John M. Lachin

Dr. Lachin is Research Professor of Biostatistics and Epidemiology, and of Statistics, formerly with tenure 1987-2014; and former Director of The Biostatistics Center from 1988-2000 and 2010-12. Dr. Lachin was the founding director of the GWU graduate program in Biostatistics and Epidemiology and mentored 15 doctoral students in Statistics, Biostatistics or Epidemiology. Dr. Lachin no longer teaches.

Dr. Lachin is a Fellow of the American Statistical Association, the Royal Statistical Society, and the Society for Clinical Trials for which he also served as President.

Dr. Lachin authored 5 books including Biostatistical Methods (Wiley, Second Edition, 2011) and Randomization in Clinical Trials (co-author, Wiley, Second Edition, 2016), and over 406 papers/articles including 321 scientific/medical and over 85 technical biostatistical papers. Per Google Scholar, as of 2/7/2019, his H-Index was 81, meaning that 81 of his peer reviewed authored papers had been cited at least 81 times. His i10-Index was 180 and his papers had received 82,629 citations. The H-index of his methodology papers alone is 30.

Per a 2020 report from Stanford University, among millions of authors world-wide since 1960, the number of citations to Dr. Lachin’s work was ranked in the top 6/100th of a percent in the fields of Endocrinology & Metabolism, Clinical Medicine, and Statistics & Probability.

His recent co-authored medical papers include the demonstration that intensive treatment of type 1 diabetes leading to near normal levels of glucose can reduce mortality and the progression or retinal and renal microvascular complications during the EDIC study, the phenomenon of metabolic memory that was first reported in a landmark paper in the New England Journal of Medicine (NEJM) in 2000 for which Dr. Lachin was the lead author. He is an author of 18 papers in the NEJM.

Recent methodological publications include description of the Wei-Lachin multivariate one-directional test for multiple outcomes and its application to composite outcomes.

During his career Dr. Lachin has been awarded over $470 million from the National Institutes of Health (since 1985) to fund the research projects for which he served as the Principal or Co-Principal Investigator.

Faculty Profile

Dr. Lachin’s Detailed Biographical Sketch

Current BSC Projects

• Principal Investigator, Data Coordinating Center for the study of the Epidemiology of Diabetes Interventions and Complications (EDIC: 1996 - 2022)
• Principal Investigator, Glycemia Reduction Approaches for Diabetes: A Comparative  Effectiveness Study (GRADE: 2012-2022), both funded by the National Institute of Diabetes, Digestive and Kidney Diseases of the National Institutes of Health.

Statistical Materials for Download

This page contains links to statistical programs and materials developed by the Biostatistics Center faculty and staff. All such programs are copyrighted by the authors. The authors do not guarantee the accuracy of the methods or the programs. These programs are free software; you can redistribute them and/or modify them under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License or any later version. This software is provided "AS IS" without warranty of any kind. The entire risk as to the quality and performance of the software is with you. Please see the GNU license text for details.

Biostatistical Methods: The Assessment of Relative Risks.

• Biostatistical Methods (First Edition)
• Biostatistical Methods (Second Edition)

Wei and Lachin Multivariate Rank Analysis.

Multivariate Rank Analysis. This is a SAS macro to implement the original Wei and Lachin (1984) multivariate Mann-Whitney analysis. The Macro also provides a number of different test statistics as described in Lachin (1992). Download. In August 2013 a minor discrepancy was detected between the program documentation and descriptions and how the output of the Mann-Whitney analysis was labeled. See the document readme.doc. The program was revised to correspond with the documentation.

References:

• Wei LJ and Lachin JM. Two-sample asymptotically distribution-free tests for incomplete multivariate observations. Journal of the American Statistical Association, 1984, 79, 653-661.
• Thall PF and Lachin JM. Analysis of recurrent events: Nonparametric methods for random interval count data. Journal of the American Statistical Association, 1988, 83, 339-347.
• Lachin JM. Some large sample distribution-free estimators and tests for multivariate partially incomplete data from two populations. Statistics in Medicine, 11, 1151-1170, 1992.

Futility Assessment Based on Conditional Power

These are SAS programs to compute the properties of a futility assessment based on a conditional power computation as described in Lachin (2005). Download

Reference:

• Lachin JM. A review of methods for futility stopping based on conditional power. Statistics in Medicine, 24, 2747-2764, 2005.

Futility Assessment Based on Interim Z-test Values.

These are SAS programs to implement a futility assessment based on a boundary on the interim Z-test value or a confidence limit as described in Lachin (2009). This does not require specification of a boundary on a conditional power value although the corresponding conditional power value is readily obtained. Download

Reference:

• Lachin JM. Futility interim monitoring with control of type I and II error probabilities using the interim Z-value or confidence limit. Clinical Trials, 2009, 6:565-573.

Parametric Survival Models For Interval Censored Data With Time-Dependent Covariates.

This SAS program fits parametric regression model to interval censored event-time data with allowance for time-dependent covariates. The SAS macro and documentation are provided. Download programs, documentation, and SAS data set here.

Reference:

• Sparling YH, Younes N, Lachin JM, and Bautista OM. Parametric survival models for interval-censored data with time-dependent covariates.Biostatistics;7, 599-614, 2006.

Robust Estimating Equations (REE).

This SAS Macro implements robust estimating equations for analyzing longitudinal data based on the methods developed by Hu and Lachin (2001). It includes GEE as a special option. The Levenberg-Marquardt algorithm has been used to improve convergence rates. This macro employs the same platform as the GEE macro developed by Ulrike Groemping (1993). The SAS macro and documentation are provided. Download

Reference:

• Hu M and Lachin JM. Application of robust estimating equations to the analysis of quantitative longitudinal data, Statistics in Medicine, 20, 411-3428, 2001.

Sample Size and Power for a Logrank Test and Cox Proportional Hazards Model with Multiple Groups and Strata, or a Quantitative Covariate with Multiple Strata.

These are the SAS programs that performed the various computations presented in the Example section as described in Lachin (2013): Download.

Reference:

• Lachin JM. Sample size and power for a logrank test and Cox proportional hazards model with multiple groups and strata, or a quantitative covariate with multiple strata. Statistics in Medicine 2013; 32:4413-4425. DOI: 10.1002/sim.5839.

Data Analysis Plan and Data Analysis Policy.

An outline of a data analysis plan that can be modified and adapted for individual use is given here: Download.

An example of Data Analysis Policy is available here: Download.

Reference:

• Best Practice: The Data Analysis Plan ? A Blueprint for Success. The American Statistical Association Conference on Statistical Practice February 20, 2015, New Orleans.

Optimal Screening Schedules For Disease Progression, With Application to Diabetic Retinopathy.

This is a technical report describing the statistical models employed for the net-benefit analysis in Nathan et al. (2016): Download.

R code: Download. 

Reference:

• Nathan DM, Bebu I, Hainsworth D, Klein R, Tamborlane W, Lorenzi G, Gubitosi-Klug R, Lachin JM. Establishing Rational individualized Screening Guidelines for Retinopathy in Type 1 Diabetes, submitted for publication, 2016.