TDAI hosts lab in residence linking data, global outreach, and verbal autopsy

Samuel Clark

Samuel Clark, TDAI affiliate and professor in the department of sociology, has been working with a team of professionals looking at the use of verbal autopsy (VA) to determine a cause of death in human populations around the world.  That team is one of four new labs in residence hosted by TDAI within Pomerene Hall, launched in January 2020.

“Verbal autopsy is a method for ascertaining the cause of death when you don’t have the ability to do traditional autopsies or medical examinations. It gets used in developing countries where vital statistics systems don’t function well,” explains Clark.

After entering the health systems field about 20 years ago, Clark has spent most of his career working in Africa on demographic and health-related issues. “What became very clear over the years is we don’t have a fundamental picture of health for these populations and the most rudimentary way to do that is to look at mortality. If you don’t have access to a traditional autopsy or medical certification then verbal autopsy is really the only option,” he said.

More specifically, verbal autopsy involves healthcare professionals interviewing individuals who were in a caregiving role for someone who has died and knows what happened in the months leading up to their death. Standard questionnaires that are used in the interview consist of 200-400 questions, including a story, in their own words, of what happened. Physicians then read the interview transcript, determine what the conditions were leading up to the death and diagnose the person retrospectively.

However, in some populations where people are dying of unknown causes there typically are not many doctors. To aid in these circumstances, Clark said, “we need another approach to doing the cause identification, which is the computer. We have tried developing a computational, statistical algorithm that processes all of the answers from the interview in an automated way and assigns a likely cause of death.”

The team develops statistical algorithms and computational tools to implement those algorithms in order to automate the assigned cause of death from the VA data. The Lab in Residence is hosted by TDAI in room 350 of Pomerene Hall.

On a global level, Clark notes that, “There is a big push by The World Bank, the UN, and other UN agencies to do something in a coordinated way to improve vital statistics and civil registration and population data.” Those agencies are targeting certain countries where these systems don’t work and are trying to improve them.

Because cause of death is so fundamental to understanding health, there is a lot of emphasis on registering death and doing VA for them to calculate something called the “burden of disease.” This is essentially just a ranking of the top causes of death in a given population so that those parts of the world can be targeted with interventions. “Over time,” says Clark, “we can use this same VA system to monitor whether or not these interventions are having any effect.”

At the moment, their team is conducting research in countries such as Bangladesh, Ghana, India, Kenya, Morocco, South Africa, and have begun collaborating with a group in Brazil, where their work is growing rapidly. “Most of our work in developing countries involves taking our software and setting it up for people to use, training people to use it, fixing any issues, and developing new tools with the ultimate goal of having the national governments use the tools in a routine way,” Clark said.

When asked how these global partnerships got started, Clark has his upbringing and education to thank. “I am Kenyan. I was born in Kenya and grew up in East Africa. My interest has always been in the regions of Africa. I did my dissertation working in Zambia and completed a five-year postdoc living in South Africa, and then decided to come to the States to do an academic career.”

Over the course of 25 years, Clark has developed collaborations with networks of field sites in Africa centered around intensive population and health surveillance systems. As he shifted to working in the United States, Clark said, “those connections in Africa naturally led to developing connections with the WHO and UN groups who work on similar things. I am a member of the advisory board at the UN, and through doing that I became connected with some of the key people doing civil registration and vital statistics strengthening over the last few years.”

As for the timeline of their Lab in Residence and the impact of COVID-19, Clark has shifted his focus a bit and enlisted his VA team to aid in Ohio’s fight against the virus. They work alongside a group of professionals from the College of Public Health and the Department of Geography at Ohio State, using four methods that are inspired by approaches used in the developing world. This has allowed them to quickly set up a new set of data sources to keep track of the epidemic.

Clark and his colleagues will continue their intensive work to aid in the fight of COVID-19 and say that the state has been very enthusiastic and receptive of their ideas.

To read more about Clark’s work on the impacts of COVID-19, click here.

Share this page
Suggested Articles
Undergrads and businesses meet in business analytics program

Fisher College of Business is home to a business analytics cluster that’s a hit both with students and with companies who sponsor the program. Students get hands-on experience creating real-world...

Arora collaborating on NYC noise pollution project

TDA affiliate Anish Arora, professor of computer science and engineering and a specialist in sensor networks, is collaborating with scientists from New York University to examine noise pollution in New...

5 Steps to Take as an Antiracist Data Scientist

Data scientists are data stewards. They collect data, store data, transform data, visualize data, and ultimately impact how data are used. In our data-driven world, stewards hold the objective responsibility...

New women in computing page features downloadable backgrounds

Ever wanted to use both a fun and educational background for your Zoom meetings? Recently added to the TDAI website is a Women In Computing page where you can read...

TDAI hosts (virtual) R programming workshops

TDAI is pleased to announce a multi-part workshop for learning about R.  R is a free and open source programming language used by many for statistical computing.  This series of...