November 30, 2020
It is no secret the breathtaking advance of the COVID-19 virus has placed pressure on every aspect of our lives, but the pressures on health professionals to accurately diagnose COVID-19 infections are in the spotlight. While false-positive results may result in unnecessary precautions and treatments, false negatives can lead to a misplaced sense of confidence and rapid spreading of the virus. Worse still, a reluctance to take symptoms seriously after a false negative can have fatal consequences.
There are currently two general testing approaches employed worldwide today: diagnostic testing, and antibody testing. COVID-19 diagnostic testing is primarily conducted using reverse transcription-polymerase chain reaction, or RT-PCR. This testing typically requires a nasal swab whereby the genetic material collected is treated in a complex series of reactions to amplify the quantities to detectable levels which are then verified against an existing COVID-19 viral signature, or against a reference of a COVID-19 antigen that focuses on specific viral proteins.
Technological advances in RT-PCR testing allow results to be generated in as little as 2-4 hours, with some platforms, results are available in about 5 minutes.
Yet another diagnostic test, rapid antigen testing, also requires either a nasal or saliva swab but has fallen upon disfavor due to a false negative rate approaching 50%.
The test itself is simple. Samples from the swab are processed to break up any viruses present and release certain viral fragments, or antigens. The sample is swiped on a credit card-sized test strip embedded with an antibody specially designed to bind with the viral antigens if they are present. If the antigens are present, either a brown stripe appears, or a positive result is detected as a fluorescent glow.
A large number of false negatives correlates to the viral load. Rapid antigen testing works best with a high viral load. Low levels during the initial stages of infection are the root cause of false negatives.
In either case, a positive result indicates an active COVID-19 infection.
Diagnostic testing can certainly detect the presence of the COVID 19 virus but does little in terms of assessing immunity. False negatives are much more likely at the onset of exposure due to the low viral load. Researchers found that the probability of a false negative result decreases from 100% on day 1 of being infected to 67% on day 4, and drops to 20% by day 8 (three days after a person begins experiencing symptoms).
Antibody testing is conducted via a serological test called an immunoassay. The most common of which is called enzyme-linked immunosorbent assay, or ELISA. These assays are highly automated and typically run several samples at one time. While the assay may complete in as little as 2-5 hours, great quantities of collected samples are pooled and tested together so delays in obtaining results may exceed 7 days simply due to the batch orientation of the testing.
Serological testing has a fundamental limitation in that blood samples drawn less than two weeks after initial infection may show no signs of the presence of COVID-19 antibodies as the response time for the body to ramp up a defense against the infection can affect the results.
Beyond the time-based sensitivity as to when specimens are collected and their effect on false positives or false negatives, the very nature of the available testing methods may lead to a variety of errors.
Summary of false-negative percentage of some COVID-19 tests are below
|Test||False Negative %||Highest Observed False Negatives|
|Rapid Antigen Testing||1.5%||50%|
Fortunately, a COVID-19 lab management software or a simply put, a COVID-19 LIMS can reduce or eliminate many of these potential sources of errors.
The collection of both dry or nasopharyngeal swabs used for most genomic-based testing has come under some scrutiny. Transport media, the solution used to stabilize samples, is important to stave off sample degradation in those cases where they must be transported for testing. In general, the transport media for COVID-19 genomic testing destroys the virus while keeping the all-important genetic material intact.
A great many samples destined for contract or large-scale testing labs are collected in kits. In many cases, kits contain sample collection containers. Samples collected must be identified to the patient from which they are collected. Multiple collection containers must be placed in a shipment box, and many shipment boxes may be loaded into larger shipment containers.
For the following, we will use the term “barcode”, but at many levels in the labeling hierarchy, some combination of human-readable labels, barcodes, and RFID chips can coexist.
At the sample collection point, a LIMS screen can be accessed to allow the correlation of the sample collected with the patient. The required COVID-19 test would likely be entered at this time as well. This is key, as we will see in a moment.
As multiple patient samples are collected, they are placed in a larger box. That box’s label must be correlated to the patient samples within the box. If yet another shipping box is required, the process of correlation with all the sample boxes and the parent box label is repeated.
All of the labels are linked in LIMS so we know with one simple scan what the contents are in the sample boxes.
Upon receipt in a testing lab, the shipping boxes are scanned. As the sample containers are removed, they can be either scanned manually or by an automated device or auto sorter. Either way, since we have linked the sample IDs to the test required, they can quickly be dispatched to the testing stations.
Testing is only as good as the competency and skills of the technicians conducting the preparation and testing of the samples. The volume of testing, internal profitability targets, staff training, tedium, and resource availability, can significantly strain organizations and open the door for an increase in procedural errors.
Automated workflows within a COVID-19 LIMS can direct technicians and analysts to execute method steps in a “cookbook” fashion, even to the extent that the process cannot advance until verification of the previous steps has occurred. Enforcing such a degree of control can greatly increase the overall quality of the testing processes. Furthermore, a LIMS helps in scheduling staff training, monitoring ongoing training, and maintaining training and competency records of staff. This enables laboratory managers to assign specimen preparation or diagnostic testing and reporting of COVID-19 specimens only to competent staff.
Analytical instruments, such as RT-PCR, must be calibrated from time to time as per a laboratory’s internal standard operating procedures or manufacturers’ guidelines to avoid erroneous testing of samples. Furthermore, the maintenance of analytical instruments must be scheduled and performed on time to prevent any chances of false positives or negatives. The calibration and maintenance results must be properly recorded, and the next set of calibration and maintenance must be pre-scheduled to prevent instrumental errors from interfering in the accurate diagnosis of samples.
Liquid handling instruments, particularly with high-throughput capabilities have been refined over the years to ensure samples are not cross-contaminated. But, what if the very test kits were contaminated before being sent out to hospitals and clinics? This in fact may have happened early in the COVID-19 crisis when a federal review uncovered the possible contamination of Centers for Disease Control (CDC) test kits during manufacturing. Another concern is instrument sensitivity and specificity, and the FDA exhaustively defines ranges for instrument sensitivity before approving the device for use or validation in post-authorization.
A COVID-19 lab management software can help schedule calibration and maintenance of analytical instruments to prevent instrument errors, thereby preventing chances of false positives or negatives. It alerts laboratory managers ahead of the scheduled calibration or maintenance date so that necessary actions can be initiated on time. Most modern instruments generate error codes when any type of misread or malfunction occurs. These errors can be expeditiously addressed if they are conveyed to a LIMS and reported so they can be categorized as transient or systematic errors. A LIMS also helps in maintaining records of all instruments available in a laboratory to perform COVID-19 diagnostic tests along with their sensitivity and specificity information.
Not all transcription errors are directly attributable to humans transposing, omitting, or simply entering incorrect information. Bearing in mind the volume of data involved with testing thousands of patients necessarily means that some form of instrument interfacing or post-processing of data is required.
For example, in Texas a coding error and a system upgrade caused a sudden reading of 124,693 cases of COVID-19 to be reported when a contract lab discovered that files sent to the Texas Department of State Health (DSHS) contained an error in a data field causing the DSHS systems to not be able to read the data. When the programming errors were fixed, the heretofore positive results were tallied.
Transcription errors are largely eliminated by direct interfacing of analytical instruments with a COVID-19 lab management software. Not only results are quickly transferred to the software, but also field-level validation of the entered results ensures that nonsensical values are immediately rejected, and remaining values fit the parameters defined.
Patients have both a need and a right to receive accurate COVID-19 test results in the shortest possible time frame. COVID-19 diagnostic laboratories need to securely deliver test results to patients in real-time to minimize the turnaround time. Quick delivery of results is also important to initiate timely clinical care of COVID-19 positive patients and to isolate them to prevent the spread of infection.
Moreover, the anxiety patients experience in awaiting test results can be greatly reduced by the rapid delivery of results to them via a patient portal where results can be delivered at the moment they are released by a COVID-19 diagnostics.
As we have seen, the accuracy of testing depends on several factors, many of which are related to the time the patient samples are taken after possible infection. While the sampling interval errors cannot be averted using a LIMS, systemic errors in terms of method execution, instrumental errors, transcription, and calculations can most certainly be reduced with a COVID-19 lab management software.