By Marie Rosenthal, MS

As states are moving to the down side of the COVID-19 curve—still treating new patients, but looking toward reopening the country—the Infectious Diseases Society of America released 15 recommendations about SARS-CoV-2 diagnostic testing for symptomatic and asymptomatic people. The hope is these guidelines will assist clinicians making decisions about using the more than 65 COVID-19 tests available.

“I think the overarching goal was to give some guidance about who should be tested, how they should be tested and when they should be tested,” explained the guideline’s lead author Kimberly E. Hanson, MD, MHS, in a briefing sponsored by the IDSA.


The most important group to test right now is still patients with symptoms, and every one of them should be tested, according to the guideline. They include patients with cough and shortness of breath or difficulty breathing, and at least two other symptoms of COVID-19: fever, chills, repeated shaking with chills, muscle pain, headache, sore throat, and new loss of taste or smell, according to the CDC.

“The top priority recommendation is to emphasize and recommend the importance of testing all patients who have clinical signs or symptoms that could be consistent with COVID,” Dr. Hanson said. “The reason this is our top priority and our top recommendation is really knowing if a patient is infected with SARS CoV-2 starts a cascade of decision making for physicians and clinicians,” such as whether the person should be hospitalized or can be quarantined at home.

Another reason to test symptomatic patients is to enter them into one of the many clinical trials testing various treatments for COVID-19, she said.

Testing symptomatic patients also helps in preventing transmission and protecting health care workers, according to Dr. Hanson, who is an associate professor of internal medicine and adjunct associate professor of pathology at the University of Utah School of Medicine, in Salt Lake City.

In addition, testing symptomatic patients helps guide public health initiatives, such as contact tracing, and allows epidemiologists to get a better handle on the rate of transmission in a community, she said.

Still Seeing Shortages

“There was a relatively easy consensus around testing of symptomatic patients,” said Angela M. Caliendo, MD, PhD, FIDSA, who also was on the guideline panel. 

One reason why the panel emphasized the testing of symptomatic patients first was that testing capabilities still vary widely, according to Dr. Caliendo, who is a professor and the executive vice chair of the Department of Medicine, Warren Alpert Medical School, Brown University, in Providence, R.I. Many health care systems and regions of the country do not have enough swabs to take samples, storage media to transport the tests to the lab or reagents to run the tests, she explained.


“We have difficulties with access to adequate tests. So if you don’t have adequate testing, you’re going to prefer to test your symptomatic patients rather than your asymptomatic patients,” Dr. Caliendo said. 

Although more testing would be better, “because resources are not limitless—there are still places in the country that really don’t have a sustained access to testing—that part of our document really was to try to help people prioritize. So if you’ve got the symptomatic testing down, your supplies are good, you can address all of those patients,” then it is time to address the asymptomatic patients, according to Dr. Hanson.

Another consideration before broad testing is laboratory capacity and human resources, added Dr. Caliendo. “I was just meeting with our Department of Health team and accessing specimens, getting results back to people, all the non-testing components are so time-consuming. Not only getting out and collecting specimens, but keeping track of the results—it’s a lot of work and that is a challenge for us,” she said.

Testing Asymptomatic People

Asymptomatic people with a known or suspected exposure to COVID-19 who work or live in long-term care facilities, as well as those who work in a hospital, correctional facility, factory or other setting experiencing a COVID-19 outbreak, should be tested if the resources are available. First responders are another group that public health experts want to see tested.

The remaining asymptomatic people were separated into three groups: people who were hospitalized for a non?COVID-19 reason, immunocompromised individuals, and those who are undergoing surgery, according to Dr. Caliendo.


The guideline about surgical candidates was informed by two issues, she explained. “One is protecting health care workers and the other is protecting the patient,” Dr. Caliendo said. “There were some data in the literature about patients that went under major surgeries who had COVID and they didn’t have a good outcome,” she said. “So, we looked at that one both from the perspective of the patient as well as the perspective of the health care worker.” 

The data were especially limited in the area of which type of specimen to use, according to Dr. Caliendo. There is a recommendation to use a sample taken with a nasopharyngeal swab, a mid-turbinate nasal swab or a nasal swab. Since the literature search for the guideline development ended a couple of weeks ago, there has been new information about different specimens, so this recommendation is likely to be updated soon, she added.

There was a lot of discussion about saliva, according to Dr. Caliendo. Being a simple specimen to collect, it would have some proponents among health care workers. However, “people defined saliva differently in the various studies. Some have patients pool saliva in their mouths and collect that, while others have them cough up a saliva specimen, which was likely mixed with sputum,” she explained. “Due to the heterogeneity of the specimen itself and the limited amount of data that we had, we did not recommend saliva as a specimen of choice, but this may be revisited as more data become available.”

Repeat testing is also addressed. If a patient has respiratory symptoms but there is a low clinical suspicion for COVID-19, and they test negative, the panel recommended against retesting. However, if there is a high clinical suspicion and the clinician feels the person should be hospitalized, the panel recommended retesting, ideally with a different type of specimen, such as sputum or a tracheal aspirate.

The problem with these specimen types is that the clinical laboratory must validate sputum or tracheal aspirates before providing testing. 

The panel reviewed only the molecular diagnostic tests, not the serology tests used for antibody detection, because there were not enough data, according to Drs. Hanson and Caliendo. However, they did talk about serology tests during the press briefing.

“Antibody tests have become a very hot topic,” Dr. Caliendo admitted, but she said data are lacking about the performance of the tests and how they should be used. There are tests that measure total antibodies; tests that measure immunoglobulin G (IgG), which appears about two weeks after infection; and tests that measure IgM, which is measurable earlier than IgG. One of the issues is no one knows what it means if the patient’s result is positive, other than they were exposed to the virus.


“The big concern that many of us have is over specificity,” Dr. Caliendo explained. “If the assay has a low specificity, say 95% which sounds like a high number, but if you have a low prevalence of infection, say less that 5%, you are going to have more false positives than true positives.”

Public health officials worry what patients will do with that information. If they think they were infected and they were not and they stop practicing social distancing, this would be a mistake, according to Dr. Caliendo.

“If the test is accurate and you have antibodies, what does that mean?” she asked. “Does that mean you are protected from future infection? We don’t know. If you recovered and you have antibody, are you still infectious? We don’t know that either.”

Another issue is the quantitative question, according to Dr. Hanson. “One of the assays that we looked at doesn’t even give you a level of antibodies; it just says they are present or absent. Do we need to be quantitative? Is there a threshold above which you are protected or below?” Which antibodies, if any, confer protection? How long does the protection last? 

“We have to be smart about how we deploy it [antibody testing], as we start to think about testing asymptomatic people in the community,” Dr. Hanson said.

Dr. Caliendo said it would be helpful if there was an independent diagnostic reference standard. “You need a definition of a positive COVID-19 case. There are some patients that we see who are symptomatic, who have many characteristics of COVID, but their molecular tests are negative. And then, if we wait two weeks and do a serology test, they are positive, which leads us to believe they were infected, but just didn’t have virus in the sample,” or the virus was not at a high enough concentration in the location tested.

The guideline panel culled all the available studies, including those that were not vetted and posted online, Dr. Hanson said, to release the 15 recommendations

. Based on the evidence, which remains limited, the guidelines are not intended to supplant the judgment of physicians in addressing the needs of patients or specific clinical situations, the IDSA said.

"The guideline represents a heroic effort by IDSA to provide 15 recommendations on COVID testing," said Debra Goff, Pharma, FCCP, who was not involved in the guidelines but asked her thoughts about them. "It is striking but not unexpected that 14 of 15 recommendations are based on very low certainty of evidence," said Dr. Goff, who is an infectious disease specialist, The Ohio State University Wexner Medical Center, in Columbus.

"As the authors describe, COVID testing is a moving target. We are learning on the fly," Dr. Goff said. "The best COVID-19 test today might fall out of favor next month. It is unprecedented for IDSA to make recommendations with so little available science. It may be unprecedented but it was necessary."