Medicine and Health Sciences Commons^™

Survival Model Predictive Accuracy And Roc Curves, Patrick Heagerty, Yingye Zheng

UW Biostatistics Working Paper Series

The predictive accuracy of a survival model can be summarized using extensions of the proportion of variation explained by the model, or R^2, commonly used for continuous response models, or using extensions of sensitivity and specificity which are commonly used for binary response models.

In this manuscript we propose new time-dependent accuracy summaries based on time-specific versions of sensitivity and specificity calculated over risk sets. We connect the accuracy summaries to a previously proposed global concordance measure which is a variant of Kendall's tau. In addition, we show how standard Cox regression output can be used to obtain estimates of …

Semiparametric Estimation Of Time-Dependent: Roc Curves For Longitudinal Marker Data, Yingye Zheng, Patrick Heagerty

UW Biostatistics Working Paper Series

One approach to evaluating the strength of association between a longitudinal marker process and a key clinical event time is through predictive regression methods such as a time-dependent covariate hazard model. For example, a time-varying covariate Cox model specifies the instantaneous risk of the event as a function of the time-varying marker and additional covariates. In this manuscript we explore a second complementary approach which characterizes the distribution of the marker as a function of both the measurement time and the ultimate event time. Our goal is to flexibly extend the standard diagnostic accuracy concepts of sensitivity and specificity to …

A Corrected Pseudo-Score Approach For Additive Hazards Model With Longitudinal Covariates Measured With Error, Xiao Song, Yijian Huang

UW Biostatistics Working Paper Series

In medical studies, it is often of interest to characterize the relationship between a time-to-event and covariates, not only time-independent but also time-dependent. Time-dependent covariates are generally measured intermittently and with error. Recent interests focus on the proportional hazards framework, with longitudinal data jointly modeled through a mixed effects model. However, approaches under this framework depend on the normality assumption of the error, and might encounter intractable numerical difficulties in practice. This motivates us to consider an alternative framework, that is, the additive hazards model, under which little has been done when time-dependent covariates are measured with error. We propose …

Adjusting For Non-Ignorable Verification Bias In Clinical Studies For Alzheimer’S Disease, Xiao-Hua Zhou, Pete Castelluccio

UW Biostatistics Working Paper Series

A common problem for comparing the relative accuracy of two screening tests for Alzheimer’s disease (D) in a two-stage design study is verification bias. If the verification bias can be assumed to be ignorable, Zhou and Higgs (2000) have proposed a maximum likelihood approach to compare the relative accuracy of screening tests in a two-stage design study. However, if the verification mechanism also depends on the unobserved disease status, the ignorable assumption does not hold. In this paper, we discuss how to use a profile likelihood approach to compare the relative accuracy of two screening tests for AD without assuming …

A New Confidence Interval For The Difference Between Two Binomial Proportions Of Paired Data, Xiao-Hua Zhou, Gengsheng Qin

UW Biostatistics Working Paper Series

Motivated by a study on comparing sensitivities and specificities of two diagnostic tests in a paired design when the sample size is small, we first derived an Edgeworth expansion for the studentized difference between two binomial proportions of paired data. The Edgeworth expansion can help us understand why the usual Wald interval for the difference has poor coverage performance in the small sample size. Based on the Edgeworth expansion, we then derived a transformation based confidence interval for the difference. The new interval removes the skewness in the Edgeworth expansion; the new interval is easy to compute, and its coverage …

Identifying Target Populations For Screening Or Not Screening Using Logic Regression, Holly Janes, Margaret S. Pepe, Charles Kooperberg, Polly Newcomb

UW Biostatistics Working Paper Series

Colorectal cancer remains a significant public health concern despite the fact that effective screening procedures exist and that the disease is treatable when detected at early stages. Numerous risk factors for colon cancer have been identified, but none are very predictive alone. We sought to determine whether there are certain combinations of risk factors that distinguish well between cases and controls, and that could be used to identify subjects at particularly high or low risk of the disease to target screening. Using data from the Seattle site of the Colorectal Cancer Family Registry (C-CFR), we fit logic regression models to …

Improved Confidence Intervals For The Sensitivity At A Fixed Level Of Specificity Of A Continuous-Scale Diagnostic Test, Xiao-Hua Zhou, Gengsheng Qin

UW Biostatistics Working Paper Series

For a continuous-scale test, it is an interest to construct a confidence interval for the sensitivity of the diagnostic test at the cut-off that yields a predetermined level of its specificity (eg. 80%, 90%, or 95%). IN this paper we proposed two new intervals for the sensitivity of a continuous-scale diagnostic test at a fixed level of specificity. We then conducted simulation studies to compare the relative performance of these two intervals with the best existing BCa bootstrap interval, proposed by Platt et al. (2000). Our simulation results showed that the newly proposed intervals are better than the BCa bootstrap …

Bootstrap Confidence Intervals For Medical Costs With Censored Observations, Hongyu Jiang, Xiao-Hua Zhou

UW Biostatistics Working Paper Series

Medical costs data with administratively censored observations often arise in cost-effectiveness studies of treatments for life threatening diseases. Mean of medical costs incurred from the start of a treatment till death or certain timepoint after the implementation of treatment is frequently of interest. In many situations, due to the skewed nature of the cost distribution and non-uniform rate of cost accumulation over time, the currently available normal approximation confidence interval has poor coverage accuracy. In this paper, we proposed a bootstrap confidence interval for the mean of medical costs with censored observations. In simulation studies, we showed that the proposed …

A Bootstrap Confidence Interval Procedure For The Treatment Effect Using Propensity Score Subclassification, Wanzhu Tu, Xiao-Hua Zhou

UW Biostatistics Working Paper Series

In the analysis of observational studies, propensity score subclassification has been shown to be a powerful method for adjusting unbalanced covariates for the purpose of causal inferences. One practical difficulty in carrying out such an analysis is to obtain a correct variance estimate for such inferences, while reducing bias in the estimate of the treatment effect due to an imbalance in the measured covariates. In this paper, we propose a bootstrap procedure for the inferences concerning the average treatment effect; our bootstrap method is based on an extension of Efron’s bias-corrected accelerated (BCa) bootstrap confidence interval to a two-sample problem. …

Estimating The Accuracy Of Polymerase Chain Reaction-Based Tests Using Endpoint Dilution, Jim Hughes, Patricia Totten

UW Biostatistics Working Paper Series

PCR-based tests for various microorganisms or target DNA sequences are generally acknowledged to be highly "sensitive" yet the concept of sensitivity is ill-defined in the literature on these tests. We propose that sensitivity should be expressed as a function of the number of target DNA molecules in the sample (or specificity when the target number is 0). However, estimating this "sensitivity curve" is problematic since it is difficult to construct samples with a fixed number of targets. Nonetheless, using serially diluted replicate aliquots of a known concentration of the target DNA sequence, we show that it is possible to disentangle …

Selecting Differentially Expressed Genes From Microarray Experiments, Margaret S. Pepe, Gary M. Longton, Garnet L. Anderson, Michel Schummer

UW Biostatistics Working Paper Series

High throughput technologies, such as gene expression arrays and protein mass spectrometry, allow one to simultaneously evaluate thousands of potential biomarkers that distinguish different tissue types. Of particular interest here is cancer versus normal organ tissues. We consider statistical methods to rank genes (or proteins) in regards to differential expression between tissues. Various statistical measures are considered and we argue that two measures related to the Receiver Operating Characteristic Curve are particularly suitable for this purpose. We also propose that sampling variability in the gene rankings be quantified and suggest using the “selection probability function”, the probability distribution of rankings …

Semiparametric Receiver Operating Characteristic Analysis To Evaluate Biomarkers For Disease, Tianxi Cai, Margaret S. Pepe

UW Biostatistics Working Paper Series

The receiver operating characteristic (ROC) curve is a popular method for characterizing the accuracy of diagnostic tests when test results are not binary. Various methodologies for estimating and comparing ROC curves have been developed. One approach, due to Pepe, uses a parametric regression model with the baseline function specified up to a finite-dimensional parameter. In this article we extend the regression models by allowing arbitrary nonparametric baseline functions. We also provide asymptotic distribution theory and procedures for making statistical inference. We illustrate our approach with dataset from a prostate cancer biomarker study. Simulation studies suggest that the extra flexibility inherent …

Semi-Parametric Regression For The Area Under The Receiver Operating Characteristic Curve, Lori E. Dodd, Margaret S. Pepe

UW Biostatistics Working Paper Series

Medical advances continue to provide new and potentially better means for detecting disease. Such is true in cancer, for example, where biomarkers are sought for early detection and where improvements in imaging methods may pick up the initial functional and molecular changes associated with cancer development. In other binary classification tasks, computational algorithms such as Neural Networks, Support Vector Machines and Evolutionary Algorithms have been applied to areas as diverse as credit scoring, object recognition, and peptide-binding prediction. Before a classifier becomes an accepted technology, it must undergo rigorous evaluation to determine its ability to discriminate between states. Characterization of …

Partial Auc Estimation And Regression, Lori E. Dodd, Margaret S. Pepe

UW Biostatistics Working Paper Series

Accurate disease diagnosis is critical for health care. New diagnostic and screening tests must be evaluated for their abilities to discriminate disease from non-diseased states. The partial area under the ROC curve (partial AUC) is a measure of diagnostic test accuracy. We present an interpretation of the partial AUC that gives rise to a new non-parametric estimator. This estimator is more robust than existing estimators, which make parametric assumptions. We show that the robustness is gained with only a moderate loss in efficiency. We describe a regression modelling framework for making inference about covariate effects on the partial AUC. Such …