No abstract

Introduction Little data exists from sub-Saharan Africa describing incidence and outcomes of sepsis in emergency units and uncertainty exists surrounding optimal management of sepsis in low-income settings. There exists limited data regarding quality care metrics for non-physician clinicians trained in emergency care. The objective of this study was to describe changes in septic patients over time and evaluate associations between sepsis care and mortality. Methods Secondary analysis of a prospective cohort of all consecutive patients seen from 2010–2019 in a rural Ugandan emergency unit staffed by non-physician clinicians was performed using an electronic database based on paper charts. Sepsis was defined as suspected infection with a quick Sequential Organ Failure Assessment score (qSOFA)≥1. Multi-variable logistic regression was used to analyze three-day mortality. Results Overall, 48,653 patient visits from 2010–2019 yielded 17,490 encounters for patients age≥18 who had suspected infection, including 10,437 with sepsis. The annual proportion of patients with sepsis decreased from 45.0%% to 21.3% and the proportion with malarial sepsis decreased from 17.7% to 2.1% during the study period. Rates of septic patients receiving quality care (“both fluids and anti-infectives”) increased over time (21.2% in 2012 to 32.0% in 2019, p<0.001), but mortality did not significantly improve (4.5% in 2012 to 6.4% in 2019, p = 0.50). The increasing quality of non-physician clinician care was not associated with reduced mortality, and treatment with “both fluids and antibiotics” was associated with increased mortality (RR = 1.55, 95%CI 1.10–2.00). Conclusion The largest study of sepsis management and outcomes ever published in both Uganda and sub-Saharan Africa showed sepsis and malarial sepsis decreasing from 2010 to 2019. The increasing quality of non-physician clinician care did not significantly reduce mortality and treatment with “both fluids and antibiotics” increased mortality. With causal associations between antibiotics and mortality deemed implausible, associations between sepsis mortality and interventions likely represent confounding by indication. Defining optimal sepsis care regionally will likely require randomized controlled trials.

Introduction:Little data exists from sub-Saharan Africa describing incidence and outcomes of sepsis in emergency units and uncertainty exists surrounding optimal management of sepsis in low-income settings. There exists limited data regarding quality care metrics for non-physician clinicians trained in emergency care.Methods:Data were obtained for patients seen from 2010-2019 in a rural Ugandan emergency unit staffed by non-physician clinicians. Sepsis was defined as suspected infection with a qSOFA score ≥ 2. Descriptive analysis was performed and a multi-variable logistic regression mortality model was created. Analysis included Wilcoxon rank-sum test, t-test, one-way ANOVA, and Fisher’s exact test.Results:Overall, 48,653 patient visits from 2010-2019 yielded 17,490 encounters for patients age≥18 who had suspected infection, including 3,323 with sepsis. Overall sepsis incidence from 2010 to 2019 decreased from 16.4% to 4.7%, and malarial sepsis incidence decreased from 4.3% to 0.1%. From 2012 to 2019, the proportion of septic patients receiving quality care (both fluids and anti-infectives) increased from 36.2% to 44.7% but observed mortality rates for non-malarial sepsis increased from 6.3% to 14.9% and predicted mortality rates increased from 8.8% to 12.0%. Higher qSOFA scores were significantly associated with higher rates of both interventions and mortality. All interventions for non-malarial sepsis were independently associated with increased relative risks for death: “fluids alone” RR=1.22 [95%CI 0.57 – 1.87]; “antibiotics alone” RR=1.25 [95%CI 0.60 – 1.91]; “both fluids and antibiotics”: RR=1.85 [95%CI 1.02 – 2.69]. Conclusion: The largest study of sepsis management and outcomes ever published in both Uganda and sub-Saharan Africa showed decreasing incidence, increasing quality of non-physician clinician care and increasing predicted and observed mortality from 2010 to 2019. With causal associations between antibiotics and mortality deemed implausible, associations between sepsis mortality and interventions likely represent confounding by indication. Defining optimal sepsis care regionally will likely require randomized controlled trials.