scholarly journals Performance of methods for meta-analysis of diagnostic test accuracy with few studies or sparse data

2015 ◽  
Vol 26 (4) ◽  
pp. 1896-1911 ◽  
Author(s):  
Yemisi Takwoingi ◽  
Boliang Guo ◽  
Richard D Riley ◽  
Jonathan J Deeks
Keyword(s):  
Hierarchical Models ◽  
Operating Characteristic ◽  
Meta Analysis ◽  
Sparse Data ◽  
Parameter Estimates ◽  
Test Accuracy ◽  
Summary Receiver Operating Characteristic ◽  
Bivariate Model ◽  
Logistic Regression Models ◽  
Meta Analyses

Hierarchical models such as the bivariate and hierarchical summary receiver operating characteristic (HSROC) models are recommended for meta-analysis of test accuracy studies. These models are challenging to fit when there are few studies and/or sparse data (for example zero cells in contingency tables due to studies reporting 100% sensitivity or specificity); the models may not converge, or give unreliable parameter estimates. Using simulation, we investigated the performance of seven hierarchical models incorporating increasing simplifications in scenarios designed to replicate realistic situations for meta-analysis of test accuracy studies. Performance of the models was assessed in terms of estimability (percentage of meta-analyses that successfully converged and percentage where the between study correlation was estimable), bias, mean square error and coverage of the 95% confidence intervals. Our results indicate that simpler hierarchical models are valid in situations with few studies or sparse data. For synthesis of sensitivity and specificity, univariate random effects logistic regression models are appropriate when a bivariate model cannot be fitted. Alternatively, an HSROC model that assumes a symmetric SROC curve (by excluding the shape parameter) can be used if the HSROC model is the chosen meta-analytic approach. In the absence of heterogeneity, fixed effect equivalent of the models can be applied.

scholarly journals Diagnostic Test Accuracy of the 4AT for Delirium Detection: A Systematic Review and Meta-Analysis

2020 ◽  
Vol 17 (20) ◽  
pp. 7515 ◽  
Author(s):  
Eunhye Jeong ◽  
Jinkyung Park ◽  
Juneyoung Lee
Keyword(s):  
Systematic Review ◽  
Diagnostic Test ◽  
Healthcare Professionals ◽  
Meta Analysis ◽  
Diagnostic Test Accuracy ◽  
Test Accuracy ◽  
Summary Receiver Operating Characteristic ◽  
Bivariate Model ◽  
Data Synthesis

Under-recognition of delirium is an international problem. For the early detection of delirium, a feasible and valid screening tool for healthcare professionals is needed. This study aimed to present a scientific reason for using the 4 ‘A’s Test (4AT) through a systematic review and meta-analysis of studies on the diagnostic test accuracy. We systematically searched articles in the EMBASE, MEDLINE, CINAHL, and PsycINFO databases and selected relevant articles on the basis of the predefined inclusion criteria. The quality of the included articles was evaluated using the Quality Assessment of the Diagnostic Accuracy Studies-2 tool. We estimated the pooled values of diagnostic test accuracy by employing the bivariate model and the hierarchical summary receiver operating characteristic (HSROC) model in data synthesis. A total of 3729 patients of 13 studies were included in the analysis. The pooled estimates of sensitivity and specificity of the 4AT were 81.5% (95% confidence interval: 70.7%, 89.0%) and 87.5% (79.5%, 92.7%), respectively. Given the 4AT’s evidence of accuracy and practicality, we suggest healthcare professionals to utilize this tool for routine screening of delirium. However, for detecting delirium in the dementia population, further work is required to evaluate the 4AT with other cut-off points or scoring methods in order for it to be more sensitive and specific.

scholarly journals Anti-DFS70 Antibodies for Differentiating Systemic Autoimmune Rheumatic Disease in Patients with Positive ANA Tests: A Systematic Review and Meta-Analysis

Diagnostics ◽  
2021 ◽  
Vol 11 (9) ◽  
pp. 1592
Author(s):  
Chiao-Feng Cheng ◽  
Ming-Chieh Shih ◽  
Ting-Yuan Lan ◽  
Ko-Jen Li
Keyword(s):  
Confidence Interval ◽  
Rheumatic Disease ◽  
Meta Analysis ◽  
Cochrane Library ◽  
Test Accuracy ◽  
Summary Receiver Operating Characteristic ◽  
Bivariate Model ◽  
Autoimmune Rheumatic Disease ◽  
Systemic Autoimmune Rheumatic Disease ◽  
Test Result

Anti-DFS70 antibodies have been proposed as a marker to exclude systemic autoimmune rheumatic disease (SARD). We conducted this systematic diagnostic test accuracy review and meta-analysis to determine the performance of anti-DFS70 antibodies in patients with a positive anti-nuclear antibody (ANA) test result to exclude SARD. We searched PubMed, Embase, Web of Science, Scopus and the Cochrane Library up to 22 February 2021, and included studies examining the diagnostic accuracy of anti-DFS70 antibodies in patients with a positive ANA test result. The results were pooled using a hierarchical bivariate model and plotted in summary receiver operating characteristic curves. R software and Stata Statistical Software were used for the statistical analysis. Eight studies with 4168 patients were included. The summary sensitivity was 0.19 (95% confidence interval: 0.12–0.28) and the specificity was 0.93 (95% confidence interval: 0.88–0.96). The area under the curve was 0.69 (95% confidence interval: 0.64–0.72). The meta-regression analysis showed that targeting only ANA-associated rheumatic disease was associated with higher specificity. In addition, the studies with a non-SARD prevalence of <80% and using a chemiluminescence assay were associated with higher specificity. Anti-DFS70 antibodies have high specificity for the exclusion of SARD among patients presenting with a positive ANA test, but the sensitivity is low.

scholarly journals A Systematic and Meta-Analysis Review on the Diagnostic Accuracy of Antibodies in the Serological Diagnosis of COVID-19.

2020 ◽  
Author(s):  
Arthur Vengesai ◽  
Herald Midzi ◽  
Maritha Kasambala ◽  
Hamlet Mutandadzi ◽  
Tariro L. Mduluza-Jok ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Point Of Care ◽  
Meta Analysis ◽  
Characteristic Curve ◽  
Test Accuracy ◽  
Rt Pcr ◽  
Summary Receiver Operating Characteristic ◽  
Serological Tests ◽  
Bivariate Model ◽  
Prevalence Studies

Abstract Background: Serological testing based on different antibody types are an alternative method being used to diagnose SARS-CoV-2 and has the potential of having higher diagnostic accuracy compared to the current gold standard RT-PCR. Therefore, the objective of this review was to evaluate the diagnostic accuracy of IgG and IgM based Point-of-care (POC) lateral flow immunoassays (LFIA), chemiluminescence enzyme immunoassay (CLIA), fluorescence enzyme-linked immunoassay (FIA) and ELISA systems that detect SARS-CoV-2 antigens.Method: A systematic literature search was carried out in PubMed, Medline complete and MedRxiv. Studies evaluating the diagnostic accuracy of serological assays for SARS-CoV-2 were eligible. Study selection and data-extraction were done by two authors independently. QUADAS-2 checklist tool was used to assess the quality of the studies. The bivariate model and the hierarchical summary receiver operating characteristic curve model were performed to evaluate the diagnostic accuracy of the serological tests. Subgroup meta-analysis analyses was performed to explore the heterogeneity. Results: The pooled sensitivity for IgG, IgM and IgG-IgM based LFIA tests were 0.5856, 0.4637 and 0.6886 respectively compared to RT-PCR method. The pooled sensitivity for IgG and IgM based CLIA tests were 0.9311 and 0.8516 respectively compared to RT-PCR. The pooled sensitivity the IgG, IgM and IgG-IgM based ELISA tests were 0.8292, 0.8388 and 0.8531 respectively compared to RT-PCR. All tests displayed high specificities ranging from 0.9693 to 0.9991. Among the evaluated tests, IgG based CLIA expressed the highest sensitivity signifying its accurate detection of the largest proportion of infections identified by RT-PCR. ELISA and CLIA tests performed better in terms of sensitivity compared to LFIA. IgG based tests performed better compared to IgM ones expect for the ELISA. Conclusions: We report that IgG-IgM based ELISA tests have the best overall diagnostic test accuracy. Moreover, irrespective of the method, a combined IgG/IgM test seems to be a better choice in terms of sensitivity than measuring either antibody type independently. Given the poor performances of the current LFIA devices there is need for more research on the development of highly sensitivity and specific POC LFIA that are adequate for most individual patient applications and attractive for large sero-prevalence studies.Systematic review registration: PROSPERO Registration Number is: CRD42020179112

Meta-Analysis and Sparse-Data Bias

2020 ◽  
Author(s):  
David B Richardson ◽  
Stephen R Cole ◽  
Rachael K Ross ◽  
Charles Poole ◽  
Haitao Chu ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Meta Analysis ◽  
Sparse Data ◽  
Maximum Likelihood Estimates ◽  
Finite Sample ◽  
Logistic Regression Models ◽  
Confidence Interval Coverage ◽  
The Individual ◽  
Meta Analyses ◽  
Sparse Data Bias

Abstract Meta-analyses are undertaken to combine information from a set of studies, often in settings where some of the individual study-specific estimates are based on relatively small study samples. Finite sample bias may occur when maximum likelihood estimates of associations are obtained by fitting logistic regression models to sparse data sets. Here we show that combining information from small studies by undertaking a meta-analytical summary of logistic regression estimates can propagate such sparse-data bias. In simulations, we illustrate 2 challenges encountered in meta-analyses of logistic regression results in settings of sparse data: 1) bias in the summary meta-analytical result and 2) confidence interval coverage that can worsen rather than improve, in terms of being less than nominal, as the number of studies in the meta-analysis increases.

scholarly journals A systematic and meta-analysis review on the diagnostic accuracy of antibodies in the serological diagnosis of COVID-19

2021 ◽  
Vol 10 (1) ◽  
Author(s):  
Arthur Vengesai ◽  
Herald Midzi ◽  
Maritha Kasambala ◽  
Hamlet Mutandadzi ◽  
Tariro L. Mduluza-Jokonya ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Point Of Care ◽  
Data Extraction ◽  
Meta Analysis ◽  
Characteristic Curve ◽  
Test Accuracy ◽  
Summary Receiver Operating Characteristic ◽  
Serological Tests ◽  
Bivariate Model ◽  
Prevalence Studies

Abstract Background Serological testing based on different antibody types are an alternative method being used to diagnose SARS-CoV-2 and has the potential of having higher diagnostic accuracy compared to the current gold standard rRT-PCR. Therefore, the objective of this review was to evaluate the diagnostic accuracy of IgG and IgM based point-of-care (POC) lateral flow immunoassay (LFIA), chemiluminescence enzyme immunoassay (CLIA), fluorescence enzyme-linked immunoassay (FIA) and ELISA systems that detect SARS-CoV-2 antigens. Method A systematic literature search was carried out in PubMed, Medline complete and MedRxiv. Studies evaluating the diagnostic accuracy of serological assays for SARS-CoV-2 were eligible. Study selection and data-extraction were performed by two authors independently. QUADAS-2 checklist tool was used to assess the quality of the studies. The bivariate model and the hierarchical summary receiver operating characteristic curve model were performed to evaluate the diagnostic accuracy of the serological tests. Subgroup meta-analysis was performed to explore the heterogeneity. Results The pooled sensitivity for IgG (n = 17), IgM (n = 16) and IgG-IgM (n = 24) based LFIA tests were 0.5856, 0.4637 and 0.6886, respectively compared to rRT-PCR method. The pooled sensitivity for IgG (n = 9) and IgM (n = 10) based CLIA tests were 0.9311 and 0.8516, respectively compared to rRT-PCR. The pooled sensitivity the IgG (n = 10), IgM (n = 11) and IgG-IgM (n = 5) based ELISA tests were 0.8292, 0.8388 and 0.8531 respectively compared to rRT-PCR. All tests displayed high specificities ranging from 0.9693 to 0.9991. Amongst the evaluated tests, IgG based CLIA expressed the highest sensitivity signifying its accurate detection of the largest proportion of infections identified by rRT-PCR. ELISA and CLIA tests performed better in terms of sensitivity compared to LFIA. IgG based tests performed better compared to IgM except for the ELISA. Conclusions We report that IgG-IgM based ELISA tests have the best overall diagnostic test accuracy. Moreover, irrespective of the method, a combined IgG/IgM test seems to be a better choice in terms of sensitivity than measuring either antibody type independently. Given the poor performances of the current LFIA devices, there is a need for more research on the development of highly sensitivity and specific POC LFIA that are adequate for most individual patient applications and attractive for large sero-prevalence studies. Systematic review registration PROSPERO CRD42020179112

scholarly journals Prognostic accuracy of shock index for severe postpartum hemorrhage in high-income country: Protocol for a systematic review and meta-analysis

2021 ◽  
Author(s):  
Yuto Makino ◽  
Kentaro Miyake ◽  
Asami Okada ◽  
Yumie Ikeda ◽  
Yohei Okada
Keyword(s):  
Systematic Review ◽  
Postpartum Hemorrhage ◽  
Meta Analysis ◽  
High Income ◽  
Test Accuracy ◽  
Shock Index ◽  
Summary Receiver Operating Characteristic ◽  
Bivariate Model ◽  
Data Synthesis ◽  
Prognostic Accuracy

Background This review aims to conduct a systematic review and meta-analysis to assess the prognostic value of shock index for prediction of severe Postpartum hemorrhage (PPH) in high-income countries. Method We will perform a systematic review and meta-analysis for diagnostic test accuracy (DTA). We will search CENTRAL, MEDLINE (Ovid), Web of Science, and other sources and include all relevant reports on the prognostic accuracy of shock index for severe PPH. The target condition is defined as severe PPH required higher-level care. Two review authors will independently screen the study eligibility and extract data from included studies. Two authors will also assess study quality using the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) tool. We will use the hierarchical models comprising both the bivariate model and the hierarchical summary receiver operating characteristic (HSROC) model for data synthesis if appropriate. We presented uncertainty of the accuracy estimates using 95% confidence intervals (CIs).

scholarly journals Diagnostic Accuracy of FDA Authorized Serology Tests to Detect SARS-CoV-2 Antibodies: A Systematic Review and Meta-analysis

2020 ◽  
Author(s):  
Rajesh Pandey ◽  
Anand Gourishankar
Keyword(s):  
Systematic Review ◽  
Diagnostic Accuracy ◽  
Receiver Operating Characteristic ◽  
Test Performance ◽  
Operating Characteristic ◽  
Meta Analysis ◽  
Time Interval ◽  
Test Accuracy ◽  
Summary Receiver Operating Characteristic ◽  
Receiver Operating

AbstractImportanceSerology tests are diagnostic and complementary to molecular tests during the COVID-19 pandemic.ObjectiveTo evaluate the diagnostic accuracy of FDA authorized serology tests for the detection of SARS-CoV-2 infection.Data sourcesA search of MEDLINE, SCOPUS, CINAHL Plus, and EMBASE up to April 4, 2020, was performed to identify studies using the “COVID 19 testing” and “meta-analysis.” FDA website was accessed for the list of tests for emergency use authorization (EUA).Study SelectionManufacturer reported serology tests published in the FDA website were selected. Two reviewers independently assessed the eligibility of the selected reports.Data extraction and synthesisThe meta-analysis was performed in accordance with the PRISMA guidelines. A bivariate analysis using the “random-effects model” was applied for pooled summary estimates of sensitivity, specificity, and the summary receiver operating characteristic curves.Main outcomes and measuresThe primary outcome was the diagnostic accuracy of the serology test for detecting SARS-CoV-2 infection. Subgroup analysis of the diagnostic accuracy with lag time between symptom onset and testing were studied.ResultsSeven manufacturer listed reports were included. The pooled sensitivity was 87% (95% CI, 78% - 93%), the pooled specificity was 100% (95% CI, 97% - 100%), and the area under the hierarchical summary receiver operating characteristic curve was 0.97. At ≤ 7 days, sensitivity was 44% (95% CI, 21% - 70%), and for 8-14 days, sensitivity was 84% (95% CI, 67 % - 94%).For blood draws ≥ 15 days after the onset of symptoms, sensitivity was 96% (95% CI, 93% - 98%). Heterogeneity was substantial, and the risk of bias was low in this analysis.Conclusions and relevanceFDA authorized serology tests demonstrate high diagnostic accuracy for SARS-CoV-2 infection (certainty of evidence: moderate). There is a wide variation in the test accuracy based on the duration between the onset of symptoms and the tests (certainty of evidence: low).Key– pointsQuestionsWhat is the pooled diagnostic accuracy of FDA authorized serology tests to detect SARS-CoV-2 antibodies?FindingsIn this systematic review and meta-analysis of seven reports from FDA authorized serology tests to detect antibodies against SARS-CoV2 antibodies (3336 patients/ samples) pooled sensitivity was 87%, and pooled specificity was almost 100%. There was a wide variation in test performance based on the duration between the onset of symptoms and the tests.MeaningFDA authorized tests are highly accurate to detect antibodies against SARS-CoV-2 antibodies if tests are performed under a similar condition, as presented in the original report. There is a wide variation in the test performance based on the time interval between the onset of symptoms to the tests.

scholarly journals A systematic and meta-analysis review on the diagnostic accuracy of antibodies in the serological diagnosis of COVID-19.

2020 ◽  
Author(s):  
Arthur Vengesai ◽  
Herald Midzi ◽  
Maritha Kasambala ◽  
Hamlet Mutandadzi ◽  
Tariro L. Mduluza-Jok ◽  
...  
Keyword(s):  
Diagnostic Accuracy ◽  
Point Of Care ◽  
Data Extraction ◽  
Meta Analysis ◽  
Characteristic Curve ◽  
Test Accuracy ◽  
Summary Receiver Operating Characteristic ◽  
Serological Tests ◽  
Bivariate Model ◽  
Prevalence Studies

Abstract Background: Serological testing based on different antibody types are an alternative method being used to diagnose SARS-CoV-2 and has the potential of having higher diagnostic accuracy compared to the current gold standard rRT-PCR. Therefore, the objective of this review was to evaluate the diagnostic accuracy of IgG and IgM based point-of-care (POC) lateral flow immunoassay (LFIA), chemiluminescence enzyme immunoassay (CLIA), fluorescence enzyme-linked immunoassay (FIA) and ELISA systems that detect SARS-CoV-2 antigens.Method: A systematic literature search was carried out in PubMed, Medline complete and MedRxiv. Studies evaluating the diagnostic accuracy of serological assays for SARS-CoV-2 were eligible. Study selection and data-extraction were performed by two authors independently. QUADAS-2 checklist tool was used to assess the quality of the studies. The bivariate model and the hierarchical summary receiver operating characteristic curve model were performed to evaluate the diagnostic accuracy of the serological tests. Subgroup meta-analysis analyses was performed to explore the heterogeneity. Results: The pooled sensitivity for IgG, IgM and IgG-IgM based LFIA tests were 0.5856, 0.4637 and 0.6886, respectively compared to rRT-PCR method. The pooled sensitivity for IgG and IgM based CLIA tests were 0.9311 and 0.8516, respectively compared to rRT-PCR. The pooled sensitivity the IgG, IgM and IgG-IgM based ELISA tests were 0.8292, 0.8388 and 0.8531 respectively compared to rRT-PCR. All tests displayed high specificities ranging from 0.9693 to 0.9991. Among the evaluated tests, IgG based CLIA expressed the highest sensitivity signifying its accurate detection of the largest proportion of infections identified by rRT-PCR. ELISA and CLIA tests performed better in terms of sensitivity compared to LFIA. IgG based tests performed better compared to IgM except for the ELISA. Conclusions: We report that IgG-IgM based ELISA tests have the best overall diagnostic test accuracy. Moreover, irrespective of the method, a combined IgG/IgM test seems to be a better choice in terms of sensitivity than measuring either antibody type independently. Given the poor performances of the current LFIA devices there is need for more research on the development of highly sensitivity and specific POC LFIA that are adequate for most individual patient applications and attractive for large sero-prevalence studies.Systematic review registration: PROSPERO Registration Number is: CRD42020179112

A general framework for the use of logistic regression models in meta-analysis

2016 ◽  
Vol 25 (6) ◽  
pp. 2858-2877 ◽  
Author(s):  
Mark C Simmonds ◽  
Julian PT Higgins
Keyword(s):  
Logistic Regression ◽  
Random Effects ◽  
Regression Models ◽  
Meta Analysis ◽  
Test Accuracy ◽  
Individual Participant Data ◽  
Logistic Regression Models ◽  
One Stage ◽  
Individual Participant ◽  
Meta Analyses

Where individual participant data are available for every randomised trial in a meta-analysis of dichotomous event outcomes, “one-stage” random-effects logistic regression models have been proposed as a way to analyse these data. Such models can also be used even when individual participant data are not available and we have only summary contingency table data. One benefit of this one-stage regression model over conventional meta-analysis methods is that it maximises the correct binomial likelihood for the data and so does not require the common assumption that effect estimates are normally distributed. A second benefit of using this model is that it may be applied, with only minor modification, in a range of meta-analytic scenarios, including meta-regression, network meta-analyses and meta-analyses of diagnostic test accuracy. This single model can potentially replace the variety of often complex methods used in these areas. This paper considers, with a range of meta-analysis examples, how random-effects logistic regression models may be used in a number of different types of meta-analyses. This one-stage approach is compared with widely used meta-analysis methods including Bayesian network meta-analysis and the bivariate and hierarchical summary receiver operating characteristic (ROC) models for meta-analyses of diagnostic test accuracy.

Sign in / Sign up

Export Citation Format

Share Document