We included 341 patients who were admitted with COVID-19 infection, with a mean age of 57.9 ± 18.3 years, 54.8% (187/341) female and 48.7% (166/341) black/African American. Of these patients, 65.7% (224/341) were unvaccinated, 8.5% (29/341) had received one dose,22% (75/341) had received two doses, and 3.8% (13/341) had received three doses. The mean BMI of the cohort was 31.7 ± 10.1 and the median Charlson weighted index of comorbidity (CWIC) score was 0.0 (IQR 0.0, 2.0). The mean length of stay among the cohort was 6.7 ± 5.7 days. Mild disease was present in 29% of patients (99/341), 26.4% (90/341) had moderate disease, 33.4% (114/341) had severe, and 11.1% (38/341) had critical COVID infection based on the WHO categorization at the time of hospitalization. Forty-two patients (12.3%) required mechanical ventilation, 46 patients (13.5%) died during admission and 22 (6.5%) required readmission to one of the study hospitals within 30 days of their initial discharge.
When comparing the characteristics of patients in the three vaccination groups (Table 1), unvaccinated patients were younger than fully vaccinated (p < 0.001) and were more likely to be black/African American (p = 0.002). While unvaccinated and partially vaccinated patients were more likely to have private insurance, fully vaccinated patients were more likely to have public insurance (Medicare) (p = 0.04). Partially vaccinated patients had significantly higher mean BMI levels (p < 0.001) than fully vaccinated and unvaccinated individuals. Obesity was present in 53% (181/341) of patients and there was no significant difference in the distribution of obesity between the three vaccination groups. Fully vaccinated patients were more likely to have heart failure (18.2%) (p = 0.01), hypertension (72.7%) (p = 0.01), renal disease (22.7%) (p < 0.001), and malignancy (14.8%) (p = 0.05) than partially vaccinated and unvaccinated patients. There was no statistically significant difference in the other comorbidities among the three groups. More than one third of partially vaccinated patients were self-reported current smokers (37.9%) compared to 12.5% of unvaccinated patients and 18.2% of fully vaccinated patients (p = 0.002).
Table 1
Characteristics of Fully Vaccinated, Partially Vaccinated, and Unvaccinated Hospitalized COVID-19 Patients
Characteristic
|
Vaccination Status
|
p-value
|
|
A. Unvaccinated
(n = 224)
|
B. Partially Vaccinated
(n = 29)
|
C. Fully Vaccinated
(n = 88)
|
|
Mean age (years)
|
54.8 ± 18.6
|
58.7 ± 15.9
|
65.4 ± 15.9
|
A vs C* <0.001
|
Sex
Male
Female
|
41.1% (92)
58.9% (132)
|
51.7% (15)
48.3% (14)
|
53.4% (47)
46.6% (41)
|
0.11
|
Race
White
Black/ African American
Other
|
39.7% (89)
52.2% (117)
8% (18)
|
37.9% (11)
58.6% (17)
3.4% (1)
|
62.5% (55)
36.4% (32)
1.1% (1)
|
0.002
|
Insurance Type
Commercial
Medicare
Medicaid
|
129 (57.6%)
81 (36.2%)
14 (6.3%)
|
15 (51.7%)
12 (41.4%)
2 (6.9%)
|
35 (39.8%)
49 (55.7%)
4 (4.5%)
|
0.041
|
Mean body mass index
|
31.1 ± 9.3
|
38.9 ± 17.1
|
30.8 ± 7.8
|
B vs. A;
B vs. C.
< 0.001
|
Obesity (n = 181)
|
114 (50.9%)
|
18 (62.1%)
|
49 (55.7%)
|
0.4
|
Median CWIC
|
0.0 (0.0, 1.75)
|
1.0 (0.0, 1.0)
|
1.0 (0.0, 3.0)
|
< 0.001
|
Prior MI (n = 43)
|
25 (11.2%)
|
2 (6.9%)
|
16 (18.2%)
|
0.2
|
CHF (n = 34)
|
16 (7.1)
|
2 (6.9%)
|
16 (18.2%)
|
0.01
|
PVD (n = 56)
|
32 (14.3)
|
5 (17.2%)
|
19 (21.6%)
|
0.3
|
CVD (n = 76)
|
44 (19.6)
|
6 (20.7)
|
26 (29.5)
|
0.2
|
Diabetes without complications (n = 76)
|
46 (20.5%)
|
10 (34.5%)
|
20 (22.7%)
|
0.2
|
Diabetes with complications (n = 30)
|
17 (7.6%)
|
2 (6.9%)
|
11 (12.5%)
|
0.4
|
Asthma (n = 28)
|
18 (8%)
|
4 (13.8%)
|
6 (6.8%)
|
0.4
|
COPD (n = 47)
|
31 (13.8%)
|
5 (17.2%)
|
11 (12.5%)
|
0.8
|
Renal disease (n = 40)
|
19 (8.5%)
|
1 (3.4%)
|
20 (22.7%)
|
< 0.001
|
Dementia (n = 21)
|
12 (5.4%)
|
1 (3.4%)
|
8 (9.1%)
|
0.4
|
Any malignancy (n = 29)
|
14 (6.3%)
|
2 (6.9%)
|
13 (14.8%)
|
0.05
|
HIV (n = 6)
|
3 (1.3%)
|
1 (3.4%)
|
2 (2.3%)
|
0.7
|
Hypertension (n = 203)
|
124 (55.4%)
|
15 (51.7%)
|
64 (72.7%)
|
0.013
|
Steroid Use at home (n = 10)
|
70% (7)
|
0.0% (0)
|
30.0% (3)
|
- -
|
Current Smoker (n = 55)
|
28 (12.5%)
|
11 (37.9%)
|
16 (18.2%)
|
0.002
|
Drug Use (n = 13)
|
9 (4%)
|
1 (3.4%)
|
3 (3.4%)
|
0.9
|
Vital signs on admission
|
|
Temperature
|
37.2 ± 1
|
37.2 ± 0.9
|
37 ± 0.7
|
0.3
|
Heart rate
|
101.1 ± 23.8
|
100.4 ± 21.9
|
91.2 ± 19.1
|
A vs C
0.002
|
Respiratory rate
|
20.3 ± 5.7
|
20.5 ± 5.3
|
19.6 ± 4.8
|
0.6
|
Oxygen saturation
|
94.3 ± 5.9
|
95 ± 4.5
|
95.7 ± 5.3
|
0.1
|
Laboratory findings on admission
|
White blood cell count
|
7.7 ± 4.6
|
7.7 ± 5.1
|
8.7 ± 6.3
|
0.3
|
Hemoglobin
|
12.8 ± 2.3
|
12.8 ± 2.9
|
12.1 ± 2.1
|
A vs C 0.05
|
Platelets count
|
228.7 ± 93.4
|
217.5 ± 81.8
|
219.9 ± 84.1
|
0.7
|
Total protein
|
7.1 ± 1.1
|
6.7 ± 1
|
6.7 ± 0.9
|
A vs C 0.01
|
Albumin
|
3.6 ± 0.6
|
3.7 ± 0.6
|
3.6 ± 0.6
|
0.9
|
C-reactive protein
|
90.1 ± 81
|
77.2 ± 68.6
|
65.4 ± 75.8
|
0.09
|
Chest imaging on admission No. (%)
|
Abnormal chest x-ray (n = 198)
|
143 (71.1%)
|
16 (64%)
|
39 (48.8%)
|
0.002
|
Abnormal CT scan (n = 118)
|
89 (92.7%)
|
7 (77.8%)
|
22 (71%)
|
0.006
|
* From multiple pairwise comparisons, Bonferroni correction of the p-value |
⸶Kruskal-Wallis test |
Abbreviations: n: Number, CWIC: Charlson weighted index of comorbidity, MI: Myocardial infarction, CHF: Congestive heart failure, PVD: Peripheral vascular disease, CVD: Cerebrovascular disease, COPD: Chronic obstructive pulmonary disease, HIV: Human immunodeficiency virus
Clinical features on admission are described in Table 1. Fully vaccinated patients had lower heart rates on admission (p = 0.002) compared to unvaccinated patients. Other vital signs including temperature, respiratory rate, oxygen saturation, and blood pressure were similar in the three groups. There were no significant differences in oxygen requirements at admission by vaccination group; however, at 24 hours, the need for O2 supplementation was 60.3% in the unvaccinated group, 55.2% in the partially vaccinated group and 43.2% in the fully vaccinated group (p = 0.024). Patients who were fully vaccinated had lower mean hemoglobin (p = 0.05) and total protein levels (p = 0.01) compared to unvaccinated patients. There were no differences observed in white blood cell counts, platelet counts, and C-reactive protein (CRP) levels between the three groups. Creatinine phosphokinase (CPK), troponin, ferritin, and d dimer levels were excluded from the analysis because many patients did not have these laboratory tests done.
Chest x-rays were done at the time of hospitalization in 88.8% (199/224) unvaccinated patients, 86.2% (25/29) partially vaccinated patients and 90.9% (80/88) fully vaccinated patients. Of 199 unvaccinated patients who had chest x rays done, 71.1% had abnormal chest x rays on admission compared to 48.8% in fully vaccinated and 64% in partially vaccinated patients (p = 0.002) (Table 1). The most common chest x ray finding was patchy bilateral infiltrates present in 44.4% (135/304) of all patients who had x rays done. A total of 135 patients had computed tomography (CT) imaging of the chest done on admission, of which 71.1% (96/135) were unvaccinated, 6.7% (9/135) were partially vaccinated, and 22.2% (30/135) were fully vaccinated, (p = 0.2). Among unvaccinated patients who had a CT scan done 92.7% had abnormal results compared to 71% in the fully vaccinated and 77.8% in the partially vaccinated patients (p = 0.006). The most common chest CT findings across all groups were ground glass opacities 34.6% (47/135) and patchy bilateral infiltrates 30.1% (41/135). Forty out of 47 patients with ground glass opacities and 32 out of 41 patients with patchy bilateral infiltrates were in the unvaccinated group.
Severe and critical disease on admission based on the WHO classification (Fig. 1.) was observed in 50.4% of unvaccinated patients compared to 48.3% of partially vaccinated and 28.4% of fully vaccinated patients (p = 0.002). Overall, fully vaccinated individuals had fewer complications and required fewer interventions during their hospital stay compared to unvaccinated and partially vaccinated individuals (Table 2.). Partially vaccinated patients were more likely to require mechanical ventilation (p = 0.003) and ICU admission (p = 0.01) compared to unvaccinated and fully vaccinated individuals. Unvaccinated patients were more likely to develop septic shock (p = 0.05) and ARDS (p = 0.03) compared to partially and fully vaccinated patients. Most unvaccinated patients required steroids during their hospital stay compared to partially vaccinated patients and fully vaccinated patients (p < 0.001). Partially vaccinated patients were more likely to receive Remdesivir compared to in the unvaccinated and in the fully vaccinated patients (p < 0.001). Unvaccinated patients were more likely to receive IL-6 inhibitors (p = 0.04) than partially and fully vaccinated patients. Higher rates of readmission were observed in fully vaccinated patients compared to unvaccinated and partially vaccinated patients; however, this difference was not statistically significant (p = 0.4). The case fatality rate in fully vaccinated patients was significantly lower compared to unvaccinated and partially vaccinated patients respectively.
Table 2
Treatment and Outcomes among the Fully Vaccinated, Partially Vaccinated, and Unvaccinated Hospitalized COVID-19 Patients
Complication
|
No. (%)
|
p value
|
Unvaccinated
(n = 224)
|
Partially vaccinated
(n = 29)
|
Fully vaccinated
(n = 88)
|
Intubation/Mechanical ventilation (n = 42)
|
34 (15.2%)
|
6 (20.7%)
|
2 (2.3%)
|
0.003
|
Intensive Care Unit admission (n = 47)
|
38 (17%)
|
5 (17.2%)
|
4 (4.5%)
|
0.01
|
Septic shock (n = 29)
|
24 (10.7%)
|
3 (10.3%)
|
2 (2.3%)
|
0.05
|
Acute Respiratory Distress Syndrome (n = 23)
|
21 (9.4%)
|
1 (3.4%)
|
1 (1.1%)
|
0.03
|
Acute Kidney Injury (n = 84)
|
61 (27.2%)
|
7 (24.1%)
|
16 (18.2%)
|
0.2
|
Renal Replacement Therapy (n = 8)
|
5 (2.2%)
|
1 (3.4%)
|
2 (2.3%)
|
0.9
|
Any Arrhythmia (n = 30)
|
22 (9.8%)
|
2 (6.9%)
|
6 (6.8%)
|
0.7
|
Atrial arrythmia (n = 26)
|
18 (8.0%)
|
2 (6.9%)
|
6 (6.8)
|
0.9
|
Steroids given (n = 222)
|
162 (72.3%)
|
19 (65.5%)
|
41 (46.6%)
|
< 0.001
|
Remdesivir given (n = 108)
|
81 (36.2%)
|
13 (44.8%)
|
14 (15.9%)
|
< 0.001
|
Interleukin-6 Inhibitor given (n = 30)
|
25 (11.2%)
|
3 (10.3%)
|
2 (2.3%)
|
0.04
|
Readmission within 30 days (n = 22)
|
12 (6.5%)
|
1 (3.8%)
|
9 (10.6%)
|
0.4
|
Death (n = 46)
|
40 (17.9%)
|
3 (10.3%)
|
3 (3.4%)
|
0.003
|
When assessing risk factors associated with the severity of COVID-19 disease (Table 3.), patients with severe disease were significantly older than those with mild disease (p = 0.004) and a patient with severe disease tended to have a higher BMI than those with mild disease (p = 0.06). Fully vaccinated patients and patients of black/African American race were significantly more likely to have mild or moderate disease at the time of hospitalization. Fully vaccinated black/ African American patients were less likely to have severe disease compared to fully vaccinated white patients (15.6% vs 36.4%, p = 0.04). In addition, higher respiratory rate on admission, higher mean WBC, and higher absolute neutrophil count, were associated with severe and critical infection compared to mild infection. There was a significant association between median creatinine on admission and severity of disease (p = 0.004). Presence of hypertension (0.06) and home steroid use (0.09) were higher among patients with more severe categories of disease compared to those with mild infection at the time of hospitalization. Serum albumin levels showed an inverse association with the severity of COVID-19 at the time of hospitalization, i.e., lower albumin levels were associated with severe and critical disease.
Table 3
Univariable Analysis of Risk Factors among Hospitalized COVID-19 Patients using WHO Severity Scale at the Time of Admission
Characteristic
|
A. Mild
n = 99
|
B. Moderate
n = 90
|
C. Severe
n = 114
|
D. Critical n = 38
|
p value
|
Mean Age (yrs.)
|
52.5 ± 22.1
|
58.4 ± 17.9
|
61.0 ± 15.4
|
61.3 ± 13.8
|
A vs. C, p = 0.004
|
Mean BMI
|
29.3 ± 7.5
|
32.0 ± 11.1
|
32.9 ± 10.1
|
33.4 ± 12.4
|
A vs. C, p = 0.06
|
Median CWIC score
|
0.0 (0.0, 2.0)
|
1.0 (0.0, 2.0)
|
0.0 (0.0, 2.0)
|
0.0 (0.0, 2.0)
|
0.32
|
Mean Respiratory Rate on Admission
|
18.6 ± 3.5
|
19.3 ± 4.1
|
21.0 ± 5.7
|
23.6 ± 8.7
|
A vs. C, p = 0.006; A vs. D, p < 0.001; B vs. D, p < 0.001, C vs. D, p = 0.06
|
Mean WBC Count
|
7.6 ± 3.6
|
8.0 ± 6.7
|
7.5 ± 4.1
|
10.4 ± 6.2
|
A vs. D, p = 0.025; C vs. D, p = 0.012
|
Mean Absolute Neutrophil Count
|
5.4 ± 3.3
|
5.3 ± 3.8
|
5.9 ± 3.8
|
8.4 ± 5.9
|
A vs. D and B vs. D, p < 0.001, C vs. D p = 0.006
|
Median Creatinine on Admission
|
0.9 (0.8, 1.2)
|
1.0 (0.8, 1.4)
|
0.9 (0.7–1.4)
|
1.3 (1.0, 1.7)
|
0.004
|
Race
White
Black
Other
|
34.3% (34)
61.6% (61)
4.0% (4)
|
45.6% (41)
51.1% (46)
3.3% (3)
|
56.1% (64)
38.6% (44)
5.3% (6)
|
42.1% (16)
39.5% (15)
18.4% (7)
|
< 0.001
|
Smoking
|
17.2% (17)
|
23.3% (21)
|
13.2% (15)
|
5.3% (2)
|
0.12
|
Hypertension
|
48.5% (48)
|
64.4% (58)
|
64.9% (74)
|
60.5% (23)
|
0.06
|
Steroid Use (Home)
|
1.0% (1)
|
1.1% (1)
|
6.1% (7)
|
2.6% (1)
|
0.09
|
Serum Albumin
|
3.9 ± 0.73
|
3.6 ± 0.52
|
3.5 ± 0.49
|
3.4 ± 0.60
|
A vs B, p = 0.012; A vs C & A vs D, p < 0.001
|
Vaccination group
None
Partial
Complete
|
53.5% (53)
10.1% (10)
36.4% (36)
|
64.4% (58)
5.6% (5)
30.0% (27)
|
71.1% (81)
8.8% (10)
20.2% (23)
|
84.2% (32)
10.5% (4)
5.3% (2)
|
0.005
|
Abbreviations: n: Number, BMI: Body mass index, CWIC: Charlson weighted index of comorbidity, WBC: White blood cells |
As stated in the methods section, results from the univariable analysis using the two-category scale were used to determine which variables to enter the multivariable logistic regression model. Variables that were included in defining the WHO severity categories were excluded as independent variables. The initial variables that went into the model included age at admission, BMI, heart rate on admission, albumin value, initial absolute lymphocyte count on admission, vaccination group, race, elevated creatinine from baseline, hypertension, and home steroid use. Using a forward stepwise likelihood ratio algorithm, the final model included the variables age at admission, BMI, albumin, vaccination group, race, and home steroid use (Table 4).
Table 4
Multivariable Analysis of Risk Factors for Breakthrough Severe COVID-19 Infections Requiring Hospitalization Among the Fully vaccinated, Partially vaccinated, and unvaccinated individuals.
Characteristic
|
Odds Ratio
|
p-value
|
95% CI
|
Age at admission
|
1.02
|
0.02
|
1.004, 1.04
|
Fully vaccinated*
|
0.19
|
< 0.001
|
0.09, 0.38
|
Black race**
|
0.39
|
0.001
|
0.22, 0.70
|
Albumin
|
0.51
|
0.007
|
0.31, 0.83
|
BMI
|
1.04
|
0.005
|
1.01, 1.08
|
Steroids use (Home)
|
7.5
|
0.03
|
1.3, 44.9
|
*Unvaccinated is the comparator category. |
**White race is the comparator category. |
Abbreviations: CI: Confidence interval, BMI: Body mass index |
From the model, after controlling for age, BMI, albumin, race and home steroid use, patients who were fully vaccinated were 5.3 times (p < 0.001) less likely to have severe/critical disease. Notably, after controlling for vaccination status and the other variables in the model, Black patients were 2.6 times less likely to have severe/critical disease (p = 0.001) and every unit increase in albumin reduced the risk of severe/critical disease by 96% (p = 0.007). Older patients (p = 0.02), patients with higher BMI values (p = 0.005) and those who were using steroids for seven consecutive days at the time of hospitalization (p = 0.03) were more likely to have severe/critical disease, after controlling for the other variables in the model.