ECHO Study Outlines How Researchers Can Combine Datasets With Different Confounders*

Author(s): Ghassan B. Hamra, Bryan Lau, Catherine Lesko, Jessie Buckley, Daniel Tancredi, Irva Hertz-Picciotto, Elizabeth Jensen

*Confounders are factors in a study that can lead to bias. These factors must be adjusted so researchers can accurately understand the information.

Who sponsored this study?

This research was supported by the Environmental influences on Child Health Outcomes (ECHO) program, Office of The Director, National Institutes of Health.

What were the study results?

Most of the time, the estimators showed similar numeric values of the estimates. The exception is when using a model known as logistic regression, which did not give similar estimates. Logistic regression provides an odds ratio, which is a non-collapsible quantity, or one that cannot be reliably combined.

What was the study's impact?

This study can help researchers understand and combine information across ECHO groups. When the same confounders are not available, or when study designs are different, researchers should avoid logistic regression. Other estimators provide estimates that can be reliably combined.

Why was this study needed?

A type of graph, called Directed Acyclic Graphs (DAG), guide decisions on choosing confounders that need to be adjusted. These graphs also suggest how to get unconfounded (unbiased) effect estimates. These estimates are based on different statistical models and do not always provide the same actual numeric values. The research team explored when they do and do not correspond.

Who was involved?

Researchers from the John Hopkins University Data Analysis Center (JHU DAC), Wake Forest University, and University of California Davis joined together to explore this problem.

What happened during the study?

All study data are simulated, and Dr. Hamra at the JHU DAC built all of the models. Dr. Hamra also looked at all of the information.

What happens next?

Researchers can make more models to explore other forms of bias that could make combining information across studies difficult.

Where can I learn more?

Researchers can get the modeling code to look into this issue if they choose to in the journal article, titled “Combining Effect Estimates Across Cohorts and Sufficient Adjustment Sets for Collaborative Research: A Simulation Study” in Epidemiology.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Published: May 1, 2021

Download this summary

ECHO Study Outlines How Researchers Can Combine Datasets With Different Confounders*

Best Practices for Conducting Clinical Trials with Indigenous Children in the United States

<< Back to Research Summaries

Best Practices for Conducting Clinical Trials with Indigenous Children in the United States

Author(s): Jennifer Shaw, Erin Semmens, May Okihiro, Johnnye Lewis, Matthew Hirschfeld, Timothy VanWagoner, Lancer Stephens, David Easa, Judith Ross, Niki Graham, Sara Watson, Edgardo Szyld, Denise Dillard, Lee Pyles, Paul Darden, John Carlson, Paul Smith, Russell McCulloh, Jessica Snowden, Sarah Adeky, Rosalyn Singleton

What was done?

The authors describe key ethical issues around conducting trials with Indigenous children. They review four case studies and provide guidance for conducting clinical trials involving Indigenous children.

What was found?

Based on their experience and a review of existing literature, the authors make three main recommendations for researchers conducting clinical trials involving Indigenous children:

Engage with Indigenous communities early and over the long-term to build trust and shared goals
Build capacity among Indigenous communities for leading and partnering on research studies
Support Indigenous community ownership of data and oversight of research conducted with Indigenous children

What do the results mean?

Clinical trials are needed to build evidence for child health interventions. Indigenous children must be included in clinical trials to reduce health disparities and improve health outcomes in these populations. These studies should be done in partnership with communities using established practices of community-engaged research.

Why was this study conducted?

The United States (US) population includes nearly 7 million Indigenous people, including:

5 million American Indian and Alaska Native (AI/AN) people, and
5 million Native Hawaiian and other Pacific Island people.

Indigenous people in the US have lower life expectancies and higher disease burdens than other groups. Indigenous children have high rates of health conditions, such as asthma, obesity, and respiratory infections, compared to the general population. Few pediatric clinical trials have included Indigenous children. However, many of these children live in rural communities where interventions are often most needed. Children can respond to medicines and other health interventions differently based on their backgrounds. This paper highlights the reasons why Indigenous children may be excluded from trials and offers suggestions for improvement.

Appreciation:

The authors thank the Environmental influences on Child Health Outcomes (ECHO) program, the Office of the Director, National Institutes of Health, for supporting this research.
You can read the full publication here: https://ajph.aphapublications.org/doi/10.2105/AJPH.2021.306372

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Download this summary

Best Practices for Conducting Clinical Trials with Indigenous Children in the United States

ECHO Review Explores Statistical Approaches for Investigating Periods of Susceptibility in Children’s Environmental Health Research

<< Back to Research Summaries

ECHO Review Explores Statistical Approaches for Investigating Periods of Susceptibility in Children's Environmental Health Research

Author(s): Jessie Buckley, Ghassan Hamra, and Joseph Braun

Who sponsored this study?

This research was supported by the Environmental influences on Child Health Outcomes (ECHO) program, Office of The Director, National Institutes of Health.

What were the study results?

We found that there are many different ways to study this topic. However, several new ways stand out as more advanced.*

*Results reported here are for a single study. Other or future studies may provide new information or different results. You should not make changes to your health without first consulting your healthcare professional.

What was the study's impact?

Learning about the ages that children are most likely to be affected by their environment is important. Knowing that information will help create programs, health practices, and policies that may help children better avoid things in their environment that can have a bad effect on them. This study finds recent improvements in ways of looking at which ages children are most affected. It also explains terms about this topic and why we need ways to study it.

Why was this study needed?

Many researchers are interested in studying the ages that children are most likely to be affected by environmental factors. There are many ways to study this topic, so we looked at several different ways to decide which ones were the best.

Who was involved?

There were no participants involved in this review. This is because we looked at different ways to use math to learn more information about a topic.

What happened during the study?

During this study, we reviewed different ways to study the age ranges that children are most likely to see effects from the environment around them. Environment is not just the outdoors, but other things about a child’s life, such as where they live, their family, what they eat, and more.

What happens next?

Our team will use this information to improve the design of future studies.

Where can I learn more?

Access the full journal article titled, “Statistical Approaches for Investigating Periods of Susceptibility in Children's Environmental Health Research.”

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Published: March 2019

Download this summary

ECHO Review Explores Statistical Approaches for Investigating Periods of Susceptibility in Children's Environmental Health Research

ECHO DISCOVERY

Co-Author Jessie Buckley presented Estimating effects of exposure mixtures on child health: Novel methods for solution-oriented ECHO research at a past ECHO Discovery webinar. You can view the presentation here.

MORE RESEARCH BY JESSIE BUCKLEY

Review of Prenatal Air Pollution Exposure and Brain Development

Author(s): Heather E. Volk, Frederica Perera, Joseph M. Braun, Samantha L. Kingsley, Kim Gray, Jessie Buckley, Jane E. Clougherty, Lisa A. Croen, Brenda Eskenazi, Megan Herting, Allan C. Just, Itai Kloog, Amy Margolis, Leslie A. McClure, Rachel Miller, Sarah Levine, Rosalind Wright

Training as an Intervention to Decrease Medical Record Abstraction Errors Multicenter Studies

<< Back to Research Summaries

Training as an Intervention to Decrease Medical Record Abstraction Errors Multicenter Studies

Author(s): Meredith Zozus, Leslie Young, Alan Simon, et al

What was done?

MRA training was delivered at the beginning of the study. The training consisted of a teaching session using an example case abstraction, followed by each trainee independently abstracting two test cases. Sixty-nine abstractors from 30 sites received the MRA training. The goal for each abstractor was to achieve an error rate no greater than 4.93%.

What was found?

Only 23% of the abstractors met their error rate goal during the training.

What do the results mean?

Study-specific MRA training can improve the quality of study data. This project had several problems. Creating the training test cases took a lot of time and effort. When the training test cases were created, they contained errors that were not discovered before training began. These errors were distracting and confusing to the abstractors. Lastly, there are many different EMR systems and it is impossible to train abstractors on all of them.

Why was this study conducted?

Searching medical records to find data for another use is called medical record abstraction (MRA). The process is prone to errors, and many people question the quality of the data. This project trained abstractors for a study that used MRA as the main source of data.

Appreciation:

The authors would like to thank the Eunice Kennedy Shriver National Institute of Child Health and Human Development and the National Institutes of Health for their support of this research.

You may learn more about this publication here: https://www.ncbi.nlm.nih.gov/pubmed/30741251

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Download this summary

Training as an Intervention to Decrease Medical Record Abstraction Errors Multicenter Studies