Case fatality ratio estimates for the 2013 – 2016 West African Ebola epidemic: application of Boosted Regression Trees for imputation
File(s)
Author(s)
Forna, Alpha
Nouvellet, Pierre
Dorigatti, Ilaria
Donnelly, Christl
Type
Journal Article
Abstract
Background
The 2013-2016 West African Ebola epidemic has been the largest to date with more than 11,000 deaths in the affected countries. The data collected have provided more insight than ever before into the case fatality ratio (CFR) and how it varies with age and other characteristics. However, the accuracy and precision of the naïve CFR remain limited because 44% of survival outcomes were unreported.
Methods
Using a Boosted Regression Tree (BRT) model, we imputed survival outcomes (i.e. survival or death) when unreported, corrected for model imperfection to estimate the CFR without imputation, with imputation and adjusted with imputation. The method allowed us to further identify and explore relevant clinical and demographic predictors of the CFR.
Results
The out-of-sample performances of our model were good: sensitivity=69.7% (95% CI 52.5%-75.6%), specificity=69.8% (95% CI 54.1%-75.6%), percentage correctly classified=69.9% (95% CI 53.7%-75.5%) and area under the ROC curve= 76.0% (95% CI 56.8%-82.1%). The adjusted CFR estimates for the 2013-2016 West African epidemic were 82.8% (95% CI 45%.6-85.6%) overall and 89.1% (95% CI 40.8%-91.6%) , 65.6% (95% CI 61.3%-69.6%) and 79.2% (95% CI 45.4-84.1) for Sierra Leone, Guinea and Liberia, respectively. We found that district, hospitalisation status, age, case classification and quarter explained 93.6% of the variance in the naïve CFR.
Conclusions
The adjusted CFR estimates improved the naïve CFR estimates obtained without imputation and were more representative. Used in conjunction with other resources, adjusted estimates will inform public health contingency planning for future Ebola epidemic, and help better allocate resources and evaluate the effectiveness of future inventions.
The 2013-2016 West African Ebola epidemic has been the largest to date with more than 11,000 deaths in the affected countries. The data collected have provided more insight than ever before into the case fatality ratio (CFR) and how it varies with age and other characteristics. However, the accuracy and precision of the naïve CFR remain limited because 44% of survival outcomes were unreported.
Methods
Using a Boosted Regression Tree (BRT) model, we imputed survival outcomes (i.e. survival or death) when unreported, corrected for model imperfection to estimate the CFR without imputation, with imputation and adjusted with imputation. The method allowed us to further identify and explore relevant clinical and demographic predictors of the CFR.
Results
The out-of-sample performances of our model were good: sensitivity=69.7% (95% CI 52.5%-75.6%), specificity=69.8% (95% CI 54.1%-75.6%), percentage correctly classified=69.9% (95% CI 53.7%-75.5%) and area under the ROC curve= 76.0% (95% CI 56.8%-82.1%). The adjusted CFR estimates for the 2013-2016 West African epidemic were 82.8% (95% CI 45%.6-85.6%) overall and 89.1% (95% CI 40.8%-91.6%) , 65.6% (95% CI 61.3%-69.6%) and 79.2% (95% CI 45.4-84.1) for Sierra Leone, Guinea and Liberia, respectively. We found that district, hospitalisation status, age, case classification and quarter explained 93.6% of the variance in the naïve CFR.
Conclusions
The adjusted CFR estimates improved the naïve CFR estimates obtained without imputation and were more representative. Used in conjunction with other resources, adjusted estimates will inform public health contingency planning for future Ebola epidemic, and help better allocate resources and evaluate the effectiveness of future inventions.
Date Issued
2020-06-15
Date Acceptance
2019-07-17
Citation
Clinical Infectious Diseases, 2020, 70 (12), pp.2476-2483
ISSN
1058-4838
Publisher
Oxford University Press (OUP)
Start Page
2476
End Page
2483
Journal / Book Title
Clinical Infectious Diseases
Volume
70
Issue
12
Copyright Statement
© The Author(s) 2019. Published by Oxford University Press for the Infectious Diseases Society of America.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
Sponsor
Medical Research Council (MRC)
National Institute for Health Research
Medical Research Council (MRC)
Grant Number
MR/R015600/1
HPRU-2012-10080
MR/K010174/1B
Subjects
imputation
infectious disease epidemiology
machine learning
survival
viral hemorrhagic disease
06 Biological Sciences
11 Medical and Health Sciences
Microbiology
Publication Status
Published
Date Publish Online
2019-07-22