INTRODUCTION —
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in late 2019 in Wuhan, a city in Hubei Province, China, and subsequently spread worldwide, causing a major global pandemic. Before the World Health Organization (WHO) declared an end to the coronavirus disease 2019 (COVID-19) global health emergency in May 2023, SARS-CoV-2 infection resulted in an estimated 15 million excess deaths in 2020 and 2021 alone. As SARS-CoV-2 transitions to endemicity, it remains an important cause of illness around the world [1].
This topic will discuss the diagnosis of COVID-19. The clinical features, epidemiology, virology, and prevention of COVID-19 are discussed elsewhere. (See "COVID-19: Clinical features" and "COVID-19: Epidemiology, virology, and prevention".)
The management of COVID-19 is also discussed in detail elsewhere:
●(See "COVID-19: Evaluation and management of adults with acute infection in the outpatient setting".)
●(See "COVID-19: Management in hospitalized adults".)
●(See "COVID-19: Infection prevention for persons with SARS-CoV-2 infection".)
DIAGNOSTIC APPROACH
Clinical suspicion — The possibility of COVID-19 should be considered in anyone with new-onset fever and/or respiratory symptoms. Although cough and dyspnea are considered the classic respiratory features of COVID-19, other respiratory symptoms such as sore throat, rhinorrhea, and nasal congestion are commonly reported and are frequently the sole presenting symptoms of COVID-19. Other common clinical manifestations include smell or taste disturbances, myalgias, and diarrhea (table 1). COVID-19 should also be considered in patients with severe lower respiratory tract illness without another clear cause. The clinical features of COVID-19 are discussed in detail separately. (See "COVID-19: Clinical features", section on 'Initial presentation' and "COVID-19: Clinical features", section on 'Imaging findings'.)
As SARS-CoV-2 is prevalent worldwide, clinicians should have a low threshold for suspicion of COVID-19. The threshold for suspicion should be particularly low if the individual resides in or has traveled to locations with high rates of community transmission, has had potential exposure to SARS-CoV-2 in an outbreak setting or as a close contact of someone with confirmed or suspected infection, or resides in a congregate setting.
There are no specific clinical features that can reliably distinguish COVID-19 from other viral respiratory infections [2]. Nevertheless, some features may warrant a higher level of clinical suspicion [3-5]. Although loss of taste and loss of smell were the symptoms most strongly associated with a positive SARS-CoV-2 test with some pre-Omicron variants [4,6,7], these symptoms appear less common with Omicron variants [8]. Development of dyspnea several days after the onset of initial symptoms is also suggestive of COVID-19 [3]. Certain chest imaging findings have also been associated with COVID-19 [9]. However, none of these findings definitively establish the diagnosis of COVID-19 without microbiologic testing.
COVID-19 should also be a diagnostic consideration in patients who present with extrapulmonary complications that have been associated with SARS-CoV-2 infection, including cardiac injury, ischemic stroke and other thromboembolic events, and inflammatory complications (eg, the multisystem inflammatory syndrome in children). (See "COVID-19: Clinical features", section on 'Acute course and complications'.)
Whom to test
Selected symptomatic patients — Because the clinical features of COVID-19 are non-specific, confirming the diagnosis requires microbiologic testing. However, confirming the diagnosis is not necessary in all symptomatic patients. Any patient with suspected viral upper respiratory infection, regardless of etiology, should follow general measures to prevent transmission to others (eg, physical distancing or masking while ill).
We suggest testing symptomatic individuals with suspected COVID-19 (see 'Clinical suspicion' above) for whom the diagnosis would have specific implications for treatment or infection control. These include:
●Hospitalized patients.
●Outpatients with risk factors for severe disease that would make them candidates for antiviral treatment if the diagnosis of COVID-19 were confirmed. These factors are discussed elsewhere. (See "COVID-19: Evaluation and management of adults with acute infection in the outpatient setting", section on 'Patient selection'.)
●Patients who have close contact with individuals who are immunocompromised, frail, or otherwise more vulnerable to infection; in such cases, a confirmed diagnosis of COVID-19 might warrant more vigilant infection prevention measures, as discussed elsewhere. (See "COVID-19: Infection prevention for persons with SARS-CoV-2 infection", section on 'Symptomatic patients'.)
For all other individuals, the decision to test is based on patient preference. Some employers may also mandate testing for infection control purposes in the workplace.
Selected asymptomatic individuals — Routine testing of asymptomatic individuals is unnecessary. However, there are certain circumstances when it is reasonable to test asymptomatic individuals following exposure to COVID-19 (eg, close contact with an individual with documented infection), such as:
●When identifying asymptomatic individuals with infection could lead to preventive measures to reduce SARS-CoV-2 transmission among vulnerable populations. This includes preventing outbreaks in congregate living facilities that house individuals at risk for severe disease (eg, long-term care facilities, correctional and detention facilities, homeless shelters) or identifying infection in individuals who have close contact with individuals at risk for severe disease.
●When identifying asymptomatic infection in an individual at high risk for severe disease might help inform management (eg, plan for early treatment if symptoms develop).
If testing is performed, the optimal time to test following exposure is uncertain because the precise time to detectable ribonucleic acid (RNA) or viral antigen following exposure likely varies by infecting variant and test used. We typically test individuals five to seven days post-exposure.
If testing is performed for infection control purposes in health care or congregate settings, serial testing (eg, on days 1, 3, and 5 following exposure) is another strategy to identify infection early [10]. (See 'In outbreak settings' below.)
In the United States, the CDC suggests not testing for new infection in asymptomatic individuals who were previously diagnosed with SARS-CoV-2 within the prior month. A nucleic acid amplification test (NAAT), specifically, should be avoided in individuals who have had SARS-CoV-2 infection within the prior three months because of the low likelihood that a repeat positive NAAT during this interval represents an active reinfection. (See 'Persistent or recurrent positive NAAT during convalescence' below.)
How to test
Choosing an initial diagnostic test — A viral test is necessary to make the diagnosis of SARS-CoV-2 infection: either an NAAT, most commonly a reverse-transcription polymerase chain reaction (RT-PCR) assay, or an antigen test.
In general, we suggest NAAT, if readily available with reasonable turnaround time (eg, within 24 to 48 hours), because of its superior sensitivity. NAATs are usually performed at a laboratory. Rapid RT-PCR tests that can be performed at the point-of-care are also available and appear to perform comparably to standard laboratory-based NAAT; rapid isothermal tests may be less sensitive than other rapid NAAT [11]. (See 'NAAT (including RT-PCR)' below.)
However, antigen tests are more accessible, more convenient, and less costly than NAAT and have a shorter turnaround time than laboratory-based NAAT. Thus, they are reasonably used more commonly than NAAT for testing at home and at point-of-care sites. Individuals who use antigen testing should be aware that the sensitivity is lower than that of NAATs and negative antigen tests performed for symptoms or recent exposure generally warrant confirmation with additional testing, as detailed elsewhere (algorithm 1). (See 'Antigen testing' below.)
Antigen tests are also preferable to NAAT for individuals who had recently (ie, in the prior three months) tested positive for SARS-CoV-2 and have indications for repeat testing (eg, new symptoms); in the few months following infection, a positive NAAT may be more likely to reflect prolonged viral RNA shedding than a new infection, but it cannot distinguish between the two. (See 'Persistent or recurrent positive NAAT during convalescence' below.)
Specimen collection — Upper respiratory samples are the primary specimens for SARS-CoV-2 viral tests.
●Specimens for NAAT – In the United States, the CDC recommends collection of one of the following specimens [12]:
•Nasopharyngeal swab specimen, collected by a health care professional
•Nasal swab specimen from both anterior nares, collected by a health care professional or by the patient on-site or at home (using a flocked or spun polyester swab)
•Nasal mid-turbinate swab, collected by a health care professional or by the patient supervised on-site (using a flocked taper swab)
•Nasal or nasopharyngeal wash/aspirate, collected by a health care professional
•Oropharyngeal swab specimen, collected by a health care professional
•Saliva specimen (1 to 5 mL) collected by the patient while supervised
The Infectious Diseases Society of America (IDSA) suggests specimens from the nasopharynx, the anterior nares plus the oropharynx, the nasal mid-turbinate region, saliva, or mouth gargle; it notes that swabs from the anterior nares or oropharynx alone are acceptable if the others are not feasible but may be less sensitive [11]. Test sensitivity with different specimens is discussed elsewhere. (See 'Accuracy' below.)
●Specimens for antigen tests – Nasopharyngeal, nasal mid-turbinate, and nasal swabs are the only recommended specimens. Antigen tests are not validated for use with oral or salivary specimens.
Patients who collect their own specimens (of any type) should be given detailed instructions on correct technique.
Additional information on testing and handling of clinical specimens can be found at the CDC website.
Infection control practices during specimen collection are discussed elsewhere. (See "COVID-19: Infection prevention for persons with SARS-CoV-2 infection", section on 'Specimen collection for respiratory viral pathogens'.)
Although SARS-CoV-2 RNA can be detected in nonrespiratory specimens (eg, stool, ocular specimens, blood), testing of these specimens has a limited role in the diagnosis of COVID-19 [13-15].
Establishing the diagnosis — The diagnosis of COVID-19 is made by direct detection of SARS-CoV-2 RNA using NAATs or by detection of viral protein using an antigen test. A positive NAAT or antigen test is generally indicative of infection and does not need to be repeated.
Interpretation and further evaluation following a negative test vary for NAAT versus antigen testing and are discussed elsewhere. (See 'NAAT interpretation' below and 'Antigen test interpretation' below.)
Identifying specific variants — In general, clinically available diagnostic tests cannot confirm the specific SARS-CoV-2 variant causing infection. Viral sequencing or dedicated multiplex PCR genotype testing is needed to do this, and these methods are not routinely available for clinical decision-making.
Certain variants result in an S or N gene target failure with some NAATs (ie, either the S or N gene is not detected but the NAAT results are positive because the other target is detected). Such results can be a proxy for infection with those particular variants if other circulating variants do not result in the same gene target failure pattern. However, this is not a definitive method of variant identification. Because different NAATs detect different regions of the S or N gene, a particular viral variant may result in S or N gene target failure with one NAAT but not another. Furthermore, laboratories do not routinely provide information on gene target failure. (See 'Impact of SARS-CoV-2 mutations/variants on test accuracy' below.)
Testing for other pathogens — If influenza and respiratory syncytial virus (RSV) are circulating in the community, it is reasonable to also test for these viruses when testing for SARS-CoV-2, as this could have management implications. Depending on the presentation (eg, symptoms consistent with pneumonia, pharyngitis), testing for other pathogens may also be warranted. These are discussed in detail elsewhere. (See "Overview of community-acquired pneumonia in adults", section on 'Microbiologic testing' and "Evaluation of acute pharyngitis in adults", section on 'Evaluation'.)
However, detection of another viral (or bacterial) pathogen does not necessarily rule out SARS-CoV-2 in locations where there is widespread transmission. Coinfection with SARS-CoV-2 and other respiratory pathogens, including influenza virus, has been described, but the reported frequency is variable [16-20].
NAAT (INCLUDING RT-PCR)
Types of NAATs — Nucleic acid amplification tests (NAATs) detect SARS-CoV-2 RNA; reverse-transcription polymerase chain reaction (RT-PCR) assays are the most common type of NAAT [21,22]. Various RT-PCR assays are used around the world; different assays amplify and detect different regions of the SARS-CoV-2 genome. Some target two or more genes, including the nucleocapsid (N), envelope (E), and spike (S) genes, and regions in the first open reading frame, including the RNA-dependent RNA polymerase (RdRp) gene [23]. Other, less common types of NAAT include isothermal amplification, CRISPR-based assays, and next-generation sequencing [24-26].
In the United States, the Food and Drug Administration (FDA) has granted emergency use authorization (EUA) for many different NAAT assays [27]; testing is performed by the Centers for Disease Control and Prevention (CDC), local public health departments, hospital laboratories, and commercial laboratories. For some assays, the patient can collect the specimen at home and submit it to a laboratory [28,29]; point-of-care NAATs, some of which can be completed at home, have also been developed, although they may be less sensitive than laboratory-based tests [24,30-33].
Specific NAAT assays have different performance characteristics and times to results (ranging from 15 minutes to several hours) and are authorized for different specimen types. For laboratory-performed tests, the turnaround time for clinicians to receive a result also depends on how often the laboratory performs the test.
Pooled specimen testing with NAATs (ie, testing several specimens together and only retesting individual specimens if the pool tests positive) can improve laboratory testing capacity [34-37]. However, this approach would not save time and resources in high-prevalence settings, since frequent retesting for pools that test positive is cumbersome; thus, it is best reserved for populations with expected low prevalence, such as asymptomatic individuals without known exposure. Clinicians should be aware that pooling specimens reduces test sensitivity (to greater extents with larger pool sizes) and may prolong the time to an individual positive result.
NAAT interpretation
Positive NAAT result — A positive nucleic acid amplification test (NAAT; eg, RT-PCR) for SARS-CoV-2 generally confirms the diagnosis of COVID-19. No additional diagnostic testing is necessary. However, additional testing may be warranted for management in hospitalized patients. This is discussed elsewhere. (See "COVID-19: Management in hospitalized adults", section on 'Evaluation'.)
Patients with COVID-19 can have detectable SARS-CoV-2 RNA in upper respiratory tract specimens for weeks after the onset of symptoms [38]; however, prolonged viral RNA detection does not necessarily indicate ongoing infectiousness [39,40]. These data and the implications for discontinuing isolation precautions are discussed in detail elsewhere. (See "COVID-19: Epidemiology, virology, and prevention", section on 'Viral shedding and period of infectiousness' and "COVID-19: Infection prevention for persons with SARS-CoV-2 infection", section on 'Duration and subsequent precautions'.)
There have also been reports of patients with COVID-19 who have positive NAATs soon after documented viral RNA clearance. Thus far, there is no evidence that this finding represents relapsed or repeat infection [41].
Negative initial NAAT result — For many individuals, a single negative NAAT result is sufficient to exclude the diagnosis of COVID-19. However, false-negative NAATs (eg, RT-PCR) from upper respiratory specimens have been well documented (see 'Accuracy' below). If initial testing is negative but the suspicion for COVID-19 remains (eg, suggestive symptoms without evident alternative cause) and confirming the presence of infection is important for management or infection control, we suggest repeating the test. The optimal timing for repeat testing is not known; it is generally performed 24 to 48 hours after the initial test. Repeat testing within 24 hours is not recommended.
In patients with evidence of lower respiratory tract illness, lower respiratory tract specimens can be an option for NAAT [11,42]. We agree with the Infectious Diseases Society of America (IDSA) recommendations to reserve lower respiratory tract specimen NAATs for hospitalized patients who have an initial negative test on an upper respiratory tract specimen but for whom suspicion for lower respiratory tract SARS-CoV-2 infection remains [11,43]. For lower respiratory tract specimens, expectorated sputum should be collected from patients with productive cough, and tracheal aspirate or bronchoalveolar lavage should be collected from patients who are intubated. Most NAATs have not received EUA for lower respiratory samples, so laboratories need to validate these specimen types, and this may not be possible for all laboratories.
In hospitalized patients suspected of having COVID-19 who have a negative SARS-CoV-2 NAAT, characteristic laboratory or imaging findings can further support the clinical diagnosis of COVID-19 and be reasons to maintain infection control precautions (see "COVID-19: Clinical features", section on 'Laboratory findings' and "COVID-19: Clinical features", section on 'Imaging findings'). It is also important to consider other potential causes of symptoms in patients with negative SARS-CoV-2 NAATs.
Infection control precautions for COVID-19 should continue while repeat evaluation is being performed. (See "COVID-19: Infection prevention for persons with SARS-CoV-2 infection".)
Indeterminate NAAT result — The interpretation of an inconclusive or indeterminate result depends on the specific nucleic acid amplification test (NAAT) performed; the clinician should confer with the performing laboratory about additional testing.
In some cases, an inconclusive or indeterminate result indicates that only one of the two or more genes that the NAAT targets was identified. These results can be considered presumptive positive results, given the high specificity of NAAT assays. If the patient is early in the disease course, repeat testing can be helpful to confirm.
Accuracy — NAATs are highly specific tests [44,45]; false-positive results are rare but have been reported with certain platforms [46]. Although NAATs have high analytic sensitivity in ideal settings (ie, they are able to accurately detect low levels of viral RNA in test samples known to contain viral RNA), clinical performance is more variable. Sensitivity of testing depends on the type and quality of the specimen obtained, the duration of illness at the time of testing, and the specific assay, as outlined below.
●Overall sensitivity – NAATs generally have high sensitivity, with low false-negative rates (less than 5 percent), although reports vary and some studies have described false-negative rates up to 40 percent [47-49]. Furthermore, estimates of sensitivity are limited by the absence of a clear reference standard for comparison. In a study of over 17,000 patients who presented to a participating emergency department in Canada between March 2020 and December 2021 and had a positive NAAT, only 434 (approximately 2.5 percent) had an initial negative test within 14 days of the positive one [49]. The overall sensitivity was estimated at 96 percent. Similarly, in a study of 626 patients who had a repeat nasopharyngeal RT-PCR test within seven days of an initial negative test at two large health centers in the United States, 3.5 percent of the repeat tests were positive [48].
●Sensitivity by specimen type – Test sensitivity may vary by type of specimen.
Among upper respiratory tract specimens, nasopharyngeal, nasal mid-turbinate, and saliva specimens have high sensitivity, whereas the sensitivity of anterior nasal and oropharyngeal swab specimens are lower [11,50-54]. In a systematic review conducted by the IDSA that included 44 studies that used NAAT results on nasopharyngeal swab specimens as the comparator, the pooled sensitivities were 92 percent for saliva, 90 percent for nasal mid-turbinate swabs, 81 percent for anterior nasal swabs, and 78 percent for oropharyngeal swabs [11]. The combination of anterior nasal and oropharyngeal swabs had a sensitivity of 87 percent. However, there was substantial heterogeneity across studies (eg, in timing of testing, inclusion of asymptomatic patients, and method of collecting specific specimens) and for some specimens, estimates of sensitivity varied widely.
Self-collected specimens can reduce the need for personal protective equipment since a health care provider is not directly collecting the specimens. Accuracy of NAAT appears to be relatively good when nasal swabs and saliva specimens are self-collected. The sensitivity of NAAT with self-collected anterior nasal or nasal mid-turbinate specimens is high, although not quite as high as that with nasopharyngeal specimens collected by a health care provider [11].
Lower respiratory tract specimens may have higher viral loads and be more likely to yield positive tests than upper respiratory tract specimens [50,55]. In a study of 205 patients with COVID-19 who were sampled at various sites, the highest rates of positive viral RNA tests were reported from bronchoalveolar lavage (95 percent, 14 of 15 specimens) and sputum (72 percent, 72 of 104 specimens), compared with oropharyngeal swab (32 percent, 126 of 398 specimens) [50].
●Sensitivity by illness duration – The likelihood of detectable SARS-CoV-2 RNA may also vary by the duration of illness [49,56,57]. In an analysis of seven studies (including two unpublished reports) that evaluated RT-PCR performance by time since symptom onset or exposure, the estimated rates of false-negative results were 100 percent on the day of exposure, 38 percent on day 5 (estimated as the first day of symptoms), 20 percent at day 8, and 66 percent at day 21 [56]. Heterogeneity across studies and assumptions made in the analysis (eg, about incubation period and time of exposure) reduce confidence in these results. One of the studies included in the analysis used a combination of RT-PCR and an immunoglobulin M (IgM) serologic test to make the diagnosis of COVID-19 and suggested that RT-PCR negative rates were <10 percent on days 1 to 3 of illness, >20 percent at day 6, and >50 percent after day 14; however, these results should also be interpreted with caution, since the serologic test used was not validated for detection of acute infection and IgM tests are frequently falsely positive [57]. Other studies have also suggested that viral RNA levels are high prior to the development of symptoms (ie, in presymptomatic patients) [58]. Nevertheless, low viral RNA levels should not be interpreted as reflecting a late stage of infection, as they may be low at the outset as well [59]. (See "COVID-19: Epidemiology, virology, and prevention", section on 'Viral shedding and period of infectiousness'.)
Data on test performance and viral RNA levels in asymptomatic patients are limited [58]. It is unknown how soon viral RNA can be detectable following exposure and infection.
●Sensitivity by assay type – There are differences in the limit of detection among the major commercial NAAT assays, and retesting samples on different platforms may yield conflicting results [44,45]. Overall, the sensitivity of rapid, point-of-care NAAT assays appears comparable to that of assays performed in the laboratory. In a systematic review conducted by the IDSA, rapid NAAT had 96 percent sensitivity compared with standard laboratory-based NAAT as the reference standard [11]. However, certain point-of-care or rapid NAAT assays may not be as sensitive as standard laboratory-based tests [11,31,33,60].
Cycle threshold — The cycle threshold (Ct) refers to the number of cycles in an RT-PCR assay needed to amplify viral RNA to reach a detectable level. The Ct value can thus indicate the relative viral RNA level in a specimen (with lower Ct values reflective of higher viral levels). Resulting laboratories generally do not provide the Ct value with the qualitative NAAT result, although it can be obtained upon request for some testing platforms. However, the clinical application of the Ct is uncertain [61]. Ct values are not standardized across RT-PCR platforms, so results cannot be compared across different tests. Furthermore, the Ct threshold at which virus can be cultured may vary by variant [62], and no clinical studies have validated use of Ct to guide management.
ANTIGEN TESTING —
Tests that detect SARS-CoV-2 antigen can be performed rapidly and at the point of care and thus are more accessible with a faster time to results than most nucleic acid amplification tests (NAATs); various home antigen tests, performed on nasal swabs, allow individuals to test themselves without presenting to medical care or a testing site. Antigen tests are typically less sensitive than NAATs [33,63,64]. Nevertheless, they can be useful when NAATs are inconvenient or unavailable or when NAAT turnaround times are too long to be clinically useful, provided that clinicians (and individuals who self-test) are aware of the possibility of false negatives and the results are interpreted based on the pretest probability of COVID-19.
Antigen test interpretation
In symptomatic patients — Antigen tests can be useful alternatives to NAAT for diagnosis of SARS-CoV-2 in symptomatic individuals [65,66]. In such cases, the test should be performed within the first five to seven days of symptoms. Although antigen testing cannot detect virus at levels as low as NAAT can, its sensitivity is greatest in the first week of symptoms, when virus replication is at its highest.
●A positive antigen test in a symptomatic individual is indicative of SARS-CoV-2 infection (algorithm 1). The false-positive rate is very low, although the positive predictive value depends on the prevalence of infection. (See 'Low-prevalence settings' below.)
●A negative antigen test in a symptomatic individual could represent a false negative, does not exclude the possibility of SARS-CoV-2 infection, and should be followed by additional testing. Individuals testing at home should repeat an antigen test at 48 hours; this is consistent with recommendations from the US Food and Drug Administration (FDA) [67]. For individuals who are willing to repeat testing with an NAAT (rather than antigen testing), we suggest this approach given the greater sensitivity of NAAT. If the repeat test (with either an antigen test or NAAT) is negative, the likelihood of COVID-19 is low, but testing can be repeated if concern for COVID-19 remains following two negative antigen tests.
Although the sensitivity of antigen testing is higher in symptomatic than in asymptomatic individuals, it remains lower than that of NAAT. Sensitivity is improved with repeating antigen testing. (See 'Accuracy' below.)
For post-exposure testing — Antigen tests can be useful to evaluate for SARS-CoV-2 following close contact with an individual with known or suspected SARS-CoV-2 infection. In order to identify new infections, the CDC recommends that individuals get tested once five days have passed since the last exposure [68]; we typically test five to seven days post-exposure.
●A positive antigen test in an asymptomatic individual following exposure is generally indicative of SARS-CoV-2 infection (algorithm 1). The false-positive rate is very low.
●A negative antigen test in an asymptomatic individual following exposure could represent a false negative and should generally be followed by additional testing. The FDA recommends that individuals testing at home repeat antigen testing two additional times, each 48 hours apart (for a total of three tests) [67]. It is also reasonable to confirm an initial negative antigen test with a follow-up NAAT instead. If all the repeat testing is negative, the likelihood of COVID-19 is low. Any positive test performed in this setting is indicative of infection.
The decision to repeat a negative antigen test should take into account the potential transmission impact of an individual with a false-negative test, which depends on other factors, including vaccination rates in the community, use of masks, and extent of exposure. Repeat testing may reasonably be deferred if the potential impact of a false-negative antigen test is low. Regardless of whether a negative test is confirmed, individuals with known exposure should self-monitor for symptoms of COVID-19 after exposure and get tested if they develop.
Among asymptomatic individuals, the sensitivity of antigen testing is lower than that among symptomatic individuals, and the predictive value of a negative antigen test may be limited [66]. Sensitivity is improved with repeating antigen testing, and three serial tests are more sensitive than two among asymptomatic individuals [69]. (See 'Accuracy' below.)
In outbreak settings — Antigen tests may be useful in outbreak settings and for repeated screening of individuals in high-risk congregate settings to quickly identify individuals with SARS-CoV-2 and isolate them; modelling studies have suggested that if the frequency of testing is high enough, even antigen tests with lower sensitivity could be successfully used to reduce cumulative infection rates [70]. If used for serial testing in these settings, positive tests indicate infection and negative antigen tests do not need to be confirmed [71]. Studies evaluating transmission outcomes after serial testing are limited, and the optimal frequency and duration of serial antigen testing to sufficiently identify infections have not been established. (See "COVID-19: Epidemiology, virology, and prevention", section on 'Screening in selected high-risk settings'.)
Low-prevalence settings — If antigen tests are used in situations when the likelihood of SARS-CoV-2 is low (eg, testing of an asymptomatic individual who has had no household contact with a patient with COVID-19 and lives in a community with low infection rates), a negative antigen test can be interpreted at face value.
Positive antigen tests generally do not warrant confirmation, even in settings of low prevalence. However, it is reasonable to confirm a positive antigen test with an NAAT when the diagnosis is suspect and prevalence is very low (eg, <1 percent); in that case, a negative NAAT would indicate no infection.
To inform duration of precautions after infection — Use of antigen testing in individuals with known infection to inform the duration of isolation and precautions is discussed elsewhere. (See "COVID-19: Infection prevention for persons with SARS-CoV-2 infection", section on 'Duration and subsequent precautions'.)
Accuracy — Systematic reviews of studies evaluating antigen test performance indicate that antigen tests have very high specificity but are generally less sensitive than NAAT; they are most sensitive in symptomatic individuals during the first week of illness [66,72,73].
As an example, in a July 2022 systematic review of 152 studies, in which 49 commercially available antigen tests were evaluated on specimens with known SARS-CoV-2 NAAT results, sensitivity was highly variable, while overall specificity exceeded 99 percent [73]. Average sensitivity was higher among symptomatic compared with asymptomatic participants (73 versus 55 percent) and was higher during the first compared with second week after symptom onset (81 versus 54 percent). Other studies have suggested that antigen test sensitivity peaks approximately three to five days after symptom onset [74-76], although there are many variables that may affect this timing, including the viral variant.
Serial antigen testing appears to improve sensitivity. In a prospective study of over 5500 individuals who underwent serial antigen and NAAT testing every 48 hours for two weeks, antigen test sensitivity on the first day of NAAT positivity was 59 percent for symptomatic individuals and 9 percent for asymptomatic individuals [77]. Sensitivity with two or three serial tests increased to 92 and 94 percent for symptomatic individuals and to 39 and 56 percent for asymptomatic individuals.
Antigen test sensitivity also improves with increasing levels of viral RNA in upper respiratory specimens (ie, as reflected by lower Ct values on NAAT). In the systematic review detailed above, average sensitivity was 94 to 97 percent for Ct value ≤25, 69 percent for Ct value 25 to 30, and 19 percent for Ct value >30 [73].
Studies have also suggested that a positive antigen test is more likely than a positive RT-PCR test to indicate that infectious virus could be cultured from specimens [78-80]. As an example, in a study of 60 individuals with SARS-CoV-2 infection who underwent daily antigen and viral culture testing, antigen test was positive on 93 percent of specimens that were culture positive but only 28 percent of specimens that were culture negative [80]. However, since infectious virus can still be isolated from antigen-negative specimens, clinicians and patients should be aware that a negative antigen test cannot be used to indicate that a person is not infectious. In other studies, the rate of positive viral culture in specimens that were antigen negative ranged from 9 to 16 percent [64,74].
Studies suggest that antigen testing reverts to negative within 5 to 9 days following initial infection for approximately half and within 14 days for most individuals [80-82], although outliers have been reported [83].
False-positive results are rare but have been reported. In a study of over 11,000 participants who underwent NAAT and antigen testing daily for over a week, only 191 (1.7 percent) had a false-positive antigen test [84].
Performance also varies by the specific antigen test. Antigen tests that have been granted EUA in the United States can be found on the FDA website and the COVID-19 Testing Toolkit site, which also lists the manufacturer reported sensitivity and specificity [85,86].
Impact of the Omicron variant on antigen test accuracy is discussed elsewhere. (See 'Impact of SARS-CoV-2 mutations/variants on test accuracy' below.)
TESTS WITH LIMITED UTILITY OR AVAILABILITY
Serology to identify prior/late infection — Serologic tests detect antibodies to SARS-CoV-2 in the blood, and those that have been adequately validated can help identify patients who previously had SARS-CoV-2 infection as well as patients with current infection who have had symptoms for three to four weeks. However, serology does not have a significant role in the diagnosis of SARS-CoV-2 infection. Because serologic tests are less likely to be reactive in the first several days to weeks of infection, they have very limited utility for diagnosis in the acute setting [22,87,88]. Furthermore, since individuals can have detectable antibodies for months following infection, serology cannot distinguish between recent and more remote infection.
If serology is performed, we suggest using IgG antibody or total antibody tests rather than IgM antibody, IgA antibody, or IgM/IgG differentiation tests because of their greater accuracy; this is consistent with recommendations from the Infectious Diseases Society of America (IDSA) [89]. If a serologic test to identify prior infection is performed in an individual who has received a spike protein COVID-19 vaccine, a test that detects antibodies to antigens other than the spike protein should be used. (See 'Testing following COVID-19 vaccination' below.)
Accuracy of serologic tests varies widely by assay. In the United States, the Food and Drug Administration (FDA) has independently evaluated the accuracy of certain serologic tests for COVID-19 and posted results on its website.
In general, detectable antibodies take several days to weeks to develop, and the time to antibody detection varies by test [57,90-97]. In a systematic review of 178 studies that evaluated the sensitivity of serologic testing by time since symptom onset in patients with COVID-19, IgM or IgG was detected in 41 percent by one week, in 75 percent by two weeks, and in 88 percent by three weeks; during convalescence, the average sensitivity of IgG up to 100 days was 90 percent [94]. However, various serologic tests, including in-house laboratory-developed tests, were used in these studies, and their sensitivities within different time frames vary substantially.
Specificity also varies by type of serologic assay. In contrast with IgG antibody and total antibody tests, IgM antibody, IgA antibody, and IgM/IgG differentiation tests generally have specificities below 99 percent [89]. Cross-reactivity with other coronaviruses and other viral pathogens is a potential concern [98].
Most studies suggest that most individuals have durable, detectable IgG levels up to eight months after infection [99-105]; however, some have reported a faster rate of antibody decline [106-109]. The duration that antibodies remain detectable likely depends on the height of the initial antibody response as well as the severity of the infection [110].
Other tests
●Breathalyzer – In the United States, the FDA authorized a breathalyzer that uses gas chromatography-mass spectrometry to detect exhaled volatile organic compounds specific to SARS-CoV-2 infection [111,112]. The contraption is the size of a small suitcase and returns results in approximately three minutes. According to the authorization documents, sensitivity and specificity were 91 and 99.3 percent among a group of asymptomatic and symptomatic individuals compared with PCR on nasopharyngeal swabs. However, different viral variants may have different gas chromatography profiles that could impact test accuracy [113]. More detailed performance data, as well as other information such as cost, are necessary to inform the optimal role of this device in COVID-19 diagnosis.
●Viral culture – For safety reasons, specimens from a patient with suspected or documented COVID-19 should not be submitted to clinical laboratories for viral culture. Viral culture is mainly reserved for research purposes. (See "COVID-19: Epidemiology, virology, and prevention", section on 'Viral shedding and period of infectiousness'.)
●Other tests of SARS-CoV-2 immunity – The clinical utility of tests that can detect cell-mediated immune responses to SARS-CoV-2, such as with an interferon-gamma release assay, is also being explored [114].
SPECIAL SITUATIONS
Persistent or recurrent positive NAAT during convalescence — Patients diagnosed with COVID-19 can have detectable SARS-CoV-2 RNA in upper respiratory tract specimens for weeks after the onset of symptoms [38]. Recurrently positive nucleic acid amplification tests (NAATs) following several negative tests have also been well documented in some patients with COVID-19.
However, prolonged or recurrent viral RNA detection does not necessarily indicate prolonged infectiousness. Isolation of virus in culture from upper respiratory specimens more than 10 days after illness onset has only rarely been documented in patients who recovered from nonsevere infection [40,115,116]. Additionally, viral RNA levels during convalescence, if detectable, are lower than those detected during acute illness, and studies have suggested that infectious virus cannot be detected below a certain viral threshold [115]. These data and the implications for discontinuing isolation precautions are discussed in detail elsewhere. (See "COVID-19: Epidemiology, virology, and prevention", section on 'Viral shedding and period of infectiousness' and "COVID-19: Infection prevention for persons with SARS-CoV-2 infection", section on 'Duration and subsequent precautions'.)
In the United States, the Centers for Disease Control and Prevention (CDC) suggests against retesting asymptomatic individuals who were previously diagnosed with SARS-CoV-2 within the prior three months with NAAT because of the low likelihood that a repeat positive NAAT during this interval represents an active infection [117].
Diagnosis of reinfection — The diagnosis of reinfection is straightforward (ie, with positive NAAT or antigen testing) when many months have elapsed between the two infections. (See 'Establishing the diagnosis' above.)
However, establishing the diagnosis of reinfection can be challenging when it occurs within a few months of the prior infection. For individuals who warrant testing within three months of an earlier infection, antigen testing is preferred. Because of the possibility of prolonged respiratory shedding of viral RNA following acute infection, a repeat positive NAAT in this setting does not necessarily indicate a new infection. If an NAAT was performed and is positive within three months of an earlier infection, features that increase the likelihood of reinfection include new symptoms consistent with COVID-19, a longer interval since the first infection, and a high viral RNA level on repeat testing (eg, Ct value <33).
Interpretation of antigen testing in this setting is the same as for the general population. (See 'Antigen test interpretation' above.)
Impact of SARS-CoV-2 mutations/variants on test accuracy — Over the course of the pandemic, several SARS-CoV-2 variants that have mutations in the spike protein have become prevalent globally. (See "COVID-19: Epidemiology, virology, and prevention", section on 'Viral evolution and variants of concern'.)
●Impact on NAAT – Some variants have spike mutations (in particular, a deletion at amino acids 69-70) that impact the ability of SARS-CoV-2 NAAT to detect the S gene that encodes the spike protein. Most NAATs will still detect SARS-CoV-2 RNA even if they fail to detect the mutated S gene because they have been designed to detect more than one gene target [118]. S gene target failure can be used as a marker for variants that contain spike protein mutations for surveillance purposes [119].
Most NAATs still detect Omicron and its subvariants; some result in S gene target failure with certain Omicron subvariants. Rare NAATs that are expected to fail to detect specific variants completely are detailed on the US Food and Drug Administration (FDA) website [120,121].
●Impact on antigen tests – Most antigen tests target nucleocapsid protein, so mutations in the spike protein would not impact the accuracy of such antigen tests. Variants in the Omicron lineage have mutations in the nucleocapsid protein compared with older variants, although antigen tests continue to detect Omicron and its subvariants.
Most available evidence suggests that the sensitivity of antigen testing relative to NAAT in detecting the Omicron variant on nasal specimens is similar to that with other SARS-CoV-2 strains (ie, overall sensitivity is modest but sensitivity for high viral levels is high [122-124]). In one study of 731 individuals in California who were tested simultaneously with anterior nasal rapid antigen testing (BinaxNOW) and anterior nasal NAAT during the initial Omicron surge, of the 296 who tested positive by NAAT, 193 had a positive antigen test (overall sensitivity 65.2 percent) [125]. However, antigen test sensitivity was higher among those whose NAAT indicated higher viral RNA levels (95 percent with Ct <30 and 82 percent with Ct <35).
Testing following COVID-19 vaccination
●Viral tests – Viral tests in individuals who have received a COVID-19 vaccine should be interpreted in the same way as in unvaccinated individuals. Vaccination does not influence the results of viral tests, so a positive viral test result in a vaccinated individual should not be attributed to the vaccine.
●Serology – Advisory organizations in the United States and elsewhere recommend against routine serologic testing following infection or vaccination to assess immune status or determine the need for vaccination [87,126]. The role of serology in such settings is limited.
COVID-19 vaccination elicits antibodies against the SARS-CoV-2 spike protein. Some serologic tests detect anti-spike antibodies, and these tests cannot distinguish between prior infection and prior vaccination (a reactive result could indicate prior infection, prior vaccination, or both). Some serologic tests only detect antibodies against nucleocapsid protein; vaccines authorized in the United States, Canada, and Europe do not elicit such antibodies, so a reactive result on a nucleocapsid protein-based serologic test in an individual who has received one of these vaccines would suggest a history of infection. However, some evidence suggests that vaccinated individuals are less likely to develop anti-nucleocapsid antibodies after infection than unvaccinated individuals [127].
Multisystem inflammatory syndrome in children — For children or adolescents with suspected multisystem inflammatory syndrome, testing for SARS-CoV-2 includes both NAAT and serology. Many children with this syndrome have had detectable antibodies to SARS-CoV-2 but a negative NAAT. The diagnosis and evaluation of multisystem inflammatory syndrome are discussed elsewhere. (See "COVID-19: Multisystem inflammatory syndrome in children (MIS-C) clinical features, evaluation, and diagnosis", section on 'Evaluation'.)
SOCIETY GUIDELINE LINKS —
Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: COVID-19 – Index of guideline topics".)
INFORMATION FOR PATIENTS —
UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: COVID-19 overview (The Basics)" and "Patient education: COVID-19 and pregnancy (The Basics)" and "Patient education: COVID-19 and children (The Basics)" and "Patient education: COVID-19 vaccines (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Clinical suspicion – The possibility of COVID-19 should be considered in patients with compatible symptoms (table 1), in particular fever and/or respiratory tract symptoms, who reside in or have traveled to areas with community transmission or who have had potential exposure to SARS-CoV-2 through an outbreak setting or close contact with a confirmed or suspected case. Clinicians should also be aware of the possibility of COVID-19 in patients with severe respiratory illness when no other etiology can be identified. (See 'Clinical suspicion' above.)
●Whom to test – We suggest testing for SARS-CoV-2 in symptomatic patients with suspected COVID-19 when the result would have specific treatment or infection control implications (eg, patients who are hospitalized, would be candidates for outpatient antiviral therapy, or have close contact with vulnerable individuals). Testing asymptomatic individuals is not routinely warranted but may be useful in selected cases following exposure to SARS-CoV-2; if this is done, we suggest testing be performed five to seven days after exposure, although the optimal timing is uncertain. (See 'Whom to test' above.)
●Choice of test – A viral test is necessary to make the diagnosis of SARS-CoV-2 infection: either a nucleic acid amplification test (NAAT, most commonly a reverse-transcription polymerase chain reaction [RT-PCR] assay), or an antigen test. We suggest NAAT, if readily available with a reasonable turnaround time, because of its superior sensitivity; however, antigen tests are more accessible, more convenient, provide results faster than laboratory-based NAAT, and they are good alternatives as long as users appreciate the potential need for repeat testing to optimize sensitivity. (See 'Choosing an initial diagnostic test' above.)
•NAAT accuracy and interpretation – NAATs detect SARS-CoV-2 RNA in patient specimens and are highly specific; they are most commonly laboratory based, although point-of-care NAATs are also available. They can detect low levels of viral RNA; the precise sensitivity of these tests in the clinical setting likely depends on the type and quality of the specimen obtained, the duration of illness at the time of testing, and the specific assay. (See 'NAAT (including RT-PCR)' above.)
A positive NAAT for SARS-CoV-2 confirms the diagnosis of COVID-19. (See 'Positive NAAT result' above.)
For most symptomatic individuals, a single negative NAAT result is sufficient to exclude the diagnosis of COVID-19. (See 'Negative initial NAAT result' above.)
•Antigen test accuracy and interpretation – Antigen tests can be performed rapidly at home or at the point of care. Positive tests can be interpreted as indicative of SARS-CoV-2 infection. Because they are less sensitive than NAAT, negative antigen tests performed for symptoms or recent exposure should generally be confirmed with repeat testing (algorithm 1). Antigen tests can also be useful for serial screening for infection in high-risk settings (eg, congregate settings). (See 'Antigen testing' above.)
●Limited role for serology – Detectable antibodies generally take several days to weeks to develop and can persist for months; thus, serologic tests have less utility for diagnosis in the acute setting and cannot distinguish recent from remote infection. (See 'Serology to identify prior/late infection' above.)
●Interpreting tests in vaccinated individuals – Vaccination status does not influence interpretation of viral test (ie, NAAT or antigen test) results. Some serologic tests target the nucleocapsid protein; only serologic tests that target the spike protein can detect antibody response to currently available vaccines, but they cannot distinguish a vaccine response from prior infection. (See 'Testing following COVID-19 vaccination' above.)