Screening Versus Diagnostic Tests for COVID-19, What’s the Difference?

Dec. 7, 2020

During the COVID-19 pandemic, clinical testing has become a hot topic. The pandemic has required enormous effort on the part of both medicine and public health, but in the United States, the way these specialties function and use testing looks very different. Understanding the differences between tests in terms of their intended use and performance is imperative in achieving positive health outcomes for both the individual and the population as a whole. 

Medicine and Public Health Have Different Roles

While the fields of public health and medicine complement one another, their primary goals and focus are not entirely the same. In medicine, the focus is on the health of the individual, with overall health also determined by the context of their family and community. In public health, the primary focus is on the health of larger populations. Medicine aims to prevent, diagnose and treat disease within individuals, while public health aims to prevent and control the spread of disease throughout a population. These differences are also apparent when comparing testing for medical and public health purposes. While public health testing and clinical diagnostics may have similar or identical procedures, their primary functions differ. It is all about context: the clinical (diagnostic) test is responsible for supporting individual patient care, while public health testing helps identify and control outbreaks, as well as supporting surveillance programs

Screening Tests

Screening tests are primarily used in asymptomatic populations to determine the likelihood of having or developing a particular disease. The primary function of these tests is to identify individuals that are at a higher risk of developing a particular disease, and provide them with either follow-up testing that can confirm the presence or absence of the disease or coaching on lifestyle changes that can promote continued wellness. Importantly, screening tests are not designed to diagnose disease or illness. These tests are ideal for the early detection of disease, encouraging healthy lifestyle changes and supporting surveillance of disease within a population. Examples of screening tests include perinatal syphilis, HIV and Group B Strep screening, the prostate specific antigen (PSA) test for prostate cancer and the tuberculin skin test (TST) for tuberculosis (TB). Screening tests almost always require a second, diagnostic test to confirm disease. For example, the tuberculin skin test simply detects if a person was infected with Mycobacterium tuberculosis at some point in their life, but it cannot discern between latent TB infection or active TB disease. A chest x-ray and culture must be performed to confirm the presence or absence of TB disease. 

The tuberculin skin test, which is used to screen for M. tuberculosis infection, and a lung X-ray, which is used to diagnose active TB.
The tuberculin skin test (left), which is used to screen for M. tuberculosis infection, and a lung X-ray (right), which is used to diagnose active pulmonary tuberculosis.
As is true with all tests, screening tests are not 100% accurate. While they can be essential components to a healthy lifestyle when used in appropriate situations, it is important to note that using screening tests for rare diseases or in individuals who are not at high risk for disease can cause more harm than good. For example, the tuberculin skin test can be falsely positive due to infection with nontuberculous Mycobacteria or because the patient previously received the bacillus Calmette-Guérin vaccine (BCG). Without performing a risk assessment or confirmatory diagnostic testing, a person without TB infection who had a positive screening test may be subject to an intense 3-9 month multidrug treatment regimen, and undue psychological stress. In contrast, if the sensitivity of the screening test is not high, then a person might be told that they do not have a disease when they actually do, which puts that individual as well as others around them at risk of contracting the disease. 

As discussed in the article Why Pretest and Posttest
Fagan’s Nomogram showing how the pretest probability and likelihood ratio lead to different post-test probabilities.
Fagan’s Nomogram showing how the pretest probability and likelihood ratio of a positive diagnostic test lead to a posttest probability of 70% (black), while the pretest probability and likelihood ratio of a negative diagnostic test lead to a posttest probability of 0.50% (red).
Source: Andrea Prinzi, modified from https://commons.wikimedia.org/wiki/File:Fagan_nomogram.svg
Probability Matter in the Time of COVID-19,the way a test result is interpreted depends on a variety of factors, several of which differ between populations being screened (typically asymptomatic) and those being tested for suspected disease (symptomatic). Fagan’s Nomogram is a helpful tool that demonstrates this for us. Pre-test probability, a factor that is determined by the frequency of disease in a community (prevalence) or the presence or absence of strong clinical suspicion of disease, will be higher when the disease is expected. Connect a line from the pre-test probability through the likelihood ratio (a value calculated from test performance factors like sensitivity and specificity), and you’re left with a value that tells you how confident you can be in a test result. The positive predictive value of a test will generally be higher when it is run on people with symptoms (higher pre-test probability) than when it is run on asymptomatic patients (lower pre-test probability).

Diagnostic Tests

Diagnostic tests are used to detect the presence or absence of disease in an individual. Since further medical decision making is based on the results of a diagnostic test, considerations of pretest probability and test reliability are very important. In medicine, diagnostic testing is only warranted if the result of the test will change management of the patient or provide meaningful clinical information about the patient’s expected outcomes or prognosis. Take for example children who present to the hospital with suspected meningitis or encephalitis. Research from a large pediatric hospital shows that after the introduction of a multiplex PCR for meningitis/encephalitis that detects herpes simplex virus (HSV) among other central nervous system pathogens, the amount of testing doubled, but the use of the antiviral agent acyclovir to empirically treat HSV stayed the same. This begs the question, if the patient will be treated or managed the same way regardless of the PCR result, why test at all? Diagnostic tests identify the cause of a patient’s illness, influence the way a patient is treated and provide the information needed to perform public health surveillance. The ideal diagnostic test would be portable, affordable, rapid, highly sensitive and specific and suitable for a variety of specimen types.

COVID-19 Screening and Surveillance Considerations

Currently, the tests being used to detect SARS-CoV-2 infection have been validated and approved for diagnostic testing, meaning that they have been validated in patients who have symptoms. While the fixed test characteristics of the current SARS-CoV-2 assays will stay the same, the way the results are used and interpreted will differ entirely depending on contextual factors, and important points should be considered if they are to be used for screening. Given that asymptomatic transmission is an important component of the pandemic, using tests for screening and surveillance could play an important role in slowing or preventing the spread of disease.

Wastewater Sampling

Although the role of SARS-CoV-2 in stool is unknown at this time, the virus can be detected in fecal matter for up to 3 weeks after illness onset. Research suggests that daily sampling of wastewater to test for SARS-CoV-2 RNA (using similar technology to that used to diagnose individual patients) provides data similar to daily testing of many individuals within a community, but is less invasive and more cost-effective. While public health interventions should not be based solely on wastewater test results, they may help provide data about areas where timely diagnostic testing is unavailable or underutilized. Additionally, testing wastewater can provide data on viral concentrations in a given area over time. This strategy is actively being used at colleges across the United States, which has helped many of them identify key trends in data that allow them to follow up with individual dorms and perform diagnostic testing on the individuals who live there.

Pooled Testing

Pooled testing, which involves combining samples from many individuals and running them on one test, may be a good option for SARS-CoV-2 screening. Pooling is most effective when the prevalence of disease is low in a community. Since screening involves testing an asymptomatic population, where the prevalence of disease is likely to be lower, pooling may serve as an acceptable option. Pooling samples can also be useful for testing symptomatic individuals, as long as prevalence is low in the symptomatic population. 

Using SARS-CoV-2 Diagnostic Tests Off-Label for Screening

The tests that are currently authorized for SARS-CoV-2 detection have been designed and validated to be used on symptomatic individuals who are suspected of having COVID-19. However, the Centers for Disease Control (CDC) and Food and Drug Administration (FDA) have provided recommendations to consider for healthcare providers and organizations who are interested in using these diagnostic tests to screen asymptomatic individuals who are not suspected of having COVID-19. The Food and Drug Administration recommends that the following be considered:
  1. A highly sensitive test should be considered for screening purposes since all components of viral dynamics are not clear between asymptomatic and symptomatic individuals. 
  2. If a highly sensitive test is not feasible or turnaround times are unacceptable, health care providers may consider less sensitive point-of-care tests. In settings such as nursing homes, repeat testing using rapid point-of-care tests may be more useful for overall infection control rather than highly-sensitive tests with long turnaround times. 
  3. Screening asymptomatic populations for COVID-19 should not be standalone, and should be a part of an infection control plan that includes additional preventative measures. The consequences of poor test performance are not negligible and should be seriously considered when using rapid tests off-label. 
The COVID-19 pandemic is unprecedented, and the challenges to curbing the spread of disease seem endless. Although these are desperate times, it is important to pause and consider the utility of the tools in our tool belt. Each type of test has something to offer, but is only truly useful when used in a way that results in more benefit than harm or risk. There is an enormous role for both public health and clinical diagnostics in the COVID-19 outbreak, as well as the methods utilized by each. Various testing strategies, as well as preventative public health measures, should be taken together to optimize the impact of each.
Read more about considerations for SARS-CoV-2 screening using rapid antigen tests.

Author: Andrea Prinzi

Andrea Prinzi
Andrea Prinzi SM(ASCP),MPH,CPH has been a clinical microbiologist in Denver, Colorado for the last 11 years. She is actively pursuing her PhD in Clinical Science at the University of Colorado Denver.