Commentary- Dr. James K. Gilman, Clinical Center CEO
As my “use by” date fast approaches I decided that I wanted to see if I could get OCMR to publish one more op-ed in the CC Newsletter. They have been indulgent to now and there is more I would like to say. This piece requires a bit of stage setting and the events come from my time in the military over twenty years ago. However, if you can stay with me through the stage setting, I will do the best I can to make any points applicable to the contemporaneous setting of the NIH. All names are excluded. However, this all really happened and will be reported to the best of my ability to remember.
The year was 2003 and a young soldier died in Iraq. The death was not attributed to battle injury or an accident. The Armed Forces Medical Examiner conducted an autopsy and ruled that the death was due to heat injury. The soldier’s remains were returned to his parents and, as in all cases where deaths occur during deployment, the family was given the choice of having a flag officer (generals and admirals all have both stars and flags) preside over the military honors of the funeral and burial. The flag officer in attendance at the soldier’s funeral got an earful from the soldier’s parents. The Army had not taken very good care of their son. They were particularly irked at the Army’s medical community.
What’s so important about the death of one soldier? Remember that Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF) were the first extended conflicts where the military had to continue to recruit soldiers, sailors, airmen, and marines during the conflicts. OIF and OEF were the first real tests of the all-volunteer force. Gulf War I did not last long enough to provide much of a test. The need to continue to recruit during conflict coupled with the 24-hour news cycle put situations like this one in a strategic context. The Army needed to respond to the concerns of this soldier’s family and to do it as quickly as possible. Public trust is one of the requirements for fulfilling the recruiting mission.
The flag officer presiding over the funeral contacted the Army Office of the Surgeon General. Non-commissioned officers from the soldier’s unit in Iraq were dispatched back to meet with the parents and a decision was made that I should accompany them. I tried to find out as much about the soldier and what happened in Iraq as I could before making the trip. I mean, data is important! I contacted the regimental surgeon and spoke with her about her impressions. I got my hands on the soldier’s medical records and the autopsy report. I managed to get some information about the soldier’s service record and determined he was a pretty average soldier.
The trip to see the family with members of the soldier’s unit was memorable. I had to spend the night in an airport and take off three times to land at the smallish regional airport once to link up with the representatives of the soldier’s unit. I had to clean up, shave, and don my Class A uniform in the men’s room of a small temporary terminal building. I finally met the members of the soldier’s unit and then began the 4- or 5-hour journey across the state to the family’s home. The home was far off any beaten path. While the culture and family tradition included a deep appreciation for military service, there was a palpable mistrust for almost all other things connected with the government.
The 3-hour meeting with the soldier’s parents was among the tensest of my professional career. Besides the family, a local physician was also present. While that might seem like a good thing, this physician was feeding the mistrust of the family. He specifically was critical of the military’s vaccination program for soldiers about to deploy. He led the family to believe that their son’s death might be attributable to one of the vaccines he was required to take.
I continued to gather all the information I could about the soldier and his family. For instance, I learned that one of the soldier’s brothers died in a traffic accident. From four children, the family was down to two. I spent a lot of time learning more about the Army’s pre-deployment vaccine policy. I found out that serious reactions that might be attributed to the vaccine could be referred to the ACIP – the Advisory Committee on Immunization Practices for a review that would not be influenced by anyone in the Department of Defense. I asked the program manager for the Army’s vaccine program, an extremely intelligent and thoughtful pharmacy officer, if we could send the soldier’s records to ACIP for review. He was initially resistant. The data did not support any connection between the vaccine and the soldier’s death. I had no other avenues to pursue. I persisted in the request to send the records to ACIP. The program manager relented primarily to appease me. The records were sent to ACIP. It took a few weeks before the ACIP report came back. As expected, ACIP determined that there was a negligible chance that the soldier’s death had anything to do with the vaccines he received.
I contacted the soldier’s mother. By this time, we were communicating be email regularly. I sent her the ACIP report. She looked at it for just a few minutes. Then came the email:
Mom: “Okay, NOW I believe you.”
I forwarded the email to the head of the vaccine program. His response was quick:
“This is still a people business, isn’t it?”
My response to him:
“Always has been and always will be.”
The number of meetings about data that I attend here at the NIH has grown progressively since I arrived 8 years ago. I do understand the importance of data to the biomedical research enterprise. We need good data. Data needs to be collected, stored, analyzed, curated, turned into usable information, and applied. Data systems need to be interoperable. Data needs to be shared freely with others who have an interest in it, who are subject matter experts, and who might also want to interpret the data. Alternative interpretations drive the dialogue that might eventually lead to consensus. On the flip side, data must be kept out of the hands of those who want it for nefarious purposes.
We are consumed with gathering more and more data. The clinical data entries in the Electronic Health Record may be the starting point. Then there are the genome and the exposome, including its impact on the genome (mutations or epigenetics). Next there are all the RNA’s, all the -omics, the social determinants of health, the microbiome(s) (each of us has at least three), and the NIH is now in a big effort to understand all the viruses in and on us (the virome). We are better at collecting data than we seem to be at a priori knowing which data we need now or might want in the future. Our default position is to want as much as possible.
Lately many of the meetings seem to be looking at data about the data. I think that is what is called metadata. I was in a meeting last week where an articulate data scientist employed metadata- specific terms that were completely unfamiliar to me. No doubt this is an age-related phenomenon.
It should be readily apparent that my retirement gig, if there is one, will not be in data science. That is not a big loss to the field. However, my concern is that we have become much more comfortable with talking about the data than talking to the people who provide the data to us. I had all the data to conclude that the young soldier’s death was not related to a vaccine in 30 minutes. But I didn’t know enough about mom’s situation – antivaxxers in her ear, another son who died tragically, environment rife with mistrust of the government – to be able to convey the information that I had in a way that would have any impact. Trust had to be earned over months of patient communication. Taking the extra step of turning the analysis of her son’s death over to the ACIP finally was the tipping point.
We need great data systems, and we need lots of great data scientists. Great systems and data scientists are expensive, so we need a lot of money too. However, our patients are much more than their data. In the NIH Clinical Center, especially in the NIH Clinical Center, we should never forget that we are in the people business. We always have been, and we always will be.