Patients with confirmed COVID-19 in Busan and those who were treated in the intensive care unit (ICU) after becoming critically ill at hospital dedicated to infectious diseases from December 1, 2020, to December 31, 2021, were included in this study. Patients with COVID-19 are divided into the following four stages according to disease severity: asymptomatic, mild, severe, and critical. Mild disease is characterized by mild symptoms without pneumonia. Severe disease is characterized by dyspnea, tachypnea (>30 breaths per minute), arterial oxygen saturation ≤93%, the ratio of partial pressure of arterial oxygen (PaO2) to the fraction of inspired oxygen (FiO2) <300, and worsening of pneumonia by >50% within 24–48 hours. Critical disease is characterized by respiratory failure, septic shock, and multiple organ failure. The definition of clinical severity has changed since the beginning of the outbreak of COVID-19. In this study, “critically ill” patients were defined as those who received non-invasive high-flow oxygen therapy, invasive mechanical ventilation, extracorporeal membrane oxygenation (ECMO), or continuous renal replacement therapy (CRRT). Patients with mild-to-moderate symptoms were defined as those who received low-dose oxygen therapy at less than 5 L/min at home, a residential treatment center, or a hospital for infectious diseases [3].

The method for confirming COVID-19 is described in the guidelines of the Korea Centers for Disease Control and Prevention (KCDC; currently, Korea Disease Control and Prevention Agency). The specimens were transferred to an in-house laboratory, a private testing institution, or the Busan Health and Environment Research Institute and tested using real-time reverse transcription polymerase chain reaction [4]. Patient epidemiological information was collected using a COVID-19 basic epidemiologic survey via telephone interviews with patients confirmed with COVID-19 at 16 public health centers in Busan. In the basic epidemiological survey, age, sex, underlying diseases, symptoms, and infection route were included. Analysis of variant virus was conducted by the KCDC, and the results were collected by Busan City. The hospitalization and discharge status of patients who deteriorated to critically ill status were obtained from the infection control offices at infectious disease hospitals.

The characteristics of all critically ill patients and those with mild-tomoderate illness were compared and summarized (Table 1). The presence of symptoms at the time of confirmation of disease was determined according to the COVID-19 Basic Epidemiologic Investigation; consequently, the symptoms that followed after confirmation were not included. The differences in the distribution of symptoms at the time of diagnosis between the two groups (Figure 1) and between patients aged up to and beyond 60 years were visualized using a forest plot (Figure 2A, B). The benchmark for determining the prevalence of the delta variant virus is based on the first introduction of the variant virus into Korea on April 22, 2021, and July 2021, when the delta variant virus began to spread [5]. For COVID-19 cases confirmed before July, the KCDC requested variant analysis and mutated gene test and were assigned to the non-epidemic group of the delta variant virus. Delta variant-related cases transmitted by contact were included in the delta variant epidemic group. Investigation of the source of infection based on the route of infection implies that contact history with any previously confirmed person is unknown; among them, those who were tested because they had symptoms, or were those who were workers in facilities vulnerable to infection and who performed periodic preemptive tests, and those who were tested by themselves were included. Patients were included if they had contact at multi-use facilities(restaurants, coffee shops, sports facilities, and entertainment facilities), used such facilities, or if they were tested for a group outbreak at such facilities.

The differences in the characteristics between the delta variant virus epidemic group and the non-epidemic group observed in 460 patients whose symptoms progressed to the point of being critically ill is summarized in Table 2. The number of days critically ill patients were hospitalized in the ICU for was determined by the duration of treatment with non-invasive high-flow oxygen therapy, ventilator, ECMO, or CRRT. Of the 460 patients with critically ill status, 375 patients complained of symptoms at the time of COVID-19 confirmation; the analysis of symptomatic patients from the onset of symptoms to critically ill status was determined from the onset of symptoms to the date of starting treatment as a critically ill patient during hospitalization. Additionally, based on the prevalence of the delta variant virus, the correlation between the number of daily new confirmed cases of COVID-19 and the number of daily hospitalized patients in ICUs during was reported during the period of this study (Figure 3). We analyzed the correlation between the daily number of new confirmed cases and the number of hospitalized patients with critically ill status during the delta variant virus non-epidemic and epidemic period, which revealed that the average duration between confirmation of COVID-19 status and critically ill status was 7.46 days and 4.67 days, respectively (Figures 4, 5). For categorical variables, frequencies and percentages were calculated, and for continuous variables, mean and standard deviation were calculated. In order to compare the difference between patients with critically ill status with and without the delta variant virus, the chi-square test was used; continuous variables were tested using the F test and equal variances (Levene t-test) or heterovariance t-test (Welch’s t-test). A simple regression analysis was performed to estimate the correlation coefficient between continuous variables. Statistical significance was set at P<0.05. The statistical program used was SAS ver. 9.4 (SAS Institute Inc., Cary, NC, USA).

This study includes the results of an epidemiological investigation conducted by the national and local health authorities directly in accordance with the ‘Law on the Prevention and Management of Infectious Diseases’ in a situation that requires urgent action for public health. This falls under Article 33(3) of the Enforcement Rules, and deliberations were not carried out because they were not subject to the institutional review board deliberations.

Between December 1, 2020, and December 31, 2021, there were 24,441 confirmed cases of COVID-19 in Busan. Of them, 460 were hospitalized and treated in the ICU, including patients who were asymptomatic at the time of confirmation, whereas 23,981 patients were treated in the general wards. In critically ill patients, 55% were men, 68.9% were aged ≥60 years, 83.7% complained of clinical symptoms at the time of confirmation, and 61.5% had underlying diseases, which was significantly higher than the proportion in patients with mild-to-moderate disease severity (P<0.001). The underlying diseases in critically ill patients included hypertension (n=107 [23.3%]), diabetes (n=73 [15.9%]), and hyperlipidemia (n=23 [5%]); while those in patients with mild-tomoderate disease included hypertension (n=2,739 [11.4%]), diabetes (n=1,647 [6.9%]), and hyperlipidemia (n=858 [3.8%]). Additionally, other diseases in critically ill patients included cardiovascular disease excluding hypertension (n=71 [15.4%]), chronic respiratory diseases including asthma (n=37 [8%]), dementia including Alzheimer’s and Parkinson’s dementia (n=33 [7.2%]), cerebrovascular disease including cerebral infarction and cerebral hemorrhage (n=23 [5%]), and cancer (n=21 [4.6%]). Other diseases in patients with mild-to-moderate disease included cerebrovascular disease including cerebral infarction and cerebral hemorrhage (n=243 [3.6%]), cardiovascular disease excluding hypertension (n=643 [2.7%]), chronic respiratory disease including asthma (n=535 [2.2%]), and cancer (n=512 [2.1%]). In terms of the distribution of symptoms according to disease severity, there was no significant difference in the prevalence of hyperlipidemia and thyroid disease between patients with critical and mild-to-moderate illness (P=0.10 and P=0.73, respectively). The prevalence rates of all other underlying diseases were significantly higher in critically ill patients (P<0.001). In patients with mild-to-moderate disease, the estimated path of the infection was found to be high in 74.9% of cases (P<0.001); there was no significant difference in the prevalence of the delta variant virus between the two groups (P=0.06) (Table 1).

We analyzed the differences in the distribution of symptoms at the time of confirmation of COVID-19 in 375 critically ill patients and 17,416 patients with mild-to-moderate disease. The rates of dyspnea (odds ratio [OR], 10.84; 95% confidence interval [CI], 8.33–14.11; P<0.001), chest discomfort (OR, 3.8; 95% CI, 2.27–6.37; P<0.001), fever (OR, 2.71; 95% CI, 2.13–3.44; P<0.001), dizziness (OR, 2.58; 95% CI, 1.36–4.89; P<0.05), loss of taste (OR, 1.86; 95% Cl, 1.28–2.69; P<0.001), sputum (OR, 1.83; 95% CI, 1.46–2.30; P<0.001), myalgia (OR, 1.67; 95% CI, 1.33–2.10; P<0.001), chills (OR, 1.57; 95% CI, 1.19–2.06; P<0.05), and cough (OR, 1.39; 95% CI, 1.14–1.69; P<0.01) were higher in critically ill patients than those in patients with mild-to-moderate symptoms. In contrast, the rates of a runny nose (OR, 0.42; 95% CI, 0.25–0.69; P<0.001), sore throat (OR, 0.75; 95% CI, 0.58–0.96; P<0.05), and nasal congestion (OR, 0.29; 95% CI, 0.12–0.71; P<0.01) were significantly lower in critically ill patients than those in patients with mild-to-moderate symptoms. The rates of fatigue (OR, 1.50; 95% CI, 0.55–4.08; P=0.42), fever (OR, 1.25; 95% CI, 0.97–1.62; P=0.08), diarrhea (OR, 1.21; 95% CI, 0.49–2.95; P=0.68), headache (OR, 0.84; 95% CI, 0.63–1.12; P=0.24), and loss of smell (OR, 0.88; 95% CI, 0.55–1.40; P=0.59) were not significantly different between the groups (Figure 1). Additionally, the distribution of the symptoms was analyzed by dividing groups of patients into those <60 years old and ≥60 years old. In patients aged <60 years, the rates of cough (OR, 1.47; 95% CI, 1.03–2.09; P<0.001), sputum (OR, 1.84; 95% CI, 1.23–2.75; P<0.001), dyspnea (OR, 14.7; 95% CI, 9.44–22.89; P<0.001) were significantly lower in critically ill patients, while chest discomfort (OR, 2.97; 95% CI, 1.08–1.88; P<0.001), myalgia (OR, 2.14; 95% CI, 1.47–3.13; P<0.001), chills (OR, 2.32; 95% CI, 1.41– 3.56; P<0.001), fever (OR, 2.89; 95% CI, 1.95–4.28; P<0.001), and loss of taste (OR, 2.65; 95% CI, 1.57–4.49; P<0.001) were significantly higher in critically ill patients than the rates in those with mild-to-moderate disease. Furthermore, the rate of a runny nose (OR, 0.23; 95% CI, 0.07– 0.72; P<0.001) was significantly lower in the critically ill group compared to that in the mild-to-moderate group (Figure 2A).

In patients aged ≥60 years, the rates of sputum (OR, 1.96; 95% CI, 1.47–2.61; P<0.001), dyspnea (OR, 7.18; 95% CI, 5.13–10.05; P<0.001), chest discomfort (OR, 2.45; 95% CI, 1.12–5.39; P<0.001), fever (OR, 4.54; 95% CI, 3.30–6.25; P<0.001), mild fever (OR, 1.97; 95% CI, 1.41– 2.75; P<0.001), and loss of taste (OR, 2.51; 95% CI, 1.45–4.34; P<0.001) were significantly higher in critically ill patients, while the rate of a runny nose (OR, 0.49; 95% CI, 0.25–0.93; P=0.03) was significantly lower in critically ill patients in comparison with the rates in patients with mildto-moderate symptoms (Figure 2B).

In the delta variant non-epidemic group, 92.1% of the patients were aged >60 years, 77.4% had underlying diseases, 73% had an identified source of infection, and 98.3% were not vaccinated (Table 2). In contrast, in the delta variant epidemic group, 61.5% were aged >60 years, 51% had underlying diseases, 50.1% had an identified source of infection, and 57.7% were not vaccinated (P<0.001, respectively). Between the delta non-epidemic and epidemic groups, hypertension was the most common underlying disease (n=32 [27.8%] and n=82 [23.7%], respectively), followed by diabetes (n=21, [18.3%] and n=8 [2.3%], respectively), with no significant differences (P>0.05 for both). Additionally, the prevalence rates of chronic respiratory diseases, including hyperlipidemia and asthma, dementia, including Alzheimer’s and Parkinson’s dementia, and cerebrovascular diseases, including cerebral infarction and cerebral hemorrhage, were significantly higher in the delta variant non-epidemic group than those in the delta variant epidemic group (P<0.001). The number of patients with symptoms at the time of COVID-19 confirmation was significantly higher in the delta variant epidemic group (88.4%) than those (60.9%) in the non-epidemic group (P<0.001). Furthermore, there was no difference in the sex distribution between the groups (P=0.63). There was no significant difference in the duration between the onset of symptoms and their progression to critically ill (P=0.17). The average age in the delta variant non-epidemic group was 74.5 years, which was significantly higher than those in the epidemic group (62.9 years) (P<0.05). Additionally, in the delta variant epidemic group, the duration of ICU stay was 11.5 days, and the duration between confirmation of disease and progression to critically ill status was 4.67 days, which were both significantly shorter than those in the non-epidemic group (16 days and 7.46 days, respectively; P<0.05).

The graph of the monthly incidence of newly confirmed cases of COVID-19 and the daily number of hospitalized patients with critical illness revealed that the concordance rate increased from mid-2021 (Figure 3). In the delta variant non-epidemic group, the correlation coefficient (r) was 0.0473 between the number of daily new confirmed cases and the number of hospitalized critically ill patients, while the coefficient of determination (R²) was 0.0022 (P=0.5), thus indicating no statistical significance (Figure 4).

In the delta variant epidemic group, the correlation coefficient (r) between the number of daily new confirmed cases and the number of hospitalized critically ill patients was very high (0.83), and the coefficient of determination (R²) was 0.7149. Consequently, the regression equation was “y=0.126x+9.436”, which indicated statistical significance (P<0.05) (Figure 5).