Predicting Mortality in Hospitalized COVID-19 Patients in Zambia: An Application of Machine Learning.

The coronavirus disease 2019 (COVID-19) has wreaked havoc globally, resulting in millions of cases and deaths. The objective of this study was to predict mortality in hospitalized COVID-19 patients in Zambia using machine learning (ML) methods based on factors that have been shown to be predictive of mortality and thereby improve pandemic preparedness. This research employed seven powerful ML models that included decision tree (DT), random forest (RF), support vector machines (SVM), logistic regression (LR), Naïve Bayes (NB), gradient boosting (GB), and XGBoost (XGB). These classifiers were trained on 1,433 hospitalized COVID-19 patients from various health facilities in Zambia. The performances achieved by these models were checked using accuracy, recall, F1-Score, area under the receiver operating characteristic curve (ROC_AUC), area under the precision-recall curve (PRC_AUC), and other metrics. The best-performing model was the XGB which had an accuracy of 92.3%, recall of 94.2%, F1-Score of 92.4%, and ROC_AUC of 97.5%. The pairwise Mann-Whitney U-test analysis showed that the second-best model (GB) and the third-best model (RF) did not perform significantly worse than the best model (XGB) and had the following: GB had an accuracy of 91.7%, recall of 94.2%, F1-Score of 91.9%, and ROC_AUC of 97.1%. RF had an accuracy of 90.8%, recall of 93.6%, F1-Score of 91.0%, and ROC_AUC of 96.8%. Other models showed similar results for the same metrics checked. The study successfully derived and validated the selected ML models and predicted mortality effectively with reasonably high performance in the stated metrics. The feature importance analysis found that knowledge of underlying health conditions about patients' hospital length of stay (LOS), white blood cell count, age, and other factors can help healthcare providers offer lifesaving services on time, improve pandemic preparedness, and decongest health facilities in Zambia and other countries with similar settings.

First 100 Persons with COVID-19 - Zambia, March 18-April 28, 2020.

Zambia is a landlocked, lower-middle income country in southern Africa, with a population of 17 million (1). The first known cases of coronavirus disease 2019 (COVID-19) in Zambia occurred in a married couple who had traveled to France and were subject to port-of-entry surveillance and subsequent remote monitoring of travelers with a history of international travel for 14 days after arrival. They were identified as having suspected cases on March 18, 2020, and tested for COVID-19 after developing respiratory symptoms during the 14-day monitoring period. In March 2020, the Zambia National Public Health Institute (ZNPHI) defined a suspected case of COVID-19 as 1) an acute respiratory illness in a person with a history of international travel during the 14 days preceding symptom onset; or 2) acute respiratory illness in a person with a history of contact with a person with laboratory-confirmed COVID-19 in the 14 days preceding symptom onset; or 3) severe acute respiratory illness requiring hospitalization; or 4) being a household or close contact of a patient with laboratory-confirmed COVID-19. This definition was adapted from World Health Organization (WHO) interim guidance issued March 20, 2020, on global surveillance for COVID-19 (2) to also include asymptomatic contacts of persons with confirmed COVID-19. Persons with suspected COVID-19 were identified through various mechanisms, including port-of-entry surveillance, contact tracing, health care worker (HCW) testing, facility-based inpatient screening, community-based screening, and calls from the public into a national hotline administered by the Disaster Management and Mitigation Unit and ZNPHI. Port-of-entry surveillance included an arrival screen consisting of a temperature scan, report of symptoms during the preceding 14 days, and collection of a history of travel and contact with persons with confirmed COVID-19 in the 14 days before arrival in Zambia, followed by daily remote telephone monitoring for 14 days. Travelers were tested for SARS-CoV-2, the virus that causes COVID-19, if they were symptomatic upon arrival or developed symptoms during the 14-day monitoring period. Persons with suspected COVID-19 were tested as soon as possible after evaluation for respiratory symptoms or within 7 days of last known exposure (i.e., travel or contact with a confirmed case). All COVID-19 diagnoses were confirmed using real-time reverse transcription-polymerase chain reaction (RT-PCR) testing (SARS-CoV-2 Nucleic Acid Detection Kit, Maccura) of nasopharyngeal specimens; all patients with confirmed COVID-19 were admitted into institutional isolation at the time of laboratory confirmation, which was generally within 36 hours. COVID-19 patients were deemed recovered and released from isolation after two consecutive PCR-negative test results ≥24 hours apart. A Ministry of Health memorandum was released on April 13, 2020, mandating testing in public facilities of 1) all persons admitted to medical and pediatric wards regardless of symptoms; 2) all patients being admitted to surgical and obstetric wards, regardless of symptoms; 3) any outpatient with fever, cough, or shortness of breath; and 4) any facility or community death in a person with respiratory symptoms, and 5) biweekly screening of all HCWs in isolation centers and health facilities where persons with COVID-19 had been evaluated. This report describes the first 100 COVID-19 cases reported in Zambia, during March 18-April 28, 2020.

First COVID-19 case in Zambia - Comparative phylogenomic analyses of SARS-CoV-2 detected in African countries.

Since its first discovery in December 2019 in Wuhan, China, COVID-19, caused by the novel coronavirus SARS-CoV-2, has spread rapidly worldwide. While African countries were relatively spared initially, the initial low incidence of COVID-19 cases was not sustained for long due to continuing travel links between China, Europe and Africa. In preparation, Zambia had applied a multisectoral national epidemic disease surveillance and response system resulting in the identification of the first case within 48 h of the individual entering the country by air travel from a trip to France. Contact tracing showed that SARS-CoV-2 infection was contained within the patient's household, with no further spread to attending health care workers or community members. Phylogenomic analysis of the patient's SARS-CoV-2 strain showed that it belonged to lineage B.1.1., sharing the last common ancestor with SARS-CoV-2 strains recovered from South Africa. At the African continental level, our analysis showed that B.1 and B.1.1 lineages appear to be predominant in Africa. Whole genome sequence analysis should be part of all surveillance and case detection activities in order to monitor the origin and evolution of SARS-CoV-2 lineages across Africa.