Use of Systemic Antibiotics in Patients with COVID-19 in Colombia: A Cross-Sectional Study

Antibiotics are frequently prescribed to patients with COVID-19. The aim was to determine the pattern of use of systemic antibiotics in a group of patients diagnosed with COVID-19 in Colombia between 2020–2022. This was a descriptive cross-sectional study designed to identify antibiotics prescription patterns for patients diagnosed with COVID-19 treated in eight clinics in Colombia. The AWaRe tool of the World Health Organization (WHO) was used to classify the antibiotics. A total of 10,916 patients were included. The median age was 57 years, and 56.4% were male. A total of 57.5% received antibiotics, especially ampicillin/sulbactam (58.8%) and clarithromycin (47.9%). Most of the antibiotics were classified as Watch (65.1%), followed by Access (32.6%) and Reserve (2.4%). Men (OR: 1.29; 95%CI: 1.17–1.43), older adults (OR: 1.67; 95%CI: 1.48–1.88), patients with dyspnea (OR: 1.26; 95%CI: 1.13–1.41), rheumatoid arthritis (OR: 1.94; 95%CI: 1.17–3.20), and high blood pressure at admission (OR: 1.45; 95%CI: 1.29–1.63), patients treated in-hospital (OR: 5.15; 95%CI: 4.59–5.77), patients admitted to the ICU (OR: 10.48; 95%CI: 8.82–12.45), patients treated with systemic glucocorticoids (OR: 3.60; 95%CI: 3.21–4.03) and vasopressors (OR: 2.10; 95%CI: 1.60–2.75), and patients who received invasive mechanical ventilation (OR: 2.37; 95%CI: 1.82–3.09) were more likely to receive a systemic antibiotic. Most of the patients diagnosed with COVID-19 received antibiotics, despite evidence showing that bacterial coinfection is rare. Antibiotics from the Watch group predominated, a practice that goes against WHO recommendations.


Introduction
Coronavirus disease 2019 , caused by severe acute respiratory syndrome virus type-2 (SARS-CoV-2), was declared a pandemic by the World Health Organization (WHO) in March 2020 [1]. As of 1 November 2022, worldwide, more than 630 million people have been infected, and more than 6.5 million people have died [2]. In Colombia, according to data from the National Institute of Health, more than 6.3 million cases have been reported, and 141 thousand people have died [3]. Most COVID-19 patients have mild or moderate illness [4][5][6], but some require hospitalization for serious respiratory complications, sepsis, acute renal failure, and disseminated intravascular coagulation, among other issues [7].
Antibiotics have been used in hospitals to empirically treat patients with suspected COVID-19, probably because the clinical and radiological characteristics of viral and bacterial pneumonia overlap [8] and initially, some studies (with inadequate methodological designs) showed benefits [9]. The use of antibiotics should ideally be reserved for patients with clinical suspicion or with paraclinical confirmation of bacterial coinfection [4][5][6]. However, an appreciable prescription of antibiotics has been documented among hospitalized patients with COVID-19 [10][11][12], and the problem is much greater in low-and middleincome countries [12]. The high use of antibiotics does not correlate with the relatively low prevalence of bacterial coinfection in patients with COVID-19 [10,11,13].
The improper and excessive use of antimicrobials is the main factor that determines the appearance of drug-resistant pathogens [14]. Therefore, the unnecessary prescription of antimicrobials to patients with COVID-19 is a major global concern due to the increased risk of antibiotic resistance, which will lead to increased morbidity, mortality, and health care and social costs [15]. Antimicrobial resistance will likely become an even more significant challenge in the post-COVID-19 pandemic era [15]. The WHO developed an instrument that seeks to improve the quality of antibiotic prescriptions to decrease the spread of resistant microorganisms and reduce adverse reactions and costs. The AWaRe tool classifies antibiotics into three groups: Access, Watch, and Reserve (Supplementary Table S1). Access antibiotics are those that should initially be used for the most common and severe infections, are narrow spectrum, and are less expensive; Watch antibiotics should be used in moderation due to the relatively high risk of resistant strains; and Reserve antibiotics are to be used for the treatment of infections by microorganisms resistant to multiple antibiotics [16].
In Colombia, a low-to middle-income country, there is limited information available that addresses this problem [17]. The country's health system offers universal coverage to the entire population through two affiliation regimes, i.e., contributory, for which costs are paid by workers and employers, and subsidized, i.e., which insures people without the ability to pay, and includes a benefit plan that involves a significant number of systemic antibiotics and antifungals. The objective of this study was to determine the pattern of use of systemic antibiotics in a group of patients diagnosed with COVID-19 in Colombia between 2020-2022.

Discussion
This study allowed the characterization of the use of systemic antibiotics in a group of patients with a confirmed diagnosis of SARS-CoV-2 infection treated in eight highly complex clinics in Colombia. The median age of these patients was similar to that found in other reports (53.0-61.0 years) [10,[17][18][19], and there was a global predominance of males, as described in other studies (51.5-69.0%) [10,17,18,[20][21][22][23][24]. The most frequently found comorbidities were high blood pressure and diabetes mellitus, findings that are consistent with what has been reported in other studies [17,[19][20][21][23][24][25]. Similarly, the symptom profile was consistent with that described in the literature [17,19,24].
Clinical practice guidelines do not recommend prophylactic antibiotic coverage for patients with COVID-19 [4,5]. Antibiotics are indicated if there is clinical suspicion or laboratory confirmation of bacterial infection [4,5]. Similarly, Colombian guidelines recommend only empirical antibiotic therapy for patients with suspected coinfection by SARS-CoV-2 and bacterial pneumonia [6]. However, it is likely that the extensive empirical use of antibiotics in these patients occurs because the clinical and radiological characteristics of COVID-19 overlap with those of bacterial infections of the respiratory tract [8]. However, receiving antibiotics in the first 48 h after admission increases the risk of developing infections during hospitalization (OR: 11.4; 95% CI: 1.5-85.0; p: <0.01) [28] and of presenting more complications in general (OR: 2.1; 95% CI: 1.8-2.4; p: <0.001) [29].
According to the AWaRe classification devised by the WHO, the antibiotics that were most used in this group of patients belonged to the Watch group. This finding is consistent with those reported in other countries [20,21,25,26,30]. For example, in Pakistan and Bangladesh, the proportion of Watch antibiotics was very high (93.4% and 79.9%, respectively) [20,21]; however, in several European countries and India, the proportion was slightly lower (69.9% and 52.4%, respectively) [25,30], with the latter being consistent with our findings. The use of Access antibiotics ranged from 2.3% and 27.7% [20,21,30]. The objective proposed by the WHO is to increase the proportion of global antibiotic consumption in the Access group by at least 60% and to reduce the use of antibiotics in the Watch and Reserve groups, which are associated with an increased risk of antimicrobial resistance [16]. However, as documented in previous studies [20,21,26,30], these objectives and recommendations have not been met. A systematic review indicated that antibiotic resistance has increased during the COVID-19 pandemic and that this is due to self-medication and the empirical administration and prescription of antibiotics by general practitioners [31].
Different variables were found to be related to the general prescription of antibiotics and the use of Watch/Reserve antibiotics. Some sociodemographic variables, such as being male and of older age, were associated with a greater probability of receiving these drugs, consistent with other studies [11,18,23]. Similarly, the presence of some comorbidities increased the risk, as evidenced in other reports [18,23,27]. Although it has been described that patients with diabetes mellitus and chronic obstructive pulmonary disease have a greater susceptibility to infections, this study did not find an increase in the use of antibiotics in them. This is probably due to the fact that the patients had adequate control of their underlying pathologies or that they were severe cases [32,33]. The prescription of antibiotics was significantly higher for patients treated in-hospital and even higher for those treated in ICUs, as well as for those who needed mechanical ventilation or vasopressor support, findings that are consistent with several studies that show that their use increases based on COVID-19 severity [11,18,20,21]. A risk relationship was also found in those who were receiving systemic glucocorticoids, a finding that is consistent with other research [34,35]. Similarly, the use of antibiotics predominated in the first months of the COVID-19 pandemic, consistent with what was reported by Ramzan et al. in Pakistan, where the use of antibiotics decreased significantly after the second peak of COVID-19 [20], and with what was found by Murillo-Zamora et al. in Mexico, where for each additional week that passed since the beginning of the pandemic, the likelihood of receiving antibiotics decreased by 2.0% [18].
Some limitations should be considered when interpreting the results of this study because access to medical records was not obtained to identify the severity and complications of COVID-19 or other clinical and paraclinical variables. The concomitant presence of other infectious pathologies could not be determined. Similarly, due to the study design, the relevance of antibiotic therapy could not be established, and the drugs prescribed outside the health system or not delivered by the dispensing company that the patients may have received are unknown. However, the study included a significant number of patients distributed in several main cities of the national territory and managed in different care services (emergency, hospitalization, or ICU).

Materials and Methods
This was a cross-sectional study of the prescription patterns of systemic antibiotics used for patients treated in hospitals with a confirmed diagnosis of COVID-19; the patients were identified from reports of confirmed cases by polymerase chain reaction (RT-PCR) or antigen tests performed by the Ospedale Group network in eight clinics located in the cities of Armenia, Barranquilla, Bogotá, Cali, Cartagena, Manizales, Pereira, and Popayán in Colombia. These are highly complex clinics that are reference centers and also treat patients referred from other cities.
From this population, patients of any age, gender, and city of residence with a first confirmed diagnosis of SARS-CoV-2 between March 6, 2020, and August 31, 2022, and treated in the emergency department, a general ward, or the ICU, were selected. With this selection, information on the use of medications was obtained through the dispensing company (Audifarma SA, Pereira, Colombia). A database was designed that allowed the following groups of patient variables to be collected: The protocol was approved by the Bioethics Committee of the Universidad Tecnológica de Pereira (Technological University of Pereira) in the category of "research without risk" (approval code: 30-070421). The principles of confidentiality of information established by the Declaration of Helsinki were respected.
The data were analyzed using the statistical package SPSS Statistics, version 26.0 for Windows (IBM, Armonk, NY, USA). A descriptive analysis was performed; qualitative variables are presented as frequencies and proportions, and quantitative variables are presented as measures of central tendency and dispersion (medians and interquartile ranges). Quantitative data were compared using the Mann-Whitney U test and categorical data were compared using X 2 test or Fisher's exact test. Multivariate binary logistic regression models were developed, which included the associated variables in the bivariate analyses as well as variables with sufficient plausibility or a reported association to identify variables that are associated with the prescription of antibiotics (yes/no) and the prescription of antibiotics from the Watch/Reserve categories (yes/no). A p < 0.05 was considered statistically significant.

Conclusions
With the findings of this study, it can be concluded that more than half of the patients diagnosed with COVID-19 received antibiotics, despite evidence indicating that bacterial coinfection is not frequent. Most of the antibiotics used were from the Watch group, a practice that goes against the WHO recommendations. The high consumption of antibiotics found among patients with COVID-19 requires the implementation of appropriate antimicrobial administration interventions. The administration of antibiotics does not prevent the unfavorable progression of viral pneumonia and does not decrease mortality. Therefore, the use of antibiotics is not recommended without clear clinical suspicion or confirmation of bacterial co-infection. However, more research is needed to better identify patients with bacterial co-infection and to be able to initiate antimicrobial therapy appropriately.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/antibiotics12020252/s1, Supplementary Table S1: Classification of antibiotics according to pharmacological group and according to AWaRe. Supplementary Table S2: Binary logistic regression of the variables related to receiving antibiotics Watch/Reserve in 10,916 patients with a diagnosis of COVID-19 treated in eight highly complex clinics in Colombia.