Dataset Information

LOGISTIC REGRESSION ANALYSIS WITH STANDARDIZED MARKERS.

ABSTRACT: Two different approaches to analysis of data from diagnostic biomarker studies are commonly employed. Logistic regression is used to fit models for probability of disease given marker values while ROC curves and risk distributions are used to evaluate classification performance. In this paper we present a method that simultaneously accomplishes both tasks. The key step is to standardize markers relative to the non-diseased population before including them in the logistic regression model. Among the advantages of this method are: (i) ensuring that results from regression and performance assessments are consistent with each other; (ii) allowing covariate adjustment and covariate effects on ROC curves to be handled in a familiar way, and (iii) providing a mechanism to incorporate important assumptions about structure in the ROC curve into the fitted risk model. We develop the method in detail for the problem of combining biomarker datasets derived from multiple studies, populations or biomarker measurement platforms, when ROC curves are similar across data sources. The methods are applicable to both cohort and case-control sampling designs. The dataset motivating this application concerns Prostate Cancer Antigen 3 (PCA3) for diagnosis of prostate cancer in patients with or without previous negative biopsy where the ROC curves for PCA3 are found to be the same in the two populations. Estimated constrained maximum likelihood and empirical likelihood estimators are derived. The estimators are compared in simulation studies and the methods are illustrated with the PCA3 dataset.

SUBMITTER: Huang Y

PROVIDER: S-EPMC3817965 | biostudies-literature | 2013 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

LOGISTIC REGRESSION ANALYSIS WITH STANDARDIZED MARKERS.

Huang Ying Y Pepe Margaret S MS Feng Ziding Z

The annals of applied statistics 20130901 3

Two different approaches to analysis of data from diagnostic biomarker studies are commonly employed. Logistic regression is used to fit models for probability of disease given marker values while ROC curves and risk distributions are used to evaluate classification performance. In this paper we present a method that simultaneously accomplishes both tasks. The key step is to standardize markers relative to the non-diseased population before including them in the logistic regression model. Among ...[more]

PMID: 24204441

Similar Datasets

Project description:BACKGROUND:Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is a complex disease, whose exact cause remains unclear. A wide range of risk factors has been proposed that helps understanding potential disease pathogenesis. However, there is little consistency for many risk factor associations, thus we undertook an exploratory study of risk factors using data from the UK ME/CFS Biobank participants. We report on risk factor associations in ME/CFS compared with multiple sclerosis participants and healthy controls. METHODS:This was a cross-sectional study of 269 people with ME/CFS, including 214 with mild/moderate and 55 with severe symptoms, 74 people with multiple sclerosis (MS), and 134 healthy controls, who were recruited from primary and secondary health services. Data were collected from participants using a standardised written questionnaire. Data analyses consisted of univariate and multivariable regression analysis (by levels of proximity to disease onset). RESULTS:A history of frequent colds (OR = 8.26, P <= 0.001) and infections (OR = 25.5, P = 0.015) before onset were the strongest factors associated with a higher risk of ME/CFS compared to healthy controls. Being single (OR = 4.41, P <= 0.001), having lower income (OR = 3.71, P <= 0.001), and a family history of anxiety is associated with a higher risk of ME/CFS compared to healthy controls only (OR = 3.77, P < 0.001). History of frequent colds (OR = 6.31, P < 0.001) and infections before disease onset (OR = 5.12, P = 0.005), being single (OR = 3.66, P = 0.003) and having lower income (OR = 3.48, P = 0.001), are associated with a higher risk of ME/CFS than MS. Severe ME/CFS cases were associated with lower age of ME/CFS onset (OR = 0.63, P = 0.022) and a family history of neurological illness (OR = 6.1, P = 0.001). CONCLUSIONS:Notable differences in risk profiles were found between ME/CFS and healthy controls, ME/CFS and MS, and mild-moderate and severe ME/CFS. However, we found some commensurate overlap in risk associations between all cohorts. The most notable difference between ME/CFS and MS in our study is a history of recent infection prior to disease onset. Even recognising that our results are limited by the choice of factors we selected to investigate, our findings are consistent with the increasing body of evidence that has been published about the potential role of infections in the pathogenesis of ME/CFS, including common colds/flu.

Dataset Information

LOGISTIC REGRESSION ANALYSIS WITH STANDARDIZED MARKERS.

Publications

LOGISTIC REGRESSION ANALYSIS WITH STANDARDIZED MARKERS.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets