Talking with Data: Why is it Crucial in the COVID-19 Era?

Naz Bilecik '22 SEAS
September 16, 2020

On July 21st, 2020, Columbia Global Centers | Istanbul, Columbia Global Centers | Rio, and the Center for Sustainable Development organized a live webinar titled "Talking with Data: Why is it Crucial in the COVID-19 Era?" The keynote was delivered by Corinne Thompson, an epidemiologist at the Bureau of Communicable Diseases in the NYC Department of Health and Mental Hygiene, and Matthew Montesano, a Data Communication Specialist at the NYC Department of Health and Mental Hygiene. The following speaker was Banu Çakır, Professor of Public Health at Hacettepe University. The event was moderated by Dr. Yanis Ben Amor, Executive Director, and Dr. Özge Karadağ Çaman, a public health researcher at the Center for Sustainable Development at Columbia University.

COVID-19 related data collection and sharing practices vary amongst the pandemic responses of countries across the world. How surveillance data is collected and shared with the public, academia, businesses, and other stakeholders in each jurisdiction can change the course of public health responses, which need to be built on public trust and tailored according to the countries' real needs. This webinar gathered experts from the U.S. and Turkey to present best practices and discuss current public health data collection and sharing needs in the COVID-19 era. Some highlights from the talk include the following:

Corinne Thompson gave insight into how data played a significant role in the fight against the coronavirus pandemic in New York City, the first large jurisdiction in the U.S. to have an explosive outbreak of COVID-19. Corinne explained that at the start of the pandemic, rapid decisions had to be made to understand the spread of the epidemic in the city. Data sharing was included as part of that discussion early on as there were so many unknowns in this particular outbreak, and people wanted to fill in the gaps.

Corinne gave an overview of how New York City does surveillance for COVID-19. She mentioned that electronic laboratory reporting is the basis of NYC's disease reporting system. Anytime someone is diagnosed with the coronavirus, it is required by law for this to be reported to the NYC Health Department. Corinne noted that testing was minimal early on, and the first case was diagnosed on February 29th, but they suspect that other cases were circulating in February. In terms of surveillance, at the beginning of the pandemic, data specialists were trying to get as much information as they could on every patient, but due to being overwhelmed, they quickly turned to just trying to maintain the basic electronic disease reporting system. Therefore, health professionals have minimal information on each case, and Corinne highlighted that this was very challenging to communicate to the public as people expected complete and timely data. 

Corrine went over the COVID-19 data web page for New York City. The data presented on these pages reflect the most recent information the Health Department has collected about people who have tested positive for COVID-19 in NYC. Corinne explained that most of the data from March, April, and early May only represent very severe cases because back then, people with mild symptoms were encouraged to stay at home and not get tested. The website shows the number of confirmed cases, hospitalizations, and deaths from COVID-19 daily since February 29th. For New York City, the first week of April was the peak week for the number of cases. Since the end of May, there were a consistent number of cases per day, and the percentage of people who tested and had a positive result was around 2%. One of the most popular elements on the website is a map that shows the number of cases, death counts, and the percent of people tested with a positive result per ZIP code of residence. Corinne explained that the webpage also provides demographic data about the pandemic, such as the number of deaths according to different age-groups, gender, and race. The data shows that Hispanic/Latino and African American residents have much higher hospitalization and death rates in NYC. 

Matthew Montesano talked about the challenges of communicating COVID-19 data to the general public. While data is useful to public health professionals, policymakers, and academics, throughout this pandemic, the general public has also paid attention to data to better understand the danger posed to them. Matthew explained that one's sense of being at risk is a critical component of health behavior change. Due to this, people needed to trust what data specialists were saying about the pandemic in order for them to drastically change their day-to-day behavior and stay home. 

Matthew explained that the way to gain people's trust is to present them with straightforward and easy to understand information. People don't learn, understand, or use data just because you throw numbers at them. Information uptake is affected by the information itself, the format in which it is delivered, and the medium through which it is delivered. During COVID-19, data specialists had essential data regarding the pandemic, and they wanted to deliver it in a format that would be widely accessible and easy to understand. The COVID-19 data web page for NYC is built on a reasonably low tech system which can be easily replicated, and it is based on free and easy to use tools that require very few specialized skills. The first tool that data specialists used to create the charts on the webpage is called Datawrapper. This product was initially built for journalists to rapidly develop and deploy visualizations with strong explanatory power. This tool lets them create simple and interactive data visualizations that can be embedded into any ordinary web page and can be built to meet accessibility requirements to ensure that everyone can view this information. Matthew explained that they also needed a way to feed these charts data and update them each day. For this, they used Github, a service designed to help people store and collaborate on software projects. The Datawrapper charts refresh by reading the data stored on Github each day. Some of the other features on Github also allow people to ask questions and engage with the data. These two tools let data specialists publish fast and accessible data that was updated each day. This way, the public trusted the information that was presented and took precautionary health measures seriously. 

Banu Çakır talked about the formulation of health policies for successfully managing the pandemic. She mentioned that in Turkey, there is a lot of data about COVID-19. Still, the findings are often not conclusive, and the information can be inaccessible and complex for a policymaker. When there is no good communication between scientists and policymakers, policymakers make decisions based solely on their own values, priorities, and competing sources of information, such as anecdotal experiences. Data communication is crucial for reliable forecasting of the pandemic and evaluating the effectiveness of various interventions. Talking with data also provides effective and long-lasting interaction and trust between scientists, policymakers, and the general public. 

Banu gave insight into how data related to COVID-19 is being reported in Turkey. The first coronavirus case recorded in Turkey was on March 11th. The Ministry of Health has been publishing daily updates on COVID-19 such as the number of tests performed on that day, positive cases, deaths, recovered patients, and other data related to the pandemic. The number of cases reported is only for patients that have gotten a PCR test and tested positive. To measure the severity of the situation, Turkey also reports the number of patients who are seriously ill and who are intubated. Banu explained that with the start of normalization efforts in Turkey, the number of individuals placed in the intensive care unit (ICU) due to the coronavirus increased. This could be caused by the upsurge of older people socializing after staying home for three months. However, Banu also pointed out that the percentage of intubated patients had not increased at this time, which might show that doctors prefer to keep their patients in the ICU instead of regular inpatient clinics. 

Banu highlighted that one challenge regarding the pandemic is that there needs to be more data collected in order to understand the long-term complications caused by the virus and how to develop immunity. She believes that collective efforts are necessary to overcome the virus, such as prospective cohorts on COVID-19 patients and multi-disciplinary team efforts at an international level.