Dataset Information

Nonparametric estimation of distributions and diagnostic accuracy based on group-tested results with differential misclassification.

ABSTRACT: This article concerns the problem of estimating a continuous distribution in a diseased or nondiseased population when only group-based test results on the disease status are available. The problem is challenging in that individual disease statuses are not observed and testing results are often subject to misclassification, with further complication that the misclassification may be differential as the group size and the number of the diseased individuals in the group vary. We propose a method to construct nonparametric estimation of the distribution and obtain its asymptotic properties. The performance of the distribution estimator is evaluated under various design considerations concerning group sizes and classification errors. The method is exemplified with data from the National Health and Nutrition Examination Survey study to estimate the distribution and diagnostic accuracy of C-reactive protein in blood samples in predicting chlamydia incidence.

SUBMITTER: Zhang W

PROVIDER: S-EPMC8581970 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Nonparametric estimation of distributions and diagnostic accuracy based on group-tested results with differential misclassification.

Zhang Wei W Liu Aiyi A Li Qizhai Q Albert Paul S PS

Biometrics 20200305 4

This article concerns the problem of estimating a continuous distribution in a diseased or nondiseased population when only group-based test results on the disease status are available. The problem is challenging in that individual disease statuses are not observed and testing results are often subject to misclassification, with further complication that the misclassification may be differential as the group size and the number of the diseased individuals in the group vary. We propose a method t ...[more]

PMID: 32083733

Similar Datasets

Project description:Purpose To assess the accuracy of refraction measurements by ClickCheckTM compared with the standard practice of subjective refraction at a tertiary level eye hospital. Design Diagnostic accuracy trial. Methods All participants, recruited consecutively, underwent auto-refraction (AR) and subjective refraction (SR) followed by refraction measurement using ClickCheckTM (CR) by a trained research assistant. Eyeglass prescriptions generated using ClickCheckTM and the resulting visual acuity (VA) was compared to SR for accuracy. Inter-rater reliability and agreement were determined using Intra-class correlation and Bland Altman analysis respectively. Results The 1,079 participants enrolled had a mean (SD) age of 39.02 (17.94) years and 56% were women. Overall, 45.3% of the participants had refractive error greater than ±0.5D. The mean (SD) spherical corrections were -0.66D (1.85) and -0.89D (2.20) in SR and CR respectively. There was high level of agreement between the spherical power measured using SR and CR (ICC: 0.940 (95% CI: 0.933 to 0.947). For the assessment of cylindrical correction, there was moderate level of agreement between SR and CR (ICC: 0.493 (0.100 to 0.715). There was moderate level of agreement between the VA measurements performed by using corrections from SR and CR (ICC: 0.577 (95% CI: 0.521–0.628). The subgroup analysis based on the age categories also showed high level of agreement for spherical corrections between the two approaches (ICC: 0.900). Bland Altman analysis showed good agreement for spherical corrections between SR and CR (Mean difference: 0.224D; 95% LoA: -1.647 D to 2.096 D) without evidence of measurement bias. Conclusions There was a high level of agreement for spherical power measurement between CR and SR. However, improvements are needed in order to accurately assess the cylindrical power. Being a portable, low-cost and easy-to-use refraction device, ClickCheckTM can be used for first level assessment of refractive errors, thereby enhancing the efficiency of refractive services, especially in low- and-middle-income countries.

Dataset Information

Nonparametric estimation of distributions and diagnostic accuracy based on group-tested results with differential misclassification.

Publications

Nonparametric estimation of distributions and diagnostic accuracy based on group-tested results with differential misclassification.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets