During 2020, a team of researchers from The Ohio State University conducted three projects to better understand the extent and effects of the COVID-19 epidemic in Ohio.
- A representative, state-wide prevalence study of adults to characterize ever- and currently-infected with the coronavirus,
- A state- and county-level excess deaths study to estimate the total mortality burden associated with the epidemic in Ohio, and
- A representative, weekly telephone survey to characterize the prevalence and trend in COVID-19 symptoms and protective behaviors.
COVID-19 Prevalence Study
The prevalence study used a sample survey that was conducted in July, 2020. Both ever-infected and currently-infected prevalence rates were low at that time. A publication describing the results appears in Annals of Epidemiology.
Existing statistical methods for the analysis of survey data were inadequate for the unusual data produced by the prevalence survey. Our team developed a new statistical approach to solve those challenges. A description of the new methods and their application to the survey data is published in PNAS, and the R code used in the analysis is available on Github,
- Estimating Seroprevalence of SARS-CoV-2 in Ohio: A Bayesian Multilevel Poststratiﬁcation Approach with Multiple Diagnostic Tests.
- Bayes Prevalence: R code to implement new prevalence estimation method.
I was invited to write a commentary for PNAS based on experience with the prevalence study, and I did that with Abigail Norris Turner who led the overall study. We highlight the need to improve response rates and to prepare a robust measurement capability to be ready for the next pandemic.
Excess Deaths Study
The excess deaths study is ongoing. For the state, results closely match those of the CDC, see CDC excess deaths report and data. County-level results reveal heterogeneity in both levels and timing of patterns of excess deaths.
Weekly Telephone Survey
Data from the weekly telephone survey were not as useful as we had hoped.
COVID-19 Routine Mortality Surveillance
In addition to the work in Ohio, I work closely with the NGO Vital Strategies on a project to develop a COVID-19 routine mortality surveillance (CRMS) tool based on the verbal autopsy method. This work is in progress with partners in Colombia and Brazil.