[ Table of Contents | Graduate School of Public Health Home | Bulletins Homepage | Next Section ]





Department of Biostatistics

Acting Chairperson: Howard E. Rockette, PhD
Department Office: 318C Parran Hall
(412) 624-3022, FAX (412) 624-2183
E-mail: SUG2@VMS.CIS.PITT.EDU


  • Purpose
    Biostatistics is one of the significant tools for the study of variation in the life cycle of living organisms and the essential tool for the quantitative study of humanity's growth, reproduction, morbidity, and mortality. The proper application of the theory of biostatistics to this study requires above all a clear understanding of the concepts underlying statistical analysis. Training focuses on the theory and techniques for collecting, analyzing, and interpreting quantitative data related to the fields of public health and medicine.


  • Major Educational Areas and Programs
    The objectives of the department are to: (1) foster an understanding of statistical logic as it applies to the quantitative study of public health problems and human biology; (2) develop and apply statistical methods to the study of new and pressing problems through research and community service; and (3) train health statisticians capable of assuming and exercising leadership in the fields of public health and medicine. Courses offered by the department cover the areas of statistical theory and methods, life table analysis, biological assay, clinical trials, sampling, design of experiments, computer programming and data processing, and multivariate analysis. The program provides for concentrations in two areas: biostatistics (MS, PhD) and public health statistics (MPH, DrPH).


  • Biostatistics
    The major objective is to train students with backgrounds in mathematics and with a strong interest in biology to apply statistical theory and methods to the solution of public health problems. Master's-level students should be able to modify statistical techniques to meet special research needs in public health and to assume middle-level positions in epidemiological, environmental, or health care research with full responsibility for the appropriateness of the statistical techniques used and validity of conclusions based on these techniques. This involves the ability to fully understand the research problem and all aspects of research on this problem. Doctoral students are trained to develop statistical and research strategies to meet needs in public health; to assume leadership roles in epidemiological, environmental, health care, and experimental research settings; and to serve as faculty at both the undergraduate and graduate levels.


  • Public Health Statistics
    This program prepares students who have a prior professional degree and/or background in health (physicians, nurses, dentists, pharmacists, veterinarians, health administrators, etc.) to understand and apply statistical methods to public health problems.


  • Faculty Research Interests
    Research interests represented in the department are:

    Dr. Ahnn: Clinical trials, stochastic processes

    Dr. Anderson: Stochastic processes, splines, clinical trials, statistical computing

    Dr. Arena: Statistical methods for environmental and occupational epidemiology, statistical
    computing

    Dr. Berhane: Nonparametric regression, smoothing techniques, longitudinal data analysis, time-series, survival analysis, statistical applications in epidemiology

    Dr. Bryant: Clinical trials design, quality control methods applicable to immunologic monitoring, empirical Bayes methods, stochastic processes, statistical methods of multiple comparison for analysis of covariance models, data analysis procedures based on empirical transforms

    Dr. Costantino: Design and analysis of clinical trials, quantitative risk assessment, statistical computing, environmental and occupational epidemiology

    Dr. Richard Day: Environmental and occupational epidemiology, psychiatric epidemiology, statistical computing, statistical consulting

    Dr. Roger S. Day: Stochastic models of tumor growth, treatment and monitoring community-based cancer epidemiology, spatial cluster detection, statistical methods in AIDS epidemiology

    Dr. Dignam: Survival analysis, clinical trials, epidemiologic methods

    Dr. Enterline: Public health statistics, biostatistical applications in environmental and occupational health

    Dr. C.C. Li: Experimental design, path analysis, population genetics, emergenesis

    Dr. Marsh: Statistical methods and applications in occupational and environmental epidemiology, survey sampling, meta-analysis, statistical computing

    Dr. Mazumdar: Statistical inference, mathematical modeling, longitudinal data analysis

    Dr. Redmond: Discrete multivariate analysis, clinical trials, statistical methods for environmental and occupational epidemiology

    Dr. Rockette: Clinical trials, occupational biostatistics, estimation theory

    Dr. Rudy: Psychometrics, repeated measures, statistical methodology for assessment of chronic pain

    Dr. Schlesselman: Statistical methods for planning and analyzing observational epidemiologic studies and controlled clinical trials, particularly those related to family planning, human reproduction, women's health, and cancer cause, treatment, and prevention

    Dr. Sereika: Survival analysis, nonparametrics, empirical Bayes estimation

    Dr. Stoffer: Time-series analyses

    Dr. Stone: Generalized linear models, survival analysis, statistical methods for epidemiology

    Dr. Weissfeld: Censored data, regression diagnostics

    Dr. Wieand: Clinical trials, nonparametric statistics

    Dr. Wilson: Statistical methods in cancer research and field biology, nonparametric statistics


  • Qualifications for Admission
    All degree programs require two semesters of calculus, a course in biology, a basic computing course, and the graduate record examination. In some cases, course deficiencies can be satisfied the first semester. The MPH and DrPH also require six credits in biology, six credits in social science, and a professional degree and/or background in health.

    Credit requirements for the master's degrees are 36 plus a thesis. Credit requirements for the doctoral programs are 72 credits beyond the bachelor's degree.

    Students are eligible for financial aid including scholarships, graduate student researcher awards, and teaching assistantships.


  • Course Offerings

    BIOST 2011 PRINCIPLES OF STATISTICAL REASONING: 2 credits; Fall Term and Summer Session I (admission by permission of instructor for Summer Session); prerequisite: college algebra or a higher-level math course with a grade of C or better)
     Acquaints students with the concepts of statistical reasoning as applied to the study of public health problems. Students learn the general principles of statistical analysis and acquire the ability to utilize a statistical software package (MINITAB) as a tool to facilitate the processing, editing, storing, displaying, analysis, and interpretation of health research-related data. Includes a weekly laboratory session to review lectures and assignments.

    BIOST 2016 INTRODUCTION TO SAMPLING: 2 credits; Spring Term; prerequisite: BIOST 2011 or 2041; or permission of instructor
     Provides a working knowledge of practical sampling methods along with an understanding of their theoretical background. Covers simple random, stratified, systematic, and simple stage cluster sampling techniques, and ratio, regression, and difference estimation. Emphasis is placed on sampling human populations in large communities.

    BIOST 2019 PUBLIC HEALTH STATISTICS: 2 credits; Fall Term; prerequisite: college algebra
    Reviews the development of biometry and public health statistics and relationships to demography and actuarial science. Introduces students to national health data systems and to life table analysis and measures of fertility, morbidity, and mortality as they apply to public health statistics.

    BIOST 2021 SPECIAL STUDIES IN BIOSTATISTICS: Credits and hours to be arranged
    Qualified students may undertake advanced work or research with the approval and under the guidance of a member of the faculty.

    BIOST 2025 BIOSTATISTICS SEMINAR: 1 credit; Fall Term through Spring Term
    Students are introduced to current health problems involving the development of appropriate biostatistical methods.

    BIOST 2040 ELEMENTS OF STOCHASTIC PROCESSES: 3 credits; Spring Term; (Scheduled for 1996-97); prerequisites: MATH 0240, BIOST 2043; or permission of instructor
    Covers generating functions and convolutions of random variables, the Poisson and compound Poisson distributions, ranching processes, random walk, and the gambler's ruin problems as well as Markov chains, and simple birth and death processes.

    BIOST 2041 INTRODUCTION TO STATISTICAL METHODS 1: 2 credits; Fall Term; prerequisite: college algebra
     Discusses techniques for the application of statistical theory to data. Topics include probability theory, estimation of parameters, and tests of hypothesis for both the discrete and continuous case.

    BIOST 2042 INTRODUCTION TO STATISTICAL METHODS 2: 2 credits; Spring Term; prerequisite: BIOST 2041; or permission of instructor
     More techniques are given of the application of statistics to data with emphasis on distribution-free and multivariate methods. Interpretation of results and concepts will be stressed.

    BIOST 2043 INTRODUCTION TO STATISTICAL THEORY 1: 3 credits; Fall Term; prerequisite: MATH 0240; or permission of instructor
     Covers joint, marginal, and conditional probabilities; moment generating and characteristic functions of simple discrete and continuous distributions; transformation of variables; convergence of random variables; law of large numbers; central limit theorem.

    BIOST 2044 INTRODUCTION TO STATISTICAL THEORY 2: 3 credits; Spring Term; prerequisite: BIOST 2043
     Covers elements of statistical inference; Rao-Cramer's inequality; fundamentals of the theory of testing statistical hypotheses; chi-square test of goodness of fit; the Neyman-Pearson fundamental lemma; likelihood ratio tests.

    BIOST 2045 ANALYSIS OF CASE-CONTROL STUDIES: 2 credits; Spring Term; prerequisites: BIOST 2041, 2042; or BIOST 2041, 2042 concurrently; or permission of instructor
     Introductory course in the design and analysis of case-control studies. The course covers measures of disease occurrence and association, classical analysis of grouped and matched case-control studies, and an introduction to the use of logistic regression for analysis of grouped and matched case-control studies.

    BIOST 2046 ANALYSIS OF COHORT STUDIES: 3 credits; Fall Term; prerequisite: BIOST 2042; and BIOST 2049 recommended
     This introductory course in the statistical modeling of cohort data emphasizes Poisson and Cox regression. Issues in the design of cohort studies are also considered. The course involves the analysis of several cohort datasets.

    BIOST 2047 INTRODUCTION TO BIOLOGICAL ASSAY: 2 credits; Spring Term (Part 2); (scheduled for 1996-97); prerequisites: BIOST 2042, 2043; or permission of instructor
     Acquaints the student with statistical techniques in biological assay, including direct and indirect assays, quantitative and quantal responses, estimation of median effective dose, and comparisons of effectiveness.

    BIOST 2048 OCCUPATIONAL BIOSTATISTICS: 2 credits; Spring Term; (alternate years: given in 1996-97); prerequisite: BIOST 2042
     Acquaints the student with the statistical methods used in the identification of occupational health problems. Includes methods of testing dose-response relationships, risk analysis, and design considerations.

    BIOST 2049 APPLIED REGRESSION ANALYSIS: 3 credits; Summer Term; prerequisite: BIOST 2042; or permission of instructor
     Deals with topics in regression analysis with special emphasis on health-related problems.

    BIOST 2051 STATISTICAL ESTIMATION THEORY: 3 credits; Fall Term; (alternate years: given in 1995-96); prerequisite: BIOST 2044; or permission of instructor
     Fisher's information; Rao-Cramer inequality and sufficient statistics; Bhattacharyya bounds; Rao-Blackwell theorem; methods of moments; the method of maximum likelihood; Newton-Raphson method and Rao's scoring for parameters; estimation of several parameters; order statistics and life-testing problems.

    BIOST 2052 MULTIVARIATE ANALYSIS: 3 credits; Fall Term; (alternate years; given in 1995-96); prerequisite: BIOST 2044; or permission of instructor
     The multivariate normal distribution; partial and multiple correlation coefficients; Wishart distribution; Hotellings T2 and Mahalanobis' D2 statistics; classificatory problems and Fisher's discriminant function; likelihood ratio tests; Wilds' criterion; multivariate analysis of variance; canonical correlations; principal components.

    BIOST 2053 NONPARAMETRIC METHODS IN STATISTICS: 3 credits; Spring Term; (alternate years; given in 1995-96); prerequisite: BIOST 2044; or permission of instructor
     Order statistics and quantiles: the U statistic and Hoeffding's theorem; ranks and mid-ranks; Kendall's and Spearman's rank correlation coefficients; the sign test; Kolmogorov-Smirnov and Cramer-Von Mises tests of goodness of fit; rank (order) test; Mann-Whitney (Wilcoxon) rank test; the several-sample problem; the Mood-Brown median test; Mood's median test; the Kruskal-Wallis H test; Friedman's test; efficiency of nonparametric tests; Pitman efficiency of tests.

    BIOST 2054 SURVIVAL ANALYSIS: 3 credits; Spring Term; (alternate years: given in 1995-96); prerequisite: BIOST 2044; or permission of instructor
     Introduces the student to the design considerations and statistical analysis of clinical trials. The course covers the theoretical aspects of various models in reliability theory and the proportional hazards model, as well as the more applied problems of interpreting specific datasets and designing large-scale trials.

    BIOST 2057 PLANNING BIOMEDICAL STUDIES: 3 credits; Spring Term; prerequisites: EPID 2110, BIOST 2046; or permission of instructor
     Students will learn statistical methods and computer software that can be used in planning observational and experimental biomedical studies. They will work as members of teams in planning such studies, and articulate through written and oral presentations the statistical issues involved in study design and analysis. Students will write statistical sections of study proposals to learn how to present the design of a study and the proposed analysis of its data in terms of the study's major scientific objectives. The course will use lectures, problem-based learning, small-group discussions, large-group discussions, student presentations in class, and preparation of student-written proposals to achieve its aims. Using previously written grant proposals as case studies, the course will cover the design of clinical trials and observational studies (cohort, case-control, sample survey).

    BIOST 2060 STATISTICAL INFERENCE: 3 credits; Spring Term; prerequisites: BIOST 2041, 2042, 2051
     Covers the theory of statistical inference at an advanced level. The objectives of the course are to provide the student with the statistical background to read papers in the statistical literature and to provide a foundation for the development of new statistical techniques.

    BIOST 2063 BAYESIAN AND EMPIRICAL BAYES STATISTICAL METHODS: 3 credits; Fall Term; (alternate years: given in 1995-96); prerequisites: BIOST 2042, 2044
     The theoretical foundations of Bayesian and empirical Bayes statistical methods are presented. The use of these methods in data analysis is illustrated with specific examples and with discussions of common data analysis issues. Contrasts and similarities between Bayesian, empirical Bayesian, and classical methods are evaluated.

    BIOST 2064 BAYES AND EMPIRICAL BAYES COMPUTATIONAL METHODS: 3 credits; Spring Term; (alternate years: given in 1995-96); prerequisite: BIOST 2063
     Students gain a solid understanding of both theory and practice with regard to the EM algorithm, Markov chain Monte Carlo sampling techniques (including Gibbs and Metropolis-Hastings methods), importance sampling, and the Laplace methods. These topics complement the Bayes theory of BIOST 2063.
    Students also gain hands-on experience with UNIX workstations and with S-Plus. This will give students access to modern robust techniques, modern programming ideas, and flexible interactive graphics.

    BIOST 2081 MATHEMATICAL METHODS FOR STATISTICS: 3 credits; Fall Term; prerequisite: 1 year of college calculus
     Covers differentiation and integration of functions of several variables, infinite sequences and series, and fundamentals of matrix algebra. Class examples and homework problems emphasize applications to probability and statistics.

    BIOST 2083 LINEAR MODELS: 3 credits; Fall Term; prerequisites: BIOST 2044, 2081; or permission of instructor
     Acquaints students with linear model techniques for analyzing both balanced and unbalanced data. The topics covered are: generalized inverses, models not of full rank, orthogonal contrasts with unbalanced data, analysis of covariance, and variance component analysis.

    BIOST 2084 DISCRETE MULTIVARIATE ANALYSIS: 3 credits; Spring Term; (alternate years: given in 1996-97); prerequisite: BIOST 2051
     Theory and methods for the analysis of discrete multivariate data using both asymptotic and exact approaches. Generalized linear models and their extensions are emphasized. Methods are derived and applied to data.

    BIOST 2085 APPLIED TIME-SERIES ANALYSIS: 3 credits; Spring Term; (alternate years); prerequisites: BIOST 2043, an introductory course in computing; or permission of instructor
     Acquaints students with basic techniques of time-series analysis. The topics covered are: estimation in the time domain, forecasting, and spectral analysis. Computer application with real datasets is stressed.

    BIOST 2087 BIOSTATISTICS CONSULTING PRACTICUM: 1 credit; Fall Term through Summer Term; admission by permission of instructor
     Provides advanced students (second-year master's and doctoral) with practical experience in consulting on the biostatistical aspects of research problems arising in the biomedical or other allied fields. Students, initially under the supervision of a faculty member, participate in discussions with investigators leading to the design and/or analysis of current research problems. A weekly group discussion session is led by the instructor.

    BIOST 2088 SPECIAL TOPICS 1
    2089 SPECIAL TOPICS 2
    2090 SPECIAL TOPICS 3
    12 total lecture hours each, Summer Session I; others as announced: 1 credit each; prerequisite: variable, depending on specific topic; permission of instructor may be required for admission
     Introduces the student to specialized topics in biostatistics in areas of current interest to the field. Three special topics courses are offered each year.

    BIOST 2092 INTRODUCTION TO COMPUTING: 1 credit; Fall Term (first session)
    Provides basic computer skills on the VAX/VMS mainframe computers. Emphasis is placed on developing the basics for data processing and analysis using an interactive statistics package (i.e., Minitab). Techniques for searching computerized library databases are also presented.

    BIOST 2093 DATA MANAGEMENT AND ANALYSIS: 1 credit; Fall Term (second session); prerequisite: BIOST 2092; or permission of instructor
     Teaches data and file management, and data description aspects of several statistical packages (i.e., BMDP, SAS, SPSS).

    BIOST 2095 INTRODUCTION TO DATABASE MANAGEMENT SYSTEMS: 2 credits; Spring Term; prerequisite: BIOST 2092; or permission of instructor
     Provide students with an understanding of the concepts and techniques used in database management. Students will learn how to design functional and efficient database systems. Applications are given for both the mainframe and microcomputer environments.

    BIOST 2097 APPLIED STATISTICAL COMPUTING IN THE HEALTH SCIENCES: 1 credit; Summer Session I (alternate years: given in 1995-96); prerequisites: proficient programming in a middle/upper-level programming language (i.e., Fortran, C, Pascal, etc.) and the use of an upper-level statistics package (i.e., BMDP, SAS, SPSS, etc.), BIOST 2041, and calculus.
     Presents the fundamentals of good programming techniques and procedures. Teaches an awareness of the strengths and limitations of computational approaches. Integration of programming techniques to biostatistical and health sciences problems. Students will be expected to perform advanced programming in a computer language of their choice.

    BIOST 3010 RESEARCH AND DISSERTATION FOR THE DOCTORAL DEGREE: Credits and hours to be arranged

    FTDR 0000 FULL-TIME DISSERTATION STUDY: Every term
    Doctoral candidates who have completed all credit requirements for the degree, including any minimum dissertation requirements, and are working full-time on their dissertations may register for this course. While the course carries no credits and no grade, students who enroll in "Full-time Dissertation Study" are considered by the University to have full-time registration status.


  • Faculty

    Acting Chairperson

    HOWARD E. ROCKETTE, Professor of Biostatistics and Acting Chair of the Department. MA, PhD, Pennsylvania State University

    Emeriti Professors

    PHILIP E. ENTERLINE, Emeritus Professor of Biostatistics. MA, PhD, American University
    CHING CHUN LI, Emeritus University Professor of Biometry and Human Genetics. PhD, Cornell University


  • Faculty With Primary Appointments

    Professors

    GARY M. MARSH, Professor of Biostatistics. MS(Hyg), PhD, University of Pittsburgh

    SATI MAZUMDAR, Professor of Biostatistics. MS, University of Calcutta; PhD, Cornell University

    CAROL K. REDMOND, Professor of Biostatistics. MS(Hyg), ScD(Hyg), University of Pittsburgh

    HARRY SAMUEL WIEAND, Professor of Biostatistics. MA, PhD, University of Maryland

    Associate Professors

    JOSEPH P. COSTANTINO, Associate Professor of Biostatistics. MPH, DrPH, University of Pittsburgh

    LISA WEISSFELD, Associate Professor of Biostatistics. MA, PhD, University of Pittsburgh

    Assistant Professors

    STEWART ANDERSON, Assistant Professor of Biostatistics. MA, University of Kansas; PhD, University of Colorado

    VINCENT C. ARENA, Assistant Professor of Biostatistics. MS(Hyg), PhD, University of Pittsburgh

    RICHARD D. DAY, Assistant Professor of Biostatistics. MS(Hyg), University of Pittsburgh; PhD, University of California

    ROGER S. DAY, Assistant Professor of Biostatistics; Associate Member, Pittsburgh Cancer Institute. MA, Wesleyan University; ScD, Harvard University

    ROSLYN A. STONE, Assistant Professor of Biostatistics. MA, University of Kansas; MS, Montana State University; PhD, University of Washington

    Research Associates

    SANG H. AHNN, Research Associate. MA, Indiana University; PhD, University of Pittsburgh

    JAMES J. DIGNAM, Research Associate. MS, PhD, University of Pittsburgh

    JOHN W. WILSON, Research Associate. MS, Cornell University; MS, University of Washington; PhD, Cornell University


  • Faculty With Secondary Appointments

    Professors

    JAMES J. SCHLESSELMAN, Visiting Professor of Clinical Epidemiology and Preventive Medicine (School of Medicine); Professor of Biostatistics. MA, MA, PhD, Princeton University

    DAVID STOFFER, Professor of Mathematics and Statistics (Faculty of Arts and Sciences); Professor of Biostatistics. PhD, University of California at Davis

    Associate Professor

    THOMAS E. RUDY, Associate Professor of Anesthesiology, Psychiatry, and Behavioral Sciences (School of Medicine); Associate Professor of Biostatistics. MA, PhD, University of Notre Dame

    Assistant Professor

    KIROS T. BERHANE, Assistant Professor of Family Medicine and Clinical Epidemiology (School of Medicine); Assistant Professor of Biostatistics. MSc, University of Guelph; PhD, University of Toronto

    Research Associate Professor

    JOHN L. BRYANT, Research Associate Professor of Biostatistics. MS, PhD, Rensselaer Polytechnic Institute

    Research Assistant Professors

    SUSAN M. SEREIKA, Research Assistant Professor/Biostatistician (School of Nursing); Research Assistant Professor of Epidemiology (Graduate School of Public Health); Research Assistant Professor of Biostatistics. MPH, PhD, University of Michigan


  • Faculty With Adjunct Appointments

    Associate Professor
    ZVI GROSSMAN, Adjunct Associate Professor of Biostatistics. MS, Technion; PhD, Tel-Aviv University





    [ Table of Contents | Graduate School of Public Health Home | Bulletins Homepage | Next Section ]