high-dimensional predictor variables such as genetic sequence and treatment history, and outcome variables such as HIV drug susceptibility phenotype, immunological, virological and clinical response. Such methodology will simultaneously accommodate multiple outcome variables, some of which may be censored, time-to-event variables.
Existing statistical methods for high-dimensional data are largely ad-hoc with poorly understood fundamental statistical properties, e.g. associated false-positive rates. Furthermore, it is not clear how existing methods can simultaneously accommodate multiple response variables. The RESAMPLING-BASED statistical methods we plan to research allow formal investigations, with well-defined overall false-positive rates, into the association between genotype and several response variables, potentially facilitating patient-specific treatment regimen decisions. Finally, we aim to provide to the public, in a user-friendly form, software that executes this methodology.
Last year's Harvard/Schering-Plough Workshop, held on June 3-4, 2004, was titled "Development and Approval of Preventive and Therapeutic Products for Infectious Diseases: Impact of Statistics". Over 170 participants from academia, industry and government attended the workshop. The first session focused on medical and statistical challenges in the development of anti-infective pharmaceutical products from the academic, industry, and regulatory perspectives. Prevention and treatment of HIV as well as development of other antimicrobial agents were discussed. During the second session innovative statistical methodologies in vaccine trials were presented, including the use of validation sets for outcomes, the augmented design, and methods for assessing the dependency of vaccine efficacy on genotypic/phenotypic properties of HIV. The third session addressed the problem of resistance in drug development. Statistical methods were introduced for prediction of phenotypic drug resistance from genotypes, for assessment of prevention measures for hospital-acquired infections, and for dealing with high-dimensional covariates and multivariate outcomes. In the fourth session modeling of SARS and anthrax epidemics and assessment of intervention strategies were described. The workshop concluded with a final session on statistical issues in the regulatory environment. The use of area under the response curve (AUC) as an endpoint and the evaluation of the safety of vaccines were discussed. The speakers of this workshop included Niko Beerenwinkel, Ron Brookmeyer, Christie Chuang-Stein, Ed Cox, Victor DeGruttola, Greg DiRienzo, Christl Donnelly, Roger Echols, Dean Follmann, Mary Foulkes, Peter Gilbert, M. Elizabeth Halloran, Michael Hughes, Daniel Kuritzkes, and Marc Lipsitch. The workshop was followed by the presentation of the year 2004 Marvin Zelen Leadership Award in Statistical Science to Professor Robert C. Elston of Case Western Reserve University.
The 2005 Harvard/Schering-Plough Symposium focused on the timely topic of safety monitoring for pharmaceutical products. Thought provoking discussion by speakers from academia, regulatory agencies and the pharmaceutical industry touched on topics ranging from pre and post-registration clinical trials to post marketing surveillance. Safety assessment poses some challenging statistical problems, since the numbers of patients needed in a clinical trial designed to assess efficacy are generally far too low to provide adequate power to detect low incidence, yet serious adverse side effects. Consequently, statistical tools such as data mining and meta-analysis play a prominent role. This year's symposium generated lots of animated audience participation as well, several times around recent high profile post-marketing safety issues that have led to product withdrawals or restrictions. The workshop began on a somber note as we remembered our dear colleague Sandy Heft who had been killed in a car accident just the previous weekend. Sandy had been particularly active in the workshop planning committee. He'd had the vision when planning started a year ago to see the importance of safety monitoring and felt passionate about getting statisticians involved in thinking about the issues. Other members of the planning committee were Louise Ryan (Chair), David Harrington and Richard Gelber from Harvard and Kenneth Koury, Harold Amkraut, Ersen Arseven, Ram Suresh, Lucy Shneyer, and Lillian Mellars from Schering-Plough. Details about this year's workshop, including videotaped presentations, can be found on the biostatistics website: http://www.biostat.harvard.edu/events/schering-plough/agenda.html.
This year, our Summer Program in Quantitative Sciences is taking place from June, 5 to July, 1 2005, hosted by the Department of Biostatistics, and in partnership with the Department of Epidemiology and the Department of Society Human Development & Health. Our 12 participants are majoring in Mathematics, Biology, Health Science & Education, Psychology, and Statistics and Computer Science. We have 1 Freshman, 2 Sophomores, 6 Juniors, and 3 Seniors. The Summer Program aims at providing a greater understanding of Public Health to the students and its importance in society, guidance on applying to and study in graduate school, and offering mentoring from Public Health professionals.
The summer students have already exciting research experience as collaborators in research on healthy eating and physical activity among minority populations, research in Hispatology and Neuroscience, studies on the perceptions and usage of health services by Indian Americans, or projects studying the relations between coping styles, stress, and health behaviors. Other students have great experience working as math or science tutors, nurse and doctor assistants, research assistant in Biology, interns in infectious disease clinics, collaborators on counseling sessions on Sexual Transmitted Diseases, lab technicians, and responsible for the maintenance of computer network services and equipment on their campus, or as collaborators on projects to provide culturally sensitive healthcare.
During the Summer Program, our students are taking an Introduction to Biostatistics course, several STATA classes, and participating in lectures such as Cancer Epidemiology, Epidemiology of Infectious Diseases, Immigration and Latino Health, Violence Against Women Prevention Practice, Biostatistics Aids Research, or Bioinformatics. They are applying these different tools to the research projects they are working on during their 4 weeks here. For these research projects, we are receiving strong support from our three participating department, plus the Department of Health Policy & Management.
Faculty and students from these departments are mentoring and guiding our Summer Students on research projects addressing clinical trials for cancer research; genetic determinants of alcoholism; racial/ethnic disparities in health status and health care; effects of social and physical occupational hazard on interleukin-6 (IL-6) secretion among several minority populations; peridontal disease and biomarkers related to CVD; and the impact of Fast Food Restaurants in a neighborhood on neighborhood characteristics. The summer students are given the opportunity to analyze and present data, work effectively in a collaborative setting, and present their research in front of a broader audience of faculty and students.
Finally, the Summer Program also aims at strengthening the skills of the students to apply successfully to graduate training and provide them guidance on choosing the graduate program best suited for their interests and backgrounds. We have organized seminars on How to Read a Paper, Mock Interviews and Statement for Graduate School GRE Practice Tests, Funding, Internships & Conferences for Undergraduate Students etc.
This year, our Summer Program is offering some new seminars on international applications of public health issues and improvement of health care systems in developing countries with the Population and International Health Department, on Health Policy and Management projects, and on the difference between career paths leading to research after MD/MPH degrees and PhDs in Public Health.
The Summer Program is not only about hard work and intense learning! We are also trying to balance the academic requirements of the program with extracurricular social activities, focusing on offering activities promoting different cultures and leisure opportunities available in the Boston area. Hopefully they'll enjoy their stay here and will want to come back!
On January 24 to 26 ,2005, the department hosted its first Short Course on "Using PBAT and FBAT to analyze family-based association studies in genetics". The course featured two department faculty, Nan Laird and Christoph Lange, and one postdoc, Kristel VanSteen. The course was first given in Sweden at the Karolinska in 2004, and subsequent courses have been given in Mexico and France. The participants at HSPH were drawn mainly from the US and Canada, but included individuals from Israel, England and Sweden. A total of 45 registered, but due to an ill-timed blizzard, only 38 attended. The total number of participants so far is close to 200. The course provides a short introduction to genetic association studies, however the focus of the course is using the software developed here at HSPH by Professors Laird, Lange and Xu (Department of EH at HSPH). A modest background in genetics is assumed. The next course at HSPH is late June 2005; additional courses are planned for Switzerland, France and Mexico. To register, visit the FBAT or PBAT websites.
Biostatistics is alive and well at Beth Israel Deaconess Medical Center; in fact, the group is expanding! Most of the biostatisticians at BIDMC are in the Division of General Medicine and Primary Care, with the majority of them having a connection to HSPH. Roger Davis (ScD '88 and current HSPH faculty), who joined the division twelve years ago, leads the biostatistics group that also includes Anna Legedza (ScD '98 and post-doc '98-'01), Long Ngo, PhD (who joined the group in September after completing his postdoc at the department), Jane Soukup (MS '83) and Donglin Li, MS. In addition, Shiva Gautam, PhD, provides general support to the BIDMC General Clinical Research Center.
The Division of General Medicine and Primary Care includes over 70 Harvard Medical School faculty members involved in teaching, clinical practice and research. The Division runs a large hospital-based primary care practice, a new women's health center and a hospital medicine program. These programs provide training for medical students and medical residents. Many members of the Division conduct research, most of which is related to chronic conditions and patient safety. The research faculty includes epidemiologists and physician investigators as well as Anna, Long and Roger. Most of the research focuses on health services projects, clinical epidemiology and clinical trials.
Anna and Roger also collaborate with the HMS Division for Research and Education in Complementary and Integrative Medical Therapies. This group conducts clinical trials on alternative medicine such as acupuncture, traditional Chinese medicine and massage therapy. Many of these trials are designed to evaluate a variety of very interesting questions to improve our understanding of placebo effects. Long collaborates with investigators in Radiology and Emergency Medicine and Anna also works with researchers in the Anesthesia Department.
The Division of General Medicine is the home of two HMS-wide faculty development fellowship programs, one in general medicine and one in alternative medicine. These fellowships are designed to provide research training to physicians who plan academic careers in these fields. Roger, Long and Anna work closely with many of the fellows in these programs and Roger teaches a course on survival analysis in the HSPH Program on Clinical Effectiveness, which most of the fellows take.
Dianne Finkelstein Promoted to Professor
The department extends its congratulations and best wishes to Dr. Dianne Finkelstein, who has just been promoted to Professor of Medicine at Harvard Medical School. Dr. Finkelstein, who holds a secondary appointment in the Department of Biostatistics, is based at the Massachusetts General Hospital.
Certificate of Distinction in Teaching
The Department of Biostatistics is delighted to announce that the following Teaching Assistants have been awarded the "Certificate of Distinction in Teaching" for the academic year 2003-2004. Each year faculty are encouraged to nominate TAs for this award in recognition of their outstanding teaching. Congratulations to all for a well-deserved honor! Andrea Cook, Heather Litman, Derek Elmerick, Amy Murphy, Laura Forsberg, James Signorovitch, Ryung Suk Kim, Chang-Heok Soh, Ming Lin, Tyler Vanderweele.
Andrea Cook Wins ENAR Award
The department extends its congratulations and best wishes to doctoral student Andrea Cook. Andrea has won one of the 2005 ENAR Distinguished Student Paper Awards. She will present her paper and receive her prize at the ENAR Spring Meeting to be held in Austin, Texas in March.
Natasa Rajicic Receives an NSF Young Researcher's Travel Award
The department extends its congratulations and best wishes to doctoral student Natasa Rajicic. Natasa has been selected to receive an NSF Young Researcher's travel award to attend and present a poster at the University of Florida's Seventh Annual Winter Workshop: Longitudinal Data Analysis. The workshop is being held January 6-8, 2005, in Gainesville, Florida. The funding covers travel expenses and a three-day stay at the workshop.
Lu Zheng Wins ENAR Award
The department extends its congratulations and best wishes to doctoral student Lu "Summer" Zheng. Summer has won one of the 2005 ENAR Distinguished Student Paper Awards. She will present her paper and receive her prize at the ENAR Spring Meeting to be held in Austin, Texas in March.
Mei-Ling Ting Lee Named "Statistician of the Year"
Each year the Boston Chapter of American Statistical Association selects a Statistician of the Year. The award is named after Fred Mosteller and is given to the statistician that has made enormous contributions and impact to our profession through excellence in teaching, significant applied and methodologic research, and dedicated service. We are delighted to announce that the Chapter has selected Dr. Mei-Ling Ting Lee as the Mosteller Statistician of the Year. A banquet will be held February 16, 2005 to honor Professor Lee. Congratulations Mei-Ling!
Dr. John Quackenbush Joins Faculty
We are delighted to announce that Dr. John Quackenbush will join the Department of Biostatistics as Professor of Computational Biology and Bioinformatics, also with an appointment at the Dana-Farber Cancer Institute. Dr. Quackenbush will be leading our efforts in genomics.
Dr. Xihong Lin Joins Faculty
We are delighted to announce that Xihong Lin will be joining the Department as a Professor of Biostatistics in August 2005. Professor Lin has broad methodological and applied interests, and will play a leadership role in our environmental health activities.
2005 Harvard Award in Psychiatric Epidemiology and Biostatistics Announcement
The Department of Biostatistics is delighted to announce that Dr. Christine Waternaux is this year's recipient of the Harvard Award in Psychiatric Epidemiology and Biostatistics. The award recognizes Dr. Waternaux's lifelong career contributions that have significantly advanced the field of Psychiatric Biostatistics. Dr. Waternaux will present the award lecture at the Harvard School of Public Health during the 2005 Fall Semester.
Andrea Cook Wins Award
The Department of Biostatistics would like to congratulate doctoral student Andrea Cook for being awarded runner-up of the student paper competition in the American Statistical Association (ASA) Section on Statistics and the Environment (ENVR). She will receive the award at the 2005 Joint Statistical Meetings (JSM) in Minneapolis this August.
David Wypij's Fulbright Travel to Brazil
David Wypij will be spending a month in Fortaleza and Rio de Janeiro, Brazil, as a Fulbright Senior Specialist, beginning Saturday, April 16. He will be teaching short courses in clinical trials and longitudinal data analysis at the Federal University of Ceara, and will be collaborating on several AIDS epidemiology projects with colleagues there. He is being hosted by Dr. Ligia Kerr-Pontes, a former HSPH Takemi Fellow, in Fortaleza and Dr. Guilherme Werneck, who earned his doctoral degree at HSPH, in Rio de Janeiro.
Lingling Li and Natasa Rajicic Inducted
Lingling Li and Natasa Rajicic were inducted into Mu Sigma Rho, the national honorary society for statistics. Mu Sigma Rho was founded in 1961 to promote and encourage scholarly activity in statistics, and to recognize outstanding achievement by students in eligible academic institutions. The Boston Chapter of the American Statistical Association (which serves members in Massachusetts, Vermont, New Hampshire and Maine) was granted affiliate chapter status in Mu Sigma Rho in 2005. Lingling and Natasa were nominated based on their outstanding achievement in statistics course in addition to their overall strong academic performance, and are members of the first class of students inducted by the Chapter.
Beth Ann Griffin Receives Award
Beth Ann Griffin was a co-recipient of the Best Student Paper Award at the Conference of the Eastern Mediterranean Region of the International Biometric Society, held in Corfu, Greece from May 10-12, 2005. Her paper was entitled "Inference for Survival Distributions Based on Interrupted Hazards".
Rebecca Betensky Receives Award
Rebecca Betensky has been awarded the 2005 Mortimer Spiegelman Award. This award, established in 1969, is presented by the Statistics Section of the American Public Health Association to honor a young statistician who has made important contributions to public health statistics. Previous awardees are listed at http://watson.hgen.pitt.edu/~dweeks/spiegelman/.
Geert Molenberghs, Ph.D., Professor, Department of Biostatistics, Limburgs Universitair Centrum was this year's winner of the Myrto Lefkopoulou Distinguished Lecturer award. This annual award was initiated in 1993 in memory of Myrto Lefkopoulou, a former beloved faculty member and student in the Department of Biostatistics. Dr. Lefkopoulou tragically died of cancer in 1992 at the age of 34 after a courageous two-year battle.
Professor Molenberghs presented his lecture on "Handling Incomplete Data in Longitudinal Studies" on September 30, 2004 in the Snyder Auditorium. In his talk, Professor Molenberghs noted that some simple but commonly used methods to handle incomplete longitudinal clinical trial data, such as complete case analyses and methods based on last observation carried forward, require restrictive assumptions and stand on a weaker theoretical foundation than likelihood-based methods developed under the missing at random (MAR) framework. Given the availability of flexible software for analyzing longitudinal sequences of unequal length, implementation of likelihood-based MAR analyses is not limited by computational considerations. While such analyses are valid under the comparatively weak assumption of MAR, the possibility of data missing not at random (MNAR) is difficult to rule out. However, he noted that MNAR analyses are, themselves, surrounded with problems and therefore, rather than ignoring MNAR analyses altogether or blindly shifting to them, their optimal place is within sensitivity analysis. The concepts developed in his talk were illustrated using data from a few case studies. The lecture was followed by a question and answer session, presentation of a plaque and a reception in Professor Molenberghs' honor.
The Department of Biostatistics, Harvard School of Public Health, has named Mark van der Laanž Division of Biostatistics, School of Public Health, University of California, Berkeley, as the 2005 Myrto Lefkopoulou Distinguished Lecturer. Professor van der Laan will present a lecture on Thursday, September 15 at Harvard School of Public Health. The title of the lecture is "History Adjusted Marginal Structural Models: Applications in AIDS Research." A reception will be held following the lecture.
The lectureship was established in perpetuity in memory of Dr. Myrto Lefkopoulou, a faculty member and graduate of Harvard School of Public Health. Dr. Lefkopoulou tragically died of cancer in 1992 at the age of 34 after a courageous two-year battle. She was deeply beloved by friends, students and faculty.
Each year the Myrto Lefkopoulou Lectureship is awarded to a promising statistician who has made contributions to either collaborative or methodologic research in the applications of statistical methods to biology or medicine and/or has shown excellence in the teaching of biostatistics. Ordinarily, the lectureship is given to a statistician within 15 years of receiving an earned doctorate.
Previous recipients of the Lefkopoulou Memorial Lectureship have been Michael Boehnke, Ronald S. Brookmeyer, Brad Carlin, Steven N. Goodman, Trevor Hastie, Hans-Georg Mueller, Giovanni Parmigiani, Kathryn Roeder, Louise Ryan, Danyu Lin, Marie Davidian, and Geert Molenberghs.
Nominations for next year's lectureship are welcome and should be sent to the Myrto Lefkopoulou Lecture Committee, Department of Biostatistics, Harvard School of Public Health, 655 Huntington Avenue, Boston, MA 02115. Nominations should include a letter of nomination and a C.V. The deadline for submission of nominations is March 15, 2006.
The Distinguished Alum Award was initiated in 2003 by the faculty of the Biostatistics Department, to "recognize an individual in government, industry, or academia, who by virtue of applications to support of research, methodology and theory, significant organizational responsibility, and teaching has impacted the theory and practice of statistical science." The award recipient is invited to the HSPH to meet with the Biostatistics students in an informal brown bag lunch, and to deliver lecture on their career and life beyond the Department at the Harvard School of Public Health. A plaque is presented and the recipient is honored at an Alumni dinner. The committee who selected the awardee this year consisted of alums and faculty from this department. The selection committee members represented several nations and graduation years from the 1960's through 1990's. Given that this was the first occurrence of this award, several excellent alums were nominated each of whom was deserving of recognition.
Stuart Baker (SD 1984) of the National Cancer Institute was selected as the first recipient of the Distinguished Alum Award. Stuart received his doctorate from our dept in 1984 under the direction of Christine Waternaux. After a postdoc in our department, he took a position at the Biometry Branch of NCI, where he currently holds as a position of Mathematical Statistician GS-14. Stuart is on the editorial board of 4 journals, including JNCI and Statistics in Medicine. He has over 50 publications, of which over 30 are in statistical methods journals, including several in Biometrics and JASA. While this shows impressive research productivity, Stuart was not selected for this award solely because of his methodological research productivity. What is most impressive about Stuart's work is that it bridges the two research communities of clinical and epidemiological researchers and biostatisticians. His letter of nomination stated, "For medical investigators simple formulas have great appeal. (To quote Albert Einstein, "Everything should be made as simple as possible-but not simpler"). Much of Dr. Baker's research has involved creating novel approaches and formulating them as crisply as possible. For analyses that require complicated methodology, medical investigators desire clear assumptions with statistics that are easy to interpret. These (characteristics) are the hallmark of many of Dr. Baker's more mathematical papers." Stuart has contributed to many research areas including: novel designs and analytic methods to reduce bias in studies involving historical controls or non-randomized screening studies; methods for analysis of non-ignorable missing data; and most recently to the analysis of surrogate endpoints and genetics data.
Stuart visited the HSPH on June 2. He participated in a "brown bag lunch" with several of our current students and gave a formal lecture in the late afternoon on "The Experience Of One Alum At The National Institutes Of Health".
The brown bag lunch provided a unique opportunity for current students to get advice about potential career paths, to discuss optimizing the Biostatistics training, and also to strengthen ties with alumni. Stuart Baker was very helpful in providing descriptions of his work responsibilities in his job at NIH. The students attending the lunch were surprised to learn that Dr. Baker spends a much of his time conducting methodological research. He reported that a major advantage of a job in government, relative to academia, is that it is not necessary to obtain grants to fund research time. However, this has a drawback in that it makes it more difficult for a statistician to move from a government position to an academic one, in that government employees do not have a record of obtaining grants. They also rarely have teaching experience. On the other hand, going from academia to government is easier. Switching from a government position to an industry position is also very feasible.
Stuart is on the editorial board for the Journal of the National Cancer Institute and Statistics in Medicine, and provided guidance to students on the editorial process, reminding them that when they are attempting to publish thesis papers, they should realize that it is quite common for a paper to be rejected, and that it is important to keep re-submitting it and not be discouraged. He also recommended considering reviewer's comments, even if the paper has been rejected, and make the appropriate changes before submission to another journal.
Dr. Baker also discussed the coursework he had during his doctoral studies. The class he wished he had taken was Clinical Trials. He felt that just having a basic understanding of the clinical trial issues and analyses would have prepared him even better for his work at NCI. There is such a basic course, taught by Dr. Ware, and the new Sequential Analysis course provides a more statistically oriented approach to clinical trials, such as early-stopping rules.
The Distinguished Alum Award events provided a great opportunity for mentoring and networking among former and current students. Students were inspired by the discussion with this former graduate from our program who has managed his career as a statistician so well. Stuart's humility, humor and poise contributed a memorable Department of Biostatistics day.
One of the measures of a department is the quality of the students it graduates. The Biostatistics Department is proud of the accomplishments of our alums, and are expanding our efforts to build an association to maintain contact with our alums, and to facilitate their networking with each other. This past year, the Department initiated a new award to be given annually to a Distinguished Alumni of the department, and hosted the first annual Alum Dinner.
A dinner was held at the Harvard Faculty Club on June 2, 2004 to honor the accomplishments of the alum of this department, and to re-acquaint the faculty and former students. The event was attended by approximately forty people, including alums who had graduated from the Department in as early as the 1980's. The event occurred on the day proceeding the opening of the Schering-Plough Workshop. In attendance were alums traveling from California, Washington state, New York City, Colorado and the UK. There were reunions of former students and faculty who had not seen each other for many years. Each alum introduced themselves, indicated their year of graduation, and described their current location and job. They had diverse paths since graduation, moving into academic, commercial and government jobs. Some continued their education, returning to school in Medicine, Biostatistics, and Epidemiology. Stuart Baker, the 2004 Distinguished Alum recipient, was introduced and honored. Marvin Zelen praised his work and congratulated him for receipt of the award. The Department looks forward to hosting this dinner annually at the time of the Distinguished Alum Award events, timed as it was this year to be prior to the Schering-Plough Workshop, so that the maximum number of our out-of-town alums can participate. Watch for announcements on the Biostatistics website for this event. The event is sponsored by the Department of Biostatistics and subsidized by donations of the faculty and alums to the Alumni Fund. Those who are interested in participating in planning activities of the Biostatistics Alumni Association should contact the department at http://www.biostat.harvard.edu/about/alumni/alumni_contact.html.
Dr. Christl A. Donnelly, Professor of Statistical Epidemiology, Department of Infectious Disease Epidemiology, Division of Primary Care and Population Health Sciences, Imperial College London, was the Department of Biostatistics's Distinguished Alum of 2005. This annual award lecture is given by an individual in government, industry or academia, who by virtue of applications to support of research, methodology and theory, significant organizational responsibility, and teaching has impacted the theory and practice of statistical science. The recipient of this award is decided by a committee of the Biostatistics Department's alums.
Christl Donnelly completed her DSc in Biostatistics from HSPH in 1992 under the direction of Drs. Nan Laird and Jim Ware. Upon graduation Christl took a position at the Dept. of Math/Stat at the University of Edinburgh and in 1995 as the Head of Statistics Unit at the Centre for Epidemiology of Infectious Disease at the University of Oxford. In 2000, Christl moved with her group to Imperial College of London, where she was promoted to Professor and has helped form a new Dept. of Infectious Diseases Epidemiology.
Christl has taken a leading role in working on the epidemiology of mad cow (BSE) disease since 1996. Her analyses of the BSE screening programs and of data on clustering of the bovine and human disease variant of this disease have directly advised policymakers in governmental and research agencies. Christl also played a key role in the analysis of the foot and mouth disease epidemic in Great Britain in 2001. She advised the government that the early disease control policy was inadequate. This led to a change in government policy to include culling of neighboring farms as well as farms with affected animals. She presented her work to policymakers, the media, farmer's representatives and veterinarians. More recently, Christl advised public health policy leaders on the SARS epidemic in Hong Kong in 2003. Her work on SARS resulted in a first author, fast track publication in Lancet (as well as on in Science) in May 2003, and served to inform the public about the novel infectious agent and the needed control efforts.
For her contributions to these epidemic research programs, Christl was recently awarded the Franco-British prize from the French Academie of Science. In addition to her statistical leadership in these epidemic research programs, she has been productive as a statistical researcher and teacher, overseeing the work of several doctoral and postdoc students over the years. Her career has been an exemplary realization of the mission of the HSPH Biostatistics Department's doctoral training program.
On June 1, 2005, Christl presented a lecture to a packed audience at HSPH on "The Science/Public Policy Interface: Fertile Ground for Biostatisticians". She was presented with a plaque for the Distinguished Alum Award, and honored at a reception attended by Biostatistics Department faculty, students and alumna.
Between November 2004 and June 2005, the department awarded 20 doctoral degrees and four master's degrees in Biostatistics. Doctor of Science degrees were awarded to John H. Page, Maria D. Shubina and Bin Zhang. Doctor of Philosophy degrees were awarded to Jared C. Christensen, Andrea J. Cook, Xuemin Fang, Dionne A. Graham, Yannis Jemiai, Ming-Chih Kao, Ryung Suk Kim, David M. Loecke, Larry F. Leon, Ming Lin, Heather J. Litman, Emily C. Martin, Abigail G. Matthews, Lisa C. Miller, Carrie G. Wager, Lu Zheng, and Sheng Zhong. Master of Science degrees were awarded to Bethany L. Hedt and Tzu-Min Yeh and Master of Arts degrees were awarded to Erica H. Bae and Jennifer L. James.
We asked graduates in our department to comment on their experiences here. The following is a selection of their responses.
1) What do you miss most about your student days at Biostats? (Or for those of you who haven't had time to recover and reflect: what do you think you will miss most?)
Patsy: I think that I will miss my fellow students in the department-they taught me a lot about working together, encouraging and supporting each other, and about statistics.
Xuemin: Interaction with other students. Harvard students learn the most from other Harvard students.
Tzu-min: My classmates.
Anonymous: Seminars in Rm 426; the friendly staff both in the department and in the rest of the school.
Lynne: I will miss the community. We supported each other through all the ups and downs of grad school - and on the basketball court. Go Weibulls! : )
Beiying: I would have to say the abundant resources and opportunities. The students have the flexibility to build up their future career paths by selectively utilizing those that work to their advantages.
Abigail: To be honest, I miss my friends the most. The environment in the department is very collegial - there is always someone to chat with and things going on.
Maria: School atmosphere where you feel yourself on the edge of new developments.
John: The freedoms associated with being a student.
Bin: Having been working in a pharmaceutical company for 6 months, I miss the luxury of being able to spend a lot of time devoted to certain Statistical methodology.
Carrie: I miss the informal rapport I had with my fellow students, especially the camaraderie of working late and eating seminar leftovers together for dinner in the kitchen. I also miss the variety and frequency of seminars and the outstanding colloquia.
Lu: Free time and access to Harvard libraries.
Ryung: Life has changed a lot since I came to Boston at year 2000. For example, as an international student, I had to start living in a very different life style. I became to be more sincere and wholehearted about my faith as Christian. And I got married last year. Overall experience over these 5 years was a big growing process for me. I'll miss it.
2) Was there a particular course or consulting experience that most influenced your view of statistics or your career direction?
Patsy: I would say that they all equally influenced my career direction.
Tzu-min: Regression, the projection...and BIO247.
Yuhyun: RAing at Microarray Core Facility at DFCI. The views from biostatisticians and biologists are quite different!!
Anonymous: EPI 207.
Lynne: TAing for Bio201 and 210 were probably the most rewarding experiences for me. Teaching, and interacting with, students who were eager to apply statistical tools in their varied respective fields forced me to gain a solid understanding of biostats and an appreciation for its power.
Beiying: I think being an RA and participating in the RA seminar has been a rewarding experience. There are probably no other courses where you get to see such a diversity of real problems being tackled by statistics, it is also where you get to know that good statistical analysis methods may not necessarily be really complicated ones.
Abigail: For me, the course I enjoyed most was statistical genetics with Kathy Lunetta. After every class I would race to my office to do the homework because I relished the problem solving. It inspired me to continue in the field and will be doing a postdoc on the genetics of alcoholism.
Maria: All courses of Marvin Zelen: Probability, stochastic processes, Statistical Science outreach; Joe Ibrahim's course on Bayesian Analysis.
John: "EPI207. Advanced Epidemiologic Methods" since that course more than any other changed the way I think about study design, and data analysis. Furthermore, it was the starting point for my thesis research.
Bin: There is no course stands out. But the collective courses definitely helped me understand statistical ideas and principles much better than before.
Carrie: There are three courses I took during the final semester of my masters program that changed my mind about pursuing doctoral studies in statistics (and convinced me to apply to the PhD program): Bayesian Statistics (the first time it was taught with Joe Ibrahim), Nonparametric Statistics (David Wypij), and Generalized Linear Models with Sudeshna Adak. I also found the Environmental workshop in the Berkshires to be very inspiring, back when George Casella was a regular attendee.
Sheng: Inference I, taught by Marvin Zelen. This is one of the best courses I have had in my 20 years of studying. It shaped my view as a statistician. If statistics is a religion, and we are just about to convert ourselves into it, which denomination shall we choose? Dr. Zelen would not preach the green statisticians into one denomination. But he exposes us to all of them, analyzes their fundamentals, compares their essences, and points out where they converge. Having taken this course, I feel like before climbing a cliff in a mountain, I have bird-viewed it. I have seen where different cliffs are facing, where they are rooted and how they are connected.
3) To which areas, if any, do you wish you had received more exposure as a student?
Lynne: A more formal course covering SAS programming and more applied courses or experiences.
Yuhyun: Theoretical statistics in advanced current topics.
Xuemin: Collaboration and conversation with pharm companies.
Beiying: More exposure to principles and practice of clinical trials. We only have an intro course in HSPH so far.
Anonymous: Consulting, experimental design, protocol review, meta-analysis.
Abigail: Unfortunately I never had time to take the Design class. It would really have been helpful to learn at least an introduction in Methods I. In addition, I wish there was a Clinical Trials class that was more oriented to statisticians.
Bin: In terms of statistics, survival analysis, Bayesian methods, and regression. Cancer biology and human genetics are two non-statistical areas.
Maria: Survival stuff.
Sheng: I wish I had taken the Semiparametric course. I was too focused on my own research. One lesson that myself learned is that, as long as one wants to stay in academia, no matter how "applied" his/her research is, efforts spent on theoretical study will all eventually pay off.
Carrie: I wish I had spent more time learning about immunology and nonclinical laboratory science. There are several excellent courses taught at HSPH that I wish I had taken because they are geared towards students who do not have the the usual pre-medical biology prerequisites.
Jared: Clinical trials. Many people in academic hospitals and research institutions and industry do so much in clinical trials. I should have done a little more of that.
John: I wish I had read the book "How to Prove It: A Structured Approach" by Daniel Velleman, prior to entering the program.
Lu: Clinical trials.
Ryung: I think courses offered in the department are great. I was unaware of the computational biology field when I came here. It was very difficult because all courses and seminars seem to assume some familiarity with the biological context of the field. I felt we didn't have an intermediate course for students like me. Now, I think Shirley's course plays this role.
4) What are you doing now (employment, travel, ...) ?
Patsy: I am currently working on projects for investigators at Boston Medical Center and Dana Farber Cancer Institute, yet plan to spend next fall traveling internationally.
Lynne: After a little vacationing back on the west coast, I'll be starting work at the Center for Biostatistics in AIDS Research - just down the hall!
Beiying: Started working as a biostatistician at a biotech company.
Abigail: I will begin a postdoc at the University of Pittsburgh in the Department of Human Genetics next week.
Sheng: I am working in Stanford University' BioX Center as a visiting scholar. I will soon join University of Illinois at Urbana Champion, as an assistant professor in Bioengineering (affiliated with Statistics and Computer Science.) Welcome to visiting me in the corn field!
Bin: I am a principal biostatistician in the department of global biostatistics & clinical technology department, Wyeth Research, in Cambridge, MA.
Lu: Taking some time off and looking for a job.
Tzu-min: Job search.
John: Assistant professor of Epidemiology at HSPH (from July 1,2005).
Yuhyun: Travel -> Postdoc
Ryung: Healing and recovering after writing up the thesis. I'll most likely work as Instructor of Neurology in Harvard Medical School. But I'm still seeking for a teaching position where I can teach undergraduate students.
Carrie: I'm working for a small company in Vermont that does design and analysis of bioassay studies for pharmaceutical companies and government agencies; our clients are worldwide. My job involves some travel, and Vermont is a great place to live and have as a home base. I am quite surprised and fortunate to have landed a job here (I found it by accident, as I was looking for a job in Australia or somewhere far away...)
Heather: Right now, I'm relaxing before I start work at New England Research Institutes (NERI) in July.
Jared: I work at Wyeth Research in Cambridge, MA.
Maria: Looking for a job and prepare for a travel, with a participation in a workshop where I have a talk.
Xuemin: Research associate for University, consultant for pharm companies.
5) How was the job search process?
Lynne: Not too stressful. The biostat faculty were very helpful in providing information and advice for MS-level jobs in the area.
Yuhyun: smooth but headaching.
Tzu-min: Hmmm...still going... I got an interview this week, and I am waiting to hear back from the employer.
Beiying: It's a relatively long process but not a really hard one. I believe everyone who has gone through it knows better what he/she wants for career better and it's a good idea to take your time and find out what's the best match.
Abigail: Job search was fine. Got several interviews, and a couple offers. My advisors helped me with the process, and even got feedback from institutions where I interviewed so I could improve my tactics.
Ryung: One of the scientist in DFCI introduced me to the Adult Oncology department.
Xuemin: My job came to me.
Sheng: I have been looking for assistant professor positions in various departments, including biostatistics, biology, and bioengineering. My research area is Bioinformatics. Because of its interdisciplinary nature, different departments interviewed me. The competition for tenure-track positions in good universities is keen. It is typical in this year for 120 or so applicants to apply for one position. People in our department are all very helpful to my job hunting. I have had strong recommendations from several faculty members in our department. I had 2 offers from biostat departments, 1 from biology and 1 from bioengineering (affiliated with statistics). After I had offers, many people gave very insightful suggestions about where to go.
Lu: Good so far. Faculties and fellow graduates are very helpful.
John: I did not search widely. I applied and got one job that was close to my ideal.
Bin: I did my own search on amstat, went to jsm career placement service and scored a few interviews. The job posting by biostats department was very helpful and the job I took was posted through biostat.
Carrie: I never really got started looking for a job, as I waited until after my defense to start looking, and then I graduated at a weird time for academic job searches. I was convinced I wanted an academic appointment (postdoc, asst. prof., or research associate), possibly abroad, but I ended up getting a job in industry working for a very small company. My job found me. My employer wasn't particularly looking to hire someone at my level or with my qualifications, but when we realized it was such a good fit, he created a position for me.
Heather: Very quick and pretty intense. I tried to put it off until I was sure I'd be graduating (but when are you ever sure?) and when I started seriously looking, it went quite fast.
Jared: Haphazard. A learning curve. Lots of mistakes, but I am happy with where I ended up. I probably did not ask enough of the right people for help.
6) What were employers looking for in terms of skills, qualities, coursework and applied experience?
Lynne: They liked to see a good amount of applied experience and an eagerness to learn new things - as each job has it's own specialties and specific skills involved. I'm sure the first few months at CBAR will entail a steep learning curve, but this program has definitely helped me learn to learn.
Yuhyun: Hm... case by case.? I think that it is not usually difficult for employers to find out our skills, qualities, and applied experience after a short conversation.
Xuemin: Communication skill is the most important aspect. Since I work a lot with pharm companies courses like "clinical trails" "survival" etc are important.
Maria: I think my problem is experience in US setting.
Beiying: In my case, I think they looked for qualities and applied experience. Being a graduate from Harvard Biostat is certainly quite reassuring to employers in terms of quality but one does need to have quite a bit of consulting and/or collaborative experiences to be more convincing.
Tzu-min: Experience is definitely good. Programming skills. Some theoretical knowledge.
Ryung: We have been working together several years to answer biological questions using microarray data.
Abigail: Mostly I searched for assistant professor jobs which wanted at least a couple publications (applied or theoretical). Preferably they would like someone with post-doctoral experience, but I was able to get some interviews for the positions. Ultimately I ended up taking a postdoc in order to publish more papers and develop more applied skills.
John: In my case: strength in Biostatistics and Epidemiology; the ability to teach, consult and collaborate; and the willingness to apply for grant funding.
Bin: Exposure to pharmaceutical industry and understand what they do. SAS programming experience is very important. Survival analysis, generalized linear model (logistic regression, e.g.), non-parametric method are especially helpful.
Carrie: My employer needed somebody with very strong computing skills (ability to write a disseminable open-source R package for bioassay analysis that could be deployed for client-specific needs over a web server) as well as a strong background in developing methods for variance component analysis with mixed models. The ten years of experience I have as a statistical programmer prior to entering my PhD program is a very important asset to the company.
Lu: More applied experience if looking for a position other than faculty.
Sheng: It varies from employer to employer. In general, academic searches mostly stress two factors: 1. does this person excel in his/her own field/research direction. 2. Does the recruiting department want to expand this field/research direction. A department chair told me: "In judging the applicant, we keep asking ourselves whether this person will eventually be tenured here."
Heather: Employers were looking for people with applied experience generally.
Jared: Experience seemed to be the trump card in the pharmaceutical industry. Business skills, presentation skills, statistical know-how all seemed to be secondary issues after experience.
7) What advice would you offer to newer students in the program?
Patsy: It would be a good idea to look over Casella and Berger before starting the programs. Really take advantage of the interesting talks offered throughout the HSPH campus--there are tons of really great and interesting research projects going on in the school!
Bethany: prepare to work hard, never be afraid to ask for help, take advantage of opportunities at school, and enjoy Biostats, HSPH and Boston!
Lynne: Study really hard, but try to keep a balanced life. It's easy to get overwhelmed at times, but always much easier to get through those times when you have outlets - like working out or just hanging out with your wonderful biostat friends . . . without books.
Yuhyun: Check the due date for the course registration! The late fee is $80.
Xuemin: Get the best out of the program, otherwise you lose in your opportunity cost. Build up your knowledge, build up your people skill, build up your confidence.
John: Get experience in consulting in a wide range of areas. This may take the form of helping colleagues in other departments with their thesis projects. Also improve your knowledge of medicine/biology.
Ryung: Enjoy and be grateful to the years as graduate students. Some colleagues and I often shared stressful feelings throughout the years and we found that a lot of feel lack of confidence. So you're not alone!
Sheng: All people in the department are so willing to give help. Go to them and seek it!
Beiying: To get more exposure to data analysis, whether through research projects or internships. I think that holds true no matter if they go on to academia or industry.
Abigail: You can never be prepared for the amount of work, but know that many have gone before you and many will after. Lean on friends and other students for help with coursework, computing and complaint sessions.
Carrie: Take the time to "get your hands dirty" working on applied projects in your substantive area. Ask many questions of the substantive researcher about what their data collection process is, as this might yield some useful insight about what you can do statistically that maybe they haven't thought of. Ultimately, it is the statisticians role to serve as the communicator between the substantive researchers and the statistical analysis performed. There is another type of efficiency in the real world that our learning in statistical efficiency does not address: practical efficiency. If you can design a study that can help a pharmaceutical company save time (==dollars), that can be at least as important as developing a better estimator.
Jared: Take advantage of opportunities to learn about academia and the pharmaceutical industry. Lots of things happen in our department, so take advantage of them and learn things.
8) What was your favorite statistics book in your first year? What's your favorite statistics book now?
Abigail: I guess Ross and Prentice (probability) was my favorite, though probably because I used it in college as well. My favorite textbook now is "Multivariate Survival Analysis" by Philip Hougaard.
Bin: My favorite book now is "the statistical analysis of failure time data" by Kalbfleisch and Prentice.
Yuhyun: Cox and Hinkley.
Anonymous: I liked "Survival Analysis: Techniques for Censored and Truncated". Data" by Klein and Moeschberger" in years 2-3. Now I like "Encyclopedia of Biostatistics."
Carrie: 1st year: Casella and Berger; Now: The Elements of Statistical Learning.
Ryung: The Elements of Statistical Learning: Data Mining, Inference, and Prediction by Trevor Hastie, Robert Tibshirani, Jerome Friedman.
Xuemin: First year: None or don't remember. Now: Applied Logistic Regression second edition by David Hosmer and Stanley Lemeshow.
Sheng: I did not read any statistics books in my first year. All I read were class notes and handouts. I like the book "The Elements of Statistical Learning: Data Mining, Inference, and Prediction".
Heather: I would say that Casella and Berger's statistical inference book was the most helpful my first year for preparing for the Qual (even though I didn't use that book in any of my classes). My favorite book now is Little and Rubin's missingness book.
John: "Statistical Inference" by Casella and Berger. Still is!
Jared: A book about Baseball Dynasties by Rob Neyer. Does that count as a statistical book? My favorites statistics books are well written introductory books. I like the Pagano & Gauverau book. My thesis required me to be constantly looking in Johnson & Wichern or Morrison.
Maria: My first statistical book(long before I enrolled in school) was U. Linnik "Least square method"(Marvin has it in English). I also liked and still like Van der Warden "Mathematical Statistics".
9) Describe your coffee consumption at HSPH:
a. Coffee-free for ___ years and counting.
b. I only drank coffee during emergencies.
c. ___ cup(s) / day.
d. Kay rings up 1 coffee (student, cash) as soon as she sees me coming.
Abigail, Patsy: B
Beiying: A. Coffee-free for 4.5 years and counting. Coffee doesn't work for me.
Bin: A for 6.
Carrie: sample(1:2,1,prob=c(1,3)/4) + #cups/morning, sample(0:2/2) #cups/afternoon (the 1/2 is for 1/2 decaf)
Jared: I don't drink coffee. Coffee-free for 31 years and loving it.
John: I drank and drink coffee only occasionally (and only socially): about 2-5 times per year
Maria: e) I drink about 5 cups of tea(real!, not herbals)
Ryung: I rarely drink coffee. :)
Sheng: I regularly drank coffee in the first year. Later on I drank coffee at a frequency of probably 1 cup per week.
Tzu-min: Coffee-free for 26 years and counting.
Xuemin: c) _1_ cup(s) / day.
Congratulations to all of our graduates and best wishes for continued success!
2004 Graduates: Renee Boynton-Jarrett, Beiying Ding, Kevin Roberts, Yuhyun Park, Denise Scholtens, Cassandra Arroyo, Ken Wilkins
2005 Graduates: Yannis Jemiai, Tzu-Min Yeh, Lisa Miller Wruck, Carrie Wager, Heather Litman, Maria Shubina, Jared Christensen, Summer Zheng, Abigail Matthews, Dionne Graham, Bin Zhang, Emily Martin, Larry Leon
Biostat now has a web page for Alumni! http://www.biostat.harvard.edu/about/alumni/
Lillian Lin (Ph.D. '80)
Lillian was appointed "Acting Chief, Statistics Section" in the Statistics and Data Management Branch, Division of HIV/AIDS Prevention (DHAP), National Center for HIV, STD, and TB Prevention at CDC in June 2002. She successfully competed for the position of "Lead, Statistics Team" (same job, different title) in February 2004. Currently she supervises the work of 16 statisticians (master's and doctoral level). The work ranges from reporting on current statistics on HIV and AIDS reports to collaborating with researchers engaged in preventing the further transmission of HIV to development of statistical methods relevant to the work of investigators in DHAP.
Susan Murray (ScD, '94)
Susan has had identical boy twins, Joshua Edward Krukonis and Benjamin Christopher Krukonis, born on April 8th 2004 at 4 lbs 11oz each. These are their first children. Susan is still working at the University of Michigan at the rank of Associate Professor and Eric has a new tenure-track joint appointment in Microbiology and at the Dental School.
Vivek Goel (MD, '90)
Julia L. Bienias (Sc.D., '93)
Susan continues her work as Senior Statistician at Rush Univ. Medical Center in Chicago. She manages an analytic group that works with researchers in the Dept. of Internal Medicine, the Rush Institute for Healthy Aging, and the Rush Alzheimer's Disease Center. She is a collaborator on several NIH-funded longitudinal studies of aging. Her research interests include longitudinal modeling and analysis of complex survey data. She can be reached at email@example.com.
Aime de Muynck (MS; '82)
Aime De Muynck, alumnus MS epidemiology 81 and MS biostatistics 82, regained his original assignment at the Institute of Tropical Medicine, Antwerp, Belgium where he chaired the Department of Epidemiology. In 1996 he was elected Chairman of the Academic Board of that same Institute. GTZ, the German Technical Cooperation, invited him in 1997 for an overseas assignment in the Health Services Academy, Islamabad, Pakistan. This is the sole public health institute of the country and is supported by the German Cooperation; it has an MPH course, and carries out field research in various aspects of health. He returned to Belgium after 4 years. But the "overseas microbe" had infected him so much that he could not resist to take up more overseas long term assignments. At present he is working as Chief Technical Advisor of the Tuberculosis Control Programme, that covers the whole state of Orissa, India. The project is sponsored by the Danish Cooperation. He can be reached at firstname.lastname@example.org.
Anne M. Stoddard (ScD; '78)
Anne recently "retired" after 25 years in the Department of Biostatistics and Epidemiology at UMass Amherst. She is now working at New England Research Institutes in Watertown where she is Director of the Center for Statistical Analysis and Research. Anne works with two other Biostat Alums, Lynn Sleeper and Elizabeth Mahoney.
Isao Kamae (MPH '88, and PhD '95)
Isao has been elected since last May as a Board of Director, the International Society of Pharmacoeconomics and Outcomes Research (called ISPOR). The ISPOR 7th Annual European Congress will come in October in Germany. They changed the name of their division in April into Applied Medical Statistics and Decision Sciences.
Matteo Bottai (ScD: '98)
Matteo accepted an offer for a position as Assistant Professor in Biostatistics, tenure-track, at the University of South Carolina, Columbia, SC.
Cathy Berkey (ScD '80)
Cathy, her husband Dennis, and their 3 children (all currently in college) are moving this summer from Weston to Worcester, MA. Dennis became President of WPI (Worcester Polytechnic Institute) on July 1. Cathy will remain at the Channing Laboratory, Harvard Medical School.
Amanda Stebbins-Sharpless (SM; '93)
Amanda recently moved to Fredericksburg VA after living in Durham NC for 11 years. She still works for Duke and will be telecommuting. She has two children: Julian is 4 1/2 and Dahlia is 20 months old.
Dan Geer (ScD '88)
Dan is VP and Chief Scientist for Verdasys, a startup in Waltham, Massachusetts, and the owner of Geer Risk Services. For the interest of his various long lost colleagues, he is happy to report that the tools of biostatistics and epidemiology are immensely valuable in the study of information security at the national scale, not to mention a fine way to generate notoriety.
Gary Lyman (MPH, '82)
Gary can be reached at:
Gary H Lyman MD MPH FRCP(Edin)
Professor of Medicine and Oncology
Director of Health Services and Outcomes Research
James P Wilmot Cancer Center
University of Rochester School of Medicine and Dentistry
A.J. Rossini (ScD, '94)
A.J. and Sally are spending the year or longer in Switzerland, on leave from the University of Washington. He is working for Novartis Pharma AG in Basel, in the Modeling & Simulation group, and Sally is Guest Professor (equiv: Research Prof) at the University of Basel. His German is improving, but Swiss German has proven to be hopeless. They are still trying to figure out what they want to do when they grow up.
John McGready (SM, '96)
John has been a biostatistics instructor at Johns Hopkins for the past 6 years - and is concurrently working on his PhD. He enjoys teaching very much and has won a few awards for doing it. John has two wonderful sons, Micah Teagan (2.8 years), and Emmet Asher (5 months).
Karen Eckstein Han (ScD, '02)
Karen is currently living in San Francisco, CA and has made some informal contacts at UCSF. She and her husband welcomed Matthew Henry Kyu Han on December 10, 2004 and he is the proud little brother of two year old Caroline.
Cheryl L. Jones (ScD, '99)
Expecting her first child (boy) August 13, 2005! Enjoying being a working biostatistician at a great company! Though it is great being home, she misses folks in the department!