October 1, 2018: Dr. Greg Simon Uses a Pie Eating Contest Analogy to Explain the Intraclass Correlation Coefficient

In a new video, Dr. Greg Simon explains the intraclass correlation coefficient (ICC) with an analogy to a pie eating contest. The ICC is a descriptive statistic that measures the correlations among members of a group, and it is an important tool for cluster-randomized pragmatic trials because this calculation helps determine the sample size needed to detect an effect.

Greg Simon from NIH Collaboratory on Vimeo.

“When we randomize treatments by doctors, clinics, or even whole health systems, we need to think about how things cluster, and the intraclass correlation coefficient is the measure of that clustering. When we think about sample sizes in pragmatic clinical trials, it’s important to understand what an intraclass correlation coefficient actually is.”

For most pragmatic trials, the ICC will be between 0 and 1. If the outcomes in a group are completely correlated (ICC=1), then all participants within the group are likely to have the same outcome. When ICC=1, sampling one participant from the cluster is as informative as sampling the whole cluster, and many clusters will be needed to detect an effect. If there is no correlation among members of the groups (ICC=0), then the available sample size for the study is essentially the number of participants.

For more on the ICC, see the Intraclass Correlation section in the Living Textbook or this working document from the Collaboratory’s Biostatistics and Study Design Core.

July 30, 2018: Registration Open for 3rd Seattle Symposium on Health Care Data Analytics

Registration is open for the 3rd Seattle Symposium on Health Care Data Analytics. The symposium will bring together biostatisticians, health informaticists, epidemiologists, and other data scientists to discuss health research and methods that involve large health care databases.

Experts involved in national research initiatives that use large health care databases will discuss methodological challenges encountered in this setting and share ideas for addressing them. Speakers will share their research on:

  • statistical approaches to learning from electronic health care data;
  • methods for precision medicine; and
  • health policy.

Space is limited, and registration is required.

The event is sponsored by the Biostatistics Unit at Kaiser Permanente Washington Health Research Institute and the Department of Biostatistics at the University of Washington.

March 21, 2018: Dr. Rob Califf to Speak on Data Science at March 23 Grand Rounds

Robert Califf, MD, former FDA Commissioner and current Vice Chancellor for Health Data Science at Duke University School of Medicine, will present at NIH Collaboratory Grand Rounds on Friday, March 23 at 1 pm ET. The webinar will be broadcast live and is open to the public. Following the presentation, Dr. Califf will answer questions from the Grand Rounds audience.

As Director of Duke Forge, Duke’s interdisciplinary center for actionable health data science, Dr. Califf is currently working on initiatives designed to harness biostatics, machine learning, and sophisticated informatics approaches to improve health and healthcare. Dr. Califf is also an adjunct professor of medicine at Stanford University and is employed by Verily Life Sciences as a scientific advisor. Verily, part of the Alphabet (Google) family of companies, is aimed at transforming the growth of health-related data into practical applications.

Dr. Califf has been a pioneer in the fields of clinical, translational, and outcomes research, and the NIH Collaboratory looks forward to hearing his thoughts on the pragmatic applications of data that will advance health and health care strategies and practice.

Topic: Data Science in the Era of Data Ubiquity

Date: Friday, March 23, 2018, 1:00-2:00 p.m. ET

Meeting Info: To check whether you have the appropriate players installed for UCF (Universal Communications Format) rich media files, go to https://dukemed.webex.com/dukemed/systemdiagnosis.php.

To join the online meeting:
Go to https://dukemed.webex.com/dukemed/j.php?MTID=m1a4a0665a615ae0382440edecedbdd33

December 7, 2017: Dr. Greg Simon Explains Individual, Cluster, and Stepped-Wedge Randomization in a New Prop Video

In a new video in the Living Textbook, Dr. Greg Simon describes the differences between individual, cluster, and stepped-wedge randomization using props, including marbles, Play-Doh, and glassware.

“In the end, it’s all about randomly assigning who gets which treatment, or who gets which treatment when, so that we’re able to make some un-biased judgement about which treatment is really better.” —Greg Simon, MD

October 18, 2017: NIH Collaboratory Core Working Group Interviews: Reflections from the Biostatistics and Study Design Core

We recently asked Dr. Liz DeLong, Chair of the Biostatistics and Study Design Core, to reflect on the first 5 years of the Core’s work and the challenges ahead. She says the biggest impact of the Core has been working with the individual Demonstration Projects to provide a sounding board to discuss statistical challenges. Further, Core members have contributed to new knowledge through manuscripts that address key methodological issues related to pragmatic clinical trials. She’s hoping the Core will continue to push the boundaries of statistical methods in the coming years.

“The statisticians on the individual trials have not only developed excellent statistical methods for their own studies, but also contributed substantively to the Core.” Dr. Liz DeLong

Download the interview (PDF).

New Biostatistical Guidance Document Available: Small-Sample Robust Variance Correction for GEE

Tools for ResearchThe NIH Collaboratory’s Biostatistics and Study Design Core has just published a new guidance document by Andrea Cook, PhD, of the Group Health Research Institute, on using small-sample robust variance correction for generalized estimating equations (GEE) for use in cluster-randomized trials. The document, which includes guidance on methods available in the SAS and Stata statistical analysis packages, is available directly from the NIH Collaboratory Knowledge Repository here (opens as PDF), or via the Biostatistical Guidance Documents page in the Living Textbook.

This guidance document is one in a series of research tools focused on detailed aspects of statistical design for conducting pragmatic clinical trials. Each document in this series provides a synthesis of current developments, discusses possible future directions, and, where appropriate, makes recommendations for application to pragmatic clinical research.


New Biostatistical Guidance Document Available – “Frailty Models in Cluster-Randomized Trials”


Tools for ResearchThe NIH Collaboratory Biostatistics/Study Design Core has released a new guidance document concerning the use of frailty models in the setting of cluster-randomized trials (CRTs). This guidance, the fifth in a series from the Core, outlines considerations affecting power calculations in frailty models, as well as issues raised by the use of logistic regression models for time-to-event versus dichotomous outcomes in CRTs .

The guidance document can be found under Biostatistical Guidance Documents on the Tools for Research page on the Living Textbook, or accessed directly here (PDF).


Collaboratory Biostatistics and Study Design Core Releases Guidance Documents


The NIH Collaboratory’s Biostatistics and Study Design Core has released the first in a series of guidance documents focusing on statistical design issues for pragmatic clinical trials. Each of the four guidance documents are intended to help researchers by providing a synthesis of current developments in the field, discuss possible future directions, and, where appropriate, make recommendations for application to pragmatic clinical research.

The guidance documents are available through the Living Textbook and can be accessed on the “Tools for Research” tab or directly here.