Rapid Decision Algorithm for Patient Triage during Ebola Outbreaks

Abstract

The low specificity of Ebola virus disease clinical signs increases the risk for nosocomial transmission to patients and healthcare workers during outbreaks. Reducing this risk requires identifying patients with a high likelihood of Ebola virus infection. Analyses of retrospective data from patients suspected of having Ebola virus infection identified 13 strong predictors and time from disease onset as constituents of a prediction score for Ebola virus disease. We also noted 4 highly predictive variables that could distinguish patients at high risk for infection, independent of their scores. External validation of this algorithm on retrospective data revealed the probability of infection continuously increased with the score.

Keywords: Algorithm; Democratic Republic of the Congo; Ebola; clinical signs; contact; epidemics; exposure; outbreaks; predictors; symptomatology; symptoms; triage.

Figure 1

Performance of a rapid decision algorithm for patient triage during Ebola outbreaks (version 4.2, Ebola virus disease [EVD] prediction score only) for different decision thresholds to predict Ebola infection in a population of EVD-suspected patients in Democratic Republic of the Congo during epidemics in 2018–2019, with and without stratification by time-to-presentation (days). A) Sensitivity; B) specificity; C) positive predictive value; D) negative predictive value; E) positive likelihood ratio; F) negative likelihood ratio.

Figure 2

Classification performance (area under the receiver operating characteristic curve) of a rapid decision algorithm for patient triage during Ebola outbreaks a population of EVD-suspected patients in Democratic Republic of the Congo, during epidemics in 2018–2019. Results were based on data from different decision thresholds to predict Ebola infection (Ebola virus disease prediction score only).

Figure 3

Final triage algorithm (version 4.2) for the evaluation of the likelihood of Ebola infection among EVD-suspected patients in Democratic Republic of the Congo, during epidemics in 2018–2019. Three risk categories of Ebola infection are defined: A category (low-risk), B category (intermediate-risk), or C category (high-risk). Left: 4 priority variables. Right: 2 sets of individual scores for each of the 13 variables, to be chosen according to time-to-presentation (thresholds at the top of columns). For methods of employing algorithm, see Appendix. EVD, Ebola virus disease; EVD+, EVD-positive.

Figure 4

EVD confirmation rates (blue line) and number of patients (bars) classified by EVD prediction score obtained by a rapid decision algorithm for patient triage during Ebola outbreaks used in a population of EVD-suspected patients in Democratic Republic of the Congo during epidemics in 2018–2019. EVD, Ebola virus disease; EVD+, EVD-positive.

Figure 5

Distribution of Ebola GP cycle threshold Ct values among EVD-positive patients and averages of Ct values (orange line) by EVD prediction score obtained by a rapid decision algorithm for patient triage during Ebola outbreaks used in a population of EVD-suspected patients in Democratic Republic of the Congo during epidemics in 2018–2019. Box plots indicate medians (horizontal black lines), interquartile range (box tops and bottoms), and 95% CIs (error bars); black dots indicate outliers. Ct, cycle threshold; EVD, Ebola virus disease; GP, glycoprotein.