Dr. Jennifer Nuzzo says data scientists and public health officials should play a greater role in preparing the nation for future crises.
Dr. Jennifer Nuzzo, senior scholar at the Center for Health Security in the Johns Hopkins Bloomberg School of Public Health, has worked on pandemic preparedness for over 20 years. Her expertise has been integral to all efforts at the Coronavirus Resource Center, particularly tracking COVID-19 testing and positivity rates. Dr. Nuzzo shared her thoughts regarding public health data during this pandemic and what she hopes will be an expanded role for public health officials and data scientists in the future. She believes that they should serve as a first line of defense against future health crises and receive the significant investment their roles deserve.
I seriously hope it is something that we continue to try to improve upon. I was disheartened to read, during this pandemic, that some places are still piecing together and faxing Excel sheets to record and transmit data. Local public health data infrastructure is the foundation of our national health surveillance capability. We must not only improve their technology, but also their data analytics capabilities. The level of data science demonstrated by the CRC is not available in all places, but it should be. This is the future of public health – big data.
The challenge is that there's a tremendous amount of money on the table, and my hope is that it can actually be used to make some long-term basic infrastructure changes. When a large pot of money becomes available it often goes to buy one-time cost items as opposed to the continued support of people. Think of it like this: We fund and maintain the military for the defense of the country. We're now seeing the consequences of not having standing public health defenses. We need to think of our public health professionals as our first line of defense against health crises.
The first thing that we started to see was a number of states not reporting on the weekends, and that number has increased over time, which makes it harder to detect trends. Over Memorial Day weekend, the media and even some government officials tweeted that the United States had the lowest number of COVID-19 cases reported in a very long time. On any given weekend, almost half of states don't report data. If you look at weekend numbers, you would think that we're doing well, but the numbers represent cases from only about half of the states.
We may soon be moving towards weekly reports from states. That makes it difficult to identify trends, because you don't know until the next week if you're in trouble. With a decreasing frequency of reports we start to lose insights, and it compromises our abilities to spot problems early. That level of resolution becomes increasingly important (hyperlink to data frequency blog) because small changes have big consequences as case numbers continue to shrink.
We do have fewer infections now, but there has been a rollback of testing even prior to this latest decrease. Testing numbers were plateauing around Thanksgiving, even as the country was headed into its biggest surge of cases ever. As states were preparing to roll out vaccines, they simply switched their testing centers over to vaccine centers. When you reduce testing capacity like that, you begin sampling a different population since the reduced geographic network of testing centers prevents many people from getting tested. Now, due to the difficulty in getting a COVID-19 test, most people getting tested are those who need a negative test result for school, work, or travel. As we have dialed back on testing and have less frequent reporting of cases, we're starting to point our telescope at a different place in the sky, and we might not be seeing an oncoming meteor.
I fundamentally reject any idea that we won't need data as the case numbers decline. The types of data and how we use them are going to change. Testing remains essential. Testing is the start of the process by which we conduct our surveillance of COVID-19. The way that we are going to test is going to change.
We're not going to eradicate this virus from the planet. I see us continuing to test people with COVID-19 symptoms because there could be dangers, including genetic variants, that we need to spot. The way that we spot new problems is by testing people. While we may start dialing back on the mass testing, that doesn't mean that we're not going to do more targeted testing and surveillance testing.
I wish we had more real-time surveillance for all diseases and conditions. I once spent some time at Taiwan’s CDC and I was so amazed by how they put their notifiable diseases and conditions up online in near real time. By comparison, our system where states collect these data, look at them on their own, and don't publish them until they're vetted, sometimes years later in a summary report, is not effective. It is critical for public health to continue to remind the public what it does and to put those data out there for all to see, to build that trust and transparency.
We should not dial back this public health data infrastructure. It is important that we continue to collect information, conduct surveillance, and ask questions that can help us better understand this and other viruses - the ones that we know now and the ones that are yet to come.