SAS Macro for Constrained Randomization: Balancing covariates in Group Randomized Trials Ashraf Chaudhary, Ph.D. & Larry Moulton, Ph.D. Department of International.

Slides:



Advertisements
Similar presentations
High Resolution studies
Advertisements

If you are viewing this slideshow within a browser window, select File/Save as… from the toolbar and save the slideshow to your computer, then open it.
FDA/Industry Workshop September, 19, 2003 Johnson & Johnson Pharmaceutical Research and Development L.L.C. 1 Uses and Abuses of (Adaptive) Randomization:
Analysis by design Statistics is involved in the analysis of data generated from an experiment. It is essential to spend time and effort in advance to.
When Should a Clinical Trial Design with Pre-Stratification be Used? Group 1.
Designing an impact evaluation: Randomization, statistical power, and some more fun…
Polynomial Curve Fitting BITS C464/BITS F464 Navneet Goyal Department of Computer Science, BITS-Pilani, Pilani Campus, India.
1 Chapter 4 Experiments with Blocking Factors The Randomized Complete Block Design Nuisance factor: a design factor that probably has an effect.
Chapter 4 Randomized Blocks, Latin Squares, and Related Designs
Stratification (Blocking) Grouping similar experimental units together and assigning different treatments within such groups of experimental units A technique.
STA305 week 31 Assessing Model Adequacy A number of assumptions were made about the model, and these need to be verified in order to use the model for.
KINE 4565: The epidemiology of injury prevention Randomized controlled trials.
2005 Hopkins Epi-Biostat Summer Institute1 Module 2: Bayesian Hierarchical Models Francesca Dominici Michael Griswold The Johns Hopkins University Bloomberg.
FIELD METHODS Strategy for Monitoring Post-fire Rehabilitation Treatments Troy Wirth and David Pyke USGS – Biological Resources Division Forest and Rangeland.
Estimation and Reporting of Heterogeneity of Treatment Effects in Observational Comparative Effectiveness Research Prepared for: Agency for Healthcare.
When Is Stratification Detrimental to a Clinical Trial Design? Part II Katherine L. Monti, Ph.D. Senior Statistical Scientist and Director of the Massachusetts.
Longitudinal Experiments Larry V. Hedges Northwestern University Prepared for the IES Summer Research Training Institute July 28, 2010.
Statistics: The Science of Learning from Data Data Collection Data Analysis Interpretation Prediction  Take Action W.E. Deming “The value of statistics.
SAMPLING AND STATISTICAL POWER Erich Battistin Kinnon Scott Erich Battistin Kinnon Scott University of Padua DECRG, World Bank University of Padua DECRG,
1 Terminating Statistical Analysis By Dr. Jason Merrick.
Motive Konza: understanding disease, since there is no apparent reason to manage native pathogens of native plants Also have background information in.
July, 2014 HQ Sampling. AGENDA 1. A brief Overview of Sampling 2. Types of Random Sampling Simple Random and Systematic Random 3. Types of Probability.
Research Designs for Complex Community Interventions for Childhood Obesity Prevention Robert W. Jeffery, Ph.D. Division of Epidemiology University of Minnesota.
Clinical Trials 2015 Practical Session 1. Q1: List three parameters (quantities) necessary for the determination of sample size (n) for a Phase III clinical.
The Mimix Command Reference Based Multiple Imputation For Sensitivity Analysis of Longitudinal Trials with Protocol Deviation Suzie Cro EMERGE.
Chapter 19 Confidence Interval for a Single Proportion.
Single-Factor Experimental Designs
EDRN Approaches to Biomarker Validation DMCC Statisticians Fred Hutchinson Cancer Research Center Margaret Pepe Ziding Feng, Mark Thornquist, Yingye Zheng,
Biostatistics Case Studies 2007 Peter D. Christenson Biostatistician Session 3: Incomplete Data in Longitudinal Studies.
Estimating Incremental Cost- Effectiveness Ratios from Cluster Randomized Intervention Trials M. Ashraf Chaudhary & M. Shoukri.
Biostatistics Case Studies 2008 Peter D. Christenson Biostatistician Session 5: Choices for Longitudinal Data Analysis.
Sample Size In Clinical Trials In The Name Of God Sample size in clinical trials.
Adaptive randomization
1 THE ROLE OF COVARIATES IN CLINICAL TRIALS ANALYSES Ralph B. D’Agostino, Sr., PhD Boston University FDA ODAC March 13, 2006.
Statistics for clinicians Biostatistics course by Kevin E. Kip, Ph.D., FAHA Professor and Executive Director, Research Center University of South Florida,
Do What Needs to Be Done Today. The secret of happy successful living is to do what needs to be done now, and not worry about the past or the future.
Medical Statistics as a science
Can Mental Health Services Reduce Juvenile Justice Involvement? Non-Experimental Evidence E. Michael Foster School of Public Health, University of North.
Development and the Role of Meta- analysis on the Topic of Inflammation Donald S. Likosky, Ph.D.
Vamsi Sundus Shawnalee. “Data collected under different conditions (i.e. treatments)  whether the conditions are different from each other and […] how.
Multi-level Models Summer Institute 2005 Francesca Dominici Michael Griswold The Johns Hopkins University Bloomberg School of Public Health.
CREATE Biostatistics Core Functions Summary of activities Challenges / coming year’s activities.
Sample Size Determination
Penn CTSI Research Seminar Clinical Trials November 10, 2015 Vernon M. Chinchilli, PhD Distinguished Professor and Chair Department of Public Health Sciences.
Analysis of Experiments
Multiple Imputation using SAS Don Miller 812 Oswald Tower
Biostatistics Case Studies Peter D. Christenson Biostatistician Session 3: Missing Data in Longitudinal Studies.
Karl W Broman Department of Biostatistics Johns Hopkins Bloomberg School of Public Health What is regression?
Constrained randomisation: some applications. Tanzania: background CRT of impact of adolescent sexual health intervention on knowledge, behaviour and.
Language Testing How to make multiple choice test.
Rerandomization to Improve Covariate Balance in Randomized Experiments Kari Lock Harvard Statistics Advisor: Don Rubin 4/28/11.
Rerandomization in Randomized Experiments Kari Lock and Don Rubin Harvard University JSM 2010.
Sampling Concepts Nursing Research. Population  Population the group you are ultimately interested in knowing more about “entire aggregation of cases.
The parametric g-formula and inverse probability weighting
Statistical Considerations University of Colorado Pragmatic TRIALS L. Miriam Dickinson, PhD.
Welcome Clinical Trials October 11, 2016 Vernon M. Chinchilli, PhD
Applied Biostatistics: Lecture 4
Analysis for Designs with Assignment of Both Clusters and Individuals
Sample Size Determination
Lecture 18 Matched Case Control Studies
Simon Thompson University of Cambridge
The 4th ICTMC & 38th Annual Meeting of SCT
اختر أي شخصية واجعلها تطير!
Chair and Discussant: Karla Hemming University of Birmingham
Another Example Consider a very popular computer game played by millions of people all over the world. The average score of the game is known to be
Rerandomization to Improve Baseline Balance in Educational Experiments
Chapter 3 Hernán & Robins Observational Studies
Yu Du, PhD Research Scientist Eli Lilly and Company
Proportion difference and confidence interval based on CMH test in stratified RCT with an example in pooled analysis of HIV trials Jacob Gong.
Presentation transcript:

SAS Macro for Constrained Randomization: Balancing covariates in Group Randomized Trials Ashraf Chaudhary, Ph.D. & Larry Moulton, Ph.D. Department of International Health Division of Disease Control and Prevention Johns Hopkins University Bloomberg School of Public Health

July 7-8, 2004 Biostatistics Core Meeting: LSHTM London2 Why Constrained Randomization? In individually randomized designs, larger sample sizes ensure balance on key variables between the trial arms. Group randomized trials are typically small with perhaps only 4-20 groups to be randomized. The groups are usually contiguous and more homogenous relevant to the groups farther apart. Spatial correlation patterns are more difficult to detect in human communities.

July 7-8, 2004 Biostatistics Core Meeting: LSHTM London3 Why Constrained Randomization? Group level randomization may lead to substantial imbalance across the trial arms. Group randomized trials are therefore susceptible to the ill effects of an ‘unlucky’ or ‘bad’ randomization. But the question here is how to randomize a small number of groups so as to avoid an ‘unlucky’ or ‘bad’ randomization.

July 7-8, 2004 Biostatistics Core Meeting: LSHTM London4 Covariate-Based Constrained Randomization Randomizing the groups to, say, ‘intervention’ and ‘control’ study arms so as to achieve a balance on some baseline covariates between the trial arms.

July 7-8, 2004 Biostatistics Core Meeting: LSHTM London5 SAS Macro - Steps Generates all possible randomizations by forming combinations of groups in each stratum. Computes means of covariates for each randomization in each arm and combine the data for the two arms. Shortlists the randomizations that satisfy balancing criteria. Generates a large number of samples, say, 100, by picking one randomization ‘randomly’ from each stratum. Retains only those samples that meet the sample level balancing criteria. As a check, counts the number of times a group appears with another group in the same study arm. Selects one randomization at random from all the short listed samples.

July 7-8, 2004 Biostatistics Core Meeting: LSHTM London6 SAS Macro - Inputs SAS dataset with the following variables s: Stratum ID group: Group ID x1, x2, x3, …: Covariates r: Number to be randomized to, say, study arm 1 SAS Macro parameters Number of covariates Names of covariates Randomization level minimum acceptable differences between treatment arms for each covariate Overall sample level minimum acceptable differences between treatment arms for each covariate Seed for random selection.

July 7-8, 2004 Biostatistics Core Meeting: LSHTM London7