Volume 26, Number 9—September 2020
Seroepidemiologic Study Designs for Determining SARS-COV-2 Transmission and Immunity
|Study type||Brief description||Questions study could answer||Issues with interpretation and representativeness||Issues with conducting during a pandemic|
||A sample of the population has serum samples collected at 1 time point
||Background cross-reactivity (if started before pandemic); current proportion of population that have been infected; proportion of population that is immune (if a correlate of protection defined); infection fatality ratio (with information on cases or deaths in the same population)
||For the different modes of collection (e.g., blood banks, residual sera, and volunteers), different issues can bias the sample included in the study that must be assessed
||Blood banks might have fewer participants, residual sera studies in hospitals might have fewer samples or over representation of severe acute respiratory syndrome coronavirus 2 infections
||The same persons are followed up over time, with serum samples collected at regular intervals, and information on disease in intervening periods
||Background cross-reactivity (if started before pandemic); ratio of asymptomatic to symptomatic infections; waning of antibody levels, correlates, and duration of protection; changes in infection dynamics over time
||Attrition can make analysis and interpretation difficult, biases in which participants are retained across sampling rounds
||Challenges in collecting and continuing cohort during outbreak; attrition
|Targeted populations||Populations with particularly high exposures, such as those around index patients or healthcare workers, have serum samples taken either cross-sectionally or in a targeted cohort||Attack rates; ratio of asymptomatic to symptomatic infections; proportion of population infected, correlates, and duration of protection||Targeted populations because healthcare workers might have different infection exposure rates and intensity from the general population||Potentially logistically difficult to collect samples in household studies|
1These first authors contributed equally to this article.
2These authors contributed equally to this article.