Le Infezioni in Medicina, n. 1, 86-95, 2022

doi: 10.53854/liim-3001-10

ORIGINAL ARTICLE

Evolution of prescribing practices and outcomes in the COVID-19 pandemic in metropolitan areas

Benjamin Diaczok1, Girish Nair2, Chun-Hui Lin3, James H. Paxton4, Amr Abbas5, Greg Barkley6, Brian O’Neil4, William O’Neil7, Kirit Patel8, Matthew Sims9, Laila Poisson3, Anupam Ashutosh Sule1

1Department of Internal Medicine, St. Joseph Mercy Oakland, Pontiac, USA;
2Department of Pulmonary and Critical Care, Beaumont Health System, Royal Oak, USA;
3Department of Public Health Sciences, Henry Ford Health System, Detroit, USA;
4Department of Emergency Medicine, Wayne State University, Detroit, USA;
5Department of Cardiology, Beaumont Health System, Sterling Heights, USA;
6Department of Neurology, Henry Ford Health System, Detroit, USA;
7Department of Cardiology, Henry Ford Health System, Detroit, USA;
8Department of Cardiology, St. Joseph Mercy Oakland, Pontiac, USA;
9Department of Infectious Diseases, Beaumont Health System, Royal Oak, USA

Article received 16 November 2021, accepted 30 January 2022

Corresponding author

Benjamin Diaczok

E-mail: Benjamin.Diaczok@stjoeshealth.org

SummaRY

Introduction: We wanted to characterize the evolution of the COVID-19 pandemic in a typical metropolitan area.

Methods: Data were extracted from the Detroit COVID-19 Consortium database for hospitalized COVID-19 patients treated in Southeast Michigan over the 12-month period from March 2020 to February 2021. Demographic and outcomes data were compared to CDC data.

Results: A total of 4,775 patients were enrolled during the study period. We divided the pandemic into three phases: Phase-1 (Spring Surge); Phase-2 (Summer Lull); and Phase-3 (Fall Spike). Changes in hydroxychloroquine, remdesivir, corticosteroid, antibiotic and anticoagulant use closely followed publication of landmark studies. Mortality in critically-ill patients decreased significantly from Phase-1 to Phase-3 (60.3% vs. 47.9%, Chisq p=0.0110). Monthly mortality of all hospitalized patients ranged between 14.8% - 21.5% during Phase-1 and 9.7 to 13.4% during Phase 3 (NS).

Discussion: The COVID–19 pandemic presented in three unique phases in Southeast Michigan. Medical systems rapidly modified treatment plans, often preceding CDC and NIH recommendations. Despite improved treatment regimens, intubation rates and mortality for hospitalized patients remained elevated.

Conclusion: Preventive measures aimed at reducing hospitalizations for COVID-19 should be emphasized.

Keywords: COVID-19, pandemic, metropolitan area, prescr1ption.

INTRODUCTION

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic has persisted into a second year. Michigan reported its first two cases of COVID-19 disease on March 10th, 2020, followed by a rapid acceleration of cases in Southeast Michigan taxing regional health care systems [1]. A lull in cases occurred over the summer months, followed by a surge of cases in the fall [2]. At its peak in December 2020, the state experienced more than 200 deaths per day [2]. As of May 13th, 2021, Michigan has reported 969,452 cases of COVID-19 including 19,515 (2.0%) deaths, making Michigan one of the top ten states for COVID-19 mortality since the pandemic began [2]. During the winter of 2020-2021, southeast Michigan remains at high risk, with an average of 26 new cases per 100,000 population daily [2]. Moreover, three counties in Southeast Michigan (Wayne, Oakland, and Macomb) constitute 40% of the state’s new cases. We used southeast Michigan as a representative area of typical metropolitan area as the population demographic mix and the rise and fall of cases was similar in all areas of the US.

As the pandemic evolved, various treatment protocols were proposed in the popular press and medical literature [3-19]. Reports published in the literature regarding the epidemiology, presentation, natural course and complications of COVID-19 disease likely influenced local patterns of medical care everywhere in the world. We sought to determine whether disease demographics evolved over the course of the pandemic, and how advances in medical knowledge affected frontline physician treatment plans and outcomes. Utilizing data derived from the Detroit COVID-19 Consortium database, we assessed treatments and outcomes of patients admitted to consortium hospitals in Southeast Michigan from March 2020 through - February 2021, including a comparison of outcomes during the initial Spring Surge to those during the Fall Spike. We also present data about the evolution of clinical and demographic characteristics of the hospitalized patient population.

PATIENTS AND METHODS

Ethics Statement. Data collection for the consortium registry was approved under IRB#13785 at Henry Ford Health System (HFHS), the Data Coordinating Center for the Detroit COVID-19 Consortium. A consortium agreement and data use agreement are in place between Trinity Health and HFHS. Secondary research for this proposal was approved by Trinity Health under IRB #2020-020. The study did not receive outside funding. Contributing authors reported no conflict of interest.

Patient population. Data were collected from patients with PCR-confirmed COVID-19 admitted to one of four participating hospitals of the Henry Ford Health System or two participating hospitals of Trinity Health. All hospitals are located in Wayne, Oakland or Macomb counties (Southeast Michigan).

Study period. The presented data were collected from March 1, 2020 through February 28, 2021.

Clinical data. Data were abstracted through automated reports of the electronic medical records at contributing hospitals. The two systems submitted Michigan Health Information Network ID numbers (MiHIN) so that data from patients receiving care at both institutions could be linked. The primary data points for this study, collected by all systems, were patient characteristics: gender, primary race (Black, White, Other), age at first admission (categorized as <40, 40-64, 65-80, or >80 years); patient medical history including co-morbidities, history of present illness (NIH illness severity (moderate, severe, critical), Sequential Organ Failure Assessment (SOFA)/ Quick (q)SOFA score, clinical features at presentation); and medication use (antibiotics, anticoagulant, hydroxychloroquine, remdesivir, steroids; etc). We stratified clinical outcomes by severity of illness and the need for mechanical ventilation. The Sequential Organ Failure Assessment (SOFA), qSOFA and a modified version of the NIH severity scale were utilized for stratification of COVID-19 disease severity. For the NIH severity scale, all hospitalized patients were classified as at least moderate severity, patients requiring supplemental oxygen at presentation were considered severe, and patients requiring intubation or vasopressors within 24 hours of admission were considered critical. We also collected data on dates of service (including first admission date (i.e., date of earliest inpatient encounter for COVID-19), dates of admission to the emergency department and inpatient ward, and discharge dates); and patient outcomes: including inpatient death (died inpatient versus discharged alive), ventilator use (yes/no), length of stay (days), discharge location (home/other). Data submitted by each member system to the Data Coordinating Center were harmonized and prepared for analysis.

Data from the CDC public use dataset (COVID-19 Case Surveillance Public Use Data with Geography.csv (N: 27,185,885) was used to show similarity of our patient population with national metropolitan patient population. The inclusion criteria for the CDC data set were as follows: Confirmed cases (N: 24, 481, 394); Hospitalized (N: 1,254, 128); Type 1 and 2 counties as defined by the CDC for considering “metropolitan” area (N:1,042,150); Time period 2020/03-2021/02 (N:917,165).

Data Summaries and Analysis. Monthly summaries classify cases based on the date of first admission. Daily summaries are based on the total hospital census for that day. Categorical data were summarized as percentages and fraction of occurrence. Continuous data were summarized as median with interquartile range (IQR). Categorical variables for spring and fall surges were compared by Chi-square test. Data summaries, analysis, and visualizations were performed with the R programming language (version 4, ggplot2 package) [20].

RESULTS

A total of 4,775 patients were enrolled during the study period. Data were analysed for epidemiology, patient demographics, clinical presentation, compliance with treatment guidelines, and clinical outcomes. The pandemic was divided into three phases: Phase-1 (March through May 2020), representing the “Spring Surge” of cases; Phase-2 (June through October 2020), representing the “Summer Lull”; and Phase-3 (November 2020 through February 2021), representing the “Fall Spike”.

The total number of hospitalizations increased sharply in Phase-1, reaching 1,043 admissions in April 2020. A sharp decline in hospitalizations occurred over the summer months (Phase-2) with monthly admissions ranging from 47 to 294. The Fall Surge (Phase-3) saw monthly admissions peak at 892 in November (Figure 1). Similar increase and decrease were seen in the national metropolitan data.

Figure 1 - Number of cases hospitalized by month in the United States (Left: COVID-19 consortium; Right: CDC).

Patient demographics

Patients aged 65- to 80-years-old accounted for 30 to 40% of hospitalizations during all three phases of the pandemic. Patients aged >80 years-old steadily increased and accounted for between 23.9%-33.7% of patients admitted during Phase 3. Patients under 40 years accounted for 7.7%-19.0% of admissions during Phase 1; 8.8%-26.1% of admissions during Phase 2 and 7.1% - 10.7% of admissions during Phase 3. Females accounted for 47.9%-64.1% of all admissions with only two months falling below 50%. This female predominance was most pronounced during Phase 2. African-Americans accounted for between 42.3%-64.0% of all admissions during Phase 1. This disproportionate rate declined to 44.6%-10.2% over Phase 2 and was more consistent with Southeast Michigan demographics. During Phase 3, the proportion of hospitalized black patients increased again to 16.5%-28.6% (Figure 2). The demographic pattern observed in our patient population was reflected on a national level in the CDC data.

Figure 2 - Demographic profile of hospitalized COVID-19 patients in the United States (Left: COVID-19 consortium; Right: CDC).

Clinical presentation of hospitalized COVID-19 patients

During Phase-1, patients generally presented with shortness of breath (33.8%-59.7%) and/or cough (14.1%-48.9%). These symptoms reached a nadir during Phase-2 with a moderate rebound during Phase-3. Body aches, chills, drenching sweats, headache, nausea and vomiting were reported in approximately 20 percent of patients during all three phases. Loss of taste and smell, while prominent in the popular press, was reported in only 13% of patients. “Asymptomatic” patients reached a peak of 40% of cases during Phase-2 (Table 1).

Patient co-morbidities

The prevalence of co-morbidities, reported monthly, varied during the pandemic. Cardiac co-morbidities included hypertension (53.3%-75.5%) hyperlipidemia (33.3%-46.8%), coronary artery disease (21.7%-34.7%), atrial fibrillation (6.7%-17.3%) and heart failure (8.4%-21.3%). Pulmonary co-morbidities included asthma (7.1%-15.8%), COPD (8.1%-9.4%) and pulmonary embolism (2.1%-7.1%). Other common co-morbidities included diabetes (23.0%-40.8%), cancer (8.5%-23.7%) and chronic kidney disease (8.5%-19.4) (Table 2).

Medical management

The evolution of clinical management was tracked to CDC guidelines, FDA recommendations, clinical trials, and high visibility publications and media announcements. We reviewed the use of hydroxychloroquine, remdesivir, corticosteroids and antibiotics use throughout the study period.

Hydroxychloroquine use rapidly accelerated during Phase-1, reflecting early reports of benefit [5, 10]. This rise preceded the Emergency Use Authorization (EUA) published by the Food & Drug Administration (FDA) on March 28th [9]. At its peak, 63.4% of the patients admitted during Phase-1 received hydroxychloroquine. This spike in treatment plateaued for a week and was followed by a rapid decline which began prior to the FDA revoking EUA on June 15 [10]. During Phases 2 and 3, there was sporadic use of hydroxychloroquine (Figure 3 Panel 1).

Remdesivir was first reported to be of benefit in vitro against COVID-19 in February of 2020 (5). A subsequent clinical trial demonstrated benefit and the FDA released an EUA on May 1st, 2020 [11]. Our data demonstrate very little use of remdesivir during Phase-1 of the pandemic. Use of remdesivir began early in Phase-2 and approached 50% of hospitalized patients by September. This rise appears to precede FDA approval of remdesivir for COVID-19 but was contemporary with published reports of efficacy (Figure 3 Panel 2) (12). Remdesivir use persisted throughout Phase 3.

The potential benefits of corticosteroid use in COVID-19 patients requiring mechanical ventilation or supplemental oxygen was reported by the RECOVERY Collaborative Group in July 2020 [4, 14]. In our study, corticosteroid use was prevalent throughout all three phases. Early in Phase 1, steroid use peaked at 56.1%, declining to 38.5% by April 2020. After the NIH recommended against steroid use, there was a sharp decline of steroids use in May [23]. After the June 16th press release by the RECOVERY trial of a mortality benefit, steroid use increased sharply again to above 50% of the patients, stabilizing during Phase 3 at 50-75% (Figure 3 Panel 3) [15].

Figure 3 - Use of hydroxychloroquine, remdesivir, corticosteroids antibiotics and anticoagulation.

Antibiotic use for COVID-19, particularly azithromycin, was reported to be of benefit in March 2020 [9]. On May 12th, the NIH stated there was insufficient data to support use of antibiotics for COVID-19 patients. Subsequent studies documenting independent use of azithromycin in late-stage hospitalized COVID-19 patients did not show benefit [16-18]. In our study, antibiotics were prescribed in almost half of all patients admitted for COVID-19 during Phase 1. Use rates varied widely during Phase 2 but levelled off to 25-30% of patients in Phase 3 (Figure 3 Panel 4).

Prophylactic and therapeutic anticoagulation of hospitalized patients evolved during the pandemic. As reports of elevated D dimer, COVID associated thromboembolic disease poor prognosis of patients with abnormal coagulation markers and benefit of anticoagulation were reported, the use of anticoagulation rose [24, 25]. During Phase 1 monthly rates of prophylaxis ranged between 25%-66.7%. During Phase 2 prophylactic anticoagulation rose. During Phase 3 monthly rates ranged between 40%-70.4%. Therapeutic doses of anticoagulation started increasing sharply after the report of increased thromboembolism [26]. During Phase 1 between 28.0%-55.2% of the patients received therapeutic doses of anticoagulation. During Phases 2 and 3 anticoagulation fluctuated between 11%-79.5% (Figure 3 Panel 5).

Clinical outcomes

All three severity scales (i.e., SOFA, qSOFA and modified NIH), had similar receiver operating characteristic curves for mortality (AUC, 0.77, 0.74, 0.77, respectively). We used the modified version of the NIH severity scale to display the severity data when comparing the three phases. Patients presenting in a critical state were more common at the start of the pandemic and this declined as the number of cases hospitalized decreased. This number has been increasing steadily as the number of hospitalizations has increased. During Phase-1, monthly intubation rates ranged from 14.1% to 24.8%. Intubation rates fell to of 0.0% in June (Phase-2) but rose steadily over the summer and plateaued at 9.2% in Phase-3 (Figure 4; Table 3).

Figure 4 - Percent of hospitalized patients that were intubated/died (Left: COVID-19 consortium) and percent mortality in all hospitalized patients/ hospitalized patients with known confirmed outcome (excluding patients with unknown or missing mortality status) (Right: CDC data).

Mortality rates paralleled intubation rates. Monthly mortality rates ranged between 14.8%-21.5% of hospitalized patients during Phase-1. Mortality declined to 2.1% early in phase-2 but slowly increased over the summer to 12.6%. By late fall (Phase-3) monthly mortality ranged between 9.7 and 13.4% (Figure 4). A similar decrease followed by increase in mortality is seen in the national CDC data. Mortality rates varied by NIH score. Monthly mortality rates for patients classified moderate (0.0%-13%) to severe (0.0%-17%) COVID remained similar for all three phases. Patients classified as critical had much higher mortality rates. Mortality in critically ill patients rose significantly from Phase-1 to III (47.9% vs 60.3% Chisq p=0.0110) (Figure 5; Table 3).

A large number (>80%) of COVID-19 hospitalization survivors were discharged to sites of care other than home at the onset of the pandemic. As the pandemic progressed, this decreased to 45.6% in May but increased again before stabilizing at a level of approximately 69-86.3% of patients from July onwards.

Figure 5 - Mortality by NIH Severity Score.

DISCUSSION

The Southeast Michigan (SE) counties of Wayne, Oakland, and Macomb have been an epicenter for the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, accounting for more than 40% of the state’s disease burden. A consortium two health care systems involving six regional hospitals collected data from patients with PCR-confirmed COVID-19 over a one-year period. The pandemic was divided into three distinct phases based on number of admissions, patient characteristics, treatments and presentations.

The Spring Surge (Phase-1) was characterized by a large volume of admissions, a high intubation (14.1%-24.8%) and mortality rate (14.8%-21.5%). The Summer Lull (Phase-2) was characterized by few admissions, and low intubations and mortality rates. Phase-3 saw a resurgence of admissions, which did not reach the peak achieved in Phase-1, and a rebound in intubation (8.0%-9.5%) and mortality (9.7%-13.4%) rates. The same pattern was seen in all metropolitan areas captured in the CDC data. This shows that our metropolitan patient population can serve as a representative sample for the national metropolitan population. Thus, the clinical characteristics, comorbidities and medication usage pattern that we observe in our patient population (features that cannot be studied in a national data set) can also be generalized to any metropolitan population in the US (and possibly anywhere in the world). Comorbidities were similar in all three phases [22].

In our data set, we found our most commonly reported symptoms - cough, dyspnoea, and fever - were similar to those reported in larger studies [22]. However, we found that the incidence of all of these common COVID-19 symptoms declined over the study period. Respiratory symptoms became less common in the later months, with an associated increase in the rate of gastrointestinal and systemic symptoms. Less than 13% of patients reported loss of taste or smell in all three Phases. The cause of these trends is unclear. Potential explanations for these findings include surges in different viral variants with a predilection for different organ systems, differences in patient susceptibility to the virus (e.g., patients predisposed to respiratory symptoms contracting the virus earlier in the study period), or differences in symptom reporting by patients or providers [26].

Between 9.9%-40.4% of Phase-2 patients had no symptoms. We believe that this lack of symptoms reflected testing of all hospital admissions, including elective surgeries and patients admitted for non-COVID related illnesses.

Demographics demonstrated slightly more than half of COVID admissions were female. African Americans represented approximately two thirds of admissions during Phase-1. This dropped to less than 20% during phase-2 with a rise to 30% during Phase-3. The cause for these disproportionate representations is not clear and requires further study.

Age distribution remained similar during all three phases with a slight increase in patients over 80 years old during Phase-3.

The reduction in intubation and mortality may be the result of increased comfort and knowledge in management, improved treatment protocols or variation in virus strains. Medical interventions evolved quickly in response to studies published in the medical literature and often preceded CDC and NIH recommendations. This rapid response suggests hospital medical leadership reviewed evolving data and modified treatment protocols based on evidence published in the literature prior to CDC, NIH or FDA recommendations.

In this data set, the qSOFA, SOFA and NIH severity scales appear to be very good predictors of mortality. Intubation and mortality rates for hospitalized patients fell from Phase-1 to Phase-3. Despite this improvement, both remained approximately 10% in Phase-3. Strategies to reduce hospitalizations should be emphasized.

The study strengths include a large number of patients representing diverse ethnic, economic, racial and sex backgrounds. Secondly, both hospital systems utilize EPIC for their electronic health records (EHR). Data were extracted from each institution’s EHR into a common database for analysis.

The study is limited by lack of rural representation. Although the study only includes one metropolitan area the authors believe the findings can be applicable to any typical metropolitan area as the patient characteristics were similar to those in the CDC database. Medical practice in participating hospitals would be typical for what would be seen in a mix of academic and community hospitals anywhere in the country. The database contains patients managed by a large number of independent physicians across multiple hospitals and hence it is generalizable.

In conclusion, the COVID-19 pandemic in Southeast Michigan can be broken down into three distinct phases. Each phase was characterized by unique demographics, presentations, treatment regimens and outcomes. Treatment regimens quickly evolved in response to medical literature and frequently preceded NIH or CDC recommendations.

Conflict of interest

None to declare.

Funding

None.

REFERENCES

[1] Haddad K. Michigan coronavirus timeline: Key dates, COVID-19 case tracking, state orders [Internet]. Detroit (MI): ClickOnDetroit; 2020 April 27. Available at: https://www.clickondetroit.com/health/2020/03/24/michigan-coronavirus-timeline-key-dates-covid-19-case-tracking-state-orders/.

[2] Allen J, Almukhtar S, Aufrichtig A, et al. Tracking coronavirus in Michigan: Latest map and case count [Internet]. New York (NY): The New York Times; 2021 May 13. Available at: https://www.nytimes.com/interactive/2021/us/michigan-covid-cases.html

[3] Libby C. Timeline: Tracking Trump alongside scientific developments on hydroxychloroquine. Available at: ABC News (go.com) /Health/timeline-tracking-trump-alongside-scientific-developments hydroxychloroquine/story?id=72170553.

[4] Roberts M. Coronavirus: Dexamethasone proves first life-saving drug - Available at: BBC News

[5] Chen Jun LD, Chen Jun LD. A pilot study of hydroxychloroquine in treatment of patients with common coronavirus disease-19 (COVID-19). J Zhejiang Univ Med Sci. 2020; 49 (1), 0-0.

[6] Manli W, Ruiyuan C, Leike Z, et al. Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro. Cell Res. 2020; 30 (3), 269-71.

[7] Grein J, Ohmagari N, Shin D, et al. Compassionate use of remdesivir for patients with severe Covid-19. N Engl J Med. 2020; 382, 2327-36.

[8] Gautret P, Lagier J-C, Parola P, et al. Hydroxychloroquine and azithromycin as a treatment of COVID-19: results of an open-label non-randomized clinical trial. Int J Antimicrob Agents. 2020; 56 (1), 105949.

[9] FDA issues Emergency Use Authorization. March 28, 2020. Available at: (https://www.fda.gov/media/136534/download).

[10] Coronavirus (COVID-19) Update: FDA revokes emergency use authorization for chloroquine and hydroxychloroquine, June 15, 2020. Available at: www.fda.gov/news-events/press-announcements/coronavirus-covid-19-update-fda-revokes-emergency-use-authorization-chloroquine.

[11] Beigel JH, Tomashek KM, Dodd LE, et al. ACTT-1 /Study Group Members. Remdesivir for the treatment of Covid-19 preliminary report. N Engl J Med. 2020; 383, 1813-26.

[12] FDA Coronavirus (COVID-19) Update: FDA Issues Emergency Use Authorization for Potential COVID-19 Treatment Available at: www.fda.gov/news-events/press-announcements/coronavirus-covid-19-update-fda-issues-emergency-use-authorization-potential-covid-19-treatment.

[13] FDA Approves First Treatment for COVID-19 Available at: fda.gov/news-events/press-announcements/fda-approves-first-treatment-covid-19.

[14] RECOVERY Collaborative Group. Dexamethasone in hospitalized patients with Covid-19. N Engl J Med. 2021; 384 (8), 693-704.

[15] Rosenberg ES, Dufort EM, Udo T, et al. Association of treatment with hydroxychloroquine or azithromycin with in-hospital mortality in patients with COVID-19 in New York State. JAMA. 2020; 323 (24), 2493-502.

[16] Furtado RHM, Berwanger O, Fonseca HA, et al. Azithromycin in addition to standard of care versus standard of care alone in the treatment of patients admitted to the hospital with severe COVID-19 in Brazil (COALITION II): a randomised clinical trial. Lancet. 2020; 396 (10256), 959-67.

[17] RECOVERY Collaborative Group. Azithromycin in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial. Lancet, 2021; 397, 10274, 605-12.

[18] International Severe Acute Respiratory and Emerging Infections Consortium (ISARIC) Michigan.gov: https://files.covid19treatmentguidelines.nih.gov/guidelines/archive/covid19treatmentguidelines-05-12-2020.pdf

[19] R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.

[21] Guan WJ, Ni Z, Hu Y, Liang, W, et al. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med. 2020; 382, 1708-20.

[22] Driggin E, Madhavan M, Bkdeli B, et al. cardiovascular considerations for patients, health care workers, and health systems during the COVID-19 pandemic. J Am Coll Cardiol. 2020; 75 (18), 2352-71.

[23] FDA Update: https://files.covid19treatmentguidelines.nih.gov/guidelines /archive/covid19treatmentguidelines-05-12-2020.pdf

[24] Tang N, Li D, Wang X, et at. Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia. J Thromb Haemost. 2020; 18, 4, 844-84.

[25] Tang N, Bai H, Chen X, et al. Anticoagulant treatment is associated with decreased mortality in severe coronavirus disease 2019 patients with coagulopathy. J Thromb Haemost. 2020; 18 (5), 1094-9.

[26] Gallo Marin B, Aghagoli G, Lavine K, et. al. Predictors of COVID-19 severity: A literature review. Rev Med Virol. 2021; 31 (1), 1-10.