Meta-analysis Of Diagnostic Accuracy Studies Research Articles

A hybrid Bayesian hierarchical model combining cohort and case-control studies for meta-analysis of diagnostic tests: Accounting for partial verification bias.

To account for between-study heterogeneity in meta-analysis of diagnostic accuracy studies, bivariate random effects models have been recommended to jointly model the sensitivities and specificities. As study design and population vary, the definition of disease status or severity could differ across studies. Consequently, sensitivity and specificity may be correlated with disease prevalence. To account for this dependence, a trivariate random effects model had been proposed. However, the proposed approach can only include cohort studies with information estimating study-specific disease prevalence. In addition, some diagnostic accuracy studies only select a subset of samples to be verified by the reference test. It is known that ignoring unverified subjects may lead to partial verification bias in the estimation of prevalence, sensitivities, and specificities in a single study. However, the impact of this bias on a meta-analysis has not been investigated. In this paper, we propose a novel hybrid Bayesian hierarchical model combining cohort and case-control studies and correcting partial verification bias at the same time. We investigate the performance of the proposed methods through a set of simulation studies. Two case studies on assessing the diagnostic accuracy of gadolinium-enhanced magnetic resonance imaging in detecting lymph node metastases and of adrenal fluorine-18 fluorodeoxyglucose positron emission tomography in characterizing adrenal masses are presented.

FDG-PET/CT and MRI for Evaluation of Pathologic Response to Neoadjuvant Chemotherapy in Patients With Breast Cancer: A Meta-Analysis of Diagnostic Accuracy Studies.

This study compared the diagnostic test accuracy of magnetic resonance imaging (MRI) with that of (18)F-fluoro-2-glucose-positron emission tomography/computed tomography (FDG-PET/CT) imaging in assessment of response to neoadjuvant chemotherapy (NAC) in breast cancer. A systematic search was performed in PubMed and EMBASE (last updated in June 2015). Studies investigating the performance of MRI and FDG-PET or FDG-PET/CT imaging during or after completion of NAC in patients with histologically proven breast cancer were eligible for inclusion. We considered only studies reporting a direct comparison between these imaging modalities to establish precise summary estimates in the same setting of patients. Pathologic response was considered as the reference standard. Two authors independently screened and selected studies that met the inclusion criteria and extracted the data. A total of 10 studies were included. The pooled estimates of sensitivity and specificity across all included studies were 0.71 and 0.77 for FDG-PET/CT (n = 535) and 0.88 and 0.55 for MRI (n = 492), respectively. Studies were subgrouped according to the time of therapy assessment. In the intra-NAC setting, FDG-PET/CT imaging outperformed MRI with fairly similar pooled sensitivity (0.91 vs. 0.89) and higher specificity (0.69 vs. 0.42). However, MRI appeared to have higher diagnostic accuracy than FDG-PET/CT imaging when performed after the completion of NAC, with significantly higher sensitivity (0.88 vs. 0.57). Analysis of the available studies of patients with breast cancer indicates that the timing of imaging for NAC-response assessment exerts a major influence on the estimates of diagnostic accuracy. FDG-PET/CT imaging outperformed MRI in intra-NAC assessment, whereas the overall performance of MRI was higher after completion of NAC, before surgery. The timing of therapy assessment imaging exerts a major influence on overall estimates of diagnostic accuracy. (18)F-fluoro-2-glucose-positron emission tomography (FDG-PET)/computed tomography (CT) imaging outperformed magnetic resonance imaging (MRI) in intra-neoadjuvant chemotherapy assessment with fairly similar pooled sensitivity and higher specificity. However, MRI appeared to be more accurate than FDG-PET/CT in predicting pathologic response when used in the post-therapy setting.

Meta-epidemiologic study showed frequent time trends in summary estimates from meta-analyses of diagnostic accuracy studies

ObjectivesTo evaluate changes over time in summary estimates from meta-analyses of diagnostic accuracy studies. Study Design and SettingWe included 48 meta-analyses from 35 MEDLINE-indexed systematic reviews published between September 2011 and January 2012 (743 diagnostic accuracy studies; 344,015 participants). Within each meta-analysis, we ranked studies by publication date. We applied random-effects cumulative meta-analysis to follow how summary estimates of sensitivity and specificity evolved over time. Time trends were assessed by fitting a weighted linear regression model of the summary accuracy estimate against rank of publication. ResultsThe median of the 48 slopes was −0.02 (−0.08 to 0.03) for sensitivity and −0.01 (−0.03 to 0.03) for specificity. Twelve of 96 (12.5%) time trends in sensitivity or specificity were statistically significant. We found a significant time trend in at least one accuracy measure for 11 of the 48 (23%) meta-analyses. ConclusionTime trends in summary estimates are relatively frequent in meta-analyses of diagnostic accuracy studies. Results from early meta-analyses of diagnostic accuracy studies should be considered with caution.

A systematic review and meta-analysis of the diagnostic accuracy of MRI for suspected ACL and meniscal tears of the knee.

To determine the diagnostic accuracy of magnetic resonance imaging (MRI) and ultrasound (US) in the diagnosis of anterior cruciate ligament (ACL), medial meniscus and lateral meniscus tears in people with suspected ACL and/or meniscal tears. MEDLINE, Web of Science and the Cochrane library were searched from inception to March 2014. All prospective studies of the diagnostic accuracy of MRI or US against arthroscopy as the reference standard were included in the systematic review. Studies with a retrospective design and those with evidence of verification bias were excluded. Methodological quality of included studies was assessed using the QUADAS-2 tool. A meta-analysis of studies evaluating MRI to calculate the pooled sensitivity and specificity for each target condition was performed using a bivariate model with random effects. Sub-group and sensitivity analysis were used to examine the effect of methodological and other study variables. There were 14 studies included in the meta-analysis of the accuracy of MRI for ACL tears, 19 studies included for medial meniscal tears and 19 studies for lateral meniscal tears. The summary estimates of sensitivity and specificity of MRI were 87% (95% CI 77-94%) and 93% (95% CI 91-96%), respectively, for ACL tears; 89% (95% CI 83-94%) and 88% (95% CI 82-93%), respectively, for medial meniscal tears; and 78% (95% CI 66-87%) and 95% (95% CI 91-97%), respectively, for lateral meniscal tears. Magnetic field strength had no significant effect on accuracy. Most studies had a high or unclear risk of bias. There were an insufficient number of studies that evaluated US to perform a meta-analysis. This study provides a systematic review and meta-analysis of diagnostic accuracy studies of MRI and applies strict exclusion criteria in relation to the risk of verification bias. The risk of bias in most studies is high or unclear in relation to the reference standard. Concerns regarding the applicability of patient selection are also present in most studies. III.

Meta-analysis of diagnostic accuracy studies in mental health

ObjectivesTo 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.MethodsWe...

Marginal models for meta-analysis of diagnostic accuracy studies in frequentist and Bayesian framework using rstan and CopulaREMADA

Current statistical procedures implemented in general software packages for pooling of diagnostic test accuracy data include hSROC (Rutter and Gatsonis 2001) regression and the bivariate normal model (BRMA) (Reitsma et al. (2005), Arends (2006) and Chu and Cole (2006)). However, these models do not report the overall mean. The hSROC model is less intuitive and therefore less popular while the bivariate normal model has difficulties estimating the correlation parameter when the number studies in the meta-analysis are small and/or when the between-study variances are relatively large. As a result, the between study variance estimates from the BRMA are upwardly biased as they are inflated to compensate for the restriction on correlation parameter (Riley et al. 2007). We present advanced statistical methods for meta-analysis of diagnostic accuracy studies and demonstrate the use of different software packages in R or with an R interface together with code, to apply different model strategies for obtaining appropriate meta-analytic parameter estimates. The focus is on the joint modelling of sensitivity and specificity using copulas as well as using the concept of sharing random components. The methods are applied to two datasets: the classical example of Glas et al. (2003) on diagnostic accuracy of telomerase as urinary tumour marker for diagnosing primary bladder cancer and a second dataset from a systematic review by Arbyn et al. (2013) that aimed at comparing sensitivity and specificity of human papillomavirus testing versus repeat cytology for triage of minor cytological cervical lesions.

Meta-analysis of studies with bivariate binary outcomes: a marginal beta-binomial model approach.

When conducting a meta-analysis of studies with bivariate binary outcomes, challenges arise when the within-study correlation and between-study heterogeneity should be taken into account. In this paper, we propose a marginal beta-binomial model for the meta-analysis of studies with binary outcomes. This model is based on the composite likelihood approach and has several attractive features compared with the existing models such as bivariate generalized linear mixed model (Chu and Cole, 2006) and Sarmanov beta-binomial model (Chen et al., 2012). The advantages of the proposed marginal model include modeling the probabilities in the original scale, not requiring any transformation of probabilities or any link function, having closed-form expression of likelihood function, and no constraints on the correlation parameter. More importantly, because the marginal beta-binomial model is only based on the marginal distributions, it does not suffer from potential misspecification of the joint distribution of bivariate study-specific probabilities. Such misspecification is difficult to detect and can lead to biased inference using currents methods. We compare the performance of the marginal beta-binomial model with the bivariate generalized linear mixed model and the Sarmanov beta-binomial model by simulation studies. Interestingly, the results show that the marginal beta-binomial model performs better than the Sarmanov beta-binomial model, whether or not the true model is Sarmanov beta-binomial, and the marginal beta-binomial model is more robust than the bivariate generalized linear mixed model under model misspecifications. Two meta-analyses of diagnostic accuracy studies and a meta-analysis of case-control studies are conducted for illustration.

Nonparametric meta-analysis for diagnostic accuracy studies.

Summarizing the information of many studies using a meta-analysis becomes more and more important, also in the field of diagnostic studies. The special challenge in meta-analysis of diagnostic accuracy studies is that in general sensitivity and specificity are co-primary endpoints. Across the studies both endpoints are correlated, and this correlation has to be considered in the analysis. The standard approach for such a meta-analysis is the bivariate logistic random effects model. An alternative approach is to use marginal beta-binomial distributions for the true positives and the true negatives, linked by copula distributions. In this article, we propose a new, nonparametric approach of analysis, which has greater flexibility with respect to the correlation structure, and always converges. In a simulation study, it becomes apparent that the empirical coverage of all three approaches is in general below the nominal level. Regarding bias, empirical coverage, and mean squared error the nonparametric model is often superior to the standard model, and comparable with the copula model. The three approaches are also applied to two example meta-analyses.

Value of contrast-enhanced ultrasonography in the differential diagnosis of enlarged lymph nodes: a meta-analysis of diagnostic accuracy studies.

To evaluate the diagnostic accuracy of contrast-enhanced ultrasonography (CEUS) in differentiating between benign and malignant enlarged lymph nodes using meta-analysis. Pubmed, Embase, SCI and Cochrane databases were searched for studies (up to September 1, 2014) reporting the diagnostic performance of CEUS in discriminating between benign and malignant lymph nodes. Inclusion criteria were: prospective study; histopathology as the reference standard; and sufficient data to construct 2?2 contingency tables. Methodological quality was assessed using Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). Patient clinical characteristics, sensitivity and specificity were extracted. The summary receiver operating characteristic curve was used to examine the accuracy of CEUS. A meta-analysis was performed to evaluate the clinical utility in identification of benign and malignant lymph nodes. Sensitivity analysis was performed after omitting outliers identified in a bivariate boxplot and publication bias was assessed with Egger testing. The pooled sensitivity, specificity and AUROC were 0.92 (95%CI, 0.85-0.96), 0.91 (95%CI, 0.82-0.95) and 0.97 (95%CI, 0.95-0.98), respectively. After omitting 3 outlier studies, heterogeneity decreased. Sensitivity analysis demonstrated no disproportionate influences of individual studies. Publication bias was not significant. CEUS is a promising diagnostic modality in differentiating between benign and malignant lymph nodes and can potentially reduce unnecessary fine-needle aspiration biopsies of benign nodes.

Differential diagnosis between metastatic and non-metastatic lymph nodes using DW-MRI: a meta-analysis of diagnostic accuracy studies.

The purpose of our meta-analysis was to assess the overall diagnostic value of diffusion-weighted magnetic resonance imaging (DW-MRI) in detecting node metastases and investigate whether the apparent diffusion coefficient (ADC) value could be used to discriminate between metastatic and non-metastatic lymph nodes in patients with primary tumors. The meta-analysis included a total of 1,748 metastatic and 6,547 non-metastatic lymph nodes from 39 studies, including 8 different tumor types with lymph node metastases. The pooled sensitivity and specificity of DW-MRI were 0.82 (95 % CI 0.76-0.87) and 0.92 (95 % CI 0.88-0.94), respectively. The positive likelihood ratio (PLR), negative likelihood ratio (NLR), and the area under the curve were 9.8 (95 % CI 6.9-14.0), 0.20 (95 % CI 0.15-0.26) and 0.93 (95 % CI 0.91-0.95), respectively. The probability of 42 % can be viewed as the cutoff pretest probability for DW-MRI to diagnosis lymph node metastases; when the more chance of metastatic increased from 42 % that the pretest probability was estimated, it was more suitable to emphasize on "ruling in," on the contrary, and when the more chance of metastatic decreased from 42 %, it was more suitable to emphasize on "ruling out." Furthermore, the mean ADC value of metastatic lymph nodes was significantly lower than that of non-metastatic (P = 0.001). DW-MRI is useful for differentiation between metastatic and non-metastatic lymph nodes. However, DW-MRI has a moderate diagnostic value for physician's decision making when PLR and NLR took into consideration, while a superior ability for nodal metastases confirmation, but an inferior ability for ruling out. In the future, large-scale, high-quality trials are necessary to evaluate, respectively, their clinical value in different tumor types with nodal metastases.

Accuracy of LightCycler(®) SeptiFast for the detection and identification of pathogens in the blood of patients with suspected sepsis: a systematic review and meta-analysis.

There is an urgent need to develop diagnostic tests to improve the detection of pathogens causing life-threatening infection (sepsis). SeptiFast is a CE-marked multi-pathogen real-time PCR system capable of detecting DNA sequences of bacteria and fungi present in blood samples within a few hours. We report here a systematic review and meta-analysis of diagnostic accuracy studies of SeptiFast in the setting of suspected sepsis. A comprehensive search strategy was developed to identify studies that compared SeptiFast with blood culture in suspected sepsis. Methodological quality was assessed using QUADAS. Heterogeneity of studies was investigated using a coupled forest plot of sensitivity and specificity and a scatter plot in receiver operator characteristic space. Bivariate model method was used to estimate summary sensitivity and specificity. From 41 phase III diagnostic accuracy studies, summary sensitivity and specificity for SeptiFast compared with blood culture were 0.68 (95% CI 0.63-0.73) and 0.86 (95% CI 0.84-0.89) respectively. Study quality was judged to be variable with important deficiencies overall in design and reporting that could impact on derived diagnostic accuracy metrics. SeptiFast appears to have higher specificity than sensitivity, but deficiencies in study quality are likely to render this body of work unreliable. Based on the evidence presented here, it remains difficult to make firm recommendations about the likely clinical utility of SeptiFast in the setting of suspected sepsis.

A Unification of Models for Meta-Analysis of Diagnostic Accuracy Studies without a Gold Standard

Several statistical methods for meta-analysis of diagnostic accuracy studies have been discussed in the presence of a gold standard. However, in practice, the selected reference test may be imperfect due to measurement error, non-existence, invasive nature, or expensive cost of a gold standard. It has been suggested that treating an imperfect reference test as a gold standard can lead to substantial bias in the estimation of diagnostic test accuracy. Recently, two models have been proposed to account for imperfect reference test, namely, a multivariate generalized linear mixed model (MGLMM) and a hierarchical summary receiver operating characteristic (HSROC) model. Both models are very flexible in accounting for heterogeneity in accuracies of tests across studies as well as the dependence between tests. In this article, we show that these two models, although with different formulations, are closely related and are equivalent in the absence of study-level covariates. Furthermore, we provide the exact relations between the parameters of these two models and assumptions under which two models can be reduced to equivalent submodels. On the other hand, we show that some submodels of the MGLMM do not have corresponding equivalent submodels of the HSROC model, and vice versa. With three real examples, we illustrate the cases when fitting the MGLMM and HSROC models leads to equivalent submodels and hence identical inference, and the cases when the inferences from two models are slightly different. Our results generalize the important relations between the bivariate generalized linear mixed model and HSROC model when the reference test is a gold standard.

A simple plaster for screening for diabetic neuropathy: A diagnostic test accuracy systematic review and meta-analysis

ObjectiveNeuropad is an adhesive indicator test applied at the plantar surface of the foot that detects sweating through color change. We examined the diagnostic accuracy of this simple plaster as triage test for screening for clinically relevant diabetic sensorimotor polyneuropathy in adult outpatients with type 1 or type 2 diabetes. Materials/MethodsSystematic review and meta-analysis of diagnostic accuracy studies. We searched Medline, Embase, Cochrane Library, Biosis Previews, Web of Science, Scopus and gray literature without date or language restrictions. We pooled estimates of sensitivity and specificity, and fitted hierarchical models to produce summary receiver operating characteristic curves. We assessed methodological quality of included studies utilizing the Quality Assessment of Diagnostic Accuracy Studies 2 tool. ResultsEighteen studies with 3470 participants met the inclusion criteria. Average sensitivity and specificity were 86% (95% CI 79 to 91) and 65% (95% CI 51 to 76) respectively. Likelihood ratios (LRs) were LR+=2.44 and LR−=0.22. Subgroup analyses per reference standard utilized provided similar estimates. Most studies were at risk of bias for patient selection and use of index or reference test, and had concerns regarding applicability due to patient selection. ConclusionThe adhesive indicator test has reasonable sensitivity and could be used for triage of diabetic neuropathy to rule out foot at risk. Patients who tested positive should be referred to specialized care to establish a definite diagnosis. There is insufficient evidence for effectiveness on patient-important outcomes and cost-effectiveness of implementation in the diagnostic pathway compared with the standard clinical examination.

Methodological issues in meta-analysis of diagnostic accuracy studies

Methodological issues in meta-analysis of diagnostic accuracy studies

The impact of including different study designs in meta-analyses of diagnostic accuracy studies

Diagnostic accuracy may be overestimated when using certain study designs; thus, the inclusion of studies using different designs in meta-analyses may have important effects on their results, and influence clinical decision making. The main aim of this study was to explore the influence of heterogeneity (based on the inclusion of different study designs) on diagnostic accuracy in a sample of published meta-analyses of diagnostic accuracy studies. We identified 30 systematic reviews which included 95 separate meta-analyses combining the results from a total of 976 individual studies. We classified each individual study according to the study design (case-control studies, clinically relevant patient series or other), and each meta-analysis according to the heterogeneity of the included studies. Furthermore, we registered how the methodological quality of the individual studies was assessed. Finally, for each meta-analysis, the summary measure of diagnostic accuracy was categorised as Good, Fair or Poor. We used logistic regression to assess the relationship between reporting good diagnostic accuracy and heterogeneity. Meta-analyses with heterogeneous populations were over three times more likely to report good diagnostic accuracy compared to meta-analyses that included only clinically relevant patient series (adjusted odds ratio 3.07 95% CI 1.16-8.11). The combination of studies that use different designs, within the same meta-analysis, may lead to higher estimates of diagnostic accuracy.

Adjusting for Partial Verification or Workup Bias in Meta-Analyses of Diagnostic Accuracy Studies

A key requirement in the design of diagnostic accuracy studies is that all study participants receive both the test under evaluation and the reference standard test. For a variety of practical and ethical reasons, sometimes only a proportion of patients receive the reference standard, which can bias the accuracy estimates. Numerous methods have been described for correcting this partial verification bias or workup bias in individual studies. In this article, the authors describe a Bayesian method for obtaining adjusted results from a diagnostic meta-analysis when partial verification or workup bias is present in a subset of the primary studies. The method corrects for verification bias without having to exclude primary studies with verification bias, thus preserving the main advantages of a meta-analysis: increased precision and better generalizability. The results of this method are compared with the existing methods for dealing with verification bias in diagnostic meta-analyses. For illustration, the authors use empirical data from a systematic review of studies of the accuracy of the immunohistochemistry test for diagnosis of human epidermal growth factor receptor 2 status in breast cancer patients.

Inhibin B and anti-Mullerian hormone as markers of persistent spermatogenesis in men with non-obstructive azoospermia: a meta-analysis of diagnostic accuracy studies

A non-invasive test, which could predict the presence of sperm during a testicular sperm extraction (TESE) procedure in men with non-obstructive azoospermia (NOA), would be of profound clinical importance. Inhibin B (Inh-B) and anti-Müllerian hormone (AMH) have been proposed as direct markers of Sertoli cell function and indirect markers of spermatogenesis. A search was conducted in the electronic databases MEDLINE, EMBASE and Cochrane Central Register of Controlled Trials from inception through June 2009. Thirty-six different studies reported data on the predictive value of one or more index markers (serum Inh-B: 32 studies, seminal Inh-B: 5 studies, serum AMH: 2 studies, seminal AMH: 4 studies) and were included in the systematic review. Nine studies, which had serum Inh-B as index marker, met the predefined criteria and were included in the meta-analysis. Serum Inh-B demonstrated a sensitivity of 0.65 (95% confidence interval [CI]: 0.56-0.74) and a specificity of 0.83 (CI: 0.64-0.93) for the prediction of the presence of sperm in TESE. When the pre-test probability of 41% was incorporated in a Fagan's nomogram, resulted in a positive post-test probability of 73% and a negative post-test probability of 23% for the presence of sperm in TESE. Serum Inh-B cannot serve as a stand-alone marker of persistent spermatogenesis in men with NOA. Although limited, evidence on serum AMH and serum/seminal AMH do not support their diagnostic value in men with NOA.

Evidence-Based Medicine, Systematic Reviews, and Guidelines in Interventional Pain Management: Part 7: Systematic Reviews and Meta-Analyses of Diagnostic Accuracy Studies

Appropriate diagnosis is essential in providing proper and effective therapy. The field of diagnostic accuracy tests is dynamic with new tests being developed at a fast pace along with improvement in technology of existing tests on a continuous basis. Well-designed diagnostic test accuracy studies can help in making appropriate health care decisions, provided that they transparently and fully report their participants, tests, methods, and results. Exaggerated and biased results from poorly designed and reported diagnostic test studies can trigger their premature dissemination and lead physicians into making incorrect treatment decisions. Consequently, a diagnostic test is useful only to the extent that it distinguishes between conditions or disorders that might otherwise be confused. Since it is unlikely that clinicians, patients, and policy makers have the time, skills, and resources to find, appraise, and interpret the evidence and incorporate it into their health care decisions, systematic reviews and meta-analysis provide an accurate and reliable synthesis of vast quantities of data. A systematic review can identify what is known and what is unknown, giving guidance for future research. Systematic reviews have been considered as a vital link in the great chain of evidence that stretches from the laboratory to the bedside by helping to separate the insignificant, unsound, or redundant deadwood from the salient and critical studies that are worthy of reflection. A dangerous discrepancy exists between experts and evidence with all types of evidence. Historically, it has been reported that in only 15% of all cases can a pathoanatomical explanation be found for patients with chronic low back pain of more than 3 months resulting in the assumption that very little can be done in our present state of ignorance to treat these patients and improve their natural histories. On the other end of the spectrum, due to lack of sound diagnostic information, excessive health care is utilized with exploding costs. The validity of all diagnostic techniques has been described with variable accuracy and reliability. Lack of understanding of reference standards and their unavailability with interventional diagnostic techniques and misinterpretation secondary to interpretation bias may adversely influence the applicability of diagnostic interventions. This manuscript provides a review of the literature, a checklist, and a flow diagram describing the preferred way to present the abstract, introduction, methods, results, and discussion sections of the report of an analysis in a systematic review of diagnostic accuracy studies.

Quantitative myocardial contrast echocardiography during pharmacological stress for diagnosis of coronary artery disease: a systematic review and meta-analysis of diagnostic accuracy studies

We conducted a meta-analysis to evaluate the accuracy of quantitative stress myocardial contrast echocardiography (MCE) in coronary artery disease (CAD). Database search was performed through January 2008. We included studies evaluating accuracy of quantitative stress MCE for detection of CAD compared with coronary angiography or single-photon emission computed tomography (SPECT) and measuring reserve parameters of A, beta, and Abeta. Data from studies were verified and supplemented by the authors of each study. Using random effects meta-analysis, we estimated weighted mean difference (WMD), likelihood ratios (LRs), diagnostic odds ratios (DORs), and summary area under curve (AUC), all with 95% confidence interval (CI). Of 1443 studies, 13 including 627 patients (age range, 38-75 years) and comparing MCE with angiography (n = 10), SPECT (n = 1), or both (n = 2) were eligible. WMD (95% CI) were significantly less in CAD group than no-CAD group: 0.12 (0.06-0.18) (P < 0.001), 1.38 (1.28-1.52) (P < 0.001), and 1.47 (1.18-1.76) (P < 0.001) for A, beta, and Abeta reserves, respectively. Pooled LRs for positive test were 1.33 (1.13-1.57), 3.76 (2.43-5.80), and 3.64 (2.87-4.78) and LRs for negative test were 0.68 (0.55-0.83), 0.30 (0.24-0.38), and 0.27 (0.22-0.34) for A, beta, and Abeta reserves, respectively. Pooled DORs were 2.09 (1.42-3.07), 15.11 (7.90-28.91), and 14.73 (9.61-22.57) and AUCs were 0.637 (0.594-0.677), 0.851 (0.828-0.872), and 0.859 (0.842-0.750) for A, beta, and Abeta reserves, respectively. Evidence supports the use of quantitative MCE as a non-invasive test for detection of CAD. Standardizing MCE quantification analysis and adherence to reporting standards for diagnostic tests could enhance the quality of evidence in this field.

Meta‐analysis of diagnostic accuracy studies accounting for disease prevalence: Alternative parameterizations and model selection

In a meta-analysis of diagnostic accuracy studies, the sensitivities and specificities of a diagnostic test may depend on the disease prevalence since the severity and definition of disease may differ from study to study due to the design and the population considered. In this paper, we extend the bivariate nonlinear random effects model on sensitivities and specificities to jointly model the disease prevalence, sensitivities and specificities using trivariate nonlinear random-effects models. Furthermore, as an alternative parameterization, we also propose jointly modeling the test prevalence and the predictive values, which reflect the clinical utility of a diagnostic test. These models allow investigators to study the complex relationship among the disease prevalence, sensitivities and specificities; or among test prevalence and the predictive values, which can reveal hidden information about test performance. We illustrate the proposed two approaches by reanalyzing the data from a meta-analysis of radiological evaluation of lymph node metastases in patients with cervical cancer and a simulation study. The latter illustrates the importance of carefully choosing an appropriate normality assumption for the disease prevalence, sensitivities and specificities, or the test prevalence and the predictive values. In practice, it is recommended to use model selection techniques to identify a best-fitting model for making statistical inference. In summary, the proposed trivariate random effects models are novel and can be very useful in practice for meta-analysis of diagnostic accuracy studies.