Survival analysis of patients with COVID-19 admitted at six hospitals in Uganda in 2021: a cohort study

Abstract

Background Assessing factors associated with mortality among COVID-19 patients could guide in developing context relevant interventions to mitigate the risk. The study aimed to describe mortality and associated factors among COVID-19 patients admitted at six health facilities in Uganda. Methods We reviewed medical records of patients admitted with COVID-19 between January 1st 2021 and December 31st 2021 in six hospitals in Uganda. Using Stata version 17.0, Kaplan Meier and Cox regression analyses were performed to describe the time to death and estimate associations between various exposures and time to death. Finally, accelerated failure time (AFT) models with a lognormal distribution were used to estimate corresponding survival time ratios. Results Out of the 1040 study participants, 234 (22.5%: 95%CI 12.9 to 36.2%) died. The mortality rate was 30.7 deaths per 1000 person days, 95% CI (26.9 to 35.0). The median survival time was 33 days, IQR (9–82). Factors associated with time to COVID-19 death included; age ≥ 60 years [adjusted hazard ratio (aHR) = 2.4, 95% CI: [1.7, 3.4]], having malaria test at admission [aHR = 2.0, 95% CI:[1.0, 3.9]], a COVID-19 severity score of severe/critical [aHR = 6.7, 95% CI:[1.5, 29.1]] and admission to a public hospital [aHR = 0.4, 95% CI:[0.3, 0.6]]. The survival time of patients aged 60 years or more is estimated to be 63% shorter than that of patients aged less than 60 years [adjusted time ratio (aTR) 0.37, 95% CI 0.24, 0.56]. The survival time of patients admitted in public hospitals was 2.5 times that of patients admitted in private hospitals [aTR 2.5 to 95%CI 1.6, 3.9]. Finally, patients with a severe or critical COVID-19 severity score had 87% shorter survival time than those with a mild score [aTR 0.13, 95% CI 0.03, 0.56]. Conclusion In-hospital mortality among COVID-19 patients was high. Factors associated with shorter survival; age ≥ 60 years, a COVID-19 severity score of severe or critical, and having malaria at admission. We therefore recommend close monitoring of COVID-19 patients that are elderly and also screening for malaria in COVID-19 admitted patients.

COVID-19 and malaria co-infection: a systematic review of clinical outcomes in endemic areas

“…RT-PCR was used to diagnose COVID-19 in all papers, and RDT and blood smears were the preferred investigations for Malaria diagnosis. Three studies used RT-PCR in addition to blood smear ( Konozy et al, 2022 ; Muyinda et al, 2022 ; Caglar et al, 2021 ). Plasmodium falciparum and vivax were identified as causative types in six studies for each, respectively ( Konozy et al, 2022 ; Guha et al, 2021 ; Huang et al, 2022 ; Boonyarangka et al, 2022 ; Forero-Peña et al, 2022 ; Achan et al, 2022 ; Hussein et al, 2022 ; Scalisi et al, 2022 ; Ahmed et al, 2022 ; Caglar et al, 2021 ; Indari et al, 2021 ; Chen et al, 2021 ), while in three studies, both species were detected ( Lee, Hong & Kim, 2022 ; Muyinda et al, 2022 ; Suarez et al, 2022 ).…”

Diabetes is associated with increased risk of death in COVID‐19 hospitalizations in Mexico 2020: A retrospective cohort study

“…1.4, CI: 1-2.2, p < 0.01). 28 In different regions of Brazil in 2020 they found a roughly 10% (p < 0.01) increased prevalence of diabetes among patients with COVID-19 compared to those suggesting diabetes could both increase risk of getting COVID-19 and the outcome associated. 37 There are studies that suggest there can be a bitemporal association between diabetes and COVID- The proposed mechanism of diabetes stems beyond just that of COVID-19, but also includes worsening outcomes and increased contraction of upper respiratory infections (URIs).…”

“…A National Cohort Study in England with ICU patients found 23% increase in mortality among diabetic patients (hazard ratio [HR]: 1.23, 95% CI: 1.14–1.42, p < 0.01) in 2020 29 . A 2021 survival analysis in Uganda found a 22.5% mortality in COVID‐19 patients hospitalized and 40% higher hazard ratios (HR: 1.4, CI: 1–2.2, p < 0.01) 28 . In different regions of Brazil in 2020 they found a roughly 10% ( p < 0.01) increased prevalence of diabetes among patients with COVID‐19 compared to those suggesting diabetes could both increase risk of getting COVID‐19 and the outcome associated 37 …”

“…27 Previous studies have looked at mortality in COVID-19, but have not looked at the specific subset of individuals in Mexico with comorbid diabetes with the level of statistical analyses utilizing both Cox proportional hazards and restricted mean survival time. 11,28,29…”

“…Patients that contract COVID‐19 have a range of symptoms ranging from asymptomatic to death 27 . Previous studies have looked at mortality in COVID‐19, but have not looked at the specific subset of individuals in Mexico with comorbid diabetes with the level of statistical analyses utilizing both Cox proportional hazards and restricted mean survival time 11,28,29 . Proper delineation of the effect size of diabetes can help improve clinical reasoning from physicians and streamline resource allocation.…”