Cone-Beam CT With Augmented Fluoroscopy Combined With Electromagnetic Navigation Bronchoscopy for Biopsy of Pulmonary Nodules.

BACKGROUND: Electromagnetic navigation bronchoscopy (ENB) has been widely adopted as a guidance technique for biopsy of peripheral lung nodules. However, ENB is limited by the lack of real-time confirmation of the biopsy devices. Intraprocedural cone-beam computed tomography (CBCT) imaging can be utilized to assess or confirm the location of biopsy devices. The aim of this study is to determine the safety and diagnostic yield (DY) of image fusion of intraprocedural CBCT data with live fluoroscopy (augmented fluoroscopy) during ENB-guided biopsy of peripheral lung nodules. METHODS: Data from 75 consecutive patients who underwent biopsy with ENB was collected retrospectively. Patients underwent CBCT imaging while temporarily suspending mechanical ventilation. CBCT data were acquired and 3-dimensional segmentation of nodules was performed using commercially available software (OncoSuite). During ENB, the segmented lesions were projected and fused with live fluoroscopy enabling real-time 3-dimensional guidance. RESULTS: A total of 93 lesions with a median size of 16.0 mm were biopsied in 75 consecutive patients. The overall DY by lesion was 83.7% (95% confidence interval, 74.8%-89.9%). Multivariate regression analysis showed no independent correlation between lesion size, lesion location, lesion visibility under standard fluoroscopy, and the presence of a bronchus sign with DY. Pneumothorax occurred in 3 patients (4%). CONCLUSION: Intraprocedural CBCT imaging with augmented fluoroscopy is feasible and effective and is associated with high DY during ENB-guided biopsies.

Overdiagnosis across medical disciplines: a scoping review.

OBJECTIVE: To provide insight into how and in what clinical fields overdiagnosis is studied and give directions for further applied and methodological research. DESIGN: Scoping review. DATA SOURCES: Medline up to August 2017. STUDY SELECTION: All English studies on humans, in which overdiagnosis was discussed as a dominant theme. DATA EXTRACTION: Studies were assessed on clinical field, study aim (ie, methodological or non-methodological), article type (eg, primary study, review), the type and role of diagnostic test(s) studied and the context in which these studies discussed overdiagnosis. RESULTS: From 4896 studies, 1851 were included for analysis. Half of all studies on overdiagnosis were performed in the field of oncology (50%). Other prevalent clinical fields included mental disorders, infectious diseases and cardiovascular diseases accounting for 9%, 8% and 6% of studies, respectively. Overdiagnosis was addressed from a methodological perspective in 20% of studies. Primary studies were the most common article type (58%). The type of diagnostic tests most commonly studied were imaging tests (32%), although these were predominantly seen in oncology and cardiovascular disease (84%). Diagnostic tests were studied in a screening setting in 43% of all studies, but as high as 75% of all oncological studies. The context in which studies addressed overdiagnosis related most frequently to its estimation, accounting for 53%. Methodology on overdiagnosis estimation and definition provided a source for extensive discussion. Other contexts of discussion included definition of disease, overdiagnosis communication, trends in increasing disease prevalence, drivers and consequences of overdiagnosis, incidental findings and genomics. CONCLUSIONS: Overdiagnosis is discussed across virtually all clinical fields and in different contexts. The variability in characteristics between studies and lack of consensus on overdiagnosis definition indicate the need for a uniform typology to improve coherence and comparability of studies on overdiagnosis.

Effect of Fibrinogen Concentrate on Intraoperative Blood Loss Among Patients With Intraoperative Bleeding During High-Risk Cardiac Surgery: A Randomized Clinical Trial.

Importance: Fibrinogen concentrate might partly restore coagulation defects and reduce intraoperative bleeding. Objective: To determine whether fibrinogen concentrate infusion dosed to achieve a plasma fibrinogen level of 2.5 g/L in high-risk cardiac surgery patients with intraoperative bleeding reduces intraoperative blood loss. Design, Setting, and Participants: A randomized, placebo-controlled, double-blind clinical trial conducted in Isala Zwolle, the Netherlands (February 2011-January 2015), involving patients undergoing elective, high-risk cardiac surgery (ie, combined coronary artery bypass graft [CABG] surgery and valve repair or replacement surgery, the replacement of multiple valves, aortic root reconstruction, or reconstruction of the ascending aorta or aortic arch) with intraoperative bleeding (blood volume between 60 and 250 mL suctioned from the thoracic cavity in a period of 5 minutes) were randomized to receive either fibrinogen concentrate or placebo. Interventions: Intravenous, single-dose administration of fibrinogen concentrate (n = 60) or placebo (n = 60), targeted to achieve a postinfusion plasma fibrinogen level of 2.5 g/L. Main Outcomes and Measures: The primary outcome was blood loss in milliliters between intervention (ie, after removal of cardiopulmonary bypass) and closure of chest. Safety variables (within 30 days) included: in-hospital mortality, myocardial infarction, cerebrovascular accident or transient ischemic attack, renal insufficiency or failure, venous thromboembolism, pulmonary embolism, and operative complications. Results: Among 120 patients (mean age; 71 [SD, 10] years, 37 women [31%]) included in the study, combined CABG and valve repair or replacement surgery comprised 72% of procedures and had a mean (SD) cardiopulmonary bypass time of 200 minutes (83) minutes. For the primary outcome, median blood loss in the fibrinogen group was 50 mL (interquartile range [IQR], 29-100 mL) compared with 70 mL (IQR, 33-145 mL) in the control group (P = .19), the absolute difference 20 mL (95% CI, -13 to 35 mL). There were 6 cases of stroke or transient ischemic attack (4 in the fibrinogen group); 4 myocardial infarctions (3 in the fibrinogen group); 2 deaths (both in the fibrinogen group); 5 cases with renal insufficiency or failure (3 in the fibrinogen group); and 9 cases with reoperative thoracotomy (4 in the fibrinogen group). Conclusions and Relevance: Among patients with intraoperative bleeding during high-risk cardiac surgery, administration of fibrinogen concentrate, compared with placebo, resulted in no significant difference in the amount of intraoperative blood loss. Trial Registration: clinicaltrials.gov Identifier: NCT01124981 and EudraCT No: 2009-018086-12.

A host-protein based assay to differentiate between bacterial and viral infections in preschool children (OPPORTUNITY): a double-blind, multicentre, validation study.

BACKGROUND: A physician is frequently unable to distinguish bacterial from viral infections. ImmunoXpert is a novel assay combining three proteins: tumour necrosis factor-related apoptosis-inducing ligand (TRAIL), interferon gamma induced protein-10 (IP-10), and C-reactive protein (CRP). We aimed to externally validate the diagnostic accuracy of this assay in differentiating between bacterial and viral infections and to compare this test with commonly used biomarkers. METHODS: In this prospective, double-blind, international, multicentre study, we recruited children aged 2-60 months with lower respiratory tract infection or clinical presentation of fever without source at four hospitals in the Netherlands and two hospitals in Israel. A panel of three experienced paediatricians adjudicated a reference standard diagnosis for all patients (ie, bacterial or viral infection) using all available clinical and laboratory information, including a 28-day follow-up assessment. The panel was masked to the assay results. We identified majority diagnosis when two of three panel members agreed on a diagnosis and unanimous diagnosis when all three panel members agreed on the diagnosis. We calculated the diagnostic performance (ie, sensitivity, specificity, positive predictive value, and negative predictive value) of the index test in differentiating between bacterial (index test positive) and viral (index test negative) infection by comparing the test classification with the reference standard outcome. FINDINGS: Between Oct 16, 2013 and March 1, 2015, we recruited 777 children, of whom 577 (mean age 21 months, 56% male) were assessed. The majority of the panel diagnosed 71 cases as bacterial infections and 435 as viral infections. In another 71 patients there was an inconclusive panel diagnosis. The assay distinguished bacterial from viral infections with a sensitivity of 86·7% (95% CI 75·8-93·1), a specificity of 91·1% (87·9-93·6), a positive predictive value of 60·5% (49·9-70·1), and a negative predictive value of 97·8% (95·6-98·9). In the more clear cases with unanimous panel diagnosis (n=354), sensitivity was 87·8% (74·5-94·7), specificity 93·0% (89·6-95·3), positive predictive value 62·1% (49·2-73·4), and negative predictive value 98·3% (96·1-99·3). INTERPRETATION: This external validation study shows the diagnostic value of a three-host protein-based assay to differentiate between bacterial and viral infections in children with lower respiratory tract infection or fever without source. This diagnostic based on CRP, TRAIL, and IP-10 has the potential to reduce antibiotic misuse in young children. FUNDING: MeMed Diagnostics.

The added value of C-reactive protein measurement in diagnosing pneumonia in primary care: a meta-analysis of individual patient data.

BACKGROUND: C-reactive protein (CRP) is increasingly being included in the diagnostic work-up for community-acquired pneumonia in primary care. Its added diagnostic value beyond signs and symptoms, however, remains unclear. We conducted a meta-analysis of individual patient data to quantify the added value of CRP measurement. METHODS: We included studies of the diagnostic accuracy of CRP in adult outpatients with suspected lower respiratory tract infection. We contacted authors of eligible studies for inclusion of data and for additional data as needed. The value of adding CRP measurement to a basic signs-and-symptoms prediction model was assessed. Outcome measures were improvement in discrimination between patients with and without pneumonia in primary care and improvement in risk classification, both within the individual studies and across studies. RESULTS: Authors of 8 eligible studies (n = 5308) provided their data sets. In all of the data sets, discrimination between patients with and without pneumonia improved after CRP measurement was added to the prediction model (extended model), with a mean improvement in the area under the curve of 0.075 (range 0.02-0.18). In a hypothetical cohort of 1000 patients, the proportion of patients without pneumonia correctly classified at low risk increased from 28% to 36% in the extended model, and the proportion with pneumonia correctly classified at high risk increased from 63% to 70%. The number of patients with pneumonia classified at low risk did not change (n = 4). Overall, the proportion of patients assigned to the intermediate-risk category decreased from 56% to 51%. INTERPRETATION: Adding CRP measurement to the diagnostic work-up for suspected pneumonia in primary care improved the discrimination and risk classification of patients. However, it still left a substantial group of patients classified at intermediate risk, in which clinical decision-making remains challenging.

No rationale for 1 variable per 10 events criterion for binary logistic regression analysis.

BACKGROUND: Ten events per variable (EPV) is a widely advocated minimal criterion for sample size considerations in logistic regression analysis. Of three previous simulation studies that examined this minimal EPV criterion only one supports the use of a minimum of 10 EPV. In this paper, we examine the reasons for substantial differences between these extensive simulation studies. METHODS: The current study uses Monte Carlo simulations to evaluate small sample bias, coverage of confidence intervals and mean square error of logit coefficients. Logistic regression models fitted by maximum likelihood and a modified estimation procedure, known as Firth's correction, are compared. RESULTS: The results show that besides EPV, the problems associated with low EPV depend on other factors such as the total sample size. It is also demonstrated that simulation results can be dominated by even a few simulated data sets for which the prediction of the outcome by the covariates is perfect ('separation'). We reveal that different approaches for identifying and handling separation leads to substantially different simulation results. We further show that Firth's correction can be used to improve the accuracy of regression coefficients and alleviate the problems associated with separation. CONCLUSIONS: The current evidence supporting EPV rules for binary logistic regression is weak. Given our findings, there is an urgent need for new research to provide guidance for supporting sample size considerations for binary logistic regression analysis.

Predicting stillbirth in a low resource setting.

BACKGROUND: Stillbirth is a major contributor to perinatal mortality and it is particularly common in low- and middle-income countries, where annually about three million stillbirths occur in the third trimester. This study aims to develop a prediction model for early detection of pregnancies at high risk of stillbirth. METHODS: This retrospective cohort study examined 6,573 pregnant women who delivered at Federal Medical Centre Bida, a tertiary level of healthcare in Nigeria from January 2010 to December 2013. Descriptive statistics were performed and missing data imputed. Multivariable logistic regression was applied to examine the associations between selected candidate predictors and stillbirth. Discrimination and calibration were used to assess the model's performance. The prediction model was validated internally and over-optimism was corrected. RESULTS: We developed a prediction model for stillbirth that comprised maternal comorbidity, place of residence, maternal occupation, parity, bleeding in pregnancy, and fetal presentation. As a secondary analysis, we extended the model by including fetal growth rate as a predictor, to examine how beneficial ultrasound parameters would be for the predictive performance of the model. After internal validation, both calibration and discriminative performance of both the basic and extended model were excellent (i.e. C-statistic basic model = 0.80 (95 % CI 0.78-0.83) and extended model = 0.82 (95 % CI 0.80-0.83)). CONCLUSION: We developed a simple but informative prediction model for early detection of pregnancies with a high risk of stillbirth for early intervention in a low resource setting. Future research should focus on external validation of the performance of this promising model.

External Validation of Prediction Models for Pneumonia in Primary Care Patients with Lower Respiratory Tract Infection: An Individual Patient Data Meta-Analysis.

BACKGROUND: Pneumonia remains difficult to diagnose in primary care. Prediction models based on signs and symptoms (S&S) serve to minimize the diagnostic uncertainty. External validation of these models is essential before implementation into routine practice. In this study all published S&S models for prediction of pneumonia in primary care were externally validated in the individual patient data (IPD) of previously performed diagnostic studies. METHODS AND FINDINGS: S&S models for diagnosing pneumonia in adults presenting to primary care with lower respiratory tract infection and IPD for validation were identified through a systematical search. Six prediction models and IPD of eight diagnostic studies (N total = 5308, prevalence pneumonia 12%) were included. Models were assessed on discrimination and calibration. Discrimination was measured using the pooled Area Under the Curve (AUC) and delta AUC, representing the performance of an individual model relative to the average dataset performance. Prediction models by van Vugt et al. and Heckerling et al. demonstrated the highest pooled AUC of 0.79 (95% CI 0.74-0.85) and 0.72 (0.68-0.76), respectively. Other models by Diehr et al., Singal et al., Melbye et al., and Hopstaken et al. demonstrated pooled AUCs of 0.65 (0.61-0.68), 0.64 (0.61-0.67), 0.56 (0.49-0.63) and 0.53 (0.5-0.56), respectively. A similar ranking was present based on the delta AUCs of the models. Calibration demonstrated close agreement of observed and predicted probabilities in the models by van Vugt et al. and Singal et al., other models lacked such correspondence. The absence of predictors in the IPD on dataset level hampered a systematical comparison of model performance and could be a limitation to the study. CONCLUSIONS: The model by van Vugt et al. demonstrated the highest discriminative accuracy coupled with reasonable to good calibration across the IPD of different study populations. This model is therefore the main candidate for primary care use.

Assessing variability in results in systematic reviews of diagnostic studies.

BACKGROUND: To describe approaches used in systematic reviews of diagnostic test accuracy studies for assessing variability in estimates of accuracy between studies and to provide guidance in this area. METHODS: Meta-analyses of diagnostic test accuracy studies published between May and September 2012 were systematically identified. Information on how the variability in results was investigated was extracted. RESULTS: Of the 53 meta-analyses included in the review, most (n=48; 91%) presented variability in diagnostic accuracy estimates visually either through forest plots or ROC plots and the majority (n=40; 75%) presented a test or statistical measure for the variability. Twenty-eight reviews (53%) tested for variability beyond chance using Cochran's Q test and 31 (58%) reviews quantified it with I(2). 7 reviews (13%) presented between-study variance estimates (τ(2)) from random effects models and 3 of these presented a prediction interval or ellipse to facilitate interpretation. Half of all the meta-analyses specified what was considered a significant amount of variability (n=24; 49%). CONCLUSIONS: Approaches to assessing variability in estimates of accuracy varied widely between diagnostic test accuracy reviews and there is room for improvement. We provide initial guidance, complemented by an overview of the currently available approaches.

Problems in detecting misfit of latent class models in diagnostic research without a gold standard were shown.

OBJECTIVES: The objective of this study was to evaluate the performance of goodness-of-fit testing to detect relevant violations of the assumptions underlying the criticized "standard" two-class latent class model. Often used to obtain sensitivity and specificity estimates for diagnostic tests in the absence of a gold reference standard, this model relies on assuming that diagnostic test errors are independent. When this assumption is violated, accuracy estimates may be biased: goodness-of-fit testing is often used to evaluate the assumption and prevent bias. STUDY DESIGN AND SETTING: We investigate the performance of goodness-of-fit testing by Monte Carlo simulation. The simulation scenarios are based on three empirical examples. RESULTS: Goodness-of-fit tests lack power to detect relevant misfit of the standard two-class latent class model at sample sizes that are typically found in empirical diagnostic studies. The goodness-of-fit tests that are based on asymptotic theory are not robust to the sparseness of data. A parametric bootstrap procedure improves the evaluation of goodness of fit in the case of sparse data. CONCLUSION: Our simulation study suggests that relevant violation of the local independence assumption underlying the standard two-class latent class model may remain undetected in empirical diagnostic studies, potentially leading to biased estimates of sensitivity and specificity.

Small-study effects and time trends in diagnostic test accuracy meta-analyses: a meta-epidemiological study.

BACKGROUND: Small-study effects and time trends have been identified in meta-analyses of randomized trials. We evaluated whether these effects are also present in meta-analyses of diagnostic test accuracy studies. METHODS: A systematic search identified test accuracy meta-analyses published between May and September 2012. In each meta-analysis, the strength of the associations between estimated accuracy of the test (diagnostic odds ratio (DOR), sensitivity, and specificity) and sample size and between accuracy estimates and time since first publication were evaluated using meta-regression models. The regression coefficients over all meta-analyses were summarized using random effects meta-analysis. RESULTS: Forty-six meta-analyses and their corresponding primary studies (N = 859) were included. There was a non-significant relative change in the DOR of 1.01 per 100 additional participants (95% CI 1.00 to 1.03; P = 0.07). In the subgroup of imaging studies, there was a relative increase in sensitivity of 1.13 per 100 additional diseased subjects (95% CI 1.05 to 1.22; P = 0.002). The relative change in DOR with time since first publication was 0.94 per 5 years (95% CI 0.80 to 1.10; P = 0.42). Sensitivity was lower in studies published later (relative change 0.89, 95% CI 0.80 to 0.99; P = 0.04). CONCLUSIONS: Small-study effects and time trends do not seem to be as pronounced in meta-analyses of test accuracy studies as they are in meta-analyses of randomized trials. Small-study effects seem to be reversed in imaging, where larger studies tend to report higher sensitivity.

The added diagnostic value of five different C-reactive protein point-of-care test devices in detecting pneumonia in primary care: A nested case-control study.

BACKGROUND: The results obtained from various point-of-care (POC) test devices for estimating C-reactive protein (CRP) levels in a laboratory setting differ when compared to a laboratory reference test. We aimed to determine whether such differences meaningfully affect the accuracy and added diagnostic value in predicting radiographic pneumonia in adults presenting with acute cough in primary care. METHODS: A nested case control study of adult patients presenting with acute cough in 12 different European countries (the Genomics to combat Resistance against Antibiotics in Community-acquired LRTI in Europe [GRACE] Network). Venous blood samples from 100 patients with and 100 patients without pneumonia were tested with five different POC CRP tests and a laboratory analyzer. Single test accuracy values and the added value of CRP to symptoms and signs were calculated. RESULTS: Single test accuracy values showed similar results for all five POC CRP tests and the laboratory analyzer. The area under the curve of the different POC CRP tests and the laboratory analyzer (range 0.79-0.80) were all comparable and higher than the clinical model without CRP (0.70). Multivariable odds ratios were the same (1.2) for all CRP tests. CONCLUSIONS: Five POC CRP test devices and the laboratory analyzer performed with similar accuracy in detecting pneumonia both as single test, and when used in addition to clinical findings. Variability in results obtained from standard CRP laboratory and POC test devices do not translate into clinically relevant differences when used for prediction of pneumonia in patients with acute cough in primary care.

Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist.

Carl Moons and colleagues provide a checklist and background explanation for critically appraising and extracting data from systematic reviews of prognostic and diagnostic prediction modelling studies. Please see later in the article for the Editors' Summary.

Systematic overview finds variation in approaches to investigating and reporting on sources of heterogeneity in systematic reviews of diagnostic studies.

OBJECTIVES: To examine how authors explore and report on sources of heterogeneity in systematic reviews of diagnostic accuracy studies. STUDY DESIGN AND SETTING: A cohort of systematic reviews of diagnostic tests was systematically identified. Data were extracted on whether an exploration of the sources of heterogeneity was undertaken, how this was done, the number and type of potential sources explored, and how results and conclusions were reported. RESULTS: Of the 65 systematic reviews, 12 did not perform a meta-analysis and eight of these gave heterogeneity between studies as a reason. Of the 53 reviews containing a meta-analysis, 40 explored potential sources of heterogeneity in a formal manner and 27 identified at least one source of heterogeneity. The reviews not investigating heterogeneity were smaller than those that did (median [interquartile range {IQR}], 8 [5-15] vs. 14 [11-19] primary studies). Twelve reviews performed a sensitivity analysis, 25 stratified analyses, and 19 metaregression. Many sources of heterogeneity were explored compared with the number of primary studies in a meta-analysis (median ratio, 1:5). Review authors placed importance on the exploration of sources of heterogeneity; 37 mentioned the exploration or the findings thereof in the abstract or conclusion of the main text.results CONCLUSION: Methods for investigating sources of heterogeneity varied widely between reviews. Based on our findings of the review, we made suggestions on what to consider and report on when exploring sources of heterogeneity in systematic reviews of diagnostic studies.

Added value of hybrid myocardial perfusion SPECT and CT coronary angiography in the diagnosis of coronary artery disease.

AIMS: Hybrid single photon emission computed tomography (SPECT)/coronary computed tomography angiography (CCTA) has only been evaluated for its diagnostic accuracy as a single test in patients suspected of significant coronary artery disease (CAD). Added value of hybrid SPECT/CCTA beyond usual clinical work-up, or use of each of these tests separately, remains unclear. We evaluated the added value of hybrid myocardial perfusion SPECT (SPECT) and CCTA, beyond pre-test likelihood and exercise stress ECG (X-ECG), in the diagnosis of CAD. METHODS AND RESULTS: Two hundred and five patients with stable angina pectoris and intermediate-to-high pre-test likelihood were prospectively included. All patients underwent clinical history and examination, X-ECG, stress and rest SPECT, coronary calcium scoring (CCS) and CCTA. Fractional flow reserve measurement <0.80 or a lesion >50% on coronary angiography (CA) served as reference standard for significant CAD. Multiple imputation was used to correct for missing test results (17-20%). Added value of hybrid SPECT/CCTA to the basic model of pre-test likelihood plus X-ECG was quantified using logistic regression analysis. Model differences were then assessed using differences in C-index and in net reclassification improvement (NRI). The basic model had a C-index of 0.73 (95%CI 0.66-0.80). This significantly increased to 0.85 (95%CI 0.80-0.91) by addition of only SPECT, to 0.90 (95%CI 0.85-0.94) when adding only CCTA, and to 0.96 (95%CI 0.92-0.99) when adding hybrid SPECT/CCTA. The accompanying NRIs were 0.82 (95%CI 0.62-1.02), 0.86 (95%CI 0.66-1.06) and 1.57 (95%CI 1.11-1.59) respectively. CONCLUSION: Current analysis resembles clinical routine of layered testing and shows that hybrid SPECT/CCTA imaging has a substantially higher yield than standalone SPECT or CCTA in the diagnostic workup of patients suspected of significant CAD.

Incorporating quality assessments of primary studies in the conclusions of diagnostic accuracy reviews: a cross-sectional study.

BACKGROUND: Drawing conclusions from systematic reviews of test accuracy studies without considering the methodological quality (risk of bias) of included studies may lead to unwarranted optimism about the value of the test(s) under study. We sought to identify to what extent the results of quality assessment of included studies are incorporated in the conclusions of diagnostic accuracy reviews. METHODS: We searched MEDLINE and EMBASE for test accuracy reviews published between May and September 2012. We examined the abstracts and main texts of these reviews to see whether and how the results of quality assessment were linked to the accuracy estimates when drawing conclusions. RESULTS: We included 65 reviews of which 53 contained a meta-analysis. Sixty articles (92%) had formally assessed the methodological quality of included studies, most often using the original QUADAS tool (n = 44, 68%). Quality assessment was mentioned in 28 abstracts (43%); with a majority (n = 21) mentioning it in the methods section. In only 5 abstracts (8%) were results of quality assessment incorporated in the conclusions. Thirteen reviews (20%) presented results of quality assessment in the main text only, without further discussion. Forty-seven reviews (72%) discussed results of quality assessment; the most frequent form was as limitations in assessing quality (n = 28). Only 6 reviews (9%) further linked the results of quality assessment to their conclusions, 3 of which did not conduct a meta-analysis due to limitations in the quality of included studies. In the reviews with a meta-analysis, 19 (36%) incorporated quality in the analysis. Eight reported significant effects of quality on the pooled estimates; in none of them these effects were factored in the conclusions. CONCLUSION: While almost all recent diagnostic accuracy reviews evaluate the quality of included studies, very few consider results of quality assessment when drawing conclusions. The practice of reporting systematic reviews of test accuracy should improve if readers not only want to be informed about the limitations in the available evidence, but also on the associated implications for the performance of the evaluated tests.

Latent class models in diagnostic studies when there is no reference standard--a systematic review.

Latent class models (LCMs) combine the results of multiple diagnostic tests through a statistical model to obtain estimates of disease prevalence and diagnostic test accuracy in situations where there is no single, accurate reference standard. We performed a systematic review of the methodology and reporting of LCMs in diagnostic accuracy studies. This review shows that the use of LCMs in such studies increased sharply in the past decade, notably in the domain of infectious diseases (overall contribution: 59%). The 64 reviewed studies used a range of differently specified parametric latent variable models, applying Bayesian and frequentist methods. The critical assumption underlying the majority of LCM applications (61%) is that the test observations must be independent within 2 classes. Because violations of this assumption can lead to biased estimates of accuracy and prevalence, performing and reporting checks of whether assumptions are met is essential. Unfortunately, our review shows that 28% of the included studies failed to report any information that enables verification of model assumptions or performance. Because of the lack of information on model fit and adequate evidence "external" to the LCMs, it is often difficult for readers to judge the validity of LCM-based inferences and conclusions reached.