Tip Sheets

Prepare yourself before covering serology testing and other COVID-19 immunity issues

By Tara Haelle

Problems with PCR testing for the COVID-19 viral infection dominated headlines during the first six weeks of the pandemic. Now that serology testing — testing for antibodies to COVID-19 — is picking up steam, there is a lot of important context and uncertainty that your audience may need.

To begin this quick primer, here are some basic concepts to understand before interviewing experts in immunity and epidemiology:

• Having antibodies against a disease does not necessarily guarantee a person is immune to that disease. While immunity typically requires neutralizing antibodies to fight a disease, even the presence of neutralizing antibodies cannot guarantee immunity.

• If we assume antibody presence can confer immunity, we do not know what the correlate of protection is for immunity — that is, what level of antibodies need to be present to provide protection. 

• It is possible to be immune to a disease without having any detectable antibodies. The cells which make antibodies (B cells) can exist as “memory” cells that remember how to make the antibodies when needed. (Speak to an expert in immunity about memory T cells and memory B cells for details.)

• We do not yet know how long immunity might last to COVID-19. 

• We do not yet know what individual or environmental factors may affect immunity development or persistence. 

• If you can gain access, reading this chapter on immunity and resistance to viruses from the book Viruses: From Understanding to Investigation may provide helpful background. 

Among the most helpful Twitter resources on serology studies is this excellent thread (unrolled here) by Natalie Dean, Ph.D., an assistant professor of biostatistics at the University of Florida. Dean, who specializes in emerging infectious diseases, offers some best practices for serology studies, drawn partly from WHO guidance.

It’s also important to understand what serology studies and results mean — and don’t mean — when it comes to herd immunity. Take the time to read this excellent thread from Carl Bergstrom on herd immunity (unrolled here), which he followed up with this add-on thread (unrolled here). 

Recent Coverage

• NPR: Study raises questions about false negatives from quick COVID-19 test

• ProPublica: What antibody studies can tell you — and more importantly, what they can’t

• Science News: What coronavirus antibody tests tell us — and what they don’t

• The Atlantic: The false hope of antibody tests

• Vox: What Covid-19 antibody tests say — and don’t say — about immunity

• New York Times: Can antibody tests help end the coronavirus pandemic?

• Lifehacker Australia: How to know if you can trust that headline-grabbing COVID-19 study

• New York Times: Let’s get real about coronavirus tests

Government and University Publications

• Johns Hopkins Center for Health Security: Developing a national strategy for serology (antibody testing) in the United States

• U.S. FDA: Important information on the use of serological (antibody) tests for COVID-19 — letter to health care providers

• U.S. FDA: Emergency Use Authorizations (scroll down for testing kits, both PCR and serology)

• WHO: “Immunity passports” in the context of COVID-19

• WHO: Population-based age-stratified seroepidemiological investigation protocol for COVID-19 virus infection

When it comes to reporting on antibody testing studies, journalists need to ask a lot of questions of the researchers conducting them, especially if the findings raise eyebrows (as many have). I put it to Twitter to find out what researchers thought journalists should ask when reporting on serology testing. Thank you to Tara C. Smith, Ryan McNamara, Nicola Low, Gideon MK, Rob Meldrum, The Skeptical Scientist, Andy, Trevor Crowell, and Cecilia Sierra for their contributions, and Angela Rasmussen for a lengthy phone interview explaining a great deal to me about immunity.

Questions to ask:

• What is the study population? 

• How were they chosen? Was the population randomly selected, or were there criteria for selection/participation? What biases might result from the criteria or selection method?

• How does this study population compare to other populations elsewhere, particularly if you are writing for a local publication? 

• What test are they using? 

• What is the sensitivity and specificity of the test? 

• How was it validated? 

• What is the positive predictive value (PPV) and negative predictive value (NPV) for the test given the specific population it was tested in?

• What is the known rate of confirmed cases in the location, based on other testing methods?

• What is the study question? 

• Does the study need to be done? (It needs to be something other than “because we can.”)

• What study design did they use? Why?

• What are the limitations of that study design?

• What other limitations exist in the study? 

• What other explanations could explain the results? 

• Has the study been peer-reviewed?

• What do other experts think of the study, whether it’s a preprint or peer-reviewed?

• Does the study support or reject our current understanding about the virus and its transmission? 

• How is the test conducted? Are there any environmental factors (sunlight, temperature, storage) that could interfere with the tests’ accuracy?

• Does the test measure IgM or IgG antibodies or some others?

• Who developed the antibody test? What is their experience in this area?

Questions they won’t have simple answers for but which you should ask:

• Do we know yet how long immunity will last?

• Do we know what antibody levels are necessary to provide protection? Or that detectable antibodies mean any protection at all? Or if detectable antibodies are even necessary for protection?

• What do these findings mean for that community?

• What do these findings mean, if anything, for other areas?