Defining high-risk emergency chief complaints: data-driven triage for low- and middle-income countries.

OBJECTIVES: Emergency medicine in low- and middle-income countries (LMICs) is hindered by lack of research into patient outcomes. Chief complaints are fundamental to emergency care but have only recently been uniquely codified for an LMIC setting in Uganda. It is not known whether chief complaints independently predict emergency unit patient outcomes. METHODS: Patient data collected in a Ugandan emergency unit between 2009-2018 were randomized into validation and derivation datasets. A recursive partitioning algorithm stratified chief complaints by three-day mortality risk in each group. The process was repeated in 10,000 bootstrap samples to create an averaged risk ranking. Based on this ranking, chief complaints were categorized as "high-risk" (>2x baseline mortality), "medium-risk" (between 2 and 0.5x baseline mortality) and "low-risk" (<0.5x baseline mortality). Risk categories were then included in a logistic regression model to determine if chief complaints independently predicted three-day mortality. RESULTS: Overall, the derivation dataset included 21,953 individuals with 7,313 in the validation dataset. In total, 43 complaints were categorized, and 12 chief complaints were identified as high-risk. When controlled for triage data including age, sex, HIV status, vital signs, level of consciousness, and number of complaints, high-risk chief complaints significantly increased three-day mortality odds (OR 2.39, 95% CI 1.95 - 2.93, p<0.001) while low-risk chief complaints significantly decreased three-day mortality odds (OR 0.16, 95% CI 0.09 - 0.29, p<0.001). CONCLUSIONS: High-risk chief complaints were identified and found to predict increased three-day mortality independent of vital signs and other data available at triage. This list can be used to expand local triage systems and inform emergency training programs. The methodology can be reproduced in other LMIC settings to reflect their local disease patterns.