Statistics for quantifying heterogeneity in univariate and bivariate meta-analyses of binary data: the case of meta-analyses of diagnostic accuracy

Heterogeneity in diagnostic meta-analyses is common because of the observational nature of diagnostic studies and the lack of standardization in the positivity criterion (cut-off value) for some tests. So far the unexplained heterogeneity across studies has been quantified by either using the I(2) statistic for a single parameter (i.e. either the sensitivity or the specificity) or visually examining the data in a receiver-operating characteristic space. In this paper, we derive improved I(2) statistics measuring heterogeneity for dichotomous outcomes, with a focus on diagnostic tests. We show that the currently used estimate of the 'typical' within-study variance proposed by Higgins and Thompson is not able to properly account for the variability of the within-study variance across studies for dichotomous variables. Therefore, when the between-study variance is large, the 'typical' within-study variance underestimates the expected within-study variance, and the corresponding I(2) is overestimated. We propose to use the expected value of the within-study variation in the construction of I(2) in cases of univariate and bivariate diagnostic meta-analyses. For bivariate diagnostic meta-analyses, we derive a bivariate version of I(2) that is able to account for the correlation between sensitivity and specificity. We illustrate the performance of these new estimators using simulated data as well as two real data sets.

A Majority of Low (1-10%) ER Positive Breast Cancers Behave Like Hormone Receptor Negative Tumors

BACKGROUND

The 2010 guidelines by ASCO-CAP have mandated that breast cancer specimens with ≥1% positively staining cells by immunohistochemistry should be considered Estrogen Receptor (ER) positive. This has led to a subclass of low-ER positive (1-10%) breast cancers. We have examined the biology and clinical behavior of these low ER staining tumors.

METHODS

We have developed a probabilistic score of the "ER-positivity" by quantitative estimation of ER related gene transcripts from FFPE specimens. Immunohistochemistry for ER was done on 240 surgically excised tumors of primary breast cancer. Relative transcript abundance of 3 house-keeping genes and 6 ER related genes were determined by q-RT PCR. A logistic regression model using 3 ER associated genes provided the best probability function, and a cut-off value was derived by ROC analysis. 144 high ER (>10%), 75 ER negative and 21 low-ER (1-10%) tumors were evaluated using the probability score and the disease specific survival was compared.

RESULTS

Half of the low-ER positive tumors were assigned to the ER negative group based on the probability score; in contrast 95% of ER negative and 92% of the high ER positive tumors were assigned to the appropriate ER group (p<0.0001). The survival of the low-ER group was intermediate between that of the high ER positive and ER negative groups (p<0.05).

CONCLUSION

Our results suggest that the newly lowered ASCO-CAP criteria for ER positivity, leads to the false categorization of biologically ER negative tumors as ER positive ones. This may have particular relevance to India, where we have a much higher proportion of ER negative tumors in general.

Xpert® MTB/RIF assay for pulmonary tuberculosis and rifampicin resistance in adults

BACKGROUND

Accurate, rapid detection of tuberculosis (TB) and TB drug resistance is critical for improving patient care and decreasing TB transmission. Xpert® MTB/RIF assay is an automated test that can detect both TB and rifampicin resistance, generally within two hours after starting the test, with minimal hands-on technical time. The World Health Organization (WHO) issued initial recommendations on Xpert® MTB/RIF in early 2011. A Cochrane Review on the diagnostic accuracy of Xpert® MTB/RIF for pulmonary TB and rifampicin resistance was published January 2013. We performed this updated Cochrane Review as part of a WHO process to develop updated guidelines on the use of the test.

OBJECTIVES

To assess the diagnostic accuracy of Xpert® MTB/RIF for pulmonary TB (TB detection), where Xpert® MTB/RIF was used as both an initial test replacing microscopy and an add-on test following a negative smear microscopy result.To assess the diagnostic accuracy of Xpert® MTB/RIF for rifampicin resistance detection, where Xpert® MTB/RIF was used as the initial test replacing culture-based drug susceptibility testing (DST).The populations of interest were adults presumed to have pulmonary, rifampicin-resistant or multidrug-resistant TB (MDR-TB), with or without HIV infection. The settings of interest were intermediate- and peripheral-level laboratories. The latter may be associated with primary health care facilities.

SEARCH METHODS

We searched for publications in any language up to 7 February 2013 in the following databases: Cochrane Infectious Diseases Group Specialized Register; MEDLINE; EMBASE; ISI Web of Knowledge; MEDION; LILACS; BIOSIS; and SCOPUS. We also searched the metaRegister of Controlled Trials (mRCT) and the search portal of the WHO International Clinical Trials Registry Platform to identify ongoing trials.

SELECTION CRITERIA

We included randomized controlled trials, cross-sectional studies, and cohort studies using respiratory specimens that allowed for extraction of data evaluating Xpert® MTB/RIF against the reference standard. We excluded gastric fluid specimens. The reference standard for TB was culture and for rifampicin resistance was phenotypic culture-based DST.

DATA COLLECTION AND ANALYSIS

For each study, two review authors independently extracted data using a standardized form. When possible, we extracted data for subgroups by smear and HIV status. We assessed the quality of studies using QUADAS-2 and carried out meta-analyses to estimate pooled sensitivity and specificity of Xpert® MTB/RIF separately for TB detection and rifampicin resistance detection. For TB detection, we performed the majority of analyses using a bivariate random-effects model and compared the sensitivity of Xpert® MTB/RIF and smear microscopy against culture as reference standard. For rifampicin resistance detection, we undertook univariate meta-analyses for sensitivity and specificity separately to include studies in which no rifampicin resistance was detected.

MAIN RESULTS

We included 27 unique studies (integrating nine new studies) involving 9557 participants. Sixteen studies (59%) were performed in low- or middle-income countries. For all QUADAS-2 domains, most studies were at low risk of bias and low concern regarding applicability.As an initial test replacing smear microscopy, Xpert® MTB/RIF pooled sensitivity was 89% [95% Credible Interval (CrI) 85% to 92%] and pooled specificity 99% (95% CrI 98% to 99%), (22 studies, 8998 participants: 2953 confirmed TB, 6045 non-TB).As an add-on test following a negative smear microscopy result, Xpert®MTB/RIF pooled sensitivity was 67% (95% CrI 60% to 74%) and pooled specificity 99% (95% CrI 98% to 99%; 21 studies, 6950 participants).For smear-positive, culture-positive TB, Xpert® MTB/RIF pooled sensitivity was 98% (95% CrI 97% to 99%; 21 studies, 1936 participants).For people with HIV infection, Xpert® MTB/RIF pooled sensitivity was 79% (95% CrI 70% to 86%; 7 studies, 1789 participants), and for people without HIV infection, it was 86% (95% CrI 76% to 92%; 7 studies, 1470 participants). Comparison with smear microscopy In comparison with smear microscopy, Xpert® MTB/RIF increased TB detection among culture-confirmed cases by 23% (95% CrI 15% to 32%; 21 studies, 8880 participants).For TB detection, if pooled sensitivity estimates for Xpert® MTB/RIF and smear microscopy are applied to a hypothetical cohort of 1000 patients where 10% of those with symptoms have TB, Xpert® MTB/RIF will diagnose 88 cases and miss 12 cases, whereas sputum microscopy will diagnose 65 cases and miss 35 cases. Rifampicin resistance For rifampicin resistance detection, Xpert® MTB/RIF pooled sensitivity was 95% (95% CrI 90% to 97%; 17 studies, 555 rifampicin resistance positives) and pooled specificity was 98% (95% CrI 97% to 99%; 24 studies, 2411 rifampicin resistance negatives). Among 180 specimens with nontuberculous mycobacteria (NTM), Xpert® MTB/RIF was positive in only one specimen that grew NTM (14 studies, 2626 participants).For rifampicin resistance detection, if the pooled accuracy estimates for Xpert® MTB/RIF are applied to a hypothetical cohort of 1000 individuals where 15% of those with symptoms are rifampicin resistant, Xpert® MTB/RIF would correctly identify 143 individuals as rifampicin resistant and miss eight cases, and correctly identify 833 individuals as rifampicin susceptible and misclassify 17 individuals as resistant. Where 5% of those with symptoms are rifampicin resistant, Xpert® MTB/RIF would correctly identify 48 individuals as rifampicin resistant and miss three cases and correctly identify 931 individuals as rifampicin susceptible and misclassify 19 individuals as resistant.

AUTHORS' CONCLUSIONS

In adults thought to have TB, with or without HIV infection, Xpert® MTB/RIF is sensitive and specific. Compared with smear microscopy, Xpert® MTB/RIF substantially increases TB detection among culture-confirmed cases. Xpert® MTB/RIF has higher sensitivity for TB detection in smear-positive than smear-negative patients. Nonetheless, this test may be valuable as an add-on test following smear microscopy in patients previously found to be smear-negative. For rifampicin resistance detection, Xpert® MTB/RIF provides accurate results and can allow rapid initiation of MDR-TB treatment, pending results from conventional culture and DST. The tests are expensive, so current research evaluating the use of Xpert® MTB/RIF in TB programmes in high TB burden settings will help evaluate how this investment may help start treatment promptly and improve outcomes.

Xpert® MTB/RIF assay for pulmonary tuberculosis and rifampicin resistance in adults

BACKGROUND

Accurate, rapid detection of tuberculosis (TB) and TB drug resistance is critical for improving patient care and decreasing TB transmission. Xpert® MTB/RIF assay is an automated test that can detect both TB and rifampicin resistance, generally within two hours after starting the test, with minimal hands-on technical time. The World Health Organization (WHO) issued initial recommendations on Xpert® MTB/RIF in early 2011. A Cochrane Review on the diagnostic accuracy of Xpert® MTB/RIF for pulmonary TB and rifampicin resistance was published January 2013. We performed this updated Cochrane Review as part of a WHO process to develop updated guidelines on the use of the test.

OBJECTIVES

To assess the diagnostic accuracy of Xpert® MTB/RIF for pulmonary TB (TB detection), where Xpert® MTB/RIF was used as both an initial test replacing microscopy and an add-on test following a negative smear microscopy result.To assess the diagnostic accuracy of Xpert® MTB/RIF for rifampicin resistance detection, where Xpert® MTB/RIF was used as the initial test replacing culture-based drug susceptibility testing (DST).The populations of interest were adults presumed to have pulmonary, rifampicin-resistant or multidrug-resistant TB (MDR-TB), with or without HIV infection. The settings of interest were intermediate- and peripheral-level laboratories. The latter may be associated with primary health care facilities.

SEARCH METHODS

We searched for publications in any language up to 7 February 2013 in the following databases: Cochrane Infectious Diseases Group Specialized Register; MEDLINE; EMBASE; ISI Web of Knowledge; MEDION; LILACS; BIOSIS; and SCOPUS. We also searched the metaRegister of Controlled Trials (mRCT) and the search portal of the WHO International Clinical Trials Registry Platform to identify ongoing trials.

SELECTION CRITERIA

We included randomized controlled trials, cross-sectional studies, and cohort studies using respiratory specimens that allowed for extraction of data evaluating Xpert® MTB/RIF against the reference standard. We excluded gastric fluid specimens. The reference standard for TB was culture and for rifampicin resistance was phenotypic culture-based DST.

DATA COLLECTION AND ANALYSIS

For each study, two review authors independently extracted data using a standardized form. When possible, we extracted data for subgroups by smear and HIV status. We assessed the quality of studies using QUADAS-2 and carried out meta-analyses to estimate pooled sensitivity and specificity of Xpert® MTB/RIF separately for TB detection and rifampicin resistance detection. For TB detection, we performed the majority of analyses using a bivariate random-effects model and compared the sensitivity of Xpert® MTB/RIF and smear microscopy against culture as reference standard. For rifampicin resistance detection, we undertook univariate meta-analyses for sensitivity and specificity separately to include studies in which no rifampicin resistance was detected.

MAIN RESULTS

We included 27 unique studies (integrating nine new studies) involving 9557 participants. Sixteen studies (59%) were performed in low- or middle-income countries. For all QUADAS-2 domains, most studies were at low risk of bias and low concern regarding applicability.As an initial test replacing smear microscopy, Xpert® MTB/RIF pooled sensitivity was 89% [95% Credible Interval (CrI) 85% to 92%] and pooled specificity 99% (95% CrI 98% to 99%), (22 studies, 8998 participants: 2953 confirmed TB, 6045 non-TB).As an add-on test following a negative smear microscopy result, Xpert®MTB/RIF pooled sensitivity was 67% (95% CrI 60% to 74%) and pooled specificity 99% (95% CrI 98% to 99%; 21 studies, 6950 participants).For smear-positive, culture-positive TB, Xpert® MTB/RIF pooled sensitivity was 98% (95% CrI 97% to 99%; 21 studies, 1936 participants).For people with HIV infection, Xpert® MTB/RIF pooled sensitivity was 79% (95% CrI 70% to 86%; 7 studies, 1789 participants), and for people without HIV infection, it was 86% (95% CrI 76% to 92%; 7 studies, 1470 participants). Comparison with smear microscopy In comparison with smear microscopy, Xpert® MTB/RIF increased TB detection among culture-confirmed cases by 23% (95% CrI 15% to 32%; 21 studies, 8880 participants).For TB detection, if pooled sensitivity estimates for Xpert® MTB/RIF and smear microscopy are applied to a hypothetical cohort of 1000 patients where 10% of those with symptoms have TB, Xpert® MTB/RIF will diagnose 88 cases and miss 12 cases, whereas sputum microscopy will diagnose 65 cases and miss 35 cases. Rifampicin resistance For rifampicin resistance detection, Xpert® MTB/RIF pooled sensitivity was 95% (95% CrI 90% to 97%; 17 studies, 555 rifampicin resistance positives) and pooled specificity was 98% (95% CrI 97% to 99%; 24 studies, 2411 rifampicin resistance negatives). Among 180 specimens with nontuberculous mycobacteria (NTM), Xpert® MTB/RIF was positive in only one specimen that grew NTM (14 studies, 2626 participants).For rifampicin resistance detection, if the pooled accuracy estimates for Xpert® MTB/RIF are applied to a hypothetical cohort of 1000 individuals where 15% of those with symptoms are rifampicin resistant, Xpert® MTB/RIF would correctly identify 143 individuals as rifampicin resistant and miss eight cases, and correctly identify 833 individuals as rifampicin susceptible and misclassify 17 individuals as resistant. Where 5% of those with symptoms are rifampicin resistant, Xpert® MTB/RIF would correctly identify 48 individuals as rifampicin resistant and miss three cases and correctly identify 931 individuals as rifampicin susceptible and misclassify 19 individuals as resistant.

AUTHORS' CONCLUSIONS

In adults thought to have TB, with or without HIV infection, Xpert® MTB/RIF is sensitive and specific. Compared with smear microscopy, Xpert® MTB/RIF substantially increases TB detection among culture-confirmed cases. Xpert® MTB/RIF has higher sensitivity for TB detection in smear-positive than smear-negative patients. Nonetheless, this test may be valuable as an add-on test following smear microscopy in patients previously found to be smear-negative. For rifampicin resistance detection, Xpert® MTB/RIF provides accurate results and can allow rapid initiation of MDR-TB treatment, pending results from conventional culture and DST. The tests are expensive, so current research evaluating the use of Xpert® MTB/RIF in TB programmes in high TB burden settings will help evaluate how this investment may help start treatment promptly and improve outcomes.

Measuring the mortality impact of breast cancer screening

OBJECTIVE

To i) estimate how large the mortality reductions would be if women were offered screening from age 50 until age 69; ii) to do so using the same trials and participation rates considered by the Canadian Task Force; iii) but to be guided in our analyses by the critical differences between cancer screening and therapeutics, by the time-pattern that characterizes the mortality reductions produced by a limited number of screens, and by the year-by-year mortality data in the appropriate segment of follow-up within each trial; and thereby iv) to avoid the serious underestimates that stem from including inappropriate segments of follow-up, i.e., too soon after study entry and too late after discontinuation of screening.

METHODS

We focused on yearly mortality rate ratios in the follow-up years where, based on the screening regimen employed, mortality deficits would be expected. Because the regimens differed from trial to trial, we did not aggregate the yearly data across trials. To avoid statistical extremes arising from the small numbers of yearly deaths in each trial, we calculated rate ratios for 3-year moving windows.

RESULTS

We were able to extract year-specific data from the reports of five of the trials. The data are limited for the most part by the few rounds of screening. Nevertheless, they suggest that screening from age 50 until age 69 would, at each age from 55 to 74, result in breast cancer mortality reductions much larger than the estimate of 21% that the Canadian Task Force report is based on.

DISCUSSION

By ignoring key features of cancer screening, several of the contemporary analyses have seriously underestimated the impact to be expected from such a program of breast cancer screening.

Incremental value of T-SPOT.TB for diagnosis of active pulmonary tuberculosis in children in a high-burden setting: a multivariable analysis

INTRODUCTION

Interferon γ release assays (IGRAs) are increasingly used for tuberculosis (TB) infection, but their incremental value beyond patient demographics, clinical signs and conventional tests for active disease has not been evaluated in children.

METHODS

The incremental value of T-SPOT.TB was assessed in 491 smear-negative children from two hospitals in Cape Town, South Africa. Bayesian model averaging was used to select the optimal set of patient demographics and clinical signs for predicting culture-confirmed TB. The added value of T-SPOT.TB over and above patient characteristics and conventional tests was measured using statistics such as the difference in the area under the receiver operating characteristic curve (AUC), the net reclassification improvement (NRI) and the integrated discrimination improvement (IDI).

RESULTS

Cough longer than 2 weeks, fever longer than 2 weeks, night sweats, malaise, history of household contact and HIV status were the most important predictors of culture-confirmed TB. Binary T-SPOT.TB results did not have incremental value when added to the baseline model with clinical predictors, chest radiography and the tuberculin skin test. The AUC difference was 3% (95% CI 0% to 7%). Using risk cut-offs of <10%, 10-30% and >30%, the NRI was 7% (95% CI -8% to 31%) but the CI included the null value. The IDI was 3% (95% CI 0% to 11%), meaning that the average predicted probability across all possible cut-offs improved marginally by 3%.

CONCLUSIONS

In a high-burden setting, the T-SPOT.TB did not have added value beyond clinical data and conventional tests for diagnosis of TB disease in smear-negative children.

Xpert® MTB/RIF assay for pulmonary tuberculosis and rifampicin resistance in adults

BACKGROUND

Accurate and rapid detection of tuberculosis (TB) and drug resistance are critical for improving patient care and decreasing the spread of TB. Xpert® MTB/RIF assay (Xpert) is a rapid, automated test that can detect both TB and rifampicin resistance, within two hours after starting the test, with minimal hands-on technical time, but is more expensive than conventional sputum microscopy.

OBJECTIVES

To assess the diagnostic accuracy of Xpert for pulmonary TB (TB detection), both where Xpert was used as an initial test replacing microscopy, and where Xpert was used as an add-on test following a negative smear microscopy result.To assess the diagnostic accuracy of Xpert for rifampicin resistance detection where Xpert was used as the initial test, replacing conventional culture-based drug susceptibility testing.The population of interest was adults suspected of having pulmonary TB or multidrug-resistant TB (MDR-TB), with or without HIV infection.

SEARCH METHODS

We performed a comprehensive search of the following databases: Cochrane Infectious Diseases Group Specialized Register; MEDLINE; EMBASE; ISI Web of Knowledge; MEDION; LILACS; BIOSIS; and SCOPUS. We also searched the metaRegister of Controlled Trials (mRCT) and the search portal of the WHO International Clinical Trials Registry Platform to identify ongoing trials. We performed searches on 25 September 2011 and we repeated them on 15 December 2011, without language restriction.

SELECTION CRITERIA

We included randomized controlled trials, cross-sectional, and cohort studies that used respiratory specimens to compare Xpert with culture for detecting TB and Xpert with conventional phenotypic drug susceptibility testing for detecting rifampicin resistance.

DATA COLLECTION AND ANALYSIS

For each study, two review authors independently extracted a set of data using a standardized data extraction form. When possible, we extracted data for subgroups by smear and HIV status. We assessed the quality of studies using the QUADAS-2 tool. We carried out meta-analyses to estimate the pooled sensitivity and specificity of Xpert separately for TB detection and rifampicin resistance detection using a bivariate random-effects model. We estimated the median pooled sensitivity and specificity and their 95% credible intervals (CrI).

MAIN RESULTS

We identified 18 unique studies as eligible for this review, including two multicentre international studies, one with five and the other with six distinct study centres. The majority of studies (55.6%) were performed in low-income and middle-income countries. In 17 of the 18 studies, Xpert was performed by trained technicians in reference laboratories.When used as an initial test replacing smear microscopy (15 studies, 7517 participants), Xpert achieved a pooled sensitivity of 88% (95% CrI 83% to 92%) and pooled specificity of 98% (95% CrI 97% to 99%). As an add-on test following a negative smear microscopy result (14 studies, 5719 participants), Xpert yielded a pooled sensitivity of 67% (95% CrI 58% to 74%) and pooled specificity of 98% (95% CrI 97% to 99%). In clinical subgroups, we found the following accuracy estimates: the pooled sensitivity was 98% (95% CrI 97% to 99%) for smear-positive, culture-positive TB and 68% (95% CrI 59% to 75%) for smear-negative, culture-positive TB (15 studies); the pooled sensitivity was 80% (95% CrI 67% to 88%) in people living with HIV and 89% (95% CrI 81% to 94%) in people without HIV infection (four studies). For rifampicin resistance detection (11 studies, 2340 participants), Xpert achieved a pooled sensitivity of 94% (95% CrI 87% to 97%) and pooled specificity of 98% (95% CrI 97% to 99%). In a separate analysis, Xpert could distinguish between TB and nontuberculous mycobacteria (NTM) in clinical samples with high accuracy: among 139 specimens with NTM, Xpert was positive in only one specimen that grew NTM.In a hypothetical cohort of 1000 individuals suspected of having rifampicin resistance (a proxy for MDR-TB), where the prevalence of rifampicin resistance is 30%, we estimated that on average Xpert would wrongly identify 14 patients as being rifampicin resistant. In comparison, where the prevalence of rifampicin resistance is only 2%, we estimated that the number of individuals wrongly identified as rifampicin resistant would increase to 20, an increase of 43%.

AUTHORS' CONCLUSIONS

This review shows that Xpert used as an initial diagnostic test for TB detection and rifampicin resistance detection in patients suspected of having TB, MDR-TB, or HIV-associated TB is sensitive and specific. Xpert may also be valuable as an add-on test following microscopy for patients who have previously been found to be smear-negative. An Xpert result that is positive for rifampicin resistance should be carefully interpreted and take into consideration the risk of MDR-TB in a given patient and the expected prevalence of MDR-TB in a given setting.Studies in this review mainly assessed sensitivity and specificity of the test when used in reference laboratories in research investigations. Most studies were performed in high TB burden countries. Ongoing use of Xpert in high TB burden countries will contribute to the evidence base on the diagnostic accuracy and clinical impact of Xpert in routine programmatic and peripheral health care settings, including settings where the test is performed at the point of care.

Serodiagnosis of tuberculosis in Asian elephants (Elephas maximus) in Southern India: a latent class analysis

BACKGROUND

Mycobacterium tuberculosis, a causative agent of chronic tuberculosis disease, is widespread among some animal species too. There is paucity of information on the distribution, prevalence and true disease status of tuberculosis in Asian elephants (Elephas maximus). The aim of this study was to estimate the sensitivity and specificity of serological tests to diagnose M. tuberculosis infection in captive elephants in southern India while simultaneously estimating sero-prevalence.

METHODOLOGY/PRINCIPAL FINDINGS

Health assessment of 600 elephants was carried out and their sera screened with a commercially available rapid serum test. Trunk wash culture of select rapid serum test positive animals yielded no animal positive for M. tuberculosis isolation. Under Indian field conditions where the true disease status is unknown, we used a latent class model to estimate the diagnostic characteristics of an existing (rapid serum test) and new (four in-house ELISA) tests. One hundred and seventy nine sera were randomly selected for screening in the five tests. Diagnostic sensitivities of the four ELISAs were 91.3-97.6% (95% Credible Interval (CI): 74.8-99.9) and diagnostic specificity were 89.6-98.5% (95% CI: 79.4-99.9) based on the model we assumed. We estimate that 53.6% (95% CI: 44.6-62.8) of the samples tested were free from infection with M. tuberculosis and 15.9% (97.5% CI: 9.8 - to 24.0) tested positive on all five tests.

CONCLUSIONS/SIGNIFICANCE

Our results provide evidence for high prevalence of asymptomatic M. tuberculosis infection in Asian elephants in a captive Indian setting. Further validation of these tests would be important in formulating area-specific effective surveillance and control measures.

TB screening in Canadian health care workers using interferon-gamma release assays

Background

While many North American healthcare institutions are switching from Tuberculin Skin Test (TST) to Interferon-gamma release assays (IGRAs), there is relatively limited data on association between occupational tuberculosis (TB) risk factors and test positivity and/or patterns of test discordance.

Methods

We recruited a cohort of Canadian health care workers (HCWs) in Montreal, and performed both TST and QuantiFERON-TB Gold In Tube (QFT) tests, and assessed risk factors and occupational exposure.

Results

In a cross-sectional analysis of baseline results, the prevalence of TST positivity using the 10 mm cut-off was 5.7% (22/388, 95%CI: 3.6–8.5%), while QFT positivity was 6.2% (24/388, 95%CI: 4–9.1%). Overall agreement between the tests was poor (kappa = 0.26), and 8.3% of HCWs had discordant test results, most frequently TST−/QFT+ (17/388, 4.4%). TST positivity was associated with total years worked in health care, non-occupational exposure to TB and BCG vaccination received after infancy or on multiple occasions. QFT positivity was associated with having worked as a HCW in a foreign country.

Conclusions

Our results suggest that LTBI prevalence as measured by either the TST or the QFT is low in this HCW population. Of concern is the high frequency of unexplainable test discordance, namely: TST−/QFT+ subjects, and the lack of any association between QFT positivity and clear-cut recent TB exposure. If these discordant results are indeed false positives, the use of QFT in lieu of TST in low TB incidence settings could result in overtreatment of uninfected individuals.

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.

Positive cultures of organ preservation fluid predict postoperative infections in solid organ transplantation recipients

OBJECTIVE

The significance of positive cultures of organ preservation fluid (OPF) in solid organ transplantation is not known. We sought to describe the microbiology and define the clinical impact of positive OPF cultures.

DESIGN

Retrospective cohort study.

SETTING

Tertiary care hospital.

PATIENTS

A consecutive sample of all solid organ transplantations at our center between July 2006 and January 2009 was reviewed. A total of 331 allografts (185 kidneys, 104 livers, 31 pancreases, and 11 hearts) met the inclusion criterion of having OPF cultures taken from the transplanted allograft.

METHODS

Organisms recovered from OPF were classified as high or low risk according to their virulence. Clinical outcomes were compared between recipients of organs with positive OPF cultures and recipients of organs with negative OPF cultures.

RESULTS

 OPF cultures were positive in 62.2% of allografts and yielded high-risk organisms in 17.8%. Normal skin flora constituted the majority of positive OPF cultures, while Enterobacteriaceae spp. and Staphylococcus aureus made up the majority of high-risk organisms. Recipients of allografts with positive OPF cultures developed more frequent bacterial infections, regardless of allograft type (relative risk, 2.39; 95% confidence interval [CI], 1.61-3.54). Moreover, isolation of a given organism in OPF samples was associated with the development of a clinical infection with the same organism, regardless of allograft type.

CONCLUSIONS

Positive cultures of OPF are common events in solid organ transplantation, frequently involve high-risk organisms, and are associated with the development of postoperative clinical bacterial infections. Further study is required to determine the optimal strategies for their prevention and management.

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.

Patient, physician, encounter, and billing characteristics predict the accuracy of syndromic surveillance case definitions

BACKGROUND

Syndromic surveillance systems are plagued by high false-positive rates. In chronic disease monitoring, investigators have identified several factors that predict the accuracy of case definitions based on diagnoses in administrative data, and some have even incorporated these predictors into novel case detection methods, resulting in a significant improvement in case definition accuracy. Based on findings from these studies, we sought to identify physician, patient, encounter, and billing characteristics associated with the positive predictive value (PPV) of case definitions for 5 syndromes (fever, gastrointestinal, neurological, rash, and respiratory (including influenza-like illness)).

METHODS

The study sample comprised 4,330 syndrome-positive visits from the claims of 1,098 randomly-selected physicians working in Quebec, Canada in 2005-2007. For each visit, physician-facilitated chart review was used to assess whether the same syndrome was present in the medical chart (gold standard). We used multivariate logistic regression analyses to estimate the association between claim-chart agreement about the presence of a syndrome and physician, patient, encounter, and billing characteristics.

RESULTS

The likelihood of the medical chart agreeing with the physician claim about the presence of a syndrome was higher when the treating physician had billed many visits for the same syndrome recently (ORper 10 visit, 1.05; 95% CI, 1.01-1.08), had a lower workload (ORper 10 claims, 0.93; 95% CI, 0.90-0.97), and when the patient was younger (ORper 5 years of age, 0.96; 95% CI, 0.94-0.97), and less socially deprived (ORmost versus least deprived, 0.76; 95% CI, 0.60-0.95).

CONCLUSIONS

Many physician, patient, encounter, and billing characteristics associated with the PPV of surveillance case definition are accessible to public health, and could be used to reduce false-positive alerts by surveillance systems, either by focusing on the data most likely to be accurate, or by adjusting the observed data for known biases in diagnosis reporting and performing surveillance using the adjusted values.

Host and pathogen factors for Clostridium difficile infection and colonization

BACKGROUND

Clostridium difficile infection is the leading cause of health care-associated diarrhea, and the bacterium can also be carried asymptomatically. The objective of this study was to identify host and bacterial factors associated with health care-associated acquisition of C. difficile infection and colonization.

METHODS

We conducted a 15-month prospective study in six Canadian hospitals in Quebec and Ontario. Demographic information, known risk factors, potential confounding factors, and weekly stool samples or rectal swabs were collected. Pulsed-field gel electrophoresis (PFGE) was performed on C. difficile isolates to determine the genotype. Levels of serum antibodies against C. difficile toxins A and B were measured.

RESULTS

A total of 4143 patients were included in the study; 117 (2.8%) and 123 (3.0%) had health care-associated C. difficile infection and colonization, respectively. Older age and use of antibiotics and proton-pump inhibitors were significantly associated with health care-associated C. difficile infection. Hospitalization in the previous 2 months; use of chemotherapy, proton-pump inhibitors, and H(2) blockers; and antibodies against toxin B were associated with health care-associated C. difficile colonization. Among patients with health care-associated C. difficile infection and those with colonization, 62.7% and 36.1%, respectively, had the North American PFGE type 1 (NAP1) strain.

CONCLUSIONS

In this study, health care-associated C. difficile infection and colonization were differentially associated with defined host and pathogen variables. The NAP1 strain was predominant among patients with C. difficile infection, whereas asymptomatic patients were more likely to be colonized with other strains. (Funded by the Consortium de Recherche sur le Clostridium difficile.).

Commercial serological tests for the diagnosis of active pulmonary and extrapulmonary tuberculosis: an updated systematic review and meta-analysis

BACKGROUND

Serological (antibody detection) tests for tuberculosis (TB) are widely used in developing countries. As part of a World Health Organization policy process, we performed an updated systematic review to assess the diagnostic accuracy of commercial serological tests for pulmonary and extrapulmonary TB with a focus on the relevance of these tests in low- and middle-income countries.

METHODS AND FINDINGS

We used methods recommended by the Cochrane Collaboration and GRADE approach for rating quality of evidence. In a previous review, we searched multiple databases for papers published from 1 January 1990 to 30 May 2006, and in this update, we add additional papers published from that period until 29 June 2010. We prespecified subgroups to address heterogeneity and summarized test performance using bivariate random effects meta-analysis. For pulmonary TB, we included 67 studies (48% from low- and middle-income countries) with 5,147 participants. For all tests, estimates were variable for sensitivity (0% to 100%) and specificity (31% to 100%). For anda-TB IgG, the only test with enough studies for meta-analysis, pooled sensitivity was 76% (95% CI 63%-87%) in smear-positive (seven studies) and 59% (95% CI 10%-96%) in smear-negative (four studies) patients; pooled specificities were 92% (95% CI 74%-98%) and 91% (95% CI 79%-96%), respectively. Compared with ELISA (pooled sensitivity 60% [95% CI 6%-65%]; pooled specificity 98% [95% CI 96%-99%]), immunochromatographic tests yielded lower pooled sensitivity (53%, 95% CI 42%-64%) and comparable pooled specificity (98%, 95% CI 94%-99%). For extrapulmonary TB, we included 25 studies (40% from low- and middle-income countries) with 1,809 participants. For all tests, estimates were variable for sensitivity (0% to 100%) and specificity (59% to 100%). Overall, quality of evidence was graded very low for studies of pulmonary and extrapulmonary TB.

CONCLUSIONS

Despite expansion of the literature since 2006, commercial serological tests continue to produce inconsistent and imprecise estimates of sensitivity and specificity. Quality of evidence remains very low. These data informed a recently published World Health Organization policy statement against serological tests. Please see later in the article for the Editors' Summary.

Bayesian sample size for diagnostic test studies in the absence of a gold standard: Comparing identifiable with non-identifiable models

Diagnostic tests rarely provide perfect results. The misclassification induced by imperfect sensitivities and specificities of diagnostic tests must be accounted for when planning prevalence studies or investigations into properties of new tests. The previous work has shown that applying a single imperfect test to estimate prevalence can often result in very large sample size requirements, and that sometimes even an infinite sample size is insufficient for precise estimation because the problem is non-identifiable. Adding a second test can sometimes reduce the sample size substantially, but infinite sample sizes can still occur as the problem remains non-identifiable. We investigate the further improvement possible when three diagnostic tests are to be applied. We first develop methods required for studies when three conditionally independent tests are available, using different Bayesian criteria. We then apply these criteria to prototypic scenarios, showing that large sample size reductions can occur compared to when only one or two tests are used. As the problem is now identifiable, infinite sample sizes cannot occur except in pathological situations. Finally, we relax the conditional independence assumption, demonstrating in this once again non-identifiable situation that sample sizes may substantially grow and possibly be infinite. We apply our methods to the planning of two infectious disease studies, the first designed to estimate the prevalence of Strongyloides infection, and the second relating to estimating the sensitivity of a new test for tuberculosis transmission. The much smaller sample sizes that are typically required when three as compared to one or two tests are used should encourage researchers to plan their studies using more than two diagnostic tests whenever possible. User-friendly software is available for both design and analysis stages greatly facilitating the use of these methods.