Predictors for adult COVID-19 hospitalized inpatient mortality rate in North West Ethiopia

Machine learning algorithms for predicting COVID-19 mortality in Ethiopia

BACKGROUND

Coronavirus disease 2019 (COVID-19), a global public health crisis, continues to pose challenges despite preventive measures. The daily rise in COVID-19 cases is concerning, and the testing process is both time-consuming and costly. While several models have been created to predict mortality in COVID-19 patients, only a few have shown sufficient accuracy. Machine learning algorithms offer a promising approach to data-driven prediction of clinical outcomes, surpassing traditional statistical modeling. Leveraging machine learning (ML) algorithms could potentially provide a solution for predicting mortality in hospitalized COVID-19 patients in Ethiopia. Therefore, the aim of this study is to develop and validate machine-learning models for accurately predicting mortality in COVID-19 hospitalized patients in Ethiopia.

METHODS

Our study involved analyzing electronic medical records of COVID-19 patients who were admitted to public hospitals in Ethiopia. Specifically, we developed seven different machine learning models to predict COVID-19 patient mortality. These models included J48 decision tree, random forest (RF), k-nearest neighborhood (k-NN), multi-layer perceptron (MLP), Naïve Bayes (NB), eXtreme gradient boosting (XGBoost), and logistic regression (LR). We then compared the performance of these models using data from a cohort of 696 patients through statistical analysis. To evaluate the effectiveness of the models, we utilized metrics derived from the confusion matrix such as sensitivity, specificity, precision, and receiver operating characteristic (ROC).

RESULTS

The study included a total of 696 patients, with a higher number of females (440 patients, accounting for 63.2%) compared to males. The median age of the participants was 35.0 years old, with an interquartile range of 18-79. After conducting different feature selection procedures, 23 features were examined, and identified as predictors of mortality, and it was determined that gender, Intensive care unit (ICU) admission, and alcohol drinking/addiction were the top three predictors of COVID-19 mortality. On the other hand, loss of smell, loss of taste, and hypertension were identified as the three lowest predictors of COVID-19 mortality. The experimental results revealed that the k-nearest neighbor (k-NN) algorithm outperformed than other machine learning algorithms, achieving an accuracy of 95.25%, sensitivity of 95.30%, precision of 92.7%, specificity of 93.30%, F1 score 93.98% and a receiver operating characteristic (ROC) score of 96.90%. These findings highlight the effectiveness of the k-NN algorithm in predicting COVID-19 outcomes based on the selected features.

CONCLUSION

Our study has developed an innovative model that utilizes hospital data to accurately predict the mortality risk of COVID-19 patients. The main objective of this model is to prioritize early treatment for high-risk patients and optimize strained healthcare systems during the ongoing pandemic. By integrating machine learning with comprehensive hospital databases, our model effectively classifies patients' mortality risk, enabling targeted medical interventions and improved resource management. Among the various methods tested, the K-nearest neighbors (KNN) algorithm demonstrated the highest accuracy, allowing for early identification of high-risk patients. Through KNN feature identification, we identified 23 predictors that significantly contribute to predicting COVID-19 mortality. The top five predictors are gender (female), intensive care unit (ICU) admission, alcohol drinking, smoking, and symptoms of headache and chills. This advancement holds great promise in enhancing healthcare outcomes and decision-making during the pandemic. By providing services and prioritizing patients based on the identified predictors, healthcare facilities and providers can improve the chances of survival for individuals. This model provides valuable insights that can guide healthcare professionals in allocating resources and delivering appropriate care to those at highest risk.

COVID-19 mortality rate and its determinants in Ethiopia: a systematic review and meta-analysis

Background

The COVID-19 mortality rate continues to be high in low-income countries like Ethiopia as the new variant's transmission expands and the countries' limited capacity to combat the disease causes severe outcomes, including deaths. The aim of this study is to determine the magnitude of the COVID-19 mortality rate and its determinants in Ethiopia.

Methods

The main electronic databases searched were PubMed, CINAHL, Google Scholar, and African journals online. The included studies' qualities were assessed independently using the Newcastle-Ottawa scale. The data was extracted in Microsoft Excel spreadsheet format. The pooled effect size and odds ratios with 95% confidence intervals across studies were determined using the random-effects model. I2 is used to estimate the percentage of overall variation across studies due to heterogeneity. Egger's test and funnel plot were used to find the published bias. A subgroup analysis was conducted. The effect of a single study on the overall estimation was determined by sensitivity analysis.

Results

A total of 21 studies with 42,307 study participants were included in the final analysis. The pooled prevalence of COVID-19 mortality was 14.44% (95% CI: 10.35-19.08%), with high significant heterogeneity (I2 = 98.92%, p < 0.001). The risk of mortality from COVID-19 disease was higher for patients with comorbidity (AHR = 1.84, 95% CI: 1.13-2.54) and cardiovascular disease (AHR = 2, 95% CI: 1.09-2.99) than their counterparts without these conditions.

Conclusion

A significant number of COVID-19 patients died in Ethiopia. COVID-19 patients with comorbidities, particularly those with cardiovascular disease, should receive special attention to reduce COVID-19 mortality.

Systematic review registration

https://www.crd.york.ac.uk/PROSPERO/, registration identifier (ID) CRD42020165740.

A Retrospective Study on the Outcome of Coronavirus Disease 2019 (COVID-19) Patients Admitted to a District General Hospital and Predictors of High Mortality

BACKGROUND

The clinical features and severity of coronavirus disease 2019 (COVID-19) vary between patients and countries. Patients with certain conditions are predisposed to poor outcomes compared with those without medical conditions, such as diabetes, dementia, and hypertension (HTN).

METHODS

The aim of this retrospective study was to assess factors associated with higher mortality in patients with COVID-19 infections and to identify the reason for hospital admission in these patients. The study was performed on patients admitted between 1 and 31 March 2020. Data collection was done retrospectively from electronic medical records.

RESULTS

There were 269 patient admissions during this period, of which 147 were included in this audit. The mean age of COVID-19-positive patients was 62.8 years and 65.9 years for COVID-19-negative patients during this period. Forty-seven patients requiring hospital admission were COVID-19 positive and 93 were COVID-19 negative. There were no COVID-19 swabs in the seven patients included in the audit. Approximately 50% of the COVID-19-positive patients presented with fever and shortness of breath (sob), followed by dyspnea and cough (seven patients). The most common comorbidity was HTN, followed by type 2 diabetes mellitus (T2DM) and ischemic heart disease (IHD). The survival rate was 72.3% in COVID-19-positive patients and 80% in COVID-19-negative patients. The average length of stay was 14.4 days for COVID-19-positive survivors compared to 7.8 days for COVID-19-negative survivors. Most patients who tested positive for COVID-19 infection received oseltamivir vaccination and antibiotics. The presence of HTN, diabetes mellitus (DM), age, and organ failure was associated with a high mortality risk.

CONCLUSION

Our study supports the findings of previous studies that diabetes, HTN, coronary artery disease, old age, and organ failure were associated with high mortality in patients admitted to hospitals with COVID-19 infections.

Mortality rate and predictors of COVID-19 inpatients in Ethiopia: a systematic review and meta-analysis

Introduction

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is an extremely rare virus that devastates the economy and claims human lives. Despite countries' urgent and tenacious public health responses to the COVID-19 pandemic, the disease is killing a large number of people. The results of prior studies have not been used by policymakers and programmers due to the presence of conflicting results. As a result, this study was conducted to fill the knowledge gap and develop a research agenda.

Objective

This study aimed to assess the mortality rate and predictors of COVID-19 hospitalized patients in Ethiopia.

Methods

Electronic databases were searched to find articles that were conducted using a retrospective cohort study design and published in English up to 2022. The data were extracted using a Microsoft Excel spreadsheet and exported to StataTM version 17.0 for further analysis. The presence of heterogeneity was assessed and presented using a forest plot. The subgroup analysis, meta-regression, and publication bias were computed to identify the source of heterogeneity. The pool COVID-19 mortality rate and its predictors were calculated and identified using the random effects meta-analysis model, respectively. The significant predictors identified were reported using a relative risk ratio and 95% confidence interval (CI).

Results

Seven studies with 31,498 participants were included. The pooled mortality rate of COVID-19 was 9.13 (95% CI: 5.38, 12.88) per 1,000 person-days of mortality-free observation. Those study participants who had chronic kidney disease had 2.29 (95% CI: 1.14, 4.60) times higher chance of experiencing mortality than their corresponding counterparts, diabetics had 2.14 (95% CI: 1.22, 3.76), HIV patients had 2.98 (95% CI: 1.26, 7.03), hypertensive patients had 1.63 (95% CI: 1.43, 1.85), and smoker had 2.35 (95% CI: 1.48, 3.73).

Conclusion

COVID-19 mortality rate was high to tackle the epidemic of the disease in Ethiopia. COVID-19 patients with chronic renal disease, diabetes, hypertension, smoking, and HIV were the significant predictors of mortality among COVID-19 patients in Ethiopia. COVID-19 patients with chronic diseases and comorbidities need special attention, close follow-up, and care from all stakeholders.

Polymerase chain reaction-positivity and predictors for SARS-CoV-2 infection among diagnosed cases' in North West Ethiopia

Background

The lack of sophisticated diagnosing equipment for polymerase chain reaction (PCR) during the incidence of variant types of COVID-19 underestimates the morbidity and mortality patterns of this pandemic. Thus, this study aimed to estimate seropositive and confirmatory predictors for COVID-19 suspected and tested cases through polymerase chain reaction (RT-PCR) in two diagnosing.

Methods

A facility-based descriptive cross-sectional study was employed among COVID-19 suspected cases from January 2, 2022, to June 9, 2022. The data were collected both using a structured interviewees and nasopharyngeal (NP) swabs. The nasal swab (NS) was analyzed in the laboratory for RNA detection of the virus using PCR. The collected data were entered into Epi Data version 4.2 and then exported to STATA (SE) version R-14 software for further analysis. multivariable logistic regression was used to assess the associated risk.

Results

A total of 285 suspected cases have participated in this study. The overall mean (±SD) age of the participants was 37.5 (±18.5) years. The majority, 174 (61.1%) of the tested groups were symptomatic when diagnosed. The positivity of RT-PCR for suspected and COVID-19 diagnosed cases were confirmed in 62/285 (21.75%). In multivariable analysis, they were aged 26-50 years (adjusted odds ratio [AOR] = 4.2, 95% confidence interval [CI] = 1.5-10.75), had comorbidity (AOR = 5.8; 95% CI = 2.1-12.2), and cigarette smokers (AOR = 13.5; 95% CI = 5.3-36.6) were significantly associated with developing COVID-19 infection.

Conclusion

More than two in every nine suspected cases were positive RT-PCR tests, and the infectivity of COVID-19 was significantly associated with age 25-50 years, comorbidities, and cigarette smoking. The deployment of high-quality diagnostic kits like RT-PCR is crucial for the early detection and risk stratification of suspected cases.

Incidence of Death and Its Predictors of COVID-19 in Amhara Region, Ethiopia: A Retrospective Follow Up Study

Background

Risk factors associated with COVID-19 incidence of death would aid to notify the most favorable management strategies, hang about undecided, Moreover, studies regarding this issue are limited in Ethiopia and no region-wise study is conducted. Hence, the study investigated the COVID-19 incidence of death and its predictors in the Amhara regional state, Ethiopia.

Methods

A facility-based retrospective survey was conducted at all Amhara regional state COVID-19 treatment centers from 13 March 2020, through 13 January 2022. Epidata version 3.1 and STATA version 14 were used for data entry and analysis, respectively. Linearized survey analysis in a stratified Cox regression model was fitted to identify independent risk factors. P-value with 95% CI for hazard ratio was used for testing the significance at alpha 0.05.

Results

A total of 28,533 study participants were analyzed in this study. Of these, 2873 (11.2%) died and 25,660 (88.8%) were recovered from COVID-19. The death rate was 11.78 per 1000 person-days of observation with a median survival time of 32 days with IQR [12, 44]. Patients with co-morbidities (AHR = 1.54: 95% CI: 1.51–1.55), patients with age <5-year (AHR = 1.69: 95% CI: 1.78–1.81) and patients with age 60+ years (AHR = 2.91: 95% CI: 1.79–3.99), patients with asymptomatic diseases condition (AHR =1.15: 95% CI: 1.01–1.19), and being male (AHR = 1.22: 95% CI: 1.18–1.27) were independent significant risk factors of death from COVID-19.

Conclusion

A relatively high incidence of death from COVID-19 was found in this study. Significant risk factors were identified as patients with age <5 years, patients with age 60+ Years, being male, patients having at least one comorbid condition, and patients with asymptomatic disease conditions. These factors should be taken into consideration for a strategy of quarantining and treating COVID-19 patients.

Malaria serosurvey among acute febrile patients come for health care seeking at the high malaria-endemic setting of North West Ethiopia

Objectives

This study aimed to assess malaria seroprevalence among acute febrile illness cases who come for health care seeking in the high malaria-endemic setting of North West Ethiopia.

Methods

Institutional-based descriptive serosurvey of malaria infections was employed among 18,386 febrile patients from September 2020 to August 2021. Data were entered using Epi Data version 4.2 and exported to STATA (SE) R-14 version statistical software for further analysis. Bi-variable and multivariable regression analyses were conducted to identify malaria infection. Finally, variables with P-value less than 0.05 were considered significant predictors for malaria infection.

Results

The mean (±standard deviation) age of participants was 48.6 (±18.4) years. The overall seroprevalence of malaria infection was estimated as 27.8% (95% confidence interval = 27.2; 28.6, standard error = 0.003). Malaria infection was significantly associated with participants being female (adjusted odds ratio = 2.9; 95% confidence interval = 1.8; 3.7, P = 0.01), age 5-29 years (adjusted odds ratio = 2.2; 95% confidence interval = 1.7; 2.8, P = 0.02), rural (adjusted odds ratio = 3.9; 95% confidence interval = 1.9; 4.4, P = 0.001), and Hgb ⩽11 mg/dL (adjusted odds ratio = 3.4; 95% confidence interval = 1.9; 5.86, P = 0.01).

Conclusion

Nearly every three to ten acute febrile cases were positive for confirmed malaria infection. The risk of malaria infection was significantly associated with respondents being female, aged 5-29 years, rural, and levels of hemoglobin were significantly associated with malaria infection.

COVID-19 Case Fatality Rate and Factors Contributing to Mortality in Ethiopia: A Systematic Review of Current Evidence

Background

The ongoing novel coronavirus disease 2019 (COVID-19) is triggering significant morbidity and mortality due to its contagious nature and absence of definitive management. In Ethiopia, despite a number of primary studies have been conducted to estimate the case fatality rate (CFR) of COVID-19, no review study has attempted to summarize the findings to better understand the nature of pandemics and the virulence of the disease.

Objective

To summarize the CFR of COVID-19 and factors contributing to mortality in Ethiopia.

Methods

PRISMA guideline was followed. PubMed, Science Direct, CINAHL, SCOPUS, Hinari, and Google Scholar were systematically searched using pre-specified keywords. Observational studies ie, cohort, cross-sectional, and case-control studies were included. The Newcastle-Ottawa scale adapted for observational studies was used to assess the quality of included studies. CFR was defined as the proportion of COVID-19 cases with the outcome of death within a given period. Factors contributing to COVID-19 mortality at p-value <0.05 were described narratively from the eligible articles.

Results

A total of 13 observational studies were included in this study. Consequently, this review confirmed the CFR of COVID-19 in Ethiopia ranges between 1–20%. Additionally, comorbid conditions, older age group, male sex, substance use, clinical manifestations (abnormal oxygen saturation level, atypical lymphocyte count, fever, and shortness of breath), disease severity, and history of surgery/trauma increased the likelihood of death from COVID-19 death.

Conclusion

This study shows that the range of CFR of COVID-19 in Ethiopia is almost equivalent to other countries, despite the country’s low testing capacity and case detection rate in reference to its total population. Comorbid diseases, older age group, male sex, cigarette smoking, alcohol drinking, clinical manifestations and disease severity, and history of surgery/trauma were factors contributing to COVID-19 mortality in Ethiopia. Therefore, given the alarming global situation and rapidly evolving large-scale pandemics, urgent interdisciplinary interventions should be implemented in those vulnerable groups to lessen the risk of mortality. Furthermore, the CFR of COVID-19 should be estimated from all treatment and rehabilitation centers in the country, as underestimation could be linked to a lack of preparedness and mitigation. A large set of prospective studies are also compulsory to better understand the CFR of COVID-19 in Ethiopia.