TDAI affiliate and professor of demography Samuel Clark has recently used his decades-long research on health issues in human populations of developing countries to help the state of Ohio in its fight against COVID-19.
Clark has enlisted the help of his verbal autopsy Lab in Residence team hosted at TDAI amongst other groups of professionals to work with him on several methods that can be utilized to measure the extent and overall impact of COVID-19 in the state of Ohio.
After spending most of his career studying human populations in developing countries around the world, Clark said, “The COVID-19 situation has provided the first opportunity I’ve ever had in my career to work on something in the United States. I am working with a small group of people from the College of Public Health and the Department of Geography here at OSU and with the State Department of Health (ODH) to develop four methods of measuring the extent and impact of COVID-19 in the state of Ohio.”
All four of these methods were transferred directly from the work Clark and his colleagues have completed in the developing world so that a new set of data sources could be created to keep track of the pandemic.
The first method used to measure the COVID-19 impact is conducting telephone interviews where their team asks if they have had a list of symptoms over the last few days. The team does this repeatedly, every week, and then relay the information back to healthcare professionals who monitor how things have changed over time.
The second method the team is implementing is an “excess deaths study,” which aims to look at the number of deaths that are occurring on a daily basis and compare those numbers to what would be predicted on the same day using data from previous years. “With mortality, the number of people who die in a given period of time is pretty stable, so those numbers are relatively predictable- so by monitoring them you can identify when something is going on by seeing a number that isn’t typical for a given day in the year,” he said.
With regards to mortality, Clark said that his team is at the point where they are getting more numbers and data, and plan on developing a tool that will, “eventually become part of a dashboard for the Department of Health,” and will track the death counts very closely. Once these numbers come out of a typical bound, it will sound an alarm for professionals to come take a look at them. The reason this is so important? “With something like COVID-19 which is brand new, we obviously don’t have testing infrastructure that is necessary to identify COVID-19 in all of the deaths that are happening, and COVID-19 isn’t the only thing that is killing people, there are various secondary effects of the epidemic and lockdown,” Clark said.
The OSU team has access to raw vital statistics data from the state and various counties. Because it takes time for the death data to propagate from where the death happened to the state level, their team is building a model of the reporting so that they can, “predict the number of deaths that haven’t hit the system yet in a particular day.” This enables the team to form a mostly accurate, up to the minute, count of deaths in Ohio.
The last two methods that Clark and his team are implementing are a combination of surveys and in-home testing. This involves a random sample of 1,200 people in the state of Ohio, including diagnosis of COVID-19 and/or see if a person has antibodies. Because the team is putting this method together very rapidly, they are utilizing basic design principles to figure out how the sampling will be analyzed after the data is received. In order to make this method successful and help with bandwidth, Clark’s team is recruiting volunteers, nurses, and those with the correct training to serve as interviewers and go out to people’s homes for testing.
Once this data comes back, the team will have, “a sample of 1,200 people that will tell us for the state of Ohio what the prevalence of COVID-19 really is, and what the prevalence is of COVID-19 susceptible people. This is the most important thing to know for our reopening.”
Clark said they will likely repeat this survey in intervals over the next year to keep track of what’s going on and even have a plan to make a much larger version of that survey that will expand to looking at subpopulations; particularly communities of people who are most susceptible.
Talking about the results of implementing these new ideas and methods, Clark said, “I think some of these things might outlast COVID-19, and it will become abundantly apparent that we should keep some of these more affordable systems going all the time. I hope that’s the outcome that comes out of this.”
To learn more about Clark’s verbal autopsy lab in residence hosted by TDAI, click here.