Trial by Data Podcast: From Government to Academia, Where is Clinical Research Working?
Featuring Dr. Warren Kibbe of the Duke Cancer Institute and Dr. Sam Volchenboum
This week we welcome Dr. Warren Kibbe to Trial by Data to discuss making change in government, the dangers of inaccessible data, and the predictive abilities of wearables when it comes to COVID-19.
Dr. Kibbe is chief for Translational Biomedical Informatics in the Department of Biostatistics and Bioinformatics and chief data officer for the Duke Cancer Institute.
Prior to joining Duke, he served as the acting deputy director of the National Cancer Institute (NCI) and director of the NCI’s Center for Biomedical Informatics and Information Technology.
At the NCI, his responsibilities included defining and implementing the NCI Data Commons, a vision for a National Cancer Research Data Ecosystem. He was instrumental in establishing the NCI’s partnership with the U.S. Department of Energy (DOE), aimed at defining and developing the next generation of high performance computing architectures to address important questions in cancer biology, treatment, development of resistance in patients, and understanding outcomes.
Most recently, Dr. Kibbe played a pivotal role in coordinating the data sharing scientific components of the Cancer Moonshot initiative.
You can find him on Twitter here.
You can listen to the full episode, and others, on iTunes, SoundCloud, and Stitcher. We appreciate your tuning in.
Listen and subscribe here: Trial by Data, presented by Litmus Health.
Each episode, we pull out some of the key themes of the conversation for our listeners.
Here are the highlights from our conversation with Dr. Kibbe:
Misinformation is on the rise. Study after study shows that misinformation moves faster than the real news does, particularly on social media. This is a big problem for government and academia alike, as misinformation continuously undermines trust in previously trusted institutions. Academics and these other institutions need to earn the trust of the public back. We need to be dedicated to making sure good information is out ahead faster than bad information. We also need to consider how to properly mix policy incentives, so that the public has access to information they can really use. Finally, as researchers and leaders in the field of science, we need to be committed to communication of our work. We need to be able to explain that misinformation is an outlier, and that there are approaches we’re taking to ensure it doesn’t become the norm.
Referenced:
The current academic system promotes bad data practices. From data hoarding to data siloing, no one in academia today gets promoted because they shared data better. Not only is this problematic for the culture of science, but it can have broader impacts on areas of public health. For example, bad data practices were one of the main problems around hydroxychloroquine research that people had already come to believe. The eventual retractions of the hydroxychloroquine were due to a lack of availability data for third-party review.
Fortunately, data practices are starting to change little by little. For example, Duke recently had a revaluation of their promotion criteria, and many of the faculty members fought to ensure that contributions to open science and data sharing were all explicitly included in guidelines for promotion. On the whole, all of us in academia and research could stand to learn from the genomics community, which has really embraced a culture of data-sharing.
Referenced:
Purdue Pharma’s settlement is pretty small when you consider how much it cost the US economy, big picture. So how should readers feel about the way the case has unfolded? The opioid crisis has cost the US economy about two trillion dollars from services, care, and the huge loss of life over the last decade. Among the many states themselves that are involved in the case, there’s still divided opinion. Some have accepted the deal and others feel it is unfair.
Referenced:
Previous research on wearables is now being brought to bear on COVID. At Duke, researcher Jessilyn Dunn had been leveraging different kinds of biosensors and wearables to see whether she could identify illness before people even realized they were sick by recognizing minor and unconscious flu and flu-like symptoms. The project was so successful that the work is now being applied to COVID through project CovIdentify. Once you sign up, the project looks at both new heart rate and accelerometer data, as well as historical data from your device, to determine a baseline of your health. Litmus has seen similar success in predicting health in our Crohn’s and ulcerative colitis study with Dr. Rubin and Takeda Pharmaceuticals. Similarly to CovIdentify’s findings, Litmus found in this study that heart rate was a dead giveaway for flares.
Referenced:
- Duke COVIDENTIFY
- Dr. David Rubin Discusses Initial Results of Takeda-Sponsored and Litmus-Powered Research at DDW
The provenance of sensors is important. Many device manufacturers are late to the game in terms of validating their sensors for clinical research. In fact, many device manufacturers’ sensors come from any number of manufacturers, usually in China, and they’re not necessarily sourced back to the original manufacturer. This is an issue because if we’re going to use these devices to make medical claims, then we need to get a better grasp on the lineage of where these sensors come from.
Trial by Data, presented by Litmus Health, is a podcast exploring the data-driven technologies and strategies shaping the future of clinical trials. We cover the most pressing issues and questions facing researchers and clinicians today, in an ever-changing landscape. Listen in as we interview leaders and innovators in the field who are at the forefront of developing and using these data-driven approaches.
