Latent Class Models in Diagnostic Studies When There is No Reference Standard--A Systematic Review

Smeden, Maarten van; Naaktgeboren, Christiana A.; Reitsma, Johannes B.; Moons, Karel G.M.; Groot, Joris A H de

doi:10.1093/aje/kwt286

Cited by 172 publications

(150 citation statements)

References 38 publications

Supporting

Mentioning

145

Contrasting

Unclassified

Order By: Relevance

“…Further studies are needed to estimate test characteristics, either defining subsets of true positive and true negative samples or applying latent class analysis [65,66].…”

Section: Discussionmentioning

confidence: 99%

Performance of Hepatitis E Virus (HEV)-antibody tests: a comparative analysis based on samples from individuals with direct contact to domestic pigs or wild boar in Germany

Sommerkorn

Schauer

Schreiner

et al. 2017

Med Microbiol Immunol

View full text Add to dashboard Cite

In Europe, Hepatitis E virus (HEV) genotype 3 causes most human infections, and domestic pigs and wild boar represent the main reservoirs. Contact to these animals represents one source of human infection. However, interpretation of studies is challenged in the absence of a serological gold standard. Hence, this study compared results of different HEV immunoassays. Plasma samples from 139 individuals who had professional contact to pigs (veterinarians, meat inspectors, slaughterhouse workers; n = 114) or hunted regularly (n = 25) were tested with assays specific for HEV IgG, HEV IgM, HEV IgA, and total HEV immunoglobulin as well as for viral RNA. Furthermore, overall HEV IgG was defined (i.e., two of three IgG assays reveal the same result) to compare serological findings. Borderline results were always quoted as positive. For IgG, apparent prevalence was higher in Wantai (48.2%) compared to Euroimmun (11.5%) and Mikrogen assays (17.3%) (p = 0.0001). The overall IgG prevalence was estimated to be 18.7%. For total Ig, Wantai (40.3%) also yielded higher prevalence than Euroimmun (15.8%) (p = 0.0001). The HEV IgM prevalence ranged from 0% (Euroimmun) to 4.3% (Mikrogen). Four percent of individuals tested IgA positive, whilst none harboured HEV RNA. Our results support previous studies that the higher IgG prevalence estimated with the Wantai assay result from a higher sensitivity of this test. However, further studies are needed to verify specificity given the challenge of defining true negative samples. The high percentage of individuals with HEV IgG observed in this study underlines that direct contact to pigs represents a risk factor for HEV infection.

show abstract

“…Further studies are needed to estimate test characteristics, either defining subsets of true positive and true negative samples or applying latent class analysis [65,66].…”

Section: Discussionmentioning

confidence: 99%

Performance of Hepatitis E Virus (HEV)-antibody tests: a comparative analysis based on samples from individuals with direct contact to domestic pigs or wild boar in Germany

Sommerkorn

Schauer

Schreiner

et al. 2017

Med Microbiol Immunol

View full text Add to dashboard Cite

show abstract

“…We used Bayesian latent class analysis to simultaneously estimate the accuracy of the 5 tests with regard to the detection of CPTB, the prevalence of CPTB, and the degree of under-and overtreatment in the cohort. Latent class analysis has successfully been used in other studies of the accuracy of diagnostic tests in the absence of a gold standard (12,(17)(18)(19). However, here we present one of the first applications of latent class analysis to prospectively collected data on CPTB.…”

mentioning

confidence: 99%

Diagnostic Test Accuracy in Childhood Pulmonary Tuberculosis: A Bayesian Latent Class Analysis

Schumacher¹,

Smeden²,

Dendukuri³

et al. 2016

American Journal of Epidemiology

View full text Add to dashboard Cite

Evaluation of tests for the diagnosis of childhood pulmonary tuberculosis (CPTB) is complicated by the absence of an accurate reference test. We present a Bayesian latent class analysis in which we evaluated the accuracy of 5 diagnostic tests for CPTB. We used data from a study of 749 hospitalized South African children suspected to have CPTB from 2009 to 2014. The following tests were used: mycobacterial culture, smear microscopy, Xpert MTB/RIF (Cepheid Inc.), tuberculin skin test (TST), and chest radiography. We estimated the prevalence of CPTB to be 27% (95% credible interval (CrI): 21, 35). The sensitivities of culture, Xpert, and smear microscopy were estimated to be 60% (95% CrI: 46, 76), 49% (95% CrI: 38, 62), and 22% (95% CrI: 16, 30), respectively; specificities of these tests were estimated in accordance with prior information and were close to 100%. Chest radiography was estimated to have a sensitivity of 64% (95% CrI: 55, 73) and a specificity of 78% (95% CrI: 73, 83). Sensitivity of the TST was estimated to be 75% (95% CrI: 61, 84), and it decreased substantially among children who were malnourished and infected with human immunodeficiency virus (56%). The specificity of the TST was 69% (95% CrI: 63%, 76%). Furthermore, it was estimated that 46% (95% CrI: 42, 49) of CPTB-negative cases and 93% (95% CrI: 82; 98) of CPTB-positive cases received antituberculosis treatment, which indicates substantial overtreatment and limited undertreatment. childhood pulmonary tuberculosis; diagnosis; latent class analysis; overtreatment; sensitivity; specificity Abbreviations: CPTB, childhood pulmonary tuberculosis; CrI, credible interval; HIV, human immunodeficiency virus; PTB, pulmonary tuberculosis; TST, tuberculin skin test.Tuberculosis in children is an important global health problem. There are an estimated 0.5 to 1 million new cases each year (1, 2), with childhood pulmonary tuberculosis (CPTB) being the most common form. One of the major challenges in diagnosing CPTB is the lack of sensitive diagnostic tests (3-6). In clinical practice, the diagnosis of CPTB therefore relies on a combination of imperfect tests, which gives rise to unknown degrees of under-or overtreatment (7,8).In recent years, new tests for CPTB have been developed, and their accuracy has been evaluated using mycobacterial culture as a reference standard (4, 9). Although culture is currently considered the best available reference standard, its sensitivity for detecting CPTB is acknowledged to be imperfect (3,4,10). The culture reference standard thus inevitably leads to true CPTB case patients being misclassified as being negative for CPTB. If these misclassifications by the reference standard are ignored, then the assessment of the test accuracy can be biased (11)(12)(13)(14).To address the problem of the lack of an accurate reference standard, multivariable diagnostic algorithms for CPTB have been proposed to combine information from multiple imperfect diagnostic tests (including tests for tuberculosis infection and clinical data) in a sy...

show abstract

“…For this study on data quality [19], the 65 volunteers labelled 269 sites from a series of images and, when attributing a land cover type, volunteers also declared how confident they were (input using a slider with a label, e.g., 'sure'). The derivation of the land cover class-specific accuracies per volunteer (producer's accuracies) and of the land-cover classes estimation from posterior probabilities were possible using a Latent Class Analysis (LCA) model estimation (see [22] for a recent review on using LCA to assess the accuracies of new 'diagnostic tests' without a gold standard in the context of medical applications).…”

Section: Land Cover Validation Examplementioning

confidence: 99%

On Data Quality Assurance and Conflation Entanglement in Crowdsourcing for Environmental Studies

Leibovici

Rosser

Hodges

et al. 2017

IJGI

View full text Add to dashboard Cite

Volunteer geographical information (VGI), either in the context of citizen science or the mining of social media, has proven to be useful in various domains including natural hazards, health status, disease epidemics, and biological monitoring. Nonetheless, the variable or unknown data quality due to crowdsourcing settings are still an obstacle for fully integrating these data sources in environmental studies and potentially in policy making. The data curation process, in which a quality assurance (QA) is needed, is often driven by the direct usability of the data collected within a data conflation process or data fusion (DCDF), combining the crowdsourced data into one view, using potentially other data sources as well. Looking at current practices in VGI data quality and using two examples, namely land cover validation and inundation extent estimation, this paper discusses the close links between QA and DCDF. It aims to help in deciding whether a disentanglement can be possible, whether beneficial or not, in understanding the data curation process with respect to its methodology for future usage of crowdsourced data. Analysing situations throughout the data curation process where and when entanglement between QA and DCDF occur, the paper explores the various facets of VGI data capture, as well as data quality assessment and purposes. Far from rejecting the usability ISO quality criterion, the paper advocates for a decoupling of the QA process and the DCDF step as much as possible while still integrating them within an approach analogous to a Bayesian paradigm.

show abstract

Latent Class Models in Diagnostic Studies When There is No Reference Standard--A Systematic Review

Cited by 172 publications

References 38 publications

Performance of Hepatitis E Virus (HEV)-antibody tests: a comparative analysis based on samples from individuals with direct contact to domestic pigs or wild boar in Germany

Performance of Hepatitis E Virus (HEV)-antibody tests: a comparative analysis based on samples from individuals with direct contact to domestic pigs or wild boar in Germany

Diagnostic Test Accuracy in Childhood Pulmonary Tuberculosis: A Bayesian Latent Class Analysis

On Data Quality Assurance and Conflation Entanglement in Crowdsourcing for Environmental Studies

Contact Info

Product

Resources

About