Dataset Information

Estimating the true accuracy of diagnostic tests for dengue infection using bayesian latent class models.

ABSTRACT:

Background

Accuracy of rapid diagnostic tests for dengue infection has been repeatedly estimated by comparing those tests with reference assays. We hypothesized that those estimates might be inaccurate if the accuracy of the reference assays is not perfect. Here, we investigated this using statistical modeling.

Methods/principal findings

Data from a cohort study of 549 patients suspected of dengue infection presenting at Colombo North Teaching Hospital, Ragama, Sri Lanka, that described the application of our reference assay (a combination of Dengue IgM antibody capture ELISA and IgG antibody capture ELISA) and of three rapid diagnostic tests (Panbio NS1 antigen, IgM antibody and IgG antibody rapid immunochromatographic cassette tests) were re-evaluated using bayesian latent class models (LCMs). The estimated sensitivity and specificity of the reference assay were 62.0% and 99.6%, respectively. Prevalence of dengue infection (24.3%), and sensitivities and specificities of the Panbio NS1 (45.9% and 97.9%), IgM (54.5% and 95.5%) and IgG (62.1% and 84.5%) estimated by bayesian LCMs were significantly different from those estimated by assuming that the reference assay was perfect. Sensitivity, specificity, PPV and NPV for a combination of NS1, IgM and IgG cassette tests on admission samples were 87.0%, 82.8%, 62.0% and 95.2%, respectively.

Conclusions

Our reference assay is an imperfect gold standard. In our setting, the combination of NS1, IgM and IgG rapid diagnostic tests could be used on admission to rule out dengue infection with a high level of accuracy (NPV 95.2%). Further evaluation of rapid diagnostic tests for dengue infection should include the use of appropriate statistical models.

SUBMITTER: Pan-ngum W

PROVIDER: S-EPMC3548900 | biostudies-literature | 2013

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Estimating the true accuracy of diagnostic tests for dengue infection using bayesian latent class models.

Pan-ngum Wirichada W Blacksell Stuart D SD Lubell Yoel Y Pukrittayakamee Sasithon S Bailey Mark S MS de Silva H Janaka HJ Lalloo David G DG Day Nicholas P J NP White Lisa J LJ Limmathurotsakul Direk D

PloS one 20130118 1

<h4>Background</h4>Accuracy of rapid diagnostic tests for dengue infection has been repeatedly estimated by comparing those tests with reference assays. We hypothesized that those estimates might be inaccurate if the accuracy of the reference assays is not perfect. Here, we investigated this using statistical modeling.<h4>Methods/principal findings</h4>Data from a cohort study of 549 patients suspected of dengue infection presenting at Colombo North Teaching Hospital, Ragama, Sri Lanka, that des ...[more]

PMID: 23349667

Similar Datasets

Project description:ObjectivePrevious studies on diagnostic accuracy of dipstick testing for leukocyte esterase (LE) and nitrite to diagnose urinary tract infection (UTI) had used urine culture, which is an imperfect gold standard. Estimates of diagnostic accuracy obtained using the classical gold standard framework might not reflect the true diagnostic accuracy of dipstick tests.MethodsWe used the dataset from a prospective, observational study conducted in the emergency department of a teaching hospital in southern India. Patients with a clinical suspicion of UTI underwent dipstick testing for LE and nitrite, urine microscopy, and urine culture. Based on the results of urine microscopy and culture, UTI was classified into definite, probable, and possible. Patients with microscopic pyuria and a positive urine culture were adjudicated as definite UTI. Unequivocal imaging evidence of emphysematous pyelonephritis or perinephric collections was also considered definite UTI. We estimated the diagnostic accuracy of LE and nitrite tests using the classical analysis (assuming definite UTI as gold standard) and two different Bayesian latent class models (LCMs; 3-tests in 1-population and 2-tests in 2-populations models).ResultsWe studied 149 patients. Overall, 64 (43%) patients had definite, 76 (51%) had probable, and 2 (1.3%) had possible UTI; 7 (4.6%) had alternate diagnoses. In classical analysis, LE was more sensitive than nitrite (87.5% versus 70.5%), while nitrite was more specific (24% versus 58%). The 3-tests in 1-population Bayesian LCM indicated a substantially better sensitivity and specificity for LE (98.1% and 47.6%) and nitrite (88.2% and 97.7%). True sensitivity and specificity of urine culture as estimated by the model was 48.7% and 73.0%. Estimates of the 2-tests in 2-populations model were in agreement with the 3-tests in 1-population model.ConclusionsBayesian LCMs indicate a clinically important improvement in the true diagnostic accuracy of urine dipstick testing for LE and nitrite. Given this, a negative dipstick LE would rule-out UTI, while a positive dipstick nitrite would rule-in UTI in our study setting. True diagnostic accuracy of urine dipstick testing for UTI in various practice settings needs reevaluation using Bayesian LCMs.

Project description:BackgroundTyphoid fevers are infections caused by the bacteria Salmonella enterica serovar Typhi (Salmonella Typhi) and Paratyphi A, B and C (Salmonella Paratyphi). Approximately 17.8 million incident cases of typhoid fever occur annually, and incidence is highest in children. The accuracy of current diagnostic tests of typhoid fever is poorly understood. We aimed to determine the comparative accuracy of available tests for the pediatric population.MethodsWe first conducted a systematic literature review to identify studies that compared diagnostic tests for typhoid fever in children (aged ≤15 years) to blood culture results. We applied a Bayesian latent-class extension to a network meta-analysis model. We modelled known diagnostic properties of bone marrow culture and the relationship between bone marrow and blood culture as informative priors in a Bayesian framework. We tested sensitivities for the proportion of negative blood samples that were false as well as bone marrow sensitivity and specificity.ResultsWe found 510 comparisons from 196 studies and 57 specific to the pediatric population. IgM-based tests outperformed their IgG-based counterparts for ELISA and Typhidot tests. The lateral flow IgG test performed comparatively well with 92% sensitivity (72% to 98% across scenario analyses) and 94% specificity. The most sensitive test of those investigated for the South Asian pediatric population was the Reverse Passive Hemagglutination Assay with 99% sensitivity (98% - 100% across scenario analyses). Adding a Widal slide test to other typhoid diagnostics did not substantially improve diagnostic performance beyond the single test alone, however, a lateral flow-based IgG rapid test combined with the typhoid/paratyphoid (TPT) assay yielded improvements in sensitivity without substantial declines in specificity and was the best performing combination test in this setting.ConclusionIn the pediatric population, lateral-flow IgG, TPT and Reverse Passive Hemagglutination tests had high diagnostic accuracy compared to other diagnostics. Combinations of tests may provide a feasible option to increase diagnostic sensitivity. South Asia has the most informed set of data on typhoid diagnostic testing accuracy, and the evidence base in other important regions needs to be expanded.

Project description:BackgroundEvaluating the accuracy of extrapulmonary tuberculosis (TB) tests is challenging due to lack of a gold standard. Latent class analysis (LCA), a statistical modeling approach, can adjust for reference tests' imperfect accuracies to produce less biased test accuracy estimates than those produced by commonly used methods like composite reference standards (CRSs). Our objective is to illustrate how Bayesian LCA can address the problem of an unavailable gold standard and demonstrate how it compares to using CRSs for extrapulmonary TB tests.MethodsWe re-analyzed a dataset of presumptive extrapulmonary TB cases in New Delhi, India, for three forms of extrapulmonary TB. Results were available for culture, smear microscopy, Xpert MTB/RIF, and a non-microbiological test, cytopathology/histopathology, or adenosine deaminase (ADA). A diagram was used to define assumed relationships between observed tests and underlying latent variables in the Bayesian LCA with input from an inter-disciplinary team. We compared the results to estimates obtained from a sequence of CRSs defined by increasing numbers of positive reference tests necessary for positive disease status.ResultsData were available from 298, 388, and 230 individuals with presumptive TB lymphadenitis, meningitis, and pleuritis, respectively. Using Bayesian LCA, estimates were obtained for accuracy of all tests and for extrapulmonary TB prevalence. Xpert sensitivity neared that of culture for TB lymphadenitis and meningitis but was lower for TB pleuritis, and specificities of all microbiological tests approached 100%. Non-microbiological tests' sensitivities were high, but specificities were only moderate, preventing disease rule-in. CRSs' only provided estimates of Xpert and these varied widely per CRS definition. Accuracy of the CRSs also varied by definition, and no CRS was 100% accurate.ConclusionUnlike CRSs, Bayesian LCA takes into account known information about test performance resulting in accuracy estimates that are easier to interpret. LCA should receive greater consideration for evaluating extrapulmonary TB diagnostic tests.

Project description:BackgroundCulture remains the diagnostic gold standard for many bacterial infections, and the method against which other tests are often evaluated. Specificity of culture is 100% if the pathogenic organism is not found in healthy subjects, but the sensitivity of culture is more difficult to determine and may be low. Here, we apply Bayesian latent class models (LCMs) to data from patients with a single Gram-negative bacterial infection and define the true sensitivity of culture together with the impact of misclassification by culture on the reported accuracy of alternative diagnostic tests.Methods/principal findingsData from published studies describing the application of five diagnostic tests (culture and four serological tests) to a patient cohort with suspected melioidosis were re-analysed using several Bayesian LCMs. Sensitivities, specificities, and positive and negative predictive values (PPVs and NPVs) were calculated. Of 320 patients with suspected melioidosis, 119 (37%) had culture confirmed melioidosis. Using the final model (Bayesian LCM with conditional dependence between serological tests), the sensitivity of culture was estimated to be 60.2%. Prediction accuracy of the final model was assessed using a classification tool to grade patients according to the likelihood of melioidosis, which indicated that an estimated disease prevalence of 61.6% was credible. Estimates of sensitivities, specificities, PPVs and NPVs of four serological tests were significantly different from previously published values in which culture was used as the gold standard.Conclusions/significanceCulture has low sensitivity and low NPV for the diagnosis of melioidosis and is an imperfect gold standard against which to evaluate alternative tests. Models should be used to support the evaluation of diagnostic tests with an imperfect gold standard. It is likely that the poor sensitivity/specificity of culture is not specific for melioidosis, but rather a generic problem for many bacterial and fungal infections.

Project description:AimsThe main focus of this study is to illustrate the importance of the statistical analysis in the evaluation of the accuracy of malaria diagnostic tests, without admitting a reference test, exploring a dataset (n=3317) collected in São Tomé and Príncipe.MethodsBayesian Latent Class Models (without and with constraints) are used to estimate the malaria infection prevalence, together with sensitivities, specificities, and predictive values of three diagnostic tests (RDT, Microscopy and PCR), in four subpopulations simultaneously based on a stratified analysis by age groups (< 5, ? 5 years old) and fever status (febrile, afebrile).ResultsIn the afebrile individuals with at least five years old, the posterior mean of the malaria infection prevalence is 3.2% with a highest posterior density interval of [2.3-4.1]. The other three subpopulations (febrile ? 5 years, afebrile or febrile children less than 5 years) present a higher prevalence around 10.3% [8.8-11.7]. In afebrile children under-five years old, the sensitivity of microscopy is 50.5% [37.7-63.2]. In children under-five, the estimated sensitivities/specificities of RDT are 95.4% [90.3-99.5]/93.8% [91.6-96.0]--afebrile--and 94.1% [87.5-99.4]/97.5% [95.5-99.3]--febrile. In individuals with at least five years old are 96.0% [91.5-99.7]/98.7% [98.1-99.2]--afebrile--and 97.9% [95.3-99.8]/97.7% [96.6-98.6]--febrile. The PCR yields the most reliable results in four subpopulations.ConclusionsThe utility of this RDT in the field seems to be relevant. However, in all subpopulations, data provide enough evidence to suggest caution with the positive predictive values of the RDT. Microscopy has poor sensitivity compared to the other tests, particularly, in the afebrile children less than 5 years. This type of findings reveals the danger of statistical analysis based on microscopy as a reference test. Bayesian Latent Class Models provide a powerful tool to evaluate malaria diagnostic tests, taking into account different groups of interest.

Dataset Information

Estimating the true accuracy of diagnostic tests for dengue infection using bayesian latent class models.

Background

Methods/principal findings

Conclusions

Publications

Estimating the true accuracy of diagnostic tests for dengue infection using bayesian latent class models.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets