The latest review, published in the British Medical Journal (BMJ), found important weaknesses in the evidence base for coVID-19 antibody tests.
The scientists involved in the study said the evidence was particularly weak for immediate testing, which is carried out directly on patients outside the laboratory, and they did not support continuing such testing as a reference.
Serological tests for COVID-19 antibodies can improve diagnosis and be a useful tool for monitoring infection levels in the population. As Prime Minister Boris Johnson has said, antibody tests are a “game changer” in responding to the epidemic, but it is important to formally assess whether there is sufficient evidence that they are accurate.
As a result, an international research team set out to determine the diagnostic accuracy of coVID-19 antibody tests.
From 1 January to 30 April 2020, they searched medical databases and preprinted servers to study the sensitivity and/or specificity of coVID-19 antibody tests versus control tests.
Sensitivity is the percentage of a person diagnosed correctly with a disease, and Specificity is the percentage of a person diagnosed correctly without a disease.
Of the 40 eligible studies, most (70%) were from China, with the remainder from the UK, US, Denmark, Spain, Sweden, Japan and Germany.
Half of the studies were not peer-reviewed, and most found a high or ambiguous risk of bias (problems in the study design that could influence the results). Only four of the studies included outpatient visits, and only two were evaluated at the care site.
When the sensitivity results for each study were put together, they ranged from 66% to 97.8%, depending on the type of test used, meaning that between 2.2% and 34% of coVID-19 patients would fall through the net in the process.
The combination specificity ranges from 96.6% to 99.7%, depending on the test used, meaning that between 3.4% and 0.3% of patients will be incorrectly diagnosed with COVID-19.
Compared with other detection methods, the comprehensive sensitivity of side-flow immunoassay (LFIA) is always low. LFIA testing is a potential real-time detection method that is currently being considered.
Based on these results, the authors explain that if an LFIA test were applied to a population with a 10% coVID-19 prevalence, for every 1,000 people tested, 31 people who never developed covid-19 would be wrongly told they were immune to the disease, and 34 people would be wrongly told they had never been infected.
The comprehensive sensitivity of the commercial kit was also lower than that of the non-commercial kit (88.2%), and the comprehensive sensitivity of the first and second week after symptom onset was lower than the second week.
The researchers noted some limitations, such as differences in the study population and possible missing studies. Advantages, on the other hand, include thorough search strategies and biased/subjective assessments.
“These observations point to significant deficiencies in the evidence for COVID-19 serological tests, particularly those marketed as point of care tests,” they wrote.
“While the scientific community should be commended for the pace of development of new serological tests, this review highlights the need for high-quality clinical research to evaluate these tools,” they conclude. “With international collaboration, research like this can be done quickly.”
The novel Coronavirus test still has many aspects worth questioning, which itself is also caused by the “stealth” and “cunning” of the virus. The results of a recent study published in the Annals of Internal Medicine suggest that asymptomatic infections may account for 45 per cent of the total number of COVID-19 cases and play an important role in early and sustained transmission of COVID-19. The report stresses the need for expanded detection and contact tracing to mitigate the pandemic. The study, by Scripps Research, analyzed public data sets on asymptomatic infections and highlighted the importance of detecting viruses on a large scale and accurately.
No related posts.