Interpretation and use of COVID-19 diagnostic tests: key resources

Despite the recent decision by U.S. health agencies to pause administration of the Johnson & Johnson vaccine to investigate rare blood clots in several young women, the pace of vaccinations against SARS-CoV-2 is accelerating. The Centers for Disease Control and Prevention reports that more than half of U.S. adults have now received at least one dose, and 33% are fully vaccinated. (In patients age 65 years and older, the respective percentages are 80% and 65%.) However, with more than 50,000 new cases being reported every day, including a small number of breakthrough infections, the need for COVID-19 testing is not going to go away any time soon.

A few weeks ago, the U.S. Food and Drug Administration announced that it had authorized several COVID-19 tests for over-the-counter use for serial screening of asymptomatic persons. Soon to arrive in national chain pharmacies and other stores, these tests and others - mostly of the rapid antigen variety - will cost between $10 and $110. A clinical and cost-effectiveness analysis in the Annals of Internal Medicine estimated that a weekly home-based antigen testing program could avert 2.8 million infections and 15,700 deaths in the U.S. over a 60-day time frame, at a cost of less than $8000 per infection averted. Not only could frequent testing make residences and workplaces safer, it could also improve the safety of in-person K-12 schooling and summer camps, since children younger than age 16 are not yet eligible to receive any of the vaccines.

The April 15 issue of AFP includes an article by Dr. William Nettleton that discusses several common questions and answers about SARS-CoV-2 diagnostic test interpretation: 

How Do Test Characteristics Such as Sensitivity, Specificity, and Percent Agreement Inform SARS-CoV-2 Diagnostic Test Interpretation?

How Does Test Timing in Relation to Symptom Onset Inform SARS-CoV-2 Test Interpretation?

How Does Pretest Probability of Disease Inform SARS-CoV-2 Test Interpretation?

How Can Posttest Probability of Disease Be Determined?

What Should Be Considered with an Unexpected Negative Test Result?

Is a Symptom- or Test-Based Approach Preferred for Discontinuing Isolation Precautions for Most Patients with COVID-19?

How Should Clinicians Counsel Patients about Serologic SARS-CoV-2 Test Results?

In addition to describing key features of molecular and antigen diagnostic tests, the article includes a useful leaf plot figure that allows clinicians to translate pretest to posttest probability of COVID-19 infection based on a positive or negative test result and the test's sensitivity and specificity.

Family physicians who need a quick refresher on clinical probability formulas and related conceptual issues can read a recent article in BMJ Evidence-Based Medicine. Those seeking a more in-depth discussion of COVID-19 test use and clinical scenarios may consult an evidence-based review that I co-authored in the Journal of the American Board of Family Medicine. Notably, we observed: "Often lost when testing is encouraged is that testing does not by itself confer health benefits. Rather, testing is useful to the extent it forms a critical link to subsequent medical or public health interventions." Although access to COVID-19 testing is absolutely necessary for an effective pandemic response, it is not sufficient in the absence of equitable interventions to support behaviors that prevent viral transmission (e.g., paid leave to support self-isolation and quarantine, laws requiring mask use in crowded indoor spaces). Finally, the American Academy of Family Physicians has a comprehensive guide to COVID-19 testing on its website.