The analysis included 2108 confirmed cases of COVID-19. The age range of the patients was 1 month to 98 years and the mean age was 47.81 ± 17.79 years. 56.8% (n = 1197) were men, 43.2% women ,and 10.9% (n = 230) were admitted to the ICU. In addition, 3.3% of the patients were the medical care staff and 0.2% were the laboratory staff. Until the end of the study period, 6.3% of the patients (n = 127) died, 92.0% (n = 1939) of the cases were discharged or recovered, 4 patients (0.2%) remained hospitalized and 32 patients (1.5%) were transferred to other locations. Table 1 and Table 2 demonstrate the clinical and laboratory characteristics of the study population at the time of admission. The median duration of hospitalization in total, for cases of death, and for cases of recovery was 4 days (IQR: 1–7), 5 days (IQR: 2–10), and 4 days (IQR: 1–6), respectively.
Table 1
Socio-demographic and clinical characteristics of the study population.
Variable
|
Total Patient
n = 2108(%)
|
Dead
n = 133(%)
|
Alive n = 1975(%)
|
p- value
|
Age
|
47.81 ± 17.78
|
59.90 ± 19.04
|
27.10 ± 17.40
|
< 0.001a
|
Gender
|
Male
|
1197(56.8)
|
82 (61.7)
|
1115(56.5)
|
0.241b
|
Female
|
911 (43.2)
|
51(38.3)
|
860 (43.5)
|
Comorbidity
|
887(42.1)
|
92(69.3)
|
795 (40.2)
|
< 0.001b
|
Chronic heart disease
|
306 (14.5)
|
56 (42.5)
|
250 (12.6)
|
< 0.001b
|
High blood pressure
|
159(7.5)
|
16 (12.0)
|
143 (89.94)
|
0.043b
|
Chronic lung disease
|
38 (1.8)
|
7(5.2)
|
31 (1.6)
|
0.002b
|
Asthma
|
78 (3.7)
|
6 (4.5)
|
72 (3.6)
|
0.609b
|
Chronic kidney disease
|
50 (2.4)
|
11 (8.3)
|
39(2.0)
|
< 0.001b
|
Diabetes
|
111(5.3)
|
13 (9.8)
|
98 (5.0)
|
0.016b
|
Cancer
|
25 (1.2)
|
6 (4.5)
|
19 (1.0)
|
< 0.001b
|
Smoking
|
43 (2.0)
|
5 (3.9)
|
38 (19.2)
|
0.147b
|
Number of breaths per minute (< 24)
|
51(2.4)
|
11 (8.9)
|
40 (2.0)
|
< 0.001b
|
Heart rate per minute (≤ 125)
|
114 (5.4)
|
6 (4.5)
|
108(5.5)
|
0.637b
|
Systolic blood pressure (mmHg)
|
127.27 ± 19.67
|
123.35 ± 19.84
|
127.51 ± 19.67
|
0.024a
|
Diastolic blood pressure
|
80.40 ± 11.64
|
77.58 ± 11.58
|
80.68± 11.93
|
0.006a
|
Fever
|
1058 (50.2)
|
56 (42.1)
|
1002 (50.7)
|
0.054b
|
Cough with sputum production
|
365 (17.3)
|
22 (15.9)
|
343 (17.4)
|
0.808b
|
Cough with hemoptysis
|
40 (1.9)
|
7 (5.3)
|
33 (1.7)
|
0.011b
|
Sore Throat
|
116 (5.5)
|
2 (1.5)
|
114 (5.8)
|
0.037b
|
Diarrhea
|
129 (6.1)
|
7(5.3)
|
122 (6.2)
|
0.670b
|
myalgia
|
163(7.7)
|
7 (5.3)
|
156 (7.9)
|
0.271b
|
Headache
|
263 (12.5)
|
3 (2.3)
|
260(13.2)
|
< 0.001b
|
Nausea and vomiting
|
211 (10.0)
|
11 (8.3)
|
200 (10.2)
|
0.490b
|
Fatigue
|
133 (6.3)
|
5 (3.8)
|
128 (6.5)
|
0.211b
|
Hospitalization in the ICU
|
230 (10.9)
|
109 (81.9)
|
121 (6.1)
|
< 0.001b
|
CPR
|
≤ 10
|
1553(73.7)
|
120 (90.2)
|
1433(72.6)
|
< 0.001b
|
|
> 10
|
555 (26.3)
|
13(9.7)
|
542 (27.4)
|
|
LDH
|
≤ 436
|
942(44.7)
|
21 (15.8)
|
921(46.6)
|
< 0.001b
|
|
> 436
|
1166 (55.3)
|
112(84.2)
|
1054 (53.4)
|
|
SpO2
|
≤ 93
|
333(15.8)
|
71 (53.4)
|
262(13.3)
|
< 0.001b
|
|
> 93
|
62(46.6)
|
1713 (86.7)
|
|
|
WBC count (x103/L)
|
< 4
|
239 (13.9)
|
8 (6.0)
|
285(14.4)
|
< 0.001b
|
4–10
|
1174 (55.7)
|
58 (43.6)
|
1116 (56.5)
|
> 10
|
641 (30.4)
|
100 (75.2)
|
541 (27.4)
|
CT(Bilateral pulmonary
infiltration)
|
1720(81.6)
|
130 (97.7)
|
1590 (80.5)
|
< 0.001b
|
Number of hospitalization days
|
5.16 ± 7.24
|
7.19 ± 6.90
|
5.08 ± 7.36
|
0.002a
|
a Independent samples T-test |
b X2 test |
Table 2
Laboratory characteristics of the study population.
Variable
|
Group
|
N
|
Mean
|
Std. Deviation
|
p-value
|
SpO2_3
|
Live
|
734
|
95.30
|
6.00
|
0.601
|
|
Dead
|
21
|
96.00
|
4.74
|
Hemoglobin (g/L)
|
Live
|
722
|
13.31
|
23.21
|
0.922
|
|
Dead
|
21
|
12.81
|
2.22
|
WBC count (x103/L)
|
Live
|
691
|
8.84
|
20.43
|
0.603
|
|
Dead
|
19
|
6.40
|
2.045
|
Hematocrit (%)
|
Live
|
726
|
40.83
|
36.36
|
0.563
|
|
Dead
|
22
|
36.33
|
15.06
|
Platelets (x103/L)
|
Live
|
733
|
215.87
|
70.83
|
0.702
|
|
Dead
|
21
|
213.02
|
76.11
|
APTT/APTR
|
Live
|
383
|
27.76
|
23.07
|
0.413
|
|
Dead
|
9
|
21.43
|
12.71
|
PT (seconds)
|
Live
|
455
|
15.07
|
8.54
|
0.692
|
|
Dead
|
12
|
14.09
|
7.27
|
INR_3
|
Live
|
507
|
1.49
|
2.33
|
0.570
|
|
Dead
|
14
|
1.13
|
0.18
|
ALT/SGPT (U/L)
|
Live
|
516
|
42.98
|
53.60
|
0.965
|
|
Dead
|
15
|
43.60
|
25.12
|
Total bilirubin (µmol/L)
|
Live
|
401
|
1.73
|
8.05
|
0.626
|
|
Dead
|
14
|
.68
|
.809
|
AST/SGOT (U/L)
|
Live
|
514
|
40.64
|
40.52
|
0.723
|
|
Dead
|
16
|
44.25
|
21.84
|
Urea (BUN) (mmol/L)
|
Live
|
577
|
18.09
|
17.92
|
0.201
|
|
Dead
|
21
|
13.06
|
7.88
|
Creatinine (µmol/L)
|
Live
|
610
|
1.71
|
6.57
|
0.651
|
|
Dead
|
20
|
1.05
|
0.33
|
Sodium (mEq/L)
|
Live
|
656
|
135.39
|
53.56
|
0.859
|
|
Dead
|
21
|
137.47
|
3.82
|
Potassium (mEq/L)
|
Live
|
607
|
6.49
|
28.72
|
0.709
|
|
Dead
|
19
|
4.03
|
0.43
|
CRP (mg/L)
|
Live
|
537
|
48.09
|
18.453
|
0.895
|
|
Dead
|
15
|
41.80
|
40.63
|
LDH (U/L)
|
Live
|
440
|
444.08
|
235.28
|
0.112
|
|
Dead
|
14
|
545.35
|
207.23
|
Troponin (ng/mL)
|
Live
|
184
|
1.03
|
9.19
|
0.784
|
|
Dead
|
6
|
0.00
|
0.00
|
ESR (mm/hr)
|
Live
|
534
|
38.16
|
29.29
|
0.113
|
|
Dead
|
19
|
49.03
|
30.44
|
The Kaplan-Meier survival plots for the prognostic factors that returned statistically significant results are presented in Fig. 2. The proportional hazards assumption is satisfied since the survival risk curves do not cross during the survey period. As can be inferred from the plot, the risk is directly proportional to the age group. Moreover, the subjects who were older than 60 years had about 25% less probability of survival after 20 days of hospitalization than those who were younger than 60 years. The subjects who were admitted to the ICU, had about 40% less probability of survival after 20 days of hospitalization than those who did not enter this unit. The subjects with cancer had an approximately 45% lower probability of survival after 20 days of hospitalization than those who did not manifest these characteristics. The patients who had hemoptysis had about 17% less probability of survival after 20 days of hospitalization than those who did not. Finally, the subjects with underlying kidney diseases were less likely to survive than those without such diseases.
In general, the 1-day survival rate was 98.4%, which decreased to 95.8%, 95.1%, 94.0%, and 93.8% on days 7, 14, 21, and 28 of hospital stay, respectively. The 1-day survival rate of patients admitted to the ICU was 91.1%, which decreased to 72.6%, 60.0%, 55.8%, and 53.7% on days 7, 14, 21, and 28 of hospital stay, respectively. With patients who died, the 1-day survival rate was 74.8%, which decreaed to 33.9%, 11.8%, 5.5%, and 1.6% on days 7, 14, 21, and 28 of hospital stay, respectively. These results showed that the survival rate of the deceased patients is lower than that of living patients. We conducted the comparison between patients in the ICU and the general wards.
A Cox proportional hazards regression model was used to determine the factors associated with the time of death. The HR from multivariable Cox proportional hazards regression models is reported in Table 3. In multivariable analyses, the following factors were associated with the risk of death: Age (P = 0.016), kidney diseases (P = 0.015), cancer (P < 0.001), hemoptysis (P < 0.001) and admission to ICU (P < 0.001).
Table 3
Variable
|
B
|
SE
|
Wald Statistic
|
df
|
P- value
|
HR
|
95.0% CI for HR
|
Lower
|
Lower
|
age level > 60
|
.55
|
.23
|
5.83
|
1
|
.016
|
1.73
|
1.109
|
2.707
|
kidney disease (yes)
|
.76
|
.31
|
5.97
|
1
|
.015
|
2.14
|
1.163
|
3.952
|
Cancer(yes)
|
2.55
|
.51
|
25.53
|
1
|
< .001
|
12.74
|
4.748
|
34.203
|
Cough with hemoptysis (yes)
|
1.49
|
.34
|
18.62
|
1
|
< .001
|
4.25
|
2.204
|
8.208
|
ICU(yes)
|
2.71
|
.30
|
83.20
|
1
|
< .001
|
15.06
|
8.408
|
26.967
|
Other variables were excluded from the model due to having probability values higher than 0.10 and lack of having a statistically significant effect at the level of 0.05. The risk of death was notably higher for people older than 60 years old. They had 1.73 (CI: 1.109–2.707) times a higher risk of death compared with the those younger than 60 years. The subjects with kidney diseases were 2.14 (CI: 1.163–3.952) times more likely to die compared with those without such diseases. Furthermore, those who had cancer were 12.74 (CI: 4.748–34.203) times more likely to die compared with those without this disease. Furthermore, patients with hemoptysis were 4.25 (CI: 2.204–8.208) times more likely to die compared with patients without these signs and symptoms. Finally, patients admitted to the ICU were 15.06 (CI: 8.408–26.967) times more likely to die compared with those who did not enter these wards.
h (t |X) = h 0(t). exp (2.72 ICU.HDU)
h (t |X) = h 0(t). exp (0.030 Age + 2.596 ICU.HDU)
h (t |X) = h 0(t). exp (0.029 Age + 1.50 Cancer + 2.602 ICU.HDU)
h (t |X) = h 0(t). exp (0.028 Age + 2.062 Cancer +-1.00 hemoptysis + 2.69 ICU.HDU)
h (t |X) = h 0(t). exp (0.55 Age + 0.76 kidney disease + 2.55 Cancer + 1.49 hemoptysis + 2.71 ICU.HDU).
Finally, the Cox regression model was estimated as the last line according to the regression coefficients of the effective variables.
Furthermre, we assessed the relation between SpO2 and underlying diseases and age in COVID-19 patients. The results showed that although none of the underlying diseases were significantly associated with SpO2, age was significantly associated with it (P < 0.001). The average age of patients with SpO2 ≤ 93 was 53.10 ± 20.54 and the average age of patients with SpO2 > 93 was 42.35 ± 16.83.