Frank E Harrell Jr, PhD

Professor of Biostatistics

Founding Chair
FrankH.jpg
E-mail:   f.harrell@vanderbilt.edu
Phone:  615.322.2001
Fax:      615.343.4924
Assistant: Ashlee Bartley
Vanderbilt University School of Medicine
Department of Biostatistics
2525 West End Ste. 11000
Nashville, TN 37203

Research Interests

  • Statistical analysis
  • Bayesian models and use in clinical trials
  • Clinical prediction
  • Statistical models
  • Statistical computing and open source software
  • Statistical graphics
  • Statistical reporting
  • Research data management
  • Cardiovascular disease
  • Medical applications of statistics
  • Clinical trials
  • Clinical safety assessment and analysis in pharmaceutical trials
  • Risk adjustment
  • Health services and outcomes research
  • Pharmaceutical research
  • Medical diagnosis and prognosis

Philosophy of Biostatistics

  • Biostatistics needs to be fully integrated into biomedical research; experimental design is all important
  • Don't be afraid of using modern methods
  • Avoid categorizing continuous variables and predicted values at all costs
  • Don't assume that anything operates linearly
  • Account for model uncertainty and avoid it when possible by using subject matter knowledge
  • Use the bootstrap routinely
  • Make the sample size a random variable when possible
  • Consider using Bayesian methods
  • Use excellent graphics, liberally

Future Directions

  • Bayesian study design

Awards

  • Fellow, American Statistical Association, August 2005
  • Distinguished Alumnus Award, School of Natural Sciences and Mathematics, University of Alabama Birmingham, February 2005
  • Mitchell Lecturer, Department of Biostatistics, Glasgow University, 2008
  • Presidential Invited Lecturer, Western North American Region of the International Biometric Society, 2012
  • WJ Dixon Award for Excellence in Statistical Consulting, American Statistical Association, 2014
  • Distinguished Service to Translational Scientists Award, Vanderbilt University Medical Center, 2016
  • 2017 Visionary Speaker, Clinical Studies Coordinating Center, University of North Carolina Department of Biostatistics, Chapel Hill

Blog | Hobbies | Pictures | Computing Tools | Bibliographic Database

Biography | Brief Bio | Very Brief Bio

Dr Harrell received a PhD in Biostatistics from the University of North Carolina in 1979 under the direction of PK Sen. He was co-managing editor of the journal Health Services and Outcomes Research Methodology from 1998-2005, is an Associate Editor of Statistics in Medicine, is on the Scientific Advisory Board of Science Translational Medicine, is on the Editorial Board of American Heart Journal, on the Policy Advisory Board of Journal of Clinical Epidemiology, a member of Faculty of 1000 Medicine, and was a long-time consultant to FDA. Previously Dr Harrell was on the FDA Cardio-Renal Advisory Committee, a consultant to the FDA Arthritis Advisory Committee, an AHCPR (now AHRQ) study section, and was the director of the SUPPORT (Study to Understand Prognoses Preferences Outcomes and Risks of Treatment) Statistical Center for nine years. SUPPORT concerned end of life care and decision making of primarily elderly patients with one of 8 end-stage diseases. Dr. Harrell is a Fellow of the American Statistical Association and winner of its 2014 WJ Dixon Award for Excellence in Statistical Consulting. He was the 2008 Mitchell Lecturer for the Department of Statistics, Glasgow University. He was the 2012 Presidential Invited Lecturer for WNAR, International Biometric Society, and the 2017 Visionary Speaker, Clinical Studies Coordinating Center, University of North Carolina Department of Biostatistics.

As reflected in his more than 240 peer-reviewed publications (5 with more than 1000 citations), Dr Harrell has devoted his career to the study of patient outcomes in general and specifically to the development of accurate prognostic and diagnostic models and models for many other patient responses. Much of Dr Harrell's work has been applied to health services and outcomes research, technology evaluation, observational databases, and clinical trials. His primary methodologic research relates to development of reliable statistical models, quantifying predictive accuracy, modeling strategies utilizing data reduction methods, estimating covariable transformations, model validation methods, penalized estimation (shrinkage), missing data imputation, clinical trials, flexible Bayesian clinical trial design, pharmaceutical safety, statistical graphics, and statistical reporting. He has researched methods to estimate how continuous predictors relate to outcomes without assuming linearity, showing the advantages of piecewise cubic polynomials or spline functions. All of this work has taken into account that a risk model's likely performance on a new subject sample should be the touchstone. He has extended Efron's bootstrap estimator of the "optimism" in a model's predictive accuracy to validate more complex survival and risk models. His book Regression Modeling Strategies with Applications to Linear Models, Logistic Regression, and Survival Analysis (2001, Springer-Verlag) contains theory, examples, and detailed case studies demonstrating the use of many modern statistical modeling tools.

Dr Harrell spent 17 years in the Biometry and Cardiology Divisions and the Clinical Research Institute of Duke University Medical Center, where he taught the Regression Modeling Strategies course to fellows in the Masters of Health Sciences in Biometry Program (now the M.S. in Clinical Investigation Program) and mentored many fellows, mostly from the Department of Medicine. In 1996 Dr. Harrell founded the Division of Biostatistics and Epidemiology in the Department of Health Evaluation Sciences at the University of Virginia. There he taught Statistical Computing and Graphics, Biostatistical Modeling, and Advanced Data Analysis and assisted in the teaching of Clinical Trials Methodology, Comprehensive Introduction to Clinical Investigation, and Statistical Thinking in Biomedical Research. He directed the Clinical Investigation Track of the M.S. in Health Evaluation Sciences program, and was a co-director in the T21 and K30 clinical investigation training grants for complementary and alternative medicine, at UVa. Overall, Dr Harrell has taught biostatistics and research methodology to hundreds of physicians since the early 1980s, and has overseen research projects of dozens of medical fellows. In 2003 Dr Harrell became the founding chair of the Department of Biostatistics at Vanderbilt, serving as the chair until September 1, 2017.

Dr Harrell has done a significant amount of consulting and collaboration on biomedical research projects in both academic medical centers and in industry. He has designed statistical aspects of numerous clinical, health services, and outcomes research studies, assisted researchers with hundreds of smaller studies, and has been a consultant to the pharmaceutical industry for over twenty years. In the latter capacity he has served on and been the reporting statistician to several Data and Safety Monitoring Boards, served on mock FDA advisory committees, provided second opinions on data analyses, and advised pharmaceutical statisticians on advanced data analysis and measurement issues. He helped to create the system of daily biostatistics consultation clinics at Vanderbilt, and assists researchers from all over the university on statistical questions two days per week.

Published Articles

To search for a specific paper click the above link, then click Search. Then for example enter in the box author:Harrell && title:multivariable and click Search again. To search by key word, use for example tag:graphics.

Books

R Software Packages

  • Hmisc: Functions for data analysis, high-level graphics, utility operations, computing sample size and power, importing datasets, imputing missing values, advanced table making, variable clustering, character string manipulation, conversion of S objects to LaTeX code, and recoding variables
  • rms: Functions for regression modeling, testing, estimation, validation, graphics, prediction, and typesetting by storing enhanced model design attributes in the fit. rms is a collection of 229 functions that assist and streamline modeling, especially for biostatistical and epidemiologic applications
  • greport: Graphical reporting for clinical trials

Presentations

Course Material

Memberships

  • American Statistical Association
  • Biometric Society (ENAR)
  • Drug Information Association
  • International Society of Clinical Biostatistics

Brief Biography

  • Professor, Dept. of Biostatistics, Vanderbilt School of Medicine
  • Expert Statistical Advisor, FDA CDER Office of Biostatistics
  • PhD Biostatistics UNC Chapel Hill
  • Faculty of Duke U for 17 years
  • Founding Chief, Div. of Biostatistics and Epidemiology, UVa
  • Founding Chair, Dept. of Biostatistics, VU, 2003-2017
  • Fellow, American Statistical Association
  • 2008 Mitchell Lecturer, Glasgow University Dept. of Statistics
  • 2012 Presidential Invited Lecturer, WNAR
  • 2014 Winner of the WJ Dixon Award from the American Statistical Association, for excellence in statistical consulting
  • 2017 Visionary Speaker, Clinical Studies Coordinating Center, University of North Carolina Department of Biostatistics, Chapel Hill
  • Distinguished alumnus award from UAB Science and Math
  • FDA expert consultant for many years
  • Member, NIH Biostatistical Methods and Research Design Study Section
  • Leader, Design, Biostatistics, and Clinical Research Ethics program for the Vanderbilt NIH CTSA
  • PI, NHLBI multinational ISCHEMIA cardiovascular trial DSMB statistical center
  • Associate Editor, Statistics in Medicine
  • Scientific Advisory Board, Science Translational Medicine
  • Author of two of the most cited articles in the history of Statistics in Medicine
  • > 240 Peer-reviewed publications, 5 with >1000 citations
  • Interests: clinical predictive models, model validation, diagnosis and prognosis, clinical trials, health services research, cardiovascular research, drug safety, missing data, Bayesian statistics, statistical computing and graphics, flexible Bayesian clinical trial design and analysis

Brief Biography Text

Dr. Harrell received his PhD in Biostatistics from the University of North Carolina in 1979. He was on the faculty of Duke University for 17 years and of the University of Virginia for 7 years. He founded the Division of Biostatistics and Epidemiology at the University of Virginia School of Medicine in 1996 and the Department of Biostatistics at Vanderbilt University in 2003. He has taught biostatistics and research methodology to hundreds of physicians since 1980 and has been a mentor or co-mentor to several physician investigators. He is an Associate Editor for Statistics in Medicine, a member of Faculty of 1000 Medicine, and a member of the policy advisory board for the Journal of Clinical Epidemiology. His specialties are development of accurate prognostic and diagnostic models, model validation, clinical trials, observational clinical research, technology evaluation, quantifying predictive accuracy, missing data imputation, clinical trials, pharmaceutical safety, flexible Bayesian design and analysis, and statistical graphics and reporting. He has worked on a large number of clinical trials.

Dr. Harrell is a Fellow of the American Statistical Association and winner of its 2014 WJ Dixon Award for Excellence in Statistical Consulting. He was the 2008 Mitchell Lecturer for the Department of Statistics, Glasgow University. He was the 2012 Presidential Invited Lecturer for WNAR, International Biometric Society and the 2017 Visionary Speaker, Clinical Studies Coordinating Center, University of North Carolina Department of Biostatistics, Chapel Hill. He is an FDA Expert Statistical Advisor and was a member of the NIH Biostatistical Methods and Research Design Study Section. He is the associate director of the Research Methods program for the Vanderbilt NIH CTSA and was the director of the Statistics and Methodology Core for the Vanderbilt Kennedy Center for Research on Human Development. He is the PI of the NHLBI multinational ISCHEMIA trial DSMB statistical center. He is the author of two of the most highly cited papers (both are on development of prognostic models) in the history of Statistics in Medicine and has > 240 peer-reviewed publications (5 with >1000 citations).

Brief Areas of Expertise

Statistical modeling and model validation; diagnosis and prognosis; statistical computing, reporting, and graphics; cardiovascular disease; clinical trials, pharmaceutical research, and safety assessment; survival analysis; missing data; bootstrap, Bayesian methods.

Very Brief Biography Text

Dr. Harrell received his PhD in Biostatistics from UNC in 1979. Since 2003 he has been Professor of Biostatistics, Vanderbilt University School of Medicine, and was the department chairman from 2003-2017. He is Expert Statistial Advisor for the Office of Biostatistics for FDA CDER. He is Associate Editor of Statistics in Medicine, a member of the Scientific Advisory Board for Science Translational Medicine, a member of the Faculty of 1000 Medicine, and a member of the policy advisory board for the Journal of Clinical Epidemiology. He is a Fellow of the American Statistical Association and winner of the Association's WJ Dixon Award for Excellence in Statistical Consulting for 2014. He was the 2017 Visionary Speaker, Clinical Studies Coordinating Center, University of North Carolina Department of Biostatistics. His specialties are development of accurate prognostic and diagnostic models, model validation, clinical trials, observational clinical research, cardiovascular research, technology evaluation, pharmaceutical safety, Bayesian methods, quantifying predictive accuracy, missing data imputation, and statistical graphics and reporting.


Projects | Citation Sandbox
Topic attachments
I Attachment Action Size Date Who Comment
FrankH.jpgjpg FrankH.jpg manage 396.5 K 29 Nov 2015 - 17:25 FrankHarrell Frank Harrell photograph
discrim.pdfpdf discrim.pdf manage 2152.1 K 22 May 2014 - 18:41 FrankHarrell Harrell FE, Lee KL: A comparison of the discrimination of discriminant analysis and logistic regression under multivariate normality. In Biostatistics: Statistics in Biomedical, Public Health, and Environmental Sciences. The Bernard G. Greenberg Volume, pp 333-343. Amsterdam: North-Holland, 1985.
logistCal.pdfpdf logistCal.pdf manage 512.1 K 28 Mar 2013 - 07:33 FrankHarrell Harrell FE, Lee KL (1987): Using logistic model calibration to assess the quality of probability predictions
smi92pro.pdfpdf smi92pro.pdf manage 971.5 K 11 Oct 2009 - 15:20 FrankHarrell Smith LR, Harrell FE, Muhlbaier LH (1992):Problems and potentials in modeling survival;AHCPR Pub. No. 92-0056
Topic revision: r120 - 18 Nov 2017, FrankHarrell
 

This site is powered by FoswikiCopyright © 2013-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback