What Researchers Need to Know for the Week of May 16, 2022

It’s been just a few weeks since the All of Us Research Program announced the availability of its first dataset, and researchers are already taking advantage of the program to make discoveries. This past week, we found that researchers have used that dataset to begin making predictive models to help identify patients in most need of additional resources at time of discharge. How else will the vast data from the All of Us program be used?
Elsewhere in the research community, we also discovered headlines revealing that the National Science Foundation is seeing declining demand for funding, a deep dive into the FDA’s latest guidance on diversity in clinical trials, and more.

All of Us Research Program News

All of Us research program celebrates four years at Emory

The All of Us effort has now enrolled more than 489,900 participants. In March, the program released its first genomic dataset, bringing around 100,000 whole-genome sequences onto its researcher workbench platform. Emory researchers played a significant role in this milestone, nurturing relationships with more than 7,000 diverse participants throughout metro Atlanta to ensure strong diversity in the dataset. Emory University, part of the SouthEast Enrollment Center (SEEC) of All of Us, has now enrolled nearly 40,000 collective participants since the launch.

Social determinants of health may help predict sepsis readmission

Researchers from the University of California, San Diego in a recent study where they used data from the National Institutes of Health’s All of Us research program cohort found that factors such as lack of insurance, economic instability, and poor transportation to obtain care were associated with 30-day readmission. They also found that including various social determinants of health improved the model’s ability to predict which sepsis patients are at risk and which patients may benefit from additional resources around the time of discharge, or post-discharge.

Diversity in Research

Five Things Researchers Need to Know about the FDA’s Updated Diversity Plans Guidance

The FDA released an updated guidance recently, and this is just one more in a number of recent government actions that point to a broader priority of improving diversity in health, and health research in particular.
Though the current FDA guidance is focused on clinical trials, researchers involved in longitudinal studies or observational studies, among other areas outside of clinical trials should still take notice. There are five key themes ranging from who the FDA’s guidance is currently targeting to how to be compliant – all covered in this article.

Research Funding

Odds improve for winning NSF grants, but drop in applications troubles some observers

A new report from the National Science Foundation (NSF) on its merit review system reveals that reduced demand has boosted success rates for research applications at the government agency. Changes in the biology directorate are the most remarkable. Demand for funding in biology has dropped by 50% over the past decade, while the odds of securing a grant for funding have doubled, from 18% to 36%.

AI & Machine Learning in Research

As problems emerge in medical AI models, research points to a new way to strip them of bias

After an explosion of excitement about the potential for machine learning in medicine, cracks in the foundation are emerging. More and more researchers are focusing on the ways that medical models can introduce algorithmic bias into health care. But in a new paper, machine learning researchers caution that such self-reflection is often ad hoc and incomplete. They argue that to get “an unbiased judgment of AI bias,” there needs to be a more routine and robust ways of analyzing how well algorithms perform. Without a standardized process, researchers will only find the bias they think to look for.