To describe and demonstrate methods for analyzing correlated binary eye data.We describe non-model based (McNemar's test, Cochran-Mantel-Haenszel test) and model-based methods (generalized linear mixed effects model, marginal model) for analyses involving both eyes. These methods were applied to: (1) CAPT (Complications of Age-related Macular Degeneration Prevention Trial) where one eye was treated and the other observed (paired design); (2) ETROP (Early Treatment for Retinopathy of Prematurity) where bilaterally affected infants had one eye treated conventionally and the other treated early and unilaterally affected infants had treatment assigned randomly; and (3) AREDS (Age-Related Eye Disease Study) where treatment was systemic and outcome was eye-specific (both eyes in the same treatment group).In the CAPT (n = 80), treatment group (30% vision loss in treated vs. 44% in observed eyes) was not statistically significant (p = 0.07) when inter-eye correlation was ignored, but was significant (p = 0.01) with McNemar's test and the marginal model. Using standard logistic regression for unfavorable vision in ETROP, standard errors and p-values were larger for person-level covariates and were smaller for ocular covariates than using models accounting for inter-eye correlation. For risk factors of geographic atrophy in AREDS, two-eye analyses accounting for inter-eye correlation yielded more power than one-eye analyses and provided larger standard errors and p-values than invalid two-eye analyses ignoring inter-eye correlation.Ignoring inter-eye correlation can lead to larger p-values for paired designs and smaller p-values when both eyes are in the same group. Marginal models or mixed effects models using the eye as the unit of analysis provide valid inference.
Description
Tutorial on Biostatistics: Statistical Analysis for Correlated Binary Eye Data
%0 Journal Article
%1 Ying:2018:Ophthalmic-Epidemiol:28532207
%A Ying, G S
%A Maguire, M G
%A Glynn, R
%A Rosner, B
%D 2018
%J Ophthalmic Epidemiol
%K CorrelatedData RandomEffects gee glmm ophthalmology sas statistics
%N 1
%P 1-12
%R 10.1080/09286586.2017.1320413
%T Tutorial on Biostatistics: Statistical Analysis for Correlated Binary Eye Data
%U https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5986179/
%V 25
%X To describe and demonstrate methods for analyzing correlated binary eye data.We describe non-model based (McNemar's test, Cochran-Mantel-Haenszel test) and model-based methods (generalized linear mixed effects model, marginal model) for analyses involving both eyes. These methods were applied to: (1) CAPT (Complications of Age-related Macular Degeneration Prevention Trial) where one eye was treated and the other observed (paired design); (2) ETROP (Early Treatment for Retinopathy of Prematurity) where bilaterally affected infants had one eye treated conventionally and the other treated early and unilaterally affected infants had treatment assigned randomly; and (3) AREDS (Age-Related Eye Disease Study) where treatment was systemic and outcome was eye-specific (both eyes in the same treatment group).In the CAPT (n = 80), treatment group (30% vision loss in treated vs. 44% in observed eyes) was not statistically significant (p = 0.07) when inter-eye correlation was ignored, but was significant (p = 0.01) with McNemar's test and the marginal model. Using standard logistic regression for unfavorable vision in ETROP, standard errors and p-values were larger for person-level covariates and were smaller for ocular covariates than using models accounting for inter-eye correlation. For risk factors of geographic atrophy in AREDS, two-eye analyses accounting for inter-eye correlation yielded more power than one-eye analyses and provided larger standard errors and p-values than invalid two-eye analyses ignoring inter-eye correlation.Ignoring inter-eye correlation can lead to larger p-values for paired designs and smaller p-values when both eyes are in the same group. Marginal models or mixed effects models using the eye as the unit of analysis provide valid inference.
@article{Ying:2018:Ophthalmic-Epidemiol:28532207,
abstract = {To describe and demonstrate methods for analyzing correlated binary eye data.We describe non-model based (McNemar's test, Cochran-Mantel-Haenszel test) and model-based methods (generalized linear mixed effects model, marginal model) for analyses involving both eyes. These methods were applied to: (1) CAPT (Complications of Age-related Macular Degeneration Prevention Trial) where one eye was treated and the other observed (paired design); (2) ETROP (Early Treatment for Retinopathy of Prematurity) where bilaterally affected infants had one eye treated conventionally and the other treated early and unilaterally affected infants had treatment assigned randomly; and (3) AREDS (Age-Related Eye Disease Study) where treatment was systemic and outcome was eye-specific (both eyes in the same treatment group).In the CAPT (n = 80), treatment group (30% vision loss in treated vs. 44% in observed eyes) was not statistically significant (p = 0.07) when inter-eye correlation was ignored, but was significant (p = 0.01) with McNemar's test and the marginal model. Using standard logistic regression for unfavorable vision in ETROP, standard errors and p-values were larger for person-level covariates and were smaller for ocular covariates than using models accounting for inter-eye correlation. For risk factors of geographic atrophy in AREDS, two-eye analyses accounting for inter-eye correlation yielded more power than one-eye analyses and provided larger standard errors and p-values than invalid two-eye analyses ignoring inter-eye correlation.Ignoring inter-eye correlation can lead to larger p-values for paired designs and smaller p-values when both eyes are in the same group. Marginal models or mixed effects models using the eye as the unit of analysis provide valid inference.},
added-at = {2018-09-27T09:26:22.000+0200},
author = {Ying, G S and Maguire, M G and Glynn, R and Rosner, B},
biburl = {https://www.bibsonomy.org/bibtex/27c2dbd041b71286d1f18c24eab94ec09/jkd},
description = {Tutorial on Biostatistics: Statistical Analysis for Correlated Binary Eye Data},
doi = {10.1080/09286586.2017.1320413},
interhash = {7eaa971ce12ca3d643a98c09d35fae6e},
intrahash = {7c2dbd041b71286d1f18c24eab94ec09},
journal = {Ophthalmic Epidemiol},
keywords = {CorrelatedData RandomEffects gee glmm ophthalmology sas statistics},
month = feb,
number = 1,
pages = {1-12},
pmid = {28532207},
timestamp = {2018-10-03T04:55:42.000+0200},
title = {Tutorial on Biostatistics: Statistical Analysis for Correlated Binary Eye Data},
url = {https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5986179/},
volume = 25,
year = 2018
}