First, confirm that you have the correct template for your paper size. This template has been tailored for output on the custom paper size (21 cm × 28.5 cm).

The study took place at Touba Ndamatou Public Hospital Hospital Establishment (EPS), located in the Touba commune, 47 km away from the Diourbel department and 193 km away from Dakar. Touba’s population, estimated at 1,500,000 in 2018, is growing exponentially due to migration encouraged by the city’s religious character and the free land available for housing (83). The Touba Ndamatou EPS has various departments. The facility did not have an OCC. Patients with COVID-19 were admitted and treated on-site in the emergency, medical, and intensive care departments. Vaccination was available on site.

This was a descriptive retrospective cross-sectional analytical study of patients hospitalized at Touba Ndamatou EPS for COVID-19 from May 1, 2020, to September 30, 2021.

We performed exhaustive sampling. Any patient hospitalized at the Touba Ndamatou EPS during the study period and whose biological confirmation was obtained by a positive COVID-19 PCR and/or positive COVID-19 RDT was included in the study. Medical records of COVID-19 patients with incomplete data were not included.

Data were collected from the medical records of patients with COVID-19 admitted to Touba Ndamatou EPS during the study period. A previously tested and validated pre-established form was used to collect the data.

Data were entered using Epi Info Version 7 software and analyzed using SPSS (Statistical Package for Social Sciences) version 21. In the descriptive analysis, qualitative variables were described by numbers and percentages, and quantitative variables by mean, standard deviation, extremes, and median. The analytical study included univariate and multivariate analysis. The univariate analysis consisted of a comparison between mortality and the other variables. The Chi2 test was used to compare proportions. The difference was statistically significant when the p-value was strictly less than 0.05. For multivariate analysis, we used the binary logistic regression method. All variables with a p-value ≤ 0.25 were used to model mortality. Bottom-up modeling was used. Adjusted ORs with their [95% CI] were determined for each variable retained in the final model. The goodness-of-fit of the model was investigated using the Hosmer and Lemeshow test to verify its adequacy.

The mean age of our study population was 65 years. Onder G et al., in their study of the “Lethality rate and characteristics of patients who died of COVID-19 in Italy”, observed an average age of 52.5 years [7] . Male predominance (65%) was reported by Donamou J. et al., Wu et al., Guan et al. and Zhou et al., respectively 82.3%, 63.7%, 58.1%, and 62% [8] - [11] . The higher frequency of risk factors for disease severity in the male population may explain this male predominance. Contrariwise, research by the Institut National de Santé Publique du Québec (INSPQ) has shown a predominance of women [12] . The most frequently encountered professions were tradesmen (3%) and religious leaders. These are at risk of being exposed to COVID-19 contamination since they are constantly in contact with people who are infected or whose COVID-19 serostatus is unknown. Touba is a deeply religious city, with many visits to religious leaders, and distancing is not often respected. In their study, Lemke C et al. showed that high-exposure occupations were health workers often considered as contact cases [13] .

Only one subject out of one hundred and fourteen (114), representing 0.87%, was vaccinated with certainty. The vaccine received was Astra Zeneca. However, to obtain immunity through vaccination, good vaccine coverage is required to block transmission of the virus, combined with the application of social distancing measures. These results could be explained by the fact that people were reluctant to take part in the vaccination program because of rumors and false beliefs about vaccines. Awareness-raising sessions within and by communities are necessary to achieve the objective of vaccination at the local level [14] .

Our study showed that the most frequent reasons for consultation were respiratory difficulties (70%), cough (40%), fever (12%), and diffuse algia (11%). Among the most frequent physical signs, we found pulmonary condensation syndrome at 97%, followed by respiratory distress syndrome at 77%, room air hypoxia at 65%, and fever at 23%. In the work by Waechter C et al. [15] , fever and cough were the main signs found. This difference could be explained by the population’s recourse to self-medication before coming to hospital, as well as the almost systematic use of analgesics such as paracetamol, which also has antipyretic effects that can mask fever. More than two–thirds of individuals had at least one comorbidity, corresponding to 68.15% of cases. Hypertension and diabetes were the most frequent comorbidities, with 35.9% and 15.7% respectively. Ka et al. also found higher proportions, with hypertension and diabetes accounting for 43.7% and 42.7% of cases respectively [16] . This predominance of these two comorbidities has also been outlined by several authors [16] - [18] . Simple to moderate forms (68%) were more frequent. The same observation was made by Desvaux et al., who found that asymptomatic forms, or those with moderate clinical signs, were the most frequent (80%) [19] . COVID-19 is often associated with biological disorders. Hyperleukocytosis was present in twenty-eight percent of patients (28%), and leukopenia in 3%. A study carried out at the OCC of FANN Teaching Hospital Center (CHNU FANN) found a predominantly neutrophilic hyperleukocytosis in 55.7% of cases [16] . Seven percent of patients had thrombocytosis. Ka et al. found a higher proportion, 24.4%, of thrombocytosis [16] . This difference may be explained by the fact that Ka et al.’s study was carried out in patients with pulmonary embolism. Dysnatremia accounted for 36%, 34% of whom had hyponatremia. In a review carried out at the Centre Medical Diamant in Lubumbashi, hypokalemia was by far the most frequent ionic disorder, accounting for 53.6% of cases [20] . A quarter of patients had impaired renal function. Studies carried out in intensive care units found up to 19% of acute kidney injury (AKI), the mechanism of which remains to be clarified [21] . A study by Zaidan et al. showed that 37.5% of cases developed AKI [22] . However, there is a significant risk of these patients hospitalized for COVID-19 developing AKI, including those who had recovered “normal” function on discharge, underlining the value of renal assessment remote from COVID-19 [23] . On chest X-ray, the signs were unspecific, generally presenting as unilateral, sometimes bilateral, diffuse alveolar condensations predominantly basal with interstitial abnormalities. Chest CT scan revealed mainly multifocal, bilateral, asymmetric ground-glass areas, with at least 50% of the lung parenchyma affected in almost all patients. Guan et al. found similar results, with 56.4% of patients presenting with ground-glass images, unilateral alveolar condensation in 41.9% and bilateral condensation in 51.8%, with interstitial abnormalities in 14.7% [9] [24] . Treatment was recorded for all patients, and therapeutic management was based on Ministry of Health recommendations. As in every country in the world, patient management at the start of the pandemic was based on oxygen therapy, prevention of thromboembolic disease, and antibiotic therapy. However, the use of antibiotics has tended to decline, due to the rarity of bacterial superinfections of COVID-19 [19] . In our study, all patients (100%) received azithromycin. Almost all patients (94%) were on oxygen therapy to combat severe acute respiratory syndrome (SARS), the incidence of which is 15% in Covid-19 patients [25] . In addition, between 50 and 85% of patients admitted to emergency departments present with hypoxemia and/or respiratory exhaustion [2] . Consequently, rapid and effective respiratory support can help reduce complications and improve survival in these critically ill patients. Mechanical ventilation for patients with COVID-19 should be managed with lung-protective strategies to minimize lung injuries associated with the ventilator and improve survival. The anti-coagulant treatment and corticosteroid therapy were nearly systematic (93% and 90%, respectively). None of the patients in our study received hydroxychloroquine. There have been several controversies regarding the effectiveness of hydroxychloroquine [26] - [28] . The prophylactic use of hydroxychloroquine had little to no effect on preventing the disease, hospitalizations, or deaths related to COVID-19. Furthermore, the efficacy of hydroxychloroquine was not proven, and its use could lead to QT interval prolongation and an increased risk of cardiac arrest [29] .

The average hospital stay was 4.54 days, with extremes of 1 and 12 days. A study in Algeria by Kefti et al. [30] found a longer hospital stay of 7.3 days. Another study by Touahri et al. [31] found results almost similar to those of Kefti et al., with an average hospital stay of 7.6 days.

A majority (71.9%) of patients had a favorable outcome. Death occurred in 19% of cases. The mortality rate reported by Taieb et al. [17] is slightly higher than ours, with a proportion of 23.04%. Mortality was statistically higher in subjects aged 60 and over, with a p-value of 0.0132. Muller M et al., in their work on “COVID-19 prognostic factors”, confirmed that patients aged 60 and over are more likely to be hospitalized and to die during the course of the disease [32] . Patients over 60 were found to have more severe forms and to die more than others. Several studies have reported that advanced age is the most important predictor of severity and death in patients with COVID-19. Indeed, advanced age can lead to immunodeficiency, as well as a predisposition to the onset of chronic non-communicable diseases (diabetes, hypertension, etc.). Hypoxia or desaturation was also a risk factor for mortality, accounting for 27% of deaths versus 5% in the non-hypoxia group (P = 0.0045). However, in their series, Ouedraogo AR et al. asserted lower results, with desaturation-related mortality of 11.2% [33] . This difference may be linked to the technical resources available in the provinces. Thus, desaturation leads to a more restrictive treatment based on oxygen therapy, but also to longer hospitalization or eventual transfer to OCC. Other factors, such as gender, comorbidities, hypokalemia, and treatment received, were not significantly associated with mortality. However, in most studies advanced age and cardiovascular comorbidities are risk factors for severity and death. It’s important to strengthen the health system by integrating the management of these comorbidities and geriatric syndrome to better prepare providers during other epidemics. Every moment, there is the possibility and risk of exposure to a new pandemic. In addition, the correct filling of files must be reinforced to take advantage of and better understand these emerging transmissible diseases. A uniform computerized file would ensure better data collection.