Population characteristics
Figure 1 illustrates the history of the SARS-CoV-2 pandemic in Romania from February 26 until April 20, including dates, number of fatalities and number of the missing subjects due to incomplete information. The Romanian COVID-19 mortality study has a sample size of 432 patients with complete data regarding sex, age and comorbidities. There are 282 (65.3%) men and 150 (34.7 %) women with a statistically significant difference between the size of these two groups (Table S1). The general mean age is 67, with an SD of 13.1, and the median value is 68 (Table S2).
Sources of infection with deaths due to COVID-19 virus were reported in 36 counties in Romania with significant case concentration in Suceava- 67 (15.51%), Arges - 44 (10.19%) and Hunedoara- 41 (9.49%). Details regarding the spatial distribution of the distinct diseases and aggregated groups of diseases are displayed in Table S6 and Table S7 in the supplementary material.
A total of 170 different conditions and 20 groups of diseases was identified in the study, with a broader diversity of 149 conditions for men over 107 conditions for women. The diseases with the highest prevalence in the sample are hypertension, obesity, diabetes and chronic kidney disease as well as diseases of the circulatory system and nutritional or metabolic disorders.
The sample size of the pneumonia study is equal to 874 persons with complete data as to sex, age and comorbidities. The sample includes 492 (56.3%) men and 382 (43.7%) women, the average age is 73.3, with an SD of 13.5, and the median age is 76.
Association between sex or age and comorbidities
One-sample proportion test indicates a statistically significant difference in the relative frequency of males and females in the sample and in the subpopulations with hypertension, diabetes, diabetes mellitus type 2, chronic kidney diseases, dependence on renal dialysis, chronic obstructive pulmonary disease, acute ischaemic heart disease or respiratory failure as displayed in Table S1. This is also valid for the subpopulations without comorbidities and for those corresponding to the aggregated group of diseases with a relative frequency of more than 5% except for diseases of the nervous system. On the other hand, the variable sex is not significantly associated with single conditions or group of illnesses except for heart failure (see Table 1).
Because age is an established risk factor for mortality, we also performed an age-stratified analysis with the lifespan split in 10-year intervals starting with age 50. The chi-squared test of association between the sex and age factors is statistically significant, with the largest male relative frequency (81.16%) between 50 and 59 years. Moreover, the factor age is also significantly associated with most of the comorbidities, with hypertension and dependence on renal dialysis following also this trend but at a significance level of 10% (see Table 2).
Table 1. Association between sex and the distinct diseases or aggregated group of diseases, as absolute numbers and as percentage (by column). In bold the p-values for the differences in percentage that reached the statistical significance with respect to the chi-squared test.
Characteristics
n,%
|
Male n=282
|
Female n=150
|
p-value
|
Diseases
|
|
|
|
Hypertension
|
113
40.07%
|
49
32.67%
|
0.16
|
Obesity
|
32
11.35%
|
21
14.00%
|
0.52
|
Diabetes
|
105
37.23%
|
48
32.00%
|
48
32%
|
0.33
|
40
14.18%
|
26
17.33%
|
26
17%
|
0.47
|
56
19.86%
|
19
12.67%
|
19
13%
|
0.081
|
9
3.19%
|
3
2.00%
|
3
2%
|
0.68
|
Diseases of the circulatory system, unspecified
|
41
14.54%
|
27
18.00%
|
0.42
|
Supraventricular tachyarrhythmia
|
16
5.67%
|
9
6.00%
|
1
|
Heart failure, unspecified
|
14
4.96%
|
16
10.67%
|
0.043
|
Cerebral ischaemic stroke
|
16
|
12
|
0.47
|
|
5.67%
|
8%
|
|
Chronic kidney disease
|
31
10.99%
|
13
8.67%
|
0.55
|
Kidney failure, unspecified
|
17
6.03%
|
9
6.00%
|
1
|
Dependence on renal dialysis
|
29
10.28%
|
14
9.33%
|
0.88
|
Chronic obstructive pulmonary disease
|
18
6.38%
|
7
4.67%
|
0.61
|
Acute ischaemic heart disease, unspecified
|
15
5.32%
|
4
2.67%
|
0.3
|
Respiratory failure
|
15
6.32%
|
4
2.67%
|
0.3
|
Aggregated group of diseases
|
|
|
|
No comorbidities
|
11
3.90%
|
2
1.33%
|
0.23
|
Diseases of the circulatory system
|
168
59.57%
|
88
58.67%
|
0.94
|
Endocrine, nutritional or metabolic diseases
|
122
43.26%
|
63
42.00%
|
0.88
|
Diseases of the genitourinary system
|
54
19.15%
|
25
16.67%
|
0.61
|
Diseases of the respiratory system
|
47
16.67%
|
20
13.33%
|
0.44
|
Diseases of the nervous system
|
35
12.41%
|
26
17.33%
|
0.21
|
Diseases of the digestive system
|
30
10.64%
|
14
9.33%
|
0.79
|
Neoplasms
|
26
9.22%
|
14
9.33%
|
1
|
Mental, behavioural or neurodevelopmental disorders
|
25
8.87%
|
11
7.33%
|
0.71
|
Factors influencing health status or contact with health services
|
37
13.12%
|
17
11.33%
|
0.7
|
Symptoms, signs or clinical findings, not elsewhere classified
|
13
4.61%
|
11
7.33%
|
0.34
|
Certain infectious or parasitic diseases
|
15
5.32%
|
5
3.33%
|
0.49
|
Table 2. Association between the factor age and the distinct diseases or group of diseases, as absolute numbers and as percentage (by column). In bold the p-values for the differences in percentage that reached the statistical significance with respect to the chi-squared test
Characteristics
n, %
|
<50
44
10.19%
|
50-59
69
15.97%
|
60-69
120
27.78%
|
70-79
128
29.93%
|
>=80
71
16.44%
|
p- value
|
Sex
|
|
|
|
|
|
|
Male
|
29
66%
|
56
81.16%
|
73
60.83%
|
80
62.5%
|
44
61.97%
|
0.05
|
Female
|
15
34.09%
|
13
19.84%
|
47
39.17%
|
48
37.5%
|
27
38.03%
|
|
Diseases
|
|
|
|
|
|
|
Hypertension
|
12
27.28%
|
21
30.43%
|
49
40.83%
|
45
35.16%
|
35
49.30%
|
0.074
|
Obesity
|
13
29.55%
|
9
13.04%
|
20
16.67%
|
8
6.25%
|
3
4.23%
|
0.000
|
Diabetes
|
Total
|
14
31.82%
|
23
33.33%
|
53
44.17%
|
48
37.5%
|
16
22.53%
|
0.045
|
Diabetes mellitus, unspecified
|
7
15.91%
|
8
11.59%
|
18
15.00%
|
24
18.75%
|
10
14.08%
|
0.74
|
Diabetes mellitus type 2
|
7
15.91%
|
12
17.39%
|
31
25.83%
|
20
15.63%
|
5
7.04%
|
0.021
|
Diabetes mellitus type 1
|
-
|
3
4.34%
|
4
3.33%
|
4
3.13%
|
1
1.40%
|
-
|
Diseases of the circulatory system, unspecified,
|
6
13.64%
|
7
10.14%
|
18
15.00%
|
23
17.97%
|
14
19.72%
|
0.53
|
Supraventricular tachyarrhythmia
|
1
2.27%
|
-
|
-
|
12
9.38%
|
10
14.08%
|
-
|
Heart failure, unspecified
|
-
|
5
7.25%
|
10
8.33%
(
|
7
5.47%
|
8
11.27%
|
-
|
Cerebral ischaemic stroke
|
2
4.55%
|
1
1.49%
|
9
7.50%
|
9
7.03%
|
7
9.86%
|
0.31
|
Acute ischaemic heart disease, unspecified
|
-
|
2
2.90%
|
7
5.83%
|
4
3.13%
|
6
8.45%
|
-
|
Chronic kidney disease
|
4
9.09%
|
3
4.35%
|
14
11.67%
|
14
10.94%
|
9
12.67%
|
0.48
|
Dependence on renal dialysis
|
3
6.82%
|
13
13.04%
|
13
10.83%
|
15
11.72%
|
3
4.23%
|
0.072
|
Kidney failure, unspecified
|
3
6.82%
|
3
4.35%
|
9
7.50%
|
8
6.25%
|
3
4.23%
|
0.87
|
Respiratory failure
|
2
4.55%
|
4
5.80%
)
|
3
2.50%
|
4
3.13%
|
6
8.45%
|
0.33
|
Chronic obstructive pulmonary disease
|
1
2.27%
|
-
|
12
10.00%
|
10
7.81%
|
2
2.82%
|
-
|
|
|
|
|
|
|
|
Aggregated group of diseases
|
|
|
|
|
|
|
Diseases of the circulatory system
|
14
31.82%
|
33
47.83%
|
74
61.67%
|
82
64.06%
|
53
74.65%
|
0.000
|
Endocrine, nutritional or metabolic diseases
|
21
47.73%
|
30
43.48%
|
63
52.50%
|
53
41.41%
|
18
25.35%
|
0.007
|
Diseases of the genitourinary system
|
7
15.91%
|
7
10.14%
|
24
20.00%
|
29
22.66%
|
12
16.90%
|
0.27
|
Diseases of the respiratory system+
|
6
13.64%
|
10
14.49%
|
18
15.00%
|
20
15.63%
|
13
18.31%
|
0.96
|
Diseases of the nervous system
|
3
6.82%
|
2
2.90%
|
18
15.00%
|
19
14.84%
|
19
26.76%
|
0.001
|
Diseases of the digestive system
|
4
9.09%
|
8
11.59%
|
18
15.00%
|
10
7.81%
|
4
5.63%
|
0.23
|
Neoplasms
|
1
2.27%
|
7
10.14%
|
12
10.00%
|
16
12.50%
|
4
5.63%
|
0.25
|
Mental, behavioural or neurodevelopmental disorders
|
3
6.82%
|
2
2.90%
|
6
5.00%
|
15
11.72%
|
10
14.08%
|
0.049
|
Factors influencing health status or contact with health services
|
4
9.09%
|
10
14.49%
|
15
12.50%
|
19
14.84%
|
6
8.45%
|
0.66
|
Symptoms, signs or clinical findings, not elsewhere classified.
|
2
4.55%
|
3
4.35%
|
6
5.00%
|
11
8.59%
|
2
2.82%
|
0.46
|
Certain infectious or parasitic diseases
|
3
6.82%
|
2
2.90%
|
6
5.00%
|
4
3.13%
|
5
7.04%
|
0.63
|
Comparison between COVID-19 and pneumonia fatalities
The comparison with pneumonia fatalities yielded that the proportion of males in the COVID-19 death sample is also significantly higher than in the pneumonia death sample as illustrated in Table 3. The age structure and the relative frequency of essential hypertension, diseases of the circulatory system, obesity, dependence of renal dialysis, and chronic kidney failure are also different between the COVID-19 and pneumonia groups. Additionally, the odds to die of COVID- 19 versus hospital pneumonia are 150% greater for men over women with similar comorbidities (see Table 4). Moreover, the adjusted odds to die of COVID-19 compared to pneumonia are 22 times higher for persons with the chronic kidney failure, doubled for persons with hypertension, and are almost equal for patients with type 2 diabetes.
Table 3. Comparison between the characteristics of the COVID-19 and pneumonia groups. In bold the p-values for the differences in the relative frequency between groups that reached the statistical significance with respect to the chi-squared test.
Characteristics
|
COVID-19
|
Pneumonia
|
p-value
|
Sample size
n, %
|
432
100%
|
711
100%
|
|
Sex
n, %
|
|
|
0.000
|
Female
|
150
35%
|
320
45%
|
|
Male
|
282
65%
|
391
54.99%
|
|
Age, mean (sd),
median
|
66.97 (13.07)
68
|
73.14 (13.68)
78
|
|
Age group
n, %
|
|
|
0.000
|
<50
|
44
10.19%
|
49
6.89%
|
|
50-59
|
69
15.97%
|
51
7.17%
|
|
60-69
|
120
27.78%
|
141
19.83%
|
|
70-79
|
128
29.63%
|
199
27.99%
|
|
>=80
|
71
16.44%
|
271
38.12%
|
|
Comorbidities
|
|
|
|
Hypertension
|
162
37.50%
|
181
25.46%
|
0.000
|
Obesity
|
56
12.96%
|
63
8.86%
|
0.03
|
Diabetes mellitus type I
|
12
2.78%
|
5
0.70%
|
0.01
|
Diabetes mellitus type II
|
75
17.36%
|
145
20.39%
|
0.21
|
Dialysis
|
43
9.95%
|
4
0.56%
|
0.000
|
Chronic kidney disease
|
43
9.95%
|
88
12.38%
|
0.21
|
Supraventricular tachyarrhythmia
|
25
5.79%
|
201
28.27%
|
0.000
|
Congestive heart failure
|
12
2.78%
|
204
28.69%
|
0.000
|
COPD
|
24
5.56%
|
72
10.13%
|
0.01
|
Heart failure
|
29
6.71%
|
27
3.80%
|
0.03
|
Kidney failure
|
26
6.02%
|
4
0.56%
|
0.000
|
Respiratory failure
|
19
4.40%
|
88
12.38%
|
0.000
|
Cerebral ischaemic stroke
|
28
6.48%
|
-
|
0.000
|
Table 4: Logistic regression for the comparison between COVID-19 and Pneumonia death groups. In bold the p-values for the covariates that reached the statistical significance
Covariates
|
Coefficient
β
|
OR
(COVID-19 vs. Pneumonia)
|
95% CI
|
p-value
|
Age
|
-0.022
|
0.978
|
[0.968-0.988]
|
0.000
|
Sex (0=female)
|
0.414
|
1.513
|
[1.145-2.000]
|
0.01
|
Hypertension
|
0.740
|
2.096
|
[1.560-2.818]
|
0.000
|
Obesity
|
0.266
|
1.305
|
[0.843-2.019]
|
0.23
|
Diabetes type 2
|
-0.355
|
0.701
|
[0.494-0.996]
|
0.05
|
Dialysis
|
3.104
|
22.290
|
[7.244-68.590]
|
0.000
|
Chronic kidney disease
|
-0.522
|
0.593
|
[0.369-0.955]
|
0.03
|
Heart failure
|
1.095
|
2.990
|
[1.612-5.546]
|
0.000
|
COPD
|
-0.730
|
0.482
|
[0.284-0.817]
|
0.01
|
Supraventricular
tachyarrhythmia
|
-1.687
|
0.185
|
[0.116-0.295]
|
0.000
|
Respiratory failure
|
-1.045
|
0.352
|
[0.203-0.609]
|
0.000
|
Intercept
|
0.884
|
2.420
|
|
0.02
|
The burden of disease and survival probability
In the next step, we extended the analysis to the coexistence of multiple health conditions, and we focused on the multimorbidity. This has a mean value of 2.109, an SD of 1.66 and a median value of 2 in our study. The multimorbidity factor with levels 0,1,2,3,4 or more than 5 for single or aggregated comorbidities is significantly associated with the factor age but not with sex, confirming the pattern displayed at the level of single conditions and group of diseases (Table S3 and Table S4).
CCI had a mean of 1.324, SD of 0.95 and median of one indicating that the severity of the underlying medical condition in the study was mainly mild. This was confirmed by the high relative frequency of 91.20% for the mild level for the factor CCI. There were no significant differences between males and females or among age groups with respect to CCI (Table S3 and Table S4). Moreover, 1-year survival probability averaged out at 81%, and 50% of the study individuals had a prognosis for it of 85%.
Co-occurrence patterns
The co-occurrence of the diseases is of great importance since the mean multimorbidity of 2.109 indicates that not single but pair of diseases are pre-conditions in our mortality study. Pearson's phi coefficient is a robust measure of the comorbidity association, and its values for diseases in general and separated for men and women are displayed in the Table S5. Different co-occurrence patterns emerge between men and women with the striking lack of the correlation between pneumonia or respiratory failure, respectively stroke and dissociative neurological symptom disorder in the female profile. The largest value of the parameter phi of 0.47 respectively 0.40 for female and 0.59 for male corresponds to kidney failure and dependence to dialysis, while the latter is also correlated with chronic kidney disease as the correlation lies by 0.30 in general and
0.38 in men. They are followed by pneumonia or respiratory failure with phi equal to 0.45 in general and to 0.55 in male and by stroke and dissociative neurological symptom disorder with a correlation coefficient of 0.41 in the whole study and of 0.49 in men. Diabetes mellitus type 2 is correlated with hypertension not only for the whole study (phi = 0.11) but also in the male (phi=0.10) and female subsample (phi=0.11). Hypertension appears mainly together with stroke with phi equal to 0.15, 0.16 respectively 0.14 for the whole study, men and women.
Finally, a comorbidity network analysis rounded off our exploratory investigation. This identified 11 clusters in the general study, 8 clusters in male, and 6 clusters in female subgroups, each including two or three diseases reinforcing the previous results about pair of diseases as pre- condition for COVID-19 fatalities (see Figure 2).