Bayesian Interval Estimation for Predictive Values from Case-Control Studies

Positive predictive and negative predictive values (PPV and NPV) are often used to assess the accuracy of binary diagnostic tests. Unlike sensitivity and specificity, PPV and NPV are functions of the accuracy of the test and the overall prevalence of the disease in the population. In many studies of performance of estimators of PPV and NPV the population prevalence is assumed known. We allow for uncertainty in the estimate of the population prevalence and via simulation explore the impact of deviations from the assumed value.     

Bayesian Inference of Odds Ratios in Misclassified Binary Data with a Validation Substudy

We propose a fully Bayesian model with a non-informative prior for analyzing misclassified binary data with a validation substudy. In addition, we derive a closed-form algorithm for drawing all parameters from the posterior distribution and making statistical inference on odds ratios. Our algorithm draws each parameter from a beta distribution, avoids the specification of initial values, and does not have convergence issues. We apply the algorithm to a data set and compare the results with those obtained by other methods. Finally, the performance of our algorithm is assessed using simulation studies.     

Bayesian estimation of logistic regression with misclassified covariates and response

Measurement error is a commonly addressed problem in psychometrics and the behavioral sciences, particularly where gold standard data either does not exist or are too expensive. The Bayesian approach can be utilized to adjust for the bias that results from measurement error in tests. Bayesian methods offer other practical advantages for the analysis of epidemiological data including the possibility of incorporating relevant prior scientific information and the ability to make inferences that do not rely on large sample assumptions. In this paper we consider a logistic regression model where both the response and a binary covariate are subject to misclassification. We assume both a continuous measure and a binary diagnostic test are available for the response variable but no gold standard test is assumed available. We consider a fully Bayesian analysis that affords such adjustments, accounting for the sources of error and correcting estimates of the regression parameters. Based on the results from our example and simulations, the models that account for misclassification produce more statistically significant results, than the models that ignore misclassification. A real data example on math disorders is considered.     

Testing for empty sets when misclassifications may be present

A test for empty sets is described. This new test is suitable for testing for mutual exclusivity, sets being nested, and many other configurations. It might be hypothesized that two medical conditions do not both occur in individuals. This can be expressed as an empty intersection of the two sets of people with each condition. An important feature of the test is the incorporation of misclassification rates into the analysis. The test utilizes the potential misclassifications for false designations in the sets that are hypothesized to be empty. The test is quite powerful. When the null hypothesis is rejected, follow-up tests on one or more sets that are included in the null hypothesis can be performed using the same new test. MatLab code is supplied.     

When a Constant Classifier is as Good as Any Linear Classifier

A classifier is constant if it classifies all examples into just one class. Call a training data set “(linearly) indiscriminate” if a constant classifier minimizes, among all linear classifiers, the misclassification rate on the training data set. General sufficient conditions are presented for the probability of getting an indiscriminate data set to be positive. Similarly, general sufficient conditions are also presented for the probability of getting an indiscriminate data set to be 0.

A small simulation study examines how our results are reflected in the behavior of logistic regression.     

Measures of mortality in the study of individuals infected with the human immunodeficiency virus in the UK haemophilia population

Two statistical issues that have arisen in the course of a study of mortality and disease related to the human immunodeficiency virus (HIV) in the haemophilia population of the UK are discussed. The first of these concerns methods of standardization for age and it is shown that, when the mortality of HIV-infected individuals with different severities of haemophilia are compared, an analysis based on the ratio of observed to national expected deaths suggests that mortality in HIV-infected individuals depends on the severity of their haemophilia. This conclusion is inappropriate and mortality in HIV-infected individuals is, in fact, similar regardless of severity of haemophilia. The second part of the paper discusses the effect of using various end points for studies of survival and progression of HIV-related disease. In the present example it was possible to calculate relative survival in HIV-infected individuals, i.e. survival after correcting for mortality expected in the absence of HIV infection. An analysis based on absolute survival gave a very similar picture of the effect of age at infection to an analysis based on relative survival, whereas an analysis based on the time to diagnosis of acquired immune deficiency syndrome (AIDS) underestimated the effect substantially and the possible alternative end point of time to AIDS or HIV-related death was shown to be subject to considerable misclassification error.     

On the influence of misclassified data on results of goodness of fit testing

In this paper we focus on the chi-square test of goodness of fit, which compares an observed discrete distribution to an expected known one. We show that the results of this test, using the common Pearson statistic, are very sensitive to misclassified observations between two or more categories. We also propose a general rule of thumb for analysing data set stability with respect to such classification errors. Practical analysis of a real example illustrates our purpose.     

Accounting for Response Misclassification and Covariate Measurement Error Using a Random Effects Logit Model

Often in longitudinal data arising out of epidemiologic studies, measurement error in covariates and/or classification errors in binary responses may be present. The goal of the present work is to develop a random effects logistic regression model that corrects for the classification errors in binary responses and/or measurement error in covariates. The analysis is carried out under a Bayesian set up. Simulation study reveals the effect of ignoring measurement error and/or classification errors on the estimates of the regression coefficients.     

Systematically misclassified binary dependent variables

ABSTRACT

When a binary dependent variable is misclassified, that is, recorded in the category other than where it really belongs, probit and logit estimates are biased and inconsistent. In some cases, the probability of misclassification may vary systematically with covariates, and thus be endogenous. In this paper, we develop an estimation approach that corrects for endogenous misclassification, validate our approach using a simulation study, and apply it to the analysis of a treatment program designed to improve family dynamics. Our results show that endogenous misclassification could lead to potentially incorrect conclusions unless corrected using an appropriate technique.     

Survival Weibull regression model for mismeasured outcomes

In some survival studies, the exact time of the event of interest is unknown, but the event is known to have occurred during a particular period of time (interval-censored data). If the diagnostic tool used to detect the event of interest is not perfectly sensitive and specific, outcomes may be mismeasured; a healthy subject may be diagnosed as sick and a sick one may be diagnosed as healthy. In such cases, traditional survival analysis methods produce biased estimates for the time-to-failure distribution parameters (Paggiaro and Torelli 2004 Paggiaro, A., and N. Torelli. 2004. The effect of classification errors in survival data analysis. Statistical Methods and Applications 13:213–25.[Crossref] , [Google Scholar]). In this context, we developed a parametric model that incorporates sensitivity and specificity into a grouped survival data analysis (a case of interval-censored data in which all subjects are tested at the same predetermined time points). Inferential aspects and properties of the methodology, such as the likelihood function and identifiability, are discussed in this article. Assuming known and non differential misclassification, Monte Carlo simulations showed that the proposed model performed well in the case of mismeasured outcomes; the estimates of the relative bias of the model were lower than those provided by the naive method that assumes perfect sensitivity and specificity. The proposed methodology is illustrated by a study related to mango tree lifetimes.