Search

Background Accurate rapid diagnostic tests for SARS‐CoV‐2 infection could contribute to clinical and public health strategies to manage the COVID‐19 pandemic. Point‐of‐care antigen and molecular tests to detect current infection could increase access to testing and early confirmation of cases, and expediate clinical and public health management decisions that may reduce transmission. Objectives To assess the diagnostic accuracy of point‐of‐care antigen and molecular‐based tests for diagnosis of SARS‐CoV‐2 infection. We consider accuracy separately in symptomatic and asymptomatic population groups. Search methods Electronic searches of the Cochrane COVID‐19 Study Register and the COVID‐19 Living Evidence Database from the University of Bern (which includes daily updates from PubMed and Embase and preprints from medRxiv and bioRxiv) were undertaken on 30 Sept 2020. We checked repositories of COVID‐19 publications and included independent evaluations from national reference laboratories, the Foundation for Innovative New Diagnostics and the Diagnostics Global Health website to 16 Nov 2020. We did not apply language restrictions. Selection criteria We included studies of people with either suspected SARS‐CoV‐2 infection, known SARS‐CoV‐2 infection or known absence of infection, or those who were being screened for infection. We included test accuracy studies of any design that evaluated commercially produced, rapid antigen or molecular tests suitable for a point‐of‐care setting (minimal equipment, sample preparation, and biosafety requirements, with results within two hours of sample collection). We included all reference standards that define the presence or absence of SARS‐CoV‐2 (including reverse transcription polymerase chain reaction (RT‐PCR) tests and established diagnostic criteria). Data collection and analysis Studies were screened independently in duplicate with disagreements resolved by discussion with a third author. Study characteristics were extracted by one author and checked by a second; extraction of study results and assessments of risk of bias and applicability (made using the QUADAS‐2 tool) were undertaken independently in duplicate. We present sensitivity and specificity with 95% confidence intervals (CIs) for each test and pooled data using the bivariate model separately for antigen and molecular‐based tests. We tabulated results by test manufacturer and compliance with manufacturer instructions for use and according to symptom status. Main results Seventy‐eight study cohorts were included (described in 64 study reports, including 20 pre‐prints), reporting results for 24,087 samples (7,415 with confirmed SARS‐CoV‐2). Studies were mainly from Europe (n = 39) or North America (n = 20), and evaluated 16 antigen and five molecular assays. We considered risk of bias to be high in 29 (37%) studies because of participant selection; in 66 (85%) because of weaknesses in the reference standard for absence of infection; and in 29 (37%) for participant flow and timing. Studies of antigen tests were of a higher methodological quality compared to studies of molecular tests, particularly regarding the risk of bias for participant selection and the index test. Characteristics of participants in 35 (45%) studies differed from those in whom the test was intended to be used and the delivery of the index test in 39 (50%) studies differed from the way in which the test was intended to be used. Nearly all studies (97%) defined the presence or absence of SARS‐CoV‐2 based on a single RT‐PCR result, and none included participants meeting case definitions for probable COVID‐19. Antigen tests Forty‐eight studies reported 58 evaluations of antigen tests. Estimates of sensitivity varied considerably between studies. There were differences between symptomatic (72.0%, 95% CI 63.7% to 79.0%; 37 evaluations; 15530 samples, 4410 cases) and asymptomatic participants (58.1%, 95% CI 40.2% to 74.1%; 12 evaluations; 1581 samples, 295 cases). Average sensitivity was higher in the first week after symptom onset (78.3%, 95% CI 71.1% to 84.1%; 26 evaluations; 5769 samples, 2320 cases) than in the second week of symptoms (51.0%, 95% CI 40.8% to 61.0%; 22 evaluations; 935 samples, 692 cases). Sensitivity was high in those with cycle threshold (Ct) values on PCR ≤25 (94.5%, 95% CI 91.0% to 96.7%; 36 evaluations; 2613 cases) compared to those with Ct values >25 (40.7%, 95% CI 31.8% to 50.3%; 36 evaluations; 2632 cases). Sensitivity varied between brands. Using data from instructions for use (IFU) compliant evaluations in symptomatic participants, summary sensitivities ranged from 34.1% (95% CI 29.7% to 38.8%; Coris Bioconcept) to 88.1% (95% CI 84.2% to 91.1%; SD Biosensor STANDARD Q). Average specificities were high in symptomatic and asymptomatic participants, and for most brands (overall summary specificity 99.6%, 95% CI 99.0% to 99.8%). At 5% prevalence using data for the most sensitive assays in symptomatic people (SD Biosensor STANDARD Q and Abbott Panbio), positive predictive values (PPVs) of 84% to 90% mean that between 1 in 10 and 1 in 6 positive results will be a false positive, and between 1 in 4 and 1 in 8 cases will be missed. At 0.5% prevalence applying the same tests in asymptomatic people would result in PPVs of 11% to 28% meaning that between 7 in 10 and 9 in 10 positive results will be false positives, and between 1 in 2 and 1 in 3 cases will be missed. No studies assessed the accuracy of repeated lateral flow testing or self‐testing. Rapid molecular assays Thirty studies reported 33 evaluations of five different rapid molecular tests. Sensitivities varied according to test brand. Most of the data relate to the ID NOW and Xpert Xpress assays. Using data from evaluations following the manufacturer’s instructions for use, the average sensitivity of ID NOW was 73.0% (95% CI 66.8% to 78.4%) and average specificity 99.7% (95% CI 98.7% to 99.9%; 4 evaluations; 812 samples, 222 cases). For Xpert Xpress, the average sensitivity was 100% (95% CI 88.1% to 100%) and average specificity 97.2% (95% CI 89.4% to 99.3%; 2 evaluations; 100 samples, 29 cases). Insufficient data were available to investigate the effect of symptom status or time after symptom onset. Authors' conclusions Antigen tests vary in sensitivity. In people with signs and symptoms of COVID‐19, sensitivities are highest in the first week of illness when viral loads are higher. The assays shown to meet appropriate criteria, such as WHO's priority target product profiles for COVID‐19 diagnostics (‘acceptable’ sensitivity ≥ 80% and specificity ≥ 97%), can be considered as a replacement for laboratory‐based RT‐PCR when immediate decisions about patient care must be made, or where RT‐PCR cannot be delivered in a timely manner. Positive predictive values suggest that confirmatory testing of those with positive results may be considered in low prevalence settings. Due to the variable sensitivity of antigen tests, people who test negative may still be infected. Evidence for testing in asymptomatic cohorts was limited. Test accuracy studies cannot adequately assess the ability of antigen tests to differentiate those who are infectious and require isolation from those who pose no risk, as there is no reference standard for infectiousness. A small number of molecular tests showed high accuracy and may be suitable alternatives to RT‐PCR. However, further evaluations of the tests in settings as they are intended to be used are required to fully establish performance in practice. Several important studies in asymptomatic individuals have been reported since the close of our search and will be incorporated at the next update of this review. Comparative studies of antigen tests in their intended use settings and according to test operator (including self‐testing) are required., Plain language summary How accurate are rapid tests for diagnosing COVID‐19? What are rapid point‐of‐care tests for COVID‐19? Rapid point‐of‐care tests aim to confirm or rule out COVID‐19 infection in people with or without COVID‐19 symptoms. They: ‐ are portable, so they can be used wherever the patient is (at the point of care); ‐ are easy to perform, with a minimum amount of extra equipment or complicated preparation steps; ‐ are less expensive than standard laboratory tests; ‐ do not require a specialist operator or setting; and ‐ provide results ‘while you wait’. We were interested in two types of commercially available, rapid point‐of‐care tests: antigen and molecular tests. Antigen tests identify proteins on the virus; they come in disposable plastic cassettes, similar to pregnancy tests. Rapid molecular tests detect the virus’s genetic material in a similar way to laboratory methods, but using smaller devices that are easy to transport or to set up outside of a specialist laboratory. Both test nose or throat samples. Why is this question important? People with suspected COVID‐19 need to know quickly whether they are infected, so that they can self‐isolate, receive treatment, and inform close contacts. Currently, COVID‐19 infection is confirmed by a laboratory test called RT‐PCR, which uses specialist equipment and often takes at least 24 hours to produce a result. Rapid point‐of‐care tests could open access to testing for many more people, with and without symptoms, potentially in locations other than healthcare settings. If they are accurate, faster diagnosis could allow people to take appropriate action more quickly, with the potential to reduce the spread of COVID‐19. What did we want to find out? We wanted to know whether commercially available, rapid point‐of‐care antigen and molecular tests are accurate enough to diagnose COVID‐19 infection reliably, and to find out if accuracy differs in people with and without symptoms. What did we do? We looked for studies that measured the accuracy of any commercially produced, rapid antigen or molecular point‐of‐care test, in people tested for COVID‐19 using RT‐PCR. People could be tested in hospital or the community. Studies could test people with or without symptoms. Tests had to use minimal equipment, be performed safely without risking infection from the sample, and have results available within two hours of the sample being collected. What we found We included 64 studies in the review. They investigated a total of 24,087 nose or throat samples; COVID‐19 was confirmed in 7415 of these samples. Studies investigated 16 different antigen tests and five different molecular tests. They took place mainly in Europe and North America. Main results Antigen tests In people with confirmed COVID‐19, antigen tests correctly identified COVID‐19 infection in an average of 72% of people with symptoms, compared to 58% of people without symptoms. Tests were most accurate when used in the first week after symptoms first developed (an average of 78% of confirmed cases had positive antigen tests). This is likely to be because people have the most virus in their system in the first days after they are infected. In people who did not have COVID‐19, antigen tests correctly ruled out infection in 99.5% of people with symptoms and 98.9% of people without symptoms. Different brands of tests varied in accuracy. Pooled results for one test (SD Biosensor STANDARD Q) met World Health Organization (WHO) standards as ‘acceptable’ for confirming and ruling out COVID‐19 in people with signs and symptoms of COVID‐19. Two more tests met the WHO acceptable standards (Abbott Panbio and BIONOTE NowCheck) in at least one study. Using summary results for SD Biosensor STANDARD Q, if 1000 people with symptoms had the antigen test, and 50 (5%) of them really had COVID‐19: ‐ 53 people would test positive for COVID‐19. Of these, 9 people (17%) would not have COVID‐19 (false positive result). ‐ 947 people would test negative for COVID‐19. Of these, 6 people (0.6%) would actually have COVID‐19 (false negative result). In people with no symptoms of COVID‐19 the number of confirmed cases is expected to be much lower than in people with symptoms. Using summary results for SD Biosensor STANDARD Q in a bigger population of 10,000 people with no symptoms, where 50 (0.5%) of them really had COVID‐19: ‐ 125 people would test positive for COVID‐19. Of these, 90 people (72%) would not have COVID‐19 (false positive result). ‐ 9,875 people would test negative for COVID‐19. Of these, 15 people (0.2%) would actually have COVID‐19 (false negative result). Molecular tests Although overall results for diagnosing and ruling out COVID‐19 were good (95.1% of infections correctly diagnosed and 99% correctly ruled out), 69% of the studies used the tests in laboratories instead of at the point‐of‐care and few studies followed test manufacturer instructions. Most of the data relate to the ID NOW and Xpert Xpress tests. We noted a large difference in COVID‐19 detection between the two tests, but we cannot be certain about whether results will remain the same in a real world setting. We could not investigate differences in people with or without symptoms, nor time from when symptoms first showed because the studies did not provide enough information about their participants. How reliable were the results of the studies? In general, studies that assessed antigen tests used more rigorous methods than those that assessed molecular tests, particularly when selecting participants and performing the tests. Sometimes studies did not perform the test on the people for whom it was intended and did not follow the manufacturers’ instructions for using the test. Sometimes the tests were not carried out at the point‐of‐care. Nearly all the studies (97%) relied on a single negative RT‐PCR result as evidence of no COVID‐19 infection. Results from different test brands varied, and few studies directly compared one test brand with another. Finally, not all studies gave enough information about their participants for us to judge how long they had had symptoms, or even whether or not they had symptoms. What does this mean? Some antigen tests are accurate enough to replace RT‐PCR when used in people with symptoms. This would be most useful when quick decisions are needed about patient care, or if RT‐PCR is not available. Antigen tests may be most useful to identify outbreaks, or to select people with symptoms for further testing with PCR, allowing self‐isolation or contact tracing and reducing the burden on laboratory services. People who receive a negative antigen test result may still be infected. Several point‐of‐care molecular tests show very high accuracy and potential for use, but more evidence of their performance when evaluated in real life settings is required. We need more evidence on rapid testing in people without symptoms, on the accuracy of repeated testing, testing in non‐healthcare settings such as schools (including self‐testing), and direct comparisons of test brands, with testers following manufacturers’ instructions. How up‐to‐date is this review? This review updates our previous review and includes evidence published up to 30 September 2020.

Background: Tuberculosis (TB) is a devastating disease for which new diagnostic tests are desperately needed. Objective: To validate promising new technologies (namely whole blood transcriptomics, proteomics, flow cytometry and qRT-PCR) and existing signatures for detection of active TB in samples obtained from individuals suspected of active TB. Design: Four sub-studies, each of which used the samples from biobank collected as part of the IDEA study, which was a prospective cohort of patients recruited with suspected TB. Setting: secondary care Participants: Adults (aged ≥ 16 years old) presenting as inpatients or outpatients at 12 NHS hospital trusts in London, Slough, Oxford, Leicester and Birmingham with suspected active TB. Interventions: New tests using either: genome-wide gene expression microarray (transcriptomics); SELDI TOF/ LC-MS (proteomics), flow cytometry, qRT-PCR. Main outcome measures: Area under the curve (AUC), sensitivity and specificity, were calculated to determine diagnostic accuracy. Positive and negative predictive values were calculated in some cases. A decision tree model was developed to calculate the incremental costs and quality-adjusted life-years (QALYs) of changing from current practice to using the novels tests. Results: The project and 4 sub-studies which assessed the previous published signatures measured using each of the new technologies, and a health economic analysis where the best performing tests were evaluated for cost effectiveness. The diagnostic accuracy of the transcriptomic tests ranged from AUC=0.81-0.84 for detecting all TB in our cohort. The performance for detecting culture confirmed TB or pulmonary TB (PTB) was better than for highly probable TB or extrapulmonary TB (EPTB) respectively, but not high enough to be clinically useful. None of the previously described serum proteomic signatures for active TB provided good diagnostic accuracy, not did the candidate rule-out tests. Four of six previously described cellular immune signatures provided a reasonable level of diagnostic accuracy (AUC = 0.78-0.92) for discriminating all TB from those with other disease (OD) and latent TB infection (LTBI) in HIV- TB suspects. Two of these assays may be useful in the IGRA+ population and can provide high positive predictive value (PPV). None of the new tests for TB can be considered cost effective. Limitations: The diagnostic performance of new tests within the HIV+ population was either underpowered or not sufficiently achieved in each sub-study. Conclusions: Overall, the diagnostic performance of all previously identified ‘signatures’ of TB was lower than previously reported. This likely reflects the nature of the cohort we used, which includes the harder to diagnose groups, such as culture unconfirmed TB, and EPTB, which were underrepresented in previous cohorts. Future work: We are yet to perform our secondary objective f deriving novel signatures of TB using out datasets. This was beyond the scope of this report. We recommend that future studies using these technologies target specific sub-types of TB, specifically those groups where new diagnostic tests are required.

Background The role of interferon-gamma release assays (IGRAs) in diagnosis of active tuberculosis (TB) is unclear, yet they are commonly used in low-TB-incidence countries. This study sought to resolve this clinical uncertainty by determining the diagnostic accuracy and role of current and second-generation IGRAs in the diagnostic assessment of suspected TB in a low-incidence setting. Methods This was a prospective cohort study of 1,060 adults with suspected TB, conducted in routine secondary care in England. Patients were tested for M. tuberculosis (Mtb) infection at baseline using current and second-generation IGRAs, the latter incorporating novel Mtb antigens, and followed up for 6-12m to establish definitive diagnoses. Sensitivity, specificity and positive and negative likelihood ratios (LRs) and predictive values (PVs) of the tests for TB were determined. Findings TB was diagnosed in 363 (43%) of 845 patients included in analyses. Sensitivity of T-SPOT.TB was 81.4% (95%CI 76.6-85.3%), higher than Quantiferon-Gold In-Tube at 67.3% (95%CI 62.0-72.1%). Second-generation IGRA had higher sensitivity than current tests, at 94.0% (95%CI 90.0–96.4%) for culture-confirmed TB and 89.2% (95%CI 85.2–92.2%) when including highly-probable TB, giving a negative LR for all TB of 0.13 (95%CI 0.10-0.19). Specificity ranged from 86.2% (95%CI 82.3-89.4%) for T-SPOT.TB to 80.0% (95%CI 75.6-83.8%) for second-generation IGRA. Interpretation Currently-available IGRAs lack sufficient accuracy for diagnostic evaluation of suspected TB. Second-generation tests, however, may have sufficiently high sensitivity, low negative LR and correspondingly high negative PV in low-incidence settings to facilitate prompt rule-out of TB.

Objectives To explain methods for data synthesis of evidence from diagnostic test accuracy (DTA) studies, and to illustrate different types of analyses that may be performed in a DTA systematic review.\ud \ud Methods We described properties of meta-analytic methods for quantitative synthesis of evidence. We used a DTA review comparing the accuracy of three screening questionnaires for bipolar disorder to illustrate application of the methods for each type of analysis.\ud \ud Results The discriminatory ability of a test is commonly expressed in terms of sensitivity (proportion of those with the condition who test positive) and specificity (proportion of those without the condition who test negative). There is a trade-off between sensitivity and specificity, as an increasing threshold for defining test positivity will decrease sensitivity and increase specificity. Methods recommended for meta-analysis of DTA studies --such as the bivariate or hierarchical summary receiver operating characteristic (HSROC) model --jointly summarise sensitivity and specificity while taking into account this threshold effect, as well as allowing for between study differences in test performance beyond what would be expected by chance. The bivariate model focuses on estimation of a summary sensitivity and specificity at a common threshold while the HSROC model focuses on the estimation of a summary curve from studies that have used different thresholds.\ud \ud Conclusions Meta-analyses of diagnostic accuracy studies can provide answers to important clinical questions. We hope this article will provide clinicians with sufficient understanding of the terminology and methods to aid interpretation of systematic reviews and facilitate better patient care.

Many abstracts of diagnostic accuracy studies are currently insufficiently informative. We extended the STARD (Standards for Reporting Diagnostic Accuracy) statement by developing a list of essential items that authors should consider when reporting diagnostic accuracy studies in journal or conference abstracts. After a literature review of published guidance for reporting biomedical studies, we identified 39 items potentially relevant to report in an abstract. We then selected essential items through a two round web based survey among the 85 members of the STARD Group, followed by discussions within an executive committee. Seventy three STARD Group members responded (86%), with 100% completion rate. STARD for Abstracts is a list of 11 quintessential items, to be reported in every abstract of a diagnostic accuracy study. We provide examples of complete reporting, and developed template text for writing informative abstracts.

[This corrects the article DOI: 10.1186/s41512-016-0001-y.].

Incomplete reporting has been identified as a major source of avoidable waste in biomedical research. Essential information is often not provided in study reports, impeding the identification, critical appraisal, and replication of studies. To improve the quality of reporting of diagnostic accuracy studies, the Standards for Reporting of Diagnostic Accuracy Studies (STARD) statement was developed. Here we present STARD 2015, an updated list of 30 essential items that should be included in every report of a diagnostic accuracy study. This update incorporates recent evidence about sources of bias and variability in diagnostic accuracy and is intended to facilitate the use of STARD. As such, STARD 2015 may help to improve completeness and transparency in reporting of diagnostic accuracy studies.

Early detection of melanoma is essential to improve survival.The additional value of dermoscopy over and above visualinspection (VI) of a suspicious skin lesion is critical to under-stand its contribution to the diagnosis of melanoma. ACochrane systematic review of the diagnostic accuracy of der-moscopy for detection of melanoma in adults was undertakenfor (i) in-person diagnosis and (ii) diagnosis based on dermo-scopic images, and to compare its accuracy with VI alone. Acomprehensive search of 10 databases up to August 2016identified studies of any design evaluating dermoscopy inadults with lesions suspicious for melanoma, compared withhistology or clinical follow-up. Two reviewers independentlyextracted data and quality assessment (using QUADAS-2). Theaccuracy was estimated using hierarchical summary ROCmethods; sensitivities and specificities were estimated forselected points on the summary receiver operating characteris-tic (SROC) curve. Overall, 106 publications were included.The detection of melanoma or intraepidermal melanocyticvariants was analysed for 27 in-person (23 487 lesions; 1737melanomas) and 60 image-based (13 475 lesions; 2851 mela-nomas) datasets. In-person dermoscopy was more accuratethan image-based interpretation [relative diagnostic odds ratio(RDOR) 4.5; 95% CI 2.3–8.5, P < 0.0001]. Dermoscopy wasmore accurate than VI alone; RDORs (i) 4.8 (95% CI: 3.1–7.4; P < 0.0001) for in-person and (ii) 5.6 (95% CI: 3.7–8.5;P < 0.0001) for image-based evaluations. Predicted increasesin sensitivity were (i) 16% (92% vs. 76%) and (ii) 34% (81%vs. 47%) at a fixed specificity of 80%. Use of a publishedalgorithm to assist dermoscopy had no significant impact onaccuracy. The accuracy was significantly higher for experi-enced observers compared with less experienced. Dermoscopyis a valuable tool to support VI of suspicious skin lesions todetect melanoma, particularly in referred populations and for

BACKGROUND: Multiparametric magnetic resonance imaging (MRI), with or without targeted biopsy, is an alternative to standard transrectal ultrasonography-guided biopsy for prostate-cancer detection in men with a raised prostate-specific antigen level who have not undergone biopsy. However, comparative evidence is limited. METHODS: In a multicenter, randomized, noninferiority trial, we assigned men with a clinical suspicion of prostate cancer who had not undergone biopsy previously to undergo MRI, with or without targeted biopsy, or standard transrectal ultrasonography-guided biopsy. Men in the MRI-targeted biopsy group underwent a targeted biopsy (without standard biopsy cores) if the MRI was suggestive of prostate cancer; men whose MRI results were not suggestive of prostate cancer were not offered biopsy. Standard biopsy was a 10-to-12-core, transrectal ultrasonography-guided biopsy. The primary outcome was the proportion of men who received a diagnosis of clinically significant cancer. Secondary outcomes included the proportion of men who received a diagnosis of clinically insignificant cancer. RESULTS: A total of 500 men underwent randomization. In the MRI-targeted biopsy group, 71 of 252 men (28%) had MRI results that were not suggestive of prostate cancer, so they did not undergo biopsy. Clinically significant cancer was detected in 95 men (38%) in the MRI-targeted biopsy group, as compared with 64 of 248 (26%) in the standard-biopsy group (adjusted difference, 12 percentage points; 95% confidence interval [CI], 4 to 20; P = 0.005). MRI, with or without targeted biopsy, was noninferior to standard biopsy, and the 95% confidence interval indicated the superiority of this strategy over standard biopsy. Fewer men in the MRI-targeted biopsy group than in the standard-biopsy group received a diagnosis of clinically insignificant cancer (adjusted difference, -13 percentage points; 95% CI, -19 to -7; P