A score test for comparing cross-sectional survival data with a fraction of non-susceptible patients and its application in clinical immunology

In cross-sectional studies of time-to-event data collected by patient examinations at a single random point in time, a fraction of them will not experience the event regardless of the length of the follow-up time. This is the case in clinical immunology studies that include a mixed population, with...

Full description

Saved in:
Bibliographic Details
Published in:PloS one 2017-06, Vol.12 (6), p.e0179896-e0179896
Main Authors: Jonas, Sarah Flora, Mbogning, Cyprien, Hässler, Signe, Broët, Philippe
Format: Article
Language:English
Subjects:
Biology and Life Sciences
Biometrics
Cross-Sectional Studies
Data processing
Development and progression
Female
Genetic aspects
Humans
Immune response
Immunogenicity
Immunoglobulins
Immunology
Interferon
Male
Medicine and Health Sciences
Methods
Models, Theoretical
Multiple sclerosis
Multiple Sclerosis - immunology
Multiple Sclerosis - pathology
Null hypothesis
Patients
Physical Sciences
Physiological aspects
Research and Analysis Methods
Sampling methods
Simulation
Statistical analysis
Statistics
Survival
Survival Analysis
Test procedures
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In cross-sectional studies of time-to-event data collected by patient examinations at a single random point in time, a fraction of them will not experience the event regardless of the length of the follow-up time. This is the case in clinical immunology studies that include a mixed population, with both immune-reactive and immune-tolerant (or non-susceptible) patients. In these cases, classical tests of current status data may perform poorly. New methods for testing these data are needed. In the two-sample comparison setting, we propose a score test for testing the null hypothesis that survival does not differ in either the non-susceptible fraction or the time-to-event distribution among the susceptible fraction. In a wide range of scenarios, simulation results show interesting improvements in power for the proposed score test compared to the logrank-type test in most of the configurations we investigated. In a cross-sectional study of drug immunogenicity among treated multiple sclerosis patients, the proposed score test reveals that gender is associated with the immunogenicity of interferon.
ISSN:1932-6203
1932-6203
DOI:10.1371/journal.pone.0179896