Recently, we’ve been running an ‘open science stars‘ series to highlight a range of great people from around the world working to advance open science practices. This week, we have something a little special for you. All previous interviews have been with students or researchers, but this story is from a physician in the United States Navy, Commander Jean-Paul Chretien! So sit back and enjoy the show.
Hi Commander Jean-Paul! For starters could you let us know a little about your background?
Thank you for interviewing me. Let me say first that throughout this interview I’m expressing my own views, not necessarily the official policy or position of the Department of Defense, Defense Health Agency, or US Government.
I’m a physician in the United States Navy, and my training is in public health, epidemiology, and informatics. I work on challenges at the intersection of health and national security, like infectious disease outbreaks and climate change.
I was drawn first to the military, before medicine, but I knew what life as a doctor is like because my parents are physicians. I wanted to be a military officer from a pretty young age. Service to country, the chance to lead, the adventure – all of that appealed to me. For college I went to the U.S. Naval Academy in Annapolis, Maryland, thinking maybe I would command a warship someday. But while I was there, studying international affairs and national security, I learned that some of the most pressing security challenges were health problems like HIV/AIDS, at the time. And I learned that in many battles and wars, diseases crippled military forces and civilian populations in war zones. Infections often caused more casualties than combat.
So I decided to go to medical school, but not to be a doctor practicing in a clinic. I wanted to be a doctor for populations, and bring medical knowledge to decisions that impact military service members, the broader American public, and, well, everyone.
When did you first hear about Open Access and Open Science? What were your first thoughts? Has there ever been a case where lack of access to information has seriously compromised your work?
When I was a student working on my MD and my PhD in epidemiology, I didn’t think about Open Access because access wasn’t a problem for me. Through my university, I could access just about any journal article I needed. But later, when I began my global health work in the U.S. military, I saw how access restrictions constrained biomedical research, patient care, and population health around the world.
At that time, I worked in a Department of Defense program that partners with dozens of countries to improve their capabilities for detecting and containing epidemics. I had collaborators around the world, public health personnel and researchers in countries with limited resources, who could not read about studies on the diseases that burden them. How can they join the global effort against infectious disease outbreaks if they can’t always access the most current and best research on those diseases?
When I began my global health work in the U.S. military, I saw how access restrictions constrained biomedical research, patient care, and population health around the world.
Then, what galvanized my commitment to open access and open science in general was the Ebola outbreak that began in West Africa in late 2013, and spread to Europe and the U.S. It’s waning now, but infections are still occurring. There have been more than 28,000 confirmed cases with around 11,000 deaths, by far the largest Ebola epidemic ever.
How can [people] join the global effort against infectious disease outbreaks if they can’t always access the most current and best research on those diseases?
When disease transmission was intense, data was essential for an effective response – public health agencies needed to know, in as close to real time as possible, where cases were occurring, what the presenting clinical symptoms were, how transmission was occurring through contact networks, where containment was succeeding and failing, and whether the virus was changing.
Many health workers and scientists participated in the response, and many shared the data they collected freely. But some did not. Journal articles would appear with months-old data, which might have been more useful earlier. Colleagues and I assessed this publication lag and data-sharing practices as part of a broader review of the Ebola outbreak, which we published in eLife, an open access journal (https://elifesciences.org/content/4/e09186v2).
As the outbreak was winding down, in September of 2015, I was invited to join a consultation convened by the World Health Organization on data-sharing during public health emergencies. The WHO brought in representatives from the biomedical research community, biomedical journals, pharmaceutical companies, funding agencies, and governments to advance new norms for rapid data release. Participants agreed that sharing data before publication should be the norm during public health emergencies (http://www.who.int/medicines/ebola-treatment/blueprint_phe_data-share-results/en/). I endorsed this view as an individual; I was not asked to nor was I authorized to consider it on behalf of the Department of Defense.
How important is it to share data and methods openly and rapidly? Do you think views on this differ between the military and institutionalised academic research?
Sharing data and methods as openly and rapidly as possible is critical during public health emergencies like infectious disease outbreaks. We saw this with Ebola; groups that shared data publicly enabled many more studies. In our review, 80 percent of published studies on Ebola transmission dynamics used only data that other researchers had already posted online. We see the importance of data-sharing now with the Zika virus outbreak. There are scientists around the world who are willing to contribute their expertise, and can provide important new insights for the response effort, if they’re able to access data that others are collecting.
Sharing data and methods as openly and rapidly as possible is critical during public health emergencies like infectious disease outbreaks. We saw this with Ebola
With biomedical research and public health programs in the military, we see similar successes and challenges in open data sharing. We do have the additional consideration of operational security – for some efforts, we must think carefully about whether releasing the data publicly could compromise a mission or national security more broadly. But across the Department of Defense, agencies are looking for opportunities to release data appropriately. This is in line with the President’s Executive Order in 2013 making open and machine-readable data the new default for government information (https://www.whitehouse.gov/the-press-office/2013/05/09/executive-order-making-open-and-machine-readable-new-default-government-).
Why do you think some research communities are resistant to the idea of openly sharing data and methods?
It’s understandable that research groups collecting data may be hesitant to release it openly. One common concern is that others will not understand the data as well as they do, and as a result might commit analytical or inferential errors. Investigators who collect data could reduce this risk by annotating their data well and providing appropriate metadata. Often, scientists are accustomed to recording and storing data in a way that makes sense to them, but may not include the additional information that others would need to use it appropriately.
And there’s long been a concern that pre-publication data release could compromise future peer-reviewed publication. But I think we’re seeing a shift in biomedical research, with some pr
ominent journals now stating clearly that pre-publication data release will not prejudice later consideration for publication. For emergency situations in particular, pre-publication data release is the expectation among some journal editors.
My personal view is that investigators should be expected to deposit all supporting data in a public archive once it’s appropriately de-identified and quality-assured (again, a personal opinion, not an institutional view from Department of Defense). If more and more within the research community do this and demand it of others, I think it will become the norm.
Often, scientists are accustomed to recording and storing data in a way that makes sense to them, but may not include the additional information that others would need to use it appropriately.
Any thoughts on the Wellcome Trust’s commitment to open data sharing in wake of the Zika crisis? Why do you think this is emphasises in times of health crisis, instead of being the default for all research?
The broad coalition that endorsed the Zika data-sharing commitment is a major advance for open science. It shows that norms are changing in the biomedical research community, and that, more and more, open sharing is the default.
True, the move to open data is most evident for public health emergencies. It makes sense to focus there initially, since the need is pressing and the case for openness compelling. But I think we can build on the emerging consensus around openness to reduce obstacles to data-sharing in biomedical research more generally.
How can ‘open advocates’, policymakers, and researchers all come together to promote a culture of open sharing?
We need to credit researchers who share data and honor data-sharing as a scientific contribution as significant as original research. At least as significant as original research, since sharing enables others to confirm previous results or make new discoveries.
We need to credit researchers who share data and honor data-sharing as a scientific contribution as significant as original research.
For this to happen, we need to be able to track use of shared datasets. For example, as a first step, journals could insist that authors who use external datasets cite them in a standardized and discoverable way, and not simply thank the data providers in the Acknowledgements section, which often happens. At the same time, researchers who develop and share datasets could identify them in a standard way, like with digital object identifiers. With agreed approaches to citing and identifying datasets, we could envision a system that comprehensively tracks use of open datasets, and supports emerging metrics around data-sharing.
This is just one small, possible part of a better approach. For this and other open data initiatives, we must engage all of the stakeholders, including academic departments, which could decide that data-sharing is a scientific contribution and consider it in faculty evaluations; journals and research sponsors, which could use their leverage to push for openness; and the public at large, which, after all, should be the primary beneficiary of biomedical research.
You work for the Armed Forces Health Surveillance Branch of the Public Health Division within the Defense Health Agency. What role does open data play in your daily operations?
Our mission at AFHSB is to provide timely, relevant, actionable, and comprehensive health surveillance information to promote, maintain, and enhance the health of military and military-associated populations. Some of the data we use to accomplish this mission cannot be open, because of privacy or security constraints. For example, we maintain a very large database called the Defense Medical Surveillance System that integrates data from the Department of Defense electronic medical record and other deployment, clinical and demographic databases which contain personal identifying information. AFHSB analysts continually use these data to assess possible health threats to our population.
But we also look beyond the military, at disease outbreaks and other health threats around the world that could impact the Department of Defense. We make extensive use of open data to do this. Our Alert and Response Operations Team, for example, scans many online, open source types of information sources daily, and combines open information with information shared in interagency and other professional networks.
What advice would you give to students looking to take their first steps into a research career?
Find a good mentor! In my mind, this is someone who is accomplished in an area that interests you, even if it’s not exactly what you had envisioned doing; is passionate about their work, and sees it in a broader context beyond the immediate research aims; and will be generous in devoting time to you.
The views expressed are my own, and do not necessarily represent the views of any part of the U.S. Government.
Thank you so much for your insight, Jean-Paul! I hope people find the military perspective on this as fascinating as we have!
Jean-Paul Chretien is a Commander in the US Navy. He currently leads the Innovation & Evaluation Team at the Armed Forces Health Surveillance Branch, Defense Health Agency, where he focuses on development of biosurveillance systems, epidemiological prediction models, and climate change impacts on health. Jean-Paul co-chairs the White House Office of Science and Technology Policy working group on Pandemic Prediction and Forecasting; is Department of Defense representative to the interagency Crosscutting Group on Climate Change and Human Health; and serves on the Board on Environment and Health of the American Meteorological Society. He is a Life Member in the Council on Foreign Relations.
Previously, he served with the US Marines in Helmand Province, Afghanistan, as the commanding general’s senior public health adviser (2011-2012); and coordinated international public health partnerships for the DoD Global Emerging Infections Surveillance and Response System (2004-2008).
Jean-Paul graduated with a BS in Political Science from the US Naval Academy, where he was Truman Scholar. He received an MD from the Johns Hopkins University School of Medicine and PhD (Epidemiology) and MHS (Biostatistics) from the Johns Hopkins Bloomberg School of Public Health. He completed a post-doctoral fellowship in Health Science Informatics at the Johns Hopkins University School of Medicine, and residency in General Preventive Medicine at the Walter Reed Army Institute of Research. He is Board-Certified in Preventive Medicine.
Jean-Paul’s awards include the William Kane Rising Star Award from the American College of Preventive Medicine; Best Publication of the Year from the International Society for Disease Surveillance; and Superior Technology Transfer Award from the US Department of Agriculture. He has published more than 40 peer-reviewed articles and book chapters. (source)