@article{ref1,
title="Imputation of ordinal outcomes: a comparison of approaches in traumatic brain injury",
journal="Journal of neurotrauma",
year="2020",
author="Kunzmann, Kevin and Wernisch, Lorenz and Richardson, Sylvia and Steyerberg, Ewout W. and Lingsma, Hester F. and Ercole, Ari and Maas, Andrew and Menon, David and Wilson, Lindsay",
volume="ePub",
number="ePub",
pages="ePub-ePub",
abstract="Loss to follow-up and missing outcomes data are important issues for longitudinal observational studies and clinical trials in traumatic brain injury. One popular solution to missing 6-month outcomes has been to use the last observation carry forward (LOCF). The purpose of the current study was to compare the performance of model-based single-imputation methods with that of the LOCF approach. We hypothesized that model-based methods would perform better as they potentially make better use of available outcome data. The CENTER-TBI study (n = 4509) included longitudinal outcome collection at 2 weeks, 3 months, 6 months, and 12 months post injury; a total of 8185 GOSe observations were included in the database. We compared imputation of 6-month outcomes using LOCF, a mixed effect model, a Gaussian process regression, and a multi-state model. Model performance was assessed via cross-validation on the subset of individuals with a valid GOSe value within 180 +/- 14 days post-injury (n = 1083). All models were fit on the entire available data after removing the 180 +/- 14 days post-injury observations from the respective test fold. The LOCF method showed lower accuracy (i.e. poorer agreement between imputed and observed values) than model-based methods of imputation, and showed a strong negative bias (i.e. it imputed lower than observed outcomes). Accuracy and bias for the three model-based approaches were similar to one another, with the multi-state model having the best overall performance. All methods of imputation showed variation across different outcome categories, with better performance for more frequent outcomes. We conclude that model-based methods of single imputation have substantial performance advantages over LOCF in addition to providing more complete outcome data.<p /> <p>Language: en</p>",
language="en",
issn="0897-7151",
doi="10.1089/neu.2019.6858",
url="http://dx.doi.org/10.1089/neu.2019.6858"
}