Biostatistics Commons^™

Combining Biomarkers By Maximizing The True Positive Rate For A Fixed False Positive Rate, Allison Meisner, Marco Carone, Margaret Pepe, Kathleen F. Kerr

UW Biostatistics Working Paper Series

Biomarkers abound in many areas of clinical research, and often investigators are interested in combining them for diagnosis, prognosis and screening. In many applications, the true positive rate for a biomarker combination at a prespecified, clinically acceptable false positive rate is the most relevant measure of predictive capacity. We propose a distribution-free method for constructing biomarker combinations by maximizing the true positive rate while constraining the false positive rate. Theoretical results demonstrate good operating characteristics for the resulting combination. In simulations, the biomarker combination provided by our method demonstrated improved operating characteristics in a variety of scenarios when compared with …

Developing Biomarker Combinations In Multicenter Studies Via Direct Maximization And Penalization, Allison Meisner, Chirag R. Parikh, Kathleen F. Kerr

UW Biostatistics Working Paper Series

When biomarker studies involve patients at multiple centers and the goal is to develop biomarker combinations for diagnosis, prognosis, or screening, we consider evaluating the predictive capacity of a given combination with the center-adjusted AUC (aAUC), a summary of conditional performance. Rather than using a general method to construct the biomarker combination, such as logistic regression, we propose estimating the combination by directly maximizing the aAUC. Furthermore, it may be desirable to have a biomarker combination with similar predictive capacity across centers. To that end, we allow for penalization of the variability in center-specific performance. We demonstrate good asymptotic properties …

Testing The Independence Hypothesis Of Accepted Mutations For Pairs Of Adjacent Amino Acids In Protein Sequences, Jyotsna Ramanan, Peter Revesz

School of Computing: Faculty Publications

Evolutionary studies usually assume that the genetic mutations are independent of each other. However, that does not imply that the observed mutations are independent of each other because it is possible that when a nucleotide is mutated, then it may be biologically beneficial if an adjacent nucleotide mutates too. With a number of decoded genes currently available in various genome libraries and online databases, it is now possible to have a large-scale computer-based study to test whether the independence assumption holds for pairs of adjacent amino acids. Hence the independence question also arises for pairs of adjacent amino acids within …

Peripheral Inflammation, Apolipoprotein E4, And Amyloid-Β Interact To Induce Cognitive And Cerebrovascular Dysfunction, Felecia M. Marottoli, Yuriko Katsumata, Kevin P. Koster, Riya Thomas, David W. Fardo, Leon M. Tai

Biostatistics Faculty Publications

Cerebrovascular dysfunction is rapidly reemerging as a major process of Alzheimer’s disease (AD). It is, therefore, crucial to delineate the roles of AD risk factors in cerebrovascular dysfunction. While apolipoprotein E4 (APOE4), Amyloid-β (Aβ), and peripheral inflammation independently induce cerebrovascular damage, their collective effects remain to be elucidated. The goal of this study was to determine the interactive effect of APOE4, Aβ, and chronic repeated peripheral inflammation on cerebrovascular and cognitive dysfunction in vivo. EFAD mice are a well-characterized mouse model that express human APOE3 (E3FAD) or APOE4 (E4FAD) and overproduce human Aβ42 via expression of …

Visualizing Lab And Phenotype Associations Using Phewas And Electronic Health Records, Brenda Emerson, Miriam Goldman, Sahiti Kolli

Honors Projects

As the digitization of patient health records is becoming more common, we are given a great opportunity to analyze these records and hopefully make discoveries about diseases or medicines. Being given large datasets of Electronic Health Records, I and two other students decided to look for novel phenotype associations with mean lab values, look to see whether the presence of a lab had associations with a phenotype, and create an interactive application to visual the associations between labs and phenotypes.

Burden Of Atopic Dermatitis In The United States: Analysis Of Healthcare Claims Data In The Commercial, Medicare, And Medi-Cal Databases, Sulena Shrestha, Raymond Miao, Li Wang, Jingdong Chao, Huseyin Yuce, Wenhui Wei

Publications and Research

Comparative data on the burden of atopic dermatitis (AD) in adults relative to the general population are limited. We performed a large-scale evaluation of the burden of disease among US adults with AD relative to matched non-AD controls, encompassing comorbidities, healthcare resource utilization (HCRU), and costs, using healthcare claims data. The impact of AD disease severity on these outcomes was also evaluated.

Biomarker Combinations For Diagnosis And Prognosis In Multicenter Studies: Principles And Methods, Allison Meisner, Chirag R. Parikh, Kathleen F. Kerr

UW Biostatistics Working Paper Series

Many investigators are interested in combining biomarkers to predict an outcome of interest or detect underlying disease. This endeavor is complicated by the fact that many biomarker studies involve data from multiple centers. Depending upon the relationship between center, the biomarkers, and the target of prediction, care must be taken when constructing and evaluating combinations of biomarkers. We introduce a taxonomy to describe the role of center and consider how a biomarker combination should be constructed and evaluated. We show that ignoring center, which is frequently done by clinical researchers, is often not appropriate. The limited statistical literature proposes using …

Mixture Models For Undiagnosed Prevalent Disease And Interval-Censored Incident Disease: Applications To A Cohort Assembled From Electronic Health Records., Li C Cheung, Qing Pan, Noorie Hyun, Mark Schiffman, Barbara Fetterman, Philip E Castle, Thomas Lorey, Hormuzd A Katki

Epidemiology Faculty Publications

For cost-effectiveness and efficiency, many large-scale general-purpose cohort studies are being assembled within large health-care providers who use electronic health records. Two key features of such data are that incident disease is interval-censored between irregular visits and there can be pre-existing (prevalent) disease. Because prevalent disease is not always immediately diagnosed, some disease diagnosed at later visits are actually undiagnosed prevalent disease. We consider prevalent disease as a point mass at time zero for clinical applications where there is no interest in time of prevalent disease onset. We demonstrate that the naive Kaplan-Meier cumulative risk estimator underestimates risks at early …

Safety Of Induced Sputum Collection In Children Hospitalized With Severe Or Very Severe Pneumonia., Andrea N Deluca, Laura L Hammitt, Julia Kim, Melissa M Higdon, Henry C Baggett, Daniel E Park, +Several Additional Authors

Epidemiology Faculty Publications

Background.: Induced sputum (IS) may provide diagnostic information about the etiology of pneumonia. The safety of this procedure across a heterogeneous population with severe pneumonia in low- and middle-income countries has not been described.

Methods.: IS specimens were obtained as part a 7-country study of the etiology of severe and very severe pneumonia in hospitalized childrenbefore, during, and after the procedure to record oxygen requirement, oxygen saturation, respiratory rate, consciousness level, and other evidence of clinical deterioration. Criteria for IS contraindications were predefined and serious adverse events (SAEs) were reported to ethics committees and a central safety monitor.

Results.: A …

Data Management And Data Quality In Perch, A Large International Case-Control Study Of Severe Childhood Pneumonia., Nora L Watson, Christine Prosperi, Amanda J Driscoll, Melissa M Higdon, Daniel E Park, Megan Sanza, Andrea N Deluca, Juliet O Awori, Doli Goswami, Emily Hammond, Lokman Hossain, Catherine Johnson, Alice Kamau, Locadiah Kuwanda, David P Moore, Omid Neyzari, Uma Onwuchekwa, David Parker, Patranuch Sapchookul

Epidemiology Faculty Publications

The Pneumonia Etiology Research for Child Health (PERCH) study is the largest multicountry etiology study of pediatric pneumonia undertaken in the past 3 decades. The study enrolled 4232 hospitalized cases and 5325 controls over 2 years across 9 research sites in 7 countries in Africa and Asia. The volume and complexity of data collection in PERCH presented considerable logistical and technical challenges. The project chose an internet-based data entry system to allow real-time access to the data, enabling the project to monitor and clean incoming data and perform preliminary analyses throughout the study. To ensure high-quality data, the project developed …

Colonization Density Of The Upper Respiratory Tract As A Predictor Of Pneumonia-Haemophilus Influenzae, Moraxella Catarrhalis, Staphylococcus Aureus, And Pneumocystis Jirovecii., Daniel E Park, Henry C Baggett, Stephen R C Howie, Qiyuan Shi, Nora L Watson, W Abdullah Brooks, Perch Study Group

Epidemiology Faculty Publications

Background.

There is limited information on the association between colonization density of upper respiratory tract colonizers and pathogen-specific pneumonia. We assessed this association for Haemophilus influenzae, Moraxella catarrhalis, Staphylococcus aureus, and Pneumocystis jirovecii. Methods.

In 7 low- and middle-income countries, nasopharyngeal/oropharyngeal swabs from children with severe pneumonia and age-frequency matched community controls were tested using quantitative polymerase chain reaction (PCR). Differences in median colonization density were evaluated using the Wilcoxon rank-sum test. Density cutoffs were determined using receiver operating characteristic curves. Cases with a pathogen identified from lung aspirate culture or PCR, pleural fluid culture or …

Microscopic Analysis And Quality Assessment Of Induced Sputum From Children With Pneumonia In The Perch Study., David R Murdoch, Susan C Morpeth, Laura L Hammitt, Amanda J Driscoll, Daniel E Park, +Several Additional Authors

Epidemiology Faculty Publications

Background.

It is standard practice for laboratories to assess the cellular quality of expectorated sputum specimens to check that they originated from the lower respiratory tract. The presence of low numbers of squamous epithelial cells (SECs) and high numbers of polymorphonuclear (PMN) cells are regarded as indicative of a lower respiratory tract specimen. However, these quality ratings have never been evaluated for induced sputum specimens from children with suspected pneumonia. Methods.

We evaluated induced sputum Gram stain smears and cultures from hospitalized children aged 1–59 months enrolled in a large study of community-acquired pneumonia. We hypothesized that a specimen representative …

Is Higher Viral Load In The Upper Respiratory Tract Associated With Severe Pneumonia? Findings From The Perch Study., Daniel R Feikin, Wei Fu, Daniel E Park, Qiyuan Shi, Melissa M Higdon, Henry C Baggett, +Several Additional Authors

Epidemiology Faculty Publications

Background.

The etiologic inference of identifying a pathogen in the upper respiratory tract (URT) of children with pneumonia is unclear. To determine if viral load could provide evidence of causality of pneumonia, we compared viral load in the URT of children with World Health Organization–defined severe and very severe pneumonia and age-matched community controls.

Methods.

In the 9 developing country sites, nasopharyngeal/oropharyngeal swabs from children with and without pneumonia were tested using quantitative real-time polymerase chain reaction for 17 viruses. The association of viral load with case status was evaluated using logistic regression. Receiver operating characteristic (ROC) curves were constructed …

Association Of C-Reactive Protein With Bacterial And Respiratory Syncytial Virus-Associated Pneumonia Among Children Aged <5 Years In The Perch Study., Melissa M Higdon, Tham Le, Katherine L O'Brien, David R Murdoch, Christine Prosperi, Henry C Baggett, Daniel E Park, +Several Additional Authors

Epidemiology Faculty Publications

Background.

Lack of a gold standard for identifying bacterial and viral etiologies of pneumonia has limited evaluation of C-reactive protein (CRP) for identifying bacterial pneumonia. We evaluated the sensitivity and specificity of CRP for identifying bacterial vs respiratory syncytial virus (RSV) pneumonia in the Pneumonia Etiology Research for Child Health (PERCH) multicenter case-control study. Methods.

We measured serum CRP levels in cases with World Health Organization–defined severe or very severe pneumonia and a subset of community controls. We evaluated the sensitivity and specificity of elevated CRP for “confirmed” bacterial pneumonia (positive blood culture or positive lung aspirate or pleural fluid …

Standardization Of Clinical Assessment And Sample Collection Across All Perch Study Sites., Jane Crawley, Christine Prosperi, Henry C Baggett, W Abdullah Brooks, Maria Deloria Knoll, Daniel E Park, +Several Additional Authors

Epidemiology Faculty Publications

Background.: Variable adherence to standardized case definitions, clinical procedures, specimen collection techniques, and laboratory methods has complicated the interpretation of previous multicenter pneumonia etiology studies. To circumvent these problems, a program of clinical standardization was embedded in the Pneumonia Etiology Research for Child Health (PERCH) study.

Methods.: Between March 2011 and August 2013, standardized training on the PERCH case definition, clinical procedures, and collection of laboratory specimens was delivered to 331 clinical staff at 9 study sites in 7 countries (The Gambia, Kenya, Mali, South Africa, Zambia, Thailand, and Bangladesh), through 32 on-site courses and a training website. Staff competency …

The Diagnostic Utility Of Induced Sputum Microscopy And Culture In Childhood Pneumonia., David R Murdoch, Susan C Morpeth, Laura L Hammitt, Amanda J Driscoll, Nora L Watson, Daniel E Park, +Several Additional Authors

Epidemiology Faculty Publications

Background.

Sputum microscopy and culture are commonly used for diagnosing the cause of pneumonia in adults but are rarely performed in children due to difficulties in obtaining specimens. Induced sputum is occasionally used to investigate lower respiratory infections in children but has not been widely used in pneumonia etiology studies. Methods.

We evaluated the diagnostic utility of induced sputum microscopy and culture in patients enrolled in the Pneumonia Etiology Research for Child Health (PERCH) study, a large study of community-acquired pneumonia in children aged 1–59 months. Comparisons were made between induced sputum samples from hospitalized children with radiographically confirmed pneumonia …

Optimal, Two Stage, Adaptive Enrichment Designs For Randomized Trials Using Sparse Linear Programming, Michael Rosenblum, Xingyuan Fang, Han Liu

Johns Hopkins University, Dept. of Biostatistics Working Papers

Adaptive enrichment designs involve preplanned rules for modifying enrollment criteria based on accruing data in a randomized trial. We focus on designs where the overall population is partitioned into two predefined subpopulations, e.g., based on a biomarker or risk score measured at baseline. The goal is to learn which populations benefit from an experimental treatment. Two critical components of adaptive enrichment designs are the decision rule for modifying enrollment, and the multiple testing procedure. We provide a general method for simultaneously optimizing these components for two stage, adaptive enrichment designs. We minimize the expected sample size under constraints on power …

Impact Of Serum Sp-A And Sp-D Levels On Comparison And Prognosis Of Idiopathic Pulmonary Fibrosis: A Systematic Review And Meta-Analysis., Kai Wang, Qing Ju, Jing Cao, Wenze Tang, Jian Zhang

Epidemiology Faculty Publications

Background and objective:

Idiopathic pulmonary fibrosis (IPF) has a poor prognosis in general; however, it is heterogeneous to detect relative biomarkers for predicting the disease progression. Serum biomarkers can be conveniently collected to detect and help to differentially diagnose IPF and predict IPF prognosis. This meta-analysis aimed to evaluate the use of serum surfactant proteins A and D (SP-A and SP-D) for differential diagnosis and prognosis of IPF.

Methods:

Relevant articles were searched in PubMed, Embase, and Chinese National Knowledge Infrastructuredatabases and reviewed by 2 independent readers. Standard mean difference (SMD) and 95% confidence interval (CI) were …

Risk Of Incident Clinical Diagnosis Of Alzheimer's Disease-Type Dementia Attributable To Pathology-Confirmed Vascular Disease, Hiroko H. Dodge, Jian Zhu, Randy Woltjer, Peter T. Nelson, David A. Bennett, Nigel J. Cairns, David W. Fardo, Jeffrey A. Kaye, Deniz-Erten Lyons, Nora Mattek, Julie A. Schneider, Lisa C. Silbert, Chengjie Xiong, Lei Yu, Frederick A. Schmitt, Richard J. Kryscio, Erin L. Abner, Smart Data Consortium

Sanders-Brown Center on Aging Faculty Publications

INTRODUCTION: The presence of cerebrovascular pathology may increase the risk of clinical diagnosis of Alzheimer's disease (AD).

METHODS: We examined excess risk of incident clinical diagnosis of AD (probable and possible AD) posed by the presence of lacunes and large infarcts beyond AD pathology using data from the Statistical Modeling of Aging and Risk of Transition study, a consortium of longitudinal cohort studies with more than 2000 autopsies. We created six mutually exclusive pathology patterns combining three levels of AD pathology (low, moderate, or high AD pathology) and two levels of vascular pathology (without lacunes and large infarcts or with …

Estimated Probability Of Becoming A Case Of Drug Dependence In Relation To Duration Of Drug-Taking Experience: A Functional Analysis Approach, Olga A. Vsevolozhskaya, James C. Anthony

Biostatistics Faculty Publications

Measured as elapsed time from first use to dependence syndrome onset, the estimated "induction interval" for cocaine is thought to be short relative to the cannabis interval, but little is known about risk of becoming dependent during first months after onset of use. Virtually all published estimates for this facet of drug dependence epidemiology are from life histories elicited years after first use. To improve estimation, we turn to new month-wise data from nationally representative samples of newly incident drug users identified via probability sampling and confidential computer-assisted self-interviews for the United States National Surveys on Drug Use and Health, …