A method for predicting individual HIV infection status in the absence of clinical information

Bayesian hierarchical latent class models for estimating diagnostic accuracy

The diagnostic accuracy of a test or rater has a crucial impact on clinical decision making. The assessment of diagnostic accuracy for multiple tests or raters also merits much attention. A Bayesian hierarchical conditional independence latent class model for estimating sensitivities and specificities for a large group of tests or raters is proposed, which is applicable to both with-gold-standard and without-gold-standard situations. Through the hierarchical structure, not only are the sensitivities and specificities of individual tests estimated, but also the diagnostic performance of the whole group of tests. For a small group of tests or raters, the proposed model is further extended by introducing pairwise covariances between tests to improve the fitting and to allow for more modeling flexibility. Correlation residual analysis is applied to detect any significant covariance between multiple tests. Just Another Gibbs Sampler (JAGS) implementation is efficiently adopted for both models. Three real data sets from literature are analyzed to explicitly illustrate the proposed methods.

Binomial regression with a misclassified covariate and outcome

Misclassification occurring in either outcome variables or categorical covariates or both is a common issue in medical science. It leads to biased results and distorted disease-exposure relationships. Moreover, it is often of clinical interest to obtain the estimates of sensitivity and specificity of some diagnostic methods even when neither gold standard nor prior knowledge about the parameters exists. We present a novel Bayesian approach in binomial regression when both the outcome variable and one binary covariate are subject to misclassification. Extensive simulation results under various scenarios and a real clinical example are given to illustrate the proposed approach. This approach is motivated and applied to a dataset from the Baylor Alzheimer's Disease and Memory Disorders Center.

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.

A Bayesian model for misclassified binary outcomes and correlated survival data with applications to breast cancer

Breast cancer patients may experience ipsilateral breast tumor relapse (IBTR) after breast conservation therapy. IBTR is classified as either true local recurrence or new ipsilateral primary tumor. The correct classification of IBTR status has significant implications in therapeutic decision-making and patient management. However, the diagnostic tests to classify IBTR are imperfect and prone to misclassification. In addition, some observed survival data (e.g., time to relapse, time from relapse to death) are strongly correlated with IBTR status. We present a Bayesian approach to model the potentially misclassified IBTR status and the correlated survival information. We conduct the inference using a Bayesian framework via Markov chain Monte Carlo simulation implemented in WinBUGS. Extensive simulation shows that the proposed method corrects biases and provides more efficient estimates for the covariate effects on the probability of IBTR and the diagnostic test accuracy. Moreover, our method provides useful subject-specific patient prognostic information. Our method is motivated by, and applied to, a dataset of 397 breast cancer patients.

GENERALIZED LATENT CLASS ANALYSIS BASED ON MODEL DOMINANCE THEORY

Latent class analysis is a popular statistical learning approach. A major challenge for learning generalized latent class is the complexity in searching the huge space of models and parameters. The computational cost is higher when the model topology is more flexible. In this paper, we propose the notion of dominance which can lead to strong pruning of the search space and significant reduction of learning complexity, and apply this notion to the Generalized Latent Class (GLC) models, a class of Bayesian networks for clustering categorical data.

GLC models can address the local dependence problem in latent class analysis by assuming a very general graph structure. However, The flexible topology of GLC leads to large increase of the learning complexity. We first propose the concept of dominance and related theoretical results which is general for all Bayesian networks. Based on dominance, we propose an efficient learning algorithm for GLC. A core technique to prune dominated models is regularization, which can eliminate dominated models, leading to significant pruning of the search space. Significant improvements on the model.

Estimating diagnostic accuracy of multiple binary tests with an imperfect reference standard

The goal in diagnostic medicine is often to estimate the diagnostic accuracy of multiple experimental tests relative to a gold standard reference. When a gold standard reference is not available, investigators commonly use an imperfect reference standard. This paper proposes methodology for estimating the diagnostic accuracy of multiple binary tests with an imperfect reference standard when information about the diagnostic accuracy of the imperfect test is available from external data sources. We propose alternative joint models for characterizing the dependence between the experimental tests and discuss the use of these models for estimating individual-test sensitivity and specificity as well as prevalence and multivariate post-test probabilities (predictive values). We show using analytical and simulation techniques that, as long as the sensitivity and specificity of the imperfect test are high, inferences on diagnostic accuracy are robust to misspecification of the joint model. The methodology is demonstrated with a study examining the diagnostic accuracy of various HIV-antibody tests for HIV.

Utility of composite reference standards and latent class analysis in evaluating the clinical accuracy of diagnostic tests for pertussis

Numerous evaluations of the clinical sensitivity and specificity of PCR and serologic assays for Bordetella pertussis have been hampered by the low sensitivity of culture, the gold standard test, which leads to biased accuracy estimates. The bias can be reduced by using statistical approaches such as the composite reference standard (CRS) (e.g., positive if culture or serology positive; negative otherwise) or latent class analysis (LCA), an internal reference standard based on a statistical model. We illustrated the benefits of the CRS and LCA approaches by reanalyzing data from a 1995 to 1996 study of cough illness among 212 patients. The accuracy of PCR in this study was evaluated using three reference standards: culture, CRS, and LCA. Using specimens obtained 0 to 34 days after cough onset, estimates of the sensitivity of PCR obtained using CRS (47%) and LCA (34%) were lower than the culture-based estimate (62%). The CRS and LCA approaches, which utilized more than one diagnostic marker of pertussis, likely produced more accurate reference standards than culture alone. In general, the CRS approach is simple, with a well-defined disease status. LCA requires statistical modeling but incorporates more indicators of disease than CRS. When three or more indicators of pertussis are available, these approaches should be used in evaluations of pertussis diagnostic tests.

Combining ratings from multiple physician reviewers helped to overcome the uncertainty associated with adverse event classification

OBJECTIVES

Adverse events (AEs) are poor patient outcomes, resulting from medical care. We performed this study to quantify the misclassification rate obtained using current AE detection methods and to evaluate the effect of combining physician AE ratings.

STUDY DESIGN AND SETTING

Three physicians independently rated poor patient outcomes. We used latent class analysis to obtain estimates for AE prevalence and reviewer accuracy. These estimates were used as a base case for four simulations of 10,000 cases rated independently by five reviewers. We assessed the effect of AE prevalence, reviewer accuracy, and the number of agreeing reviewers on the probability that cases were correctly classified as an AE.

RESULTS

Reviewer sensitivity and specificity for AE classification were 0.86 and 0.94, respectively. When prevalence was 3%, the positive predictive value that an AE occurred when a single reviewer classified the case as such was 31%, whereas when 2/3 reviewers did so it was 51%. The positive predictive values of ratings for AE occurrence increased with AE prevalence, reviewer accuracy, and the number of reviewers.

CONCLUSION

Current methods of AE detection overestimate the risk of AE. Uncertainty regarding the presence of an AE can be overcome by increasing the number of reviews.

Production of HIV-1 p24 protein in transgenic tobacco plants

The production of antigens for vaccines in plants has the potential as a safe and cost-effective alternative to traditional production systems. Toward the development of a plant-based expression system for the production of human immunodeficiency virus type I (HIV-1) p24 capsid protein, the p24 gene was introduced into the genome of tobacco plants using Agrobacterium tumefaciens-mediated gene transfer. Southern blot analyses confirmed the presence of the p24 coding sequence within the genome of transgenic lines. Western blot analysis of protein extracts from transgenic plants identified plant-expressed p24 protein that cross-reacted with a p24-specific monoclonal antibody, thus confirming the maintenance of antigenicity. Quantification of the p24 protein using enzyme-linked immunosorbent assay (ELISA) estimated yields of approx 3.5 mg per g of soluble leaf protein. Similar accumulation levels of p24 were also detected in T1 plants, confirming that the p24 gene is transmitted stably. Our results indicate that plant-based transgenic expression represents a viable means of producing p24 for the development of HIV vaccine and for use in HIV diagnostic procedures.

Structural asymmetry as a dermatoscopic indicator of malignant melanoma--a latent class analysis of sensitivity and classification errors

Asymmetry of pigmented skin lesions is an important indicator of possible malignant melanoma and contributes substantially to the diagnosis of melanoma in the dermatoscopic ABCD rule for melanocytic lesions and other algorithms. However, it may be observer dependent. Dermatoscopic asymmetry cannot be assessed objectively and no golden standard of asymmetry diagnosis exists. The aim of this study was to assess the sensitivity of axis (a-) symmetry using latent class analysis. We analysed ratings from four experts in dermatoscopy of 232 pigmented lesions by latent class analysis (LCA). Possible ratings were 'no asymmetry', 'asymmetry in one axis' and 'asymmetry in two axes'. A subset of melanocytic lesions (blue naevi excluded) was analysed. Based on LCA, the asymmetry of the singular lesion was determined. The sensitivity of 'no asymmetry' was 40-77%, 40-70% for one-axis asymmetry, and 77-92% for two-axes asymmetry. Overestimation of asymmetry was more common than underestimation. Melanomas were significantly more asymmetric than pigmented naevi, atypical naevi and papillomas, but not basal cell cancers. Analysis of the melanocytic subset gave similar results. The median asymmetry of malignant melanomas (1.67, interquartile range 1.81-1.99) was higher than for melanocytic naevi. In conclusion, asymmetry and symmetry are important criteria for diagnosing or excluding malignant melanoma using the dermatoscopic ABCD rule, risk stratification and other diagnostic rules. Using LCA, we minimized observer dependence in the assessment of axis (a-) symmetry. LCA, besides conceptualizing the diagnostic process, enables the assignment of lesions to their true diagnostic class.

Latent class analysis of human herpesvirus 8 assay performance and infection prevalence in sub-saharan Africa and Malta

Human herpesvirus 8 (HHV-8) is thought to be highly prevalent in Mediterranean countries and sub-Saharan Africa, where it causes Kaposi's sarcoma in a small proportion of infected immunocompetent persons. However, the lack of serological tests with established accuracy has hindered our understanding of the prevalence, risk factors and natural history of HHV-8 infection. We tested 837 subjects from Congo, Botswana (mostly young adults) and Malta (elderly adults), using an immunofluorescence assay and 2 enzyme immunoassays (EIAs, to viral proteins K8.1 and orf65). Each assay found HHV-8 seroprevalence to be high (49-87%) in the African populations and generally lower (9-54%) in Malta. However, there was only modest agreement among tests regarding which subjects were seropositive (3-way kappa, 0.05-0.34). We used latent class analysis to model this lack of agreement, estimating each test's sensitivity and specificity and each population's HHV-8 prevalence. Using this approach, the K8.1 EIA had consistently high sensitivity (91-100%) and specificity (92-100%) across populations, suggesting that it might be useful for epidemiological studies. Compared with the K8.1 EIA, both the immunofluorescence assay and the orf65 EIA had more variable sensitivity (80-100% and 58-87%, respectively) and more variable specificity (57-100% and 48-85%, respectively). HHV-8 prevalence was 7% among elderly Maltese adults. Prevalence was much higher (82%) in Congo, consistent with very high Kaposi's sarcoma incidence there. Prevalence was also high in Botswana (87% in Sans, an indigenous group, and 76% in Bantus), though Kaposi's sarcoma is not common, suggesting that additional co-factors besides HHV-8 are needed for development of Kaposi's sarcoma.

Diagnostic test analyses in search of their gold standard: latent class analyses with random effects

We review methods for analysing the performance of several diagnostic tests when patients must be classified as having a disease or not, when no gold standard is available. For latent class analysis (LCA) to provide consistent estimates of sensitivity, specificity and prevalence, traditionally 'independent errors conditional on disease status' have been assumed. Recent approaches derive estimators under more flexible assumptions. However, all likelihood-based approaches suffer from the sparseness of tables generated by this type of data; an issue which is often ignored. In light of this, we examine the potential and limitations of LCAs of diagnostic tests. We are guided by a data set of visceral leishmaniasis tests. In the example, LCA estimates suggest that the traditional reference test, parasitology, has poor sensitivity and underestimates prevalence. From a technical standpoint, including more test results in one analysis yields increasing degrees of sparseness in the table which are seen to lead to discordant values of asymptotically equivalent test statistics and eventually lack of convergence of the LCA algorithm. We suggest some strategies to cope with this.

Evaluation of diagnostic tests without gold standards

This paper reviews statistical methods developed to estimate the sensitivity and specificity of screening or diagnostic tests when the fallible tests are not evaluated against a gold standard. It gives a brief summary of the earlier historical developments and focuses on the more recent methods. It covers Bayesian approaches and longitudinal studies with repeated testing. In particular, it reviews the procedures that do not require the assumption of independence between tests conditional on the true disease status.

Mother-to-infant transmission of human immunodeficiency virus type 1: association with prematurity or low anti-gp120

In a prospective study of pregnant women infected with human immunodeficiency virus type 1 (HIV-1) in Brooklyn, New York, USA, 16 (29%) of 55 evaluable infants were infected with HIV-1. 9 infants had paediatric acquired immunodeficiency syndrome, 6 had less severe clinical manifestations of HIV-1 infection, and 1 was symptom-free but was seropositive for HIV-1 beyond 15 months of age. The 10 infants born at 37 weeks of gestation or earlier were at higher risk of HIV-1 infection than infants born at 38 weeks of gestation or later (60% vs 22%) but the median age at appearance of disease was approximately 5 months in both groups. The HIV-1 transmission rate was not associated with predelivery levels of maternal T cells, anti-p24, or neutralising antibodies but it was higher, among full-term infants, for those with mothers in the lowest third of the distribution of anti-gp120 levels (53%). On immunoblot, transmitting mothers lacked a gp120 band but not other bands. Protection was not associated with antibody to recombinant peptides from the hypervariable region of the major neutralising gp120 epitope, and the anti-gp120 endpoint dilution titre was similar in transmitting and non-transmitting mothers. Mothers of uninfected full-term infants appear to confer immunological protection against HIV-1 infection of their offspring by way of a high-affinity antibody to a gp120 epitope, whose specificity has importance for vaccine development and possibly perinatal immunotherapy.

A prospective study of human immunodeficiency virus type 1 infection and the development of AIDS in subjects with hemophilia

We evaluated a multicenter cohort of 1219 subjects with hemophilia or related disorders prospectively, focusing on 319 subjects with documented dates of seroconversion to human immunodeficiency virus type 1 (HIV-1). The incidence rate of the acquired immunodeficiency syndrome (AIDS) after seroconversion was 2.67 per 100 person-years and was directly related to age (from 0.83 in persons 1 to 11 years old up to 5.66 in persons 35 to 70 years old; Ptrend = 0.00003). The annual incidence of AIDS ranged from zero during the first year after seroconversion to 7 percent during the eighth year, with eight-year cumulative rates (+/- SE) of 13.3 +/- 5.3 percent for ages 1 to 17, 26.8 +/- 6.4 percent for ages 18 to 34, and 43.7 +/- 16.4 percent for ages 35 to 70. Serial immunologic and virologic markers (total numbers of CD4 lymphocytes, presence of serum interferon or HIV-1 p24 antigen, and low or absent serum levels of anti-p24 or anti-gp120) predicted a high risk for the subsequent development of AIDS. Adults 35 to 70 years old had a higher incidence of low CD4 counts than younger subjects (P less than or equal to 0.005), whereas adolescents had a low rate of anti-p24 loss (P = 0.0007) and subjects 1 to 17 years old had a lower incidence of AIDS after loss of anti-p24 (P = 0.03). These findings not only demonstrate that the risk of AIDS is related directly to age but also suggest that older adults are disproportionately affected during the earlier phases of HIV disease, that adolescents may have a low replication rate of HIV, and that children and adolescents may tolerate severe immunodeficiency better because they have fewer other infections or because of some unmeasured, age-dependent cofactor or immune alteration in the later phase of HIV disease.