A statistical method was used for the meta-analysis of tests for latent TB in the absence of a gold standard, combining random-effect and latent-class methods to estimate test accuracy

Meta-analysis of dichotomous and ordinal tests with an imperfect gold standard

Standard methods for the meta-analysis of medical tests, without assuming a gold standard, are limited to dichotomous data. Multivariate probit models are used to analyse correlated dichotomous data, and can be extended to model ordinal data. Within the context of an imperfect gold standard, they have previously been used for the analysis of dichotomous and ordinal test data from a single study, and for the meta-analysis of dichotomous tests. However, they have not previously been used for the meta-analysis of ordinal tests. In this article, we developed a Bayesian multivariate probit latent class model for the simultaneous meta-analysis of ordinal and dichotomous tests without assuming a gold standard, which also allows one to obtain summary estimates of joint test accuracy. We fitted the models using the software Stan, which uses a state-of-the-art Hamiltonian Monte Carlo algorithm, and we applied the models to a dataset in which studies evaluated the accuracy of tests, and test combinations, for deep vein thrombosis. We demonstrate the issues with dichotomising ordinal test accuracy data in the presence of an imperfect gold standard, before applying and comparing several variations of our proposed model which do not require the data to be dichotomised. The models proposed will allow researchers to more appropriately meta-analyse ordinal and dichotomous tests without a gold standard, potentially leading to less biased estimates of test accuracy. This may lead to a better understanding of which tests, and test combinations, should be used for any given medical condition.

Copas-like selection model to correct publication bias in systematic review of diagnostic test studies

The accuracy of a diagnostic test, which is often quantified by a pair of measures such as sensitivity and specificity, is critical for medical decision making. Separate studies of an investigational diagnostic test can be combined through meta-analysis; however, such an analysis can be threatened by publication bias. To the best of our knowledge, there is no existing method that accounts for publication bias in the meta-analysis of diagnostic tests involving bivariate outcomes. In this paper, we extend the Copas selection model from univariate outcomes to bivariate outcomes for the correction of publication bias when the probability of a study being published can depend on its sensitivity, specificity, and the associated standard errors. We develop an expectation-maximization algorithm for the maximum likelihood estimation under the proposed selection model. We investigate the finite sample performance of the proposed method through simulation studies and illustrate the method by assessing a meta-analysis of 17 published studies of a rapid diagnostic test for influenza.

A Bayesian Hierarchical Summary Receiver Operating Characteristic Model for Network Meta-analysis of Diagnostic Tests

In studies evaluating the accuracy of diagnostic tests, three designs are commonly used, crossover, randomized, and non-comparative. Existing methods for meta-analysis of diagnostic tests mainly consider the simple cases in which the reference test in all or none of the studies can be considered a gold standard test, and in which all studies use either a randomized or non-comparative design. The proliferation of diagnostic instruments and the diversity of study designs create a need for more general methods to combine studies that include or do not include a gold standard test and that use various designs. This paper extends the Bayesian hierarchical summary receiver operating characteristic model to network meta-analysis of diagnostic tests to simultaneously compare multiple tests within a missing data framework. The method accounts for correlations between multiple tests and for heterogeneity between studies. It also allows different studies to include different subsets of diagnostic tests and provides flexibility in the choice of summary statistics. The model is evaluated using simulations and illustrated using real data on tests for deep vein thrombosis, with sensitivity analyses.

Tuberculin skin test and QuantiFERON-Gold In Tube assay for diagnosis of latent TB infection among household contacts of pulmonary TB patients in high TB burden setting

BACKGROUND

World Health Organization (WHO) recommends systematic screening of high-risk populations, including household contacts (HHCs) of adult pulmonary tuberculosis (TB) patients, as a key strategy for elimination of TB. QuantiFERON-TB Gold In-Tube (QFT-GIT) assay and tuberculin skin test (TST) are two commonly used tools for the detection of latent tuberculosis infection (LTBI) but may yield differential results, affecting eligibility for TB preventive therapy.

MATERIALS AND METHODS

A prospective cohort study of adult pulmonary TB patients and their HHCs were recruited in 2 cities of India, Pune and Chennai. HHCs underwent QFT-GIT (QIAGEN) and TST (PPD SPAN 2TU/5TU). A positive QFT-GIT was defined as value ≥0.35 IU/ml and a positive TST as an induration of ≥5 mm. A secondary outcome of TST induration ≥10mm was explored. Proportion positive by either or both assays, discordant positives and negatives were calculated; test concordance was assessed using percentage agreement and kappa statistics; and risk factors for concordance and discordance including age categories were assessed using logistic regression. Sensitivity and specificity was estimated by latent class model.

RESULTS

Of 1048 HHCs enrolled, 869 [median (IQR) age: 27 years (15-40)] had both TST and QFT-GIT results available and prevalence of LTBI by QFT-GIT was 54% [95% CI (51, 57)], by TST was 55% [95% CI (52, 58)], by either test was 74% [95% CI (71, 77) and by both tests was 35% [95% CI (31, 38)]. Discordance of TST+/QFT-GIT- was 21% while TST-/QFT-GIT+ was 26%. Poor to fair agreement occurred with TST 5mm or 10mm cutoff (60 and 61% agreement with kappa value of 0.20 and 0.25 respectively). Test agreement varied by age, TST strength and induration cut-off. In multivariate analysis, span PPD was a risk factor for QFT-GIT+ and TST- while absence of BCG scar was for TST+ and QFT-GIT-. Being employed and exposure to TB case outside the household case were associated with positivity by both the tests. Sensitivity of TST and QFT-GIT to diagnose LTBI was 77% and 69%. Probability of having LTBI was >90% when both tests were positive irrespective of exposure gradient.

CONCLUSION

Prevalence of LTBI among HHCs of adult pulmonary TB patients in India is very high and varies by test type, age, and exposure gradient. In our high TB burden setting, a strategy to treat all HHCs or a targeted strategy whereby an exposure index is used should be assessed in future preventive therapy and vaccine studies as HHCs have several factors that place them at high risk for progression to TB disease.

Classifying Injuries in Young Children as Abusive or Accidental: Reliability and Accuracy of an Expert Panel Approach

OBJECTIVE

To assess interrater reliability and accuracy of an expert panel in classifying injuries of patients as abusive or accidental based on comprehensive case information.

STUDY DESIGN

Data came from a prospective, observational, multicenter study investigating bruising characteristics of children younger than 4 years. We enrolled 2166 patients with broad ranges of illnesses and injuries presenting to one of 5 pediatric emergency departments in whom bruises were identified during examination. We collected comprehensive data regarding current and past injuries and illnesses, and provided deidentified, standardized case information to a 9-member multidisciplinary panel of experts with extensive experience in pediatric injury. Each panelist classified cases using a 5-level ordinal scale ranging from definite abuse to definite accident. Panelists also assessed whether report to child protective services (CPS) was warranted. We calculated reliability coefficients for likelihood of abuse and decision to report to CPS.

RESULTS

The interrater reliability of the panelists was high. The Kendall coefficient (95% CI) for the likelihood of abuse was 0.89 (0.87, 0.91) and the kappa coefficient for the decision to report to CPS was 0.91 (0.87, 0.94). Reliability of pairs and subgroups of panelists were similarly high. A panel composite classification was nearly perfectly accurate in a subset of cases having definitive, corroborated injury status.

CONCLUSIONS

A panel of experts with different backgrounds but common expertise in pediatric injury is a reliable and accurate criterion standard for classifying pediatric injuries as abusive or accidental in a sample of children presenting to a pediatric emergency department.

Interferon-gamma release assay for the diagnosis of latent tuberculosis infection: A latent-class analysis

BACKGROUND

Accurate diagnosis and subsequent treatment of latent tuberculosis infection (LTBI) is essential for TB elimination. However, the absence of a gold standard test for diagnosing LTBI makes assessment of the true prevalence of LTBI and the accuracy of diagnostic tests challenging. Bayesian latent class models can be used to make inferences about disease prevalence and the sensitivity and specificity of diagnostic tests using data on the concordance between tests. We performed the largest meta-analysis to date aiming to evaluate the performance of tuberculin skin test (TST) and interferon-gamma release assays (IGRAs) for LTBI diagnosis in various patient populations using Bayesian latent class modelling.

METHODS

Systematic search of PubMeb, Embase and African Index Medicus was conducted without date and language restrictions on September 11, 2017 to identify studies that compared the performance of TST and IGRAs for LTBI diagnosis. Two IGRA methods were considered: QuantiFERON-TB Gold In Tube (QFT-GIT) and T-SPOT.TB. Studies were included if they reported 2x2 agreement data between TST and QFT-GIT or T-SPOT.TB. A Bayesian latent class model was developed to estimate the sensitivity and specificity of TST and IGRAs in various populations, including immune-competent adults, immune-compromised adults and children. A TST cut-off value of 10 mm was used for immune-competent subjects and 5 mm for immune-compromised individuals.

FINDINGS

A total of 157 studies were included in the analysis. In immune-competent adults, the sensitivity of TST and QFT-GIT were estimated to be 84% (95% credible interval [CrI] 82-85%) and 52% (50-53%), respectively. The specificity of QFT-GIT was 97% (96-97%) in non-BCG-vaccinated and 93% (92-94%) in BCG-vaccinated immune-competent adults. The estimated figures for TST were 100% (99-100%) and 79% (76-82%), respectively. T-SPOT.TB has comparable specificity (97% for both tests) and better sensitivity (68% versus 52%) than QFT-GIT in immune-competent adults. In immune-compromised adults, both TST and QFT-GIT display low sensitivity but high specificity. QFT-GIT and TST are equally specific (98% for both tests) in non-BCG-vaccinated children; however, QFT-GIT is more specific than TST (98% versus 82%) in BCG-vaccinated group. TST is more sensitive than QFT-GIT (82% versus 73%) in children.

CONCLUSIONS

This study is the first to assess the utility of TST and IGRAs for LTBI diagnosis in different population groups using all available data with Bayesian latent class modelling. Our results challenge the current beliefs about the performance of LTBI screening tests, and have important implications for LTBI screening policy and practice. We estimated that the performance of IGRAs is not as reliable as previously measured in the general population. However, IGRAs are not or minimally affected by BCG and should be the preferred tests in this setting. Adoption of IGRAs in settings where BCG is widely administered will allow for a more accurate identification and treatment of LTBI.

Evaluating the accuracy and economic value of a new test in the absence of a perfect reference test

BACKGROUND

Streptococcus pneumoniae (SP) pneumonia is often treated empirically as diagnosis is challenging because of the lack of a perfect test. Using BinaxNOW-SP, a urinary antigen test, as an add-on to standard cultures may not only increase diagnostic yield but also increase costs.

OBJECTIVE

To estimate the sensitivity and specificity of BinaxNOW-SP and subsequently estimate the cost-effectiveness of adding BinaxNOW-SP to the diagnostic work-up.

DESIGN

We fit a Bayesian latent-class meta-analysis model to obtain estimates of BinaxNOW-SP accuracy that adjust for the imperfect accuracy of culture. Meta-analysis results were combined with information on prevalence of SP pneumonia to estimate the number of patients who are correctly classified under competing diagnostic strategies. Taking into consideration the cost of antibiotics, we determined the incremental cost of adding BinaxNOW-SP to the work-up per case correctly diagnosed.

RESULTS

The BinaxNOW-SP test had a pooled sensitivity of 0.74 (95% credible interval [CrI], 0.67-0.83) and a pooled specificity of 0.96 (95% CrI, 0.92-0.99). An overall increase in diagnostic accuracy of 6.2% due to the addition of BinaxNOW-SP corresponded to an incremental cost per case correctly classified of $582 Canadian dollars.

CONCLUSIONS

The methods we have described allow us to evaluate the accuracy and economic value of a new test in the absence of a perfect reference test using an evidence-based approach.

Summary diagnostic validity of commonly used maternal major depression disorder case finding instruments in the United States: A meta-analysis

INTRODUCTION

Major Depression Disorder (MDD) is common among mothers of young children. However, its detection remains low in primary-care and community-based settings in part due to the uncertainty regarding the validity of existing case-finding instruments. We conducted meta-analyses to estimate the diagnostic validity of commonly used maternal MDD case finding instruments in the United States.

METHODS

We systematically searched three electronic bibliographic databases PubMed, PsycINFO, and EMBASE from 1994 to 2015 to identify relevant published literature. Study eligibility and quality were evaluated using the Standards for the Reporting of Diagnostic Accuracy studies and Quality Assessment of Diagnostic Accuracy Studies guidelines, respectively. Pooled sensitivity and specificity of case-finding instruments were generated using Bayesian hierarchical summary receiver operating models.

RESULTS

Overall, 1130 articles were retrieved and 74 articles were selected for full-text review. Twelve articles examining six maternal MDD case-finding instruments met the eligibility criteria and were included in our meta-analyses. Pooled sensitivity and specificity estimates were highest for the BDI-II (91%; 95% Bayesian Credible Interval (BCI): 68%; 99% and 89%; 95% BCI: 62%; 98% respectively) and EPDS10 (74%; 95% BCI: 46%; 91% and 97%; 95% BCI: 84%; 99% respectively) during the antepartum and postpartum periods respectively.

LIMITATION

No meta-regression was conducted to examine the impact of study-level characteristics on the results.

DISCUSSION

Diagnostic performance varied among instruments and between peripartum periods. These findings suggest the need for a judicious selection of maternal MDD case-finding instruments depending on the study population and target periods of assessment.

Comparative diagnostic accuracy of ganglion cell-inner plexiform and retinal nerve fiber layer thickness measures by Cirrus and Spectralis optical coherence tomography in relapsing-remitting multiple sclerosis

Objective. To estimate sensitivity and specificity of several optical coherence tomography (OCT) measurements for detecting retinal thickness changes in patients with relapsing-remitting multiple sclerosis (RRMS), such as macular ganglion cell-inner plexiform layer (GCIPL) thickness measured with Cirrus (OCT) and peripapillary retinal nerve fiber layer (pRNFL) thickness measured with Cirrus and Spectralis OCT. Methods. Seventy patients (140 eyes) with RRMS and seventy matched healthy subjects underwent pRNFL and GCIPL thickness analysis using Cirrus OCT and pRNFL using Spectralis OCT. A prospective, cross-sectional evaluation of sensitivities and specificities was performed using latent class analysis due to the absence of a gold standard. Results. GCIPL measures had higher sensitivity and specificity than temporal pRNFL measures obtained with both OCT devices. Average GCIPL thickness was significantly more sensitive than temporal pRNFL by Cirrus (96.34% versus 58.41%) and minimum GCIPL thickness was significantly more sensitive than temporal pRNFL by Spectralis (96.41% versus 69.69%). Generalised estimating equation analysis revealed that age (P = 0.030), optic neuritis antecedent (P = 0.001), and disease duration (P = 0.002) were significantly associated with abnormal results in average GCIPL thickness. Conclusion. Average and minimum GCIPL measurements had significantly better sensitivity to detect retinal thickness changes in RRMS than temporal pRNFL thickness measured by Cirrus and Spectralis OCT, respectively.

Estimation of diagnostic test accuracy without full verification: a review of latent class methods

The performance of a diagnostic test is best evaluated against a reference test that is without error. For many diseases, this is not possible, and an imperfect reference test must be used. However, diagnostic accuracy estimates may be biased if inaccurately verified status is used as the truth. Statistical models have been developed to handle this situation by treating disease as a latent variable. In this paper, we conduct a systematized review of statistical methods using latent class models for estimating test accuracy and disease prevalence in the absence of complete verification.

Prospective head-to-head study comparing 2 commercial interferon gamma release assays for the diagnosis of tuberculous uveitis

PURPOSE

To perform a head-to-head comparison of 2 commercially available interferon-gamma release assays, QuantiFERON-TB Gold In-Tube (Cellestis, Chadstone, Victoria, Australia) and T-SPOT.TB (Oxford Immunotech, Abingdon, Oxfordshire, UK), in the diagnosis of tuberculous uveitis.

DESIGN

Prospective cohort to study diagnostic accuracy.

METHODS

We recruited consecutive new patients who presented with uveitis to a tertiary institution over a 2-year period. All patients underwent complete ocular examination and systemic evaluation, including T-SPOT.TB, QuantiFERON-TB Gold In-Tube, and tuberculin skin test. Patients were followed-up for a minimum of 1 year after completion of antituberculous therapy where indicated. The main outcome measures were the sensitivity, specificity and accuracy of each test, estimated using Bayesian latent class analysis (presented with 95% Bayesian credible intervals) (Crl). Prior information was obtained from published meta-analyses for diagnostic tests: QuantiFERON Gold In-Tube sensitivity (0.64, 0.59-0.69) and specificity (0.99, 0.99-1.00); T-SPOT. tuberculosis sensitivity (0.50, 0.33-0.67) and specificity (0.91, 0.88-0.93).

RESULTS

From our study in patients with uveitis, QuantiFERON-TB Gold In-Tube was more specific but slightly less sensitive (sensitivity: 0.64, 0.60-0.69; specificity: 0.995, 0.988-0.999) than T-SPOT.TB (sensitivity: 0.67, 0.60-0.74; specificity: 0.91, 0.88-0.93). However, QuantiFERON Gold In-Tube was significantly more accurate in identifying true-positive tuberculous uveitis cases than was T-SPOT.TB among discordant cases (QuantiFERON Gold In-Tube positive 98% vs T-SPOT.TB positive 76%; ratio 1.28, 95% Crl: 1.11-1.72, ie, 95% Crl >1.0, statistically significant).

CONCLUSION

Based on statistical decision theory, our head-to-head study suggests that QuantiFERON-TB Gold In-Tube is the first-line test that should be performed in preference to T-SPOT.TB (and the tuberculin skin test) for diagnosing tuberculous uveitis.

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.

Using a web-based application to define the accuracy of diagnostic tests when the gold standard is imperfect

BACKGROUND

Estimates of the sensitivity and specificity for new diagnostic tests based on evaluation against a known gold standard are imprecise when the accuracy of the gold standard is imperfect. Bayesian latent class models (LCMs) can be helpful under these circumstances, but the necessary analysis requires expertise in computational programming. Here, we describe open-access web-based applications that allow non-experts to apply Bayesian LCMs to their own data sets via a user-friendly interface.

METHODS/PRINCIPAL FINDINGS

Applications for Bayesian LCMs were constructed on a web server using R and WinBUGS programs. The models provided (http://mice.tropmedres.ac) include two Bayesian LCMs: the two-tests in two-population model (Hui and Walter model) and the three-tests in one-population model (Walter and Irwig model). Both models are available with simplified and advanced interfaces. In the former, all settings for Bayesian statistics are fixed as defaults. Users input their data set into a table provided on the webpage. Disease prevalence and accuracy of diagnostic tests are then estimated using the Bayesian LCM, and provided on the web page within a few minutes. With the advanced interfaces, experienced researchers can modify all settings in the models as needed. These settings include correlation among diagnostic test results and prior distributions for all unknown parameters. The web pages provide worked examples with both models using the original data sets presented by Hui and Walter in 1980, and by Walter and Irwig in 1988. We also illustrate the utility of the advanced interface using the Walter and Irwig model on a data set from a recent melioidosis study. The results obtained from the web-based applications were comparable to those published previously.

CONCLUSIONS

The newly developed web-based applications are open-access and provide an important new resource for researchers worldwide to evaluate new diagnostic tests.

Statistical methods for multivariate meta-analysis of diagnostic tests: An overview and tutorial

In this article, we present an overview and tutorial of statistical methods for meta-analysis of diagnostic tests under two scenarios: (1) when the reference test can be considered a gold standard and (2) when the reference test cannot be considered a gold standard. In the first scenario, we first review the conventional summary receiver operating characteristics approach and a bivariate approach using linear mixed models. Both approaches require direct calculations of study-specific sensitivities and specificities. We next discuss the hierarchical summary receiver operating characteristics curve approach for jointly modeling positivity criteria and accuracy parameters, and the bivariate generalized linear mixed models for jointly modeling sensitivities and specificities. We further discuss the trivariate generalized linear mixed models for jointly modeling prevalence, sensitivities and specificities, which allows us to assess the correlations among the three parameters. These approaches are based on the exact binomial distribution and thus do not require an ad hoc continuity correction. Lastly, we discuss a latent class random effects model for meta-analysis of diagnostic tests when the reference test itself is imperfect for the second scenario. A number of case studies with detailed annotated SAS code in MIXED and NLMIXED procedures are presented to facilitate the implementation of these approaches.

Cost-effectiveness of latent tuberculosis screening before steroid therapy for idiopathic nephrotic syndrome in children

BACKGROUND

Guidelines differ on screening recommendations for latent tuberculosis infection (LTBI) prior to immunosuppressive therapy. We aimed to determine the most cost-effective LTBI screening strategy before long-term steroid therapy in a child with new-onset idiopathic nephrotic syndrome.

STUDY DESIGN

Markov state-transition model.

SETTING & POPULATION

5-year-old boy with new-onset idiopathic nephrotic syndrome.

MODEL, PERSPECTIVE, & TIMEFRAME

The Markov model took a societal perspective over a lifetime horizon.

INTERVENTION

3 strategies were compared: universal tuberculin skin testing (TST), targeted screening using a risk-factor questionnaire, and no screening. A secondary model included the newer interferon γ release assays (IGRAs), requiring only one visit and having greater specificity than TST.

OUTCOMES

Marginal cost-effectiveness ratios (2010 US dollars) with effectiveness measured as quality-adjusted life-years (QALYs).

RESULTS

At an LTBI prevalence of 1.1% (the average US childhood prevalence in our base case), a no-screening strategy dominated ($2,201; 29.3356 QALYs) targeted screening ($2,218; 29.3356 QALYs) and universal TST ($2,481; 29.3347 QALYs). At a prevalence >10.3%, targeted screening with a risk-factor questionnaire was the most cost-effective option. Higher than a prevalence of 58.5%, universal TST was preferred. In the secondary model, targeted screening with a questionnaire followed by IGRA testing was cost-effective compared with no screening in the base case when the LTBI prevalence was >4.9%.

LIMITATIONS

There is no established gold standard for the diagnosis of LTBI. Results of any modeling task are limited by the accuracy of available data.

CONCLUSIONS

Prior to starting steroid therapy, only patients in areas with a high prevalence of LTBI will benefit from universal TST. As more evidence becomes available about the use of IGRA testing in children, the assay may become a component of cost-effective screening protocols in populations with a higher burden of LTBI.

Performance of interferon-gamma release assays in patients with inflammatory bowel disease: a systematic review and meta-analysis

BACKGROUND

Guidelines mandate screening for latent tuberculosis infection (LTBI) prior to anti-tumor necrosis factor (anti-TNF) therapy in patients with inflammatory bowel disease (IBD). However, many are already on immunosuppressive therapy (IST) that may affect the precision of the Tuberculin skin test (TST). Our aim was to assess the performance of the new interferon-gamma release assays (IGRAs) to detect LTBI in patients with IBD.

METHODS

MEDLINE and EMBASE were searched (up to June 2011) to identify studies evaluating the performance of IGRAs (QuantiFERON-TB Gold [QFT-2G], QuantiFERON-TB Gold In-Tube [QFT-3G] and T-SPOT.TB) in individuals with IBD. Forest plots and pooled estimates using random effects models were created where applicable.

RESULTS

Nine unique studies encompassing 1309 patients with IBD were included for analysis. The pooled concordance between the TST and QFT-2G/QFT-3G was 85% (95% confidence interval [CI] 77%-90%). The concordance of the TST and TSPOT.TB was 72% (95% CI 64%-78%). Studies assessing agreement reported more IGRA-/TST+ results versus IGRA+/TST- results. The pooled percentage of indeterminate results was 5% (95% CI 2%-9%) for QFT-2G/QFT-3G. TSPOT.TB showed similar results. Both positive QFT-2G/QFT-3G results (pooled odds ratio [OR] 0.37, 95% CI 0.16-0.87) and positive TST results (pooled OR 0.28, 95% CI 0.10-0.80) were significantly influenced by IST (both P = 0.02).

CONCLUSIONS

While it remains difficult to determine superiority between the IGRAs and the TST, both are negatively affected by IST. Therefore, screening prior to initiation of IST should be considered. Nevertheless, it is imperative that all patients receive screening prior to anti-TNF therapy.

Bayesian meta-analysis of the accuracy of a test for tuberculous pleuritis in the absence of a gold standard reference

Absence of a perfect reference test is an acknowledged source of bias in diagnostic studies. In the case of tuberculous pleuritis, standard reference tests such as smear microscopy, culture and biopsy have poor sensitivity. Yet meta-analyses of new tests for this disease have always assumed the reference standard is perfect, leading to biased estimates of the new test's accuracy. We describe a method for joint meta-analysis of sensitivity and specificity of the diagnostic test under evaluation, while considering the imperfect nature of the reference standard. We use a Bayesian hierarchical model that takes into account within- and between-study variability. We show how to obtain pooled estimates of sensitivity and specificity, and how to plot a hierarchical summary receiver operating characteristic curve. We describe extensions of the model to situations where multiple reference tests are used, and where index and reference tests are conditionally dependent. The performance of the model is evaluated using simulations and illustrated using data from a meta-analysis of nucleic acid amplification tests (NAATs) for tuberculous pleuritis. The estimate of NAAT specificity was higher and the sensitivity lower compared to a model that assumed that the reference test was perfect.

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.

Interferon-gamma release assay as a diagnostic test for tuberculosis-associated uveitis

BACKGROUND

To study the use of interferon-gamma release assay (IFN-γ) (IGRAs) as a diagnostic test for tuberculosis (TB)-associated uveitis (TAU).

DESIGN

Prospective cohort study.

PARTICIPANTS

Consecutive new patients (n=162) with clinical ocular signs suggestive of TAU, seen >1 year period at a single tertiary center.

METHODS

All subjects underwent investigations to rule out underlying disease, including T-SPOT.TB and tuberculin skin test (TST). Twenty-one subjects with underlying disease and three with interdeterminate T-SPOT.TB results were excluded. Those with T-SPOT.TB- or TST-positive results were referred to infectious diseases physician for evaluation. Anti-TB therapy (ATT) was prescribed if required. Patients' treatment response and recurrence were monitored for six months after completion of ATT, if given; or 1 year if no ATT was given.

MAIN OUTCOME MEASURE

Diagnosis of TAU.

RESULTS

Mean age of study cohort (n=138) was 46.8 ± 15.3 years. Majority were Chinese (n=80, 58.0%) and female (n=75, 54.3%). TST was more sensitive than T-SPOT.TB (72.0% vs 36.0%); but T-SPOT.TB was more specific (75.0% vs 51.1%) for diagnosing TAU. Patients with either a T-SPOT.TB (1.44; 95% confidence intervals (CI), 0.86-2.42) or TST (1.47; 95% CI, 1.12-1.94)-positive result are more likely to have TAU. The accuracy of diagnosing TAU increases when both tests are used in combination (area under the receiver operator curve=0.665; 95% CI, 0.533-0.795). Patients with both tests positive are 2.16 (95% CI, 1.23-3.80) times more likely to have TAU. Negative T-SPOT.TB or TST results do not exclude TAU (negative likelihood ratios <1.0).

CONCLUSIONS

We recommend using a combination of clinical signs, IGRA, and TST to diagnose TAU.

Should interferon gamma release assays become the standard method for screening patients for Mycobacterium tuberculosis infections in the United States?

The Centers for Disease Control and Prevention recently published updated guidelines for the use of interferon gamma release assays (IGRAs) to detect Mycobacterium tuberculosis. This document gives a balanced analysis of the strengths and weaknesses of IGRAs. To date, these assays have not been widely adopted in the United States by clinical laboratories. We have asked two experts, Thomas Alexander of Summa Health Care, who has adopted an IGRA for M. tuberculosis detection in his laboratory, and Melissa Miller of UNC Hospitals, who has evaluated one but has not chosen to adopt it, to explain how each reached this decision based on their experience with the test and the data that have been published concerning IGRA.