Abstract
Aim
To compare the characteristics and clinical course of patients with coronavirus disease (COVID-19) according to the healthcare level of the admitted hospital, to provide an insight into determining the appropriate level of care for each patient.
Methods
This retrospective, observational study utilized data from the COVID-19 Registry Japan (COVIREGI-JP), the largest Japanese registry of hospitalized patients with COVID-19. Datasets were obtained from reports filed as of May 31, 2022.
Results
A total of 59,707 patients (2004 in the primary care group, 41,420 in the secondary care group, and 16,283 in the tertiary care group) from 585 facilities were included in the analysis. Patients with established risk factors for severe disease, such as old age and the presence of comorbidities, were treated at higher care facilities and had poorer initial conditions and in-hospital clinical course, as well as higher mortality. Analysis of the fatality rates for each complication suggested that patients with complications requiring procedures (e.g. pleural effusions, myocardial ischemia, and arrhythmia) may have better survival rates in facilities with specialist availability. The number of deaths and severe COVID-19 cases in this study were notably less than those reported overseas.
Conclusion
Our results showed that more difficult COVID-19 cases with poor outcomes were treated at higher care level facilities in Japan. Attending to possible complications may be useful for selecting an appropriate treatment hospital. Healthcare providers need to maintain a broad perspective on the distribution of medical resources.
Keywords: COVID-19, Medical resources, Healthcare policy, Level of care, Japan, Registry
1. Introduction
The surges of the coronavirus disease (COVID-19) pandemic continue to burden the Japanese healthcare system. Data suggests that newer strains of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2; the causative agent of COVID-19), such as the Omicron variant, cause less severe illness and death, but are more transmissible, leading to a sudden surge in cases. Under such circumstances, it is important to consider the allocation of limited medical resources. Additionally, even when the respiratory infection itself is mild, significant risks of poor outcomes remain owing to worsening comorbidities or the development of new complications. With different degrees of accessibility to specialized treatment in each healthcare facility, it is often difficult for healthcare providers to determine the appropriate level of medical service for the patient. The “COVID-19 Registry Japan” (COVIREGI-JP), a nationwide COVID-19 inpatient registry, provides a comprehensive analysis of Japanese COVID-19 cases among various healthcare facilities [1]. However, as no analysis has been based on the facility's level of care, the inter-hospital differences in patient characteristics and their clinical course remain unclear. Here, we compared the characteristics and outcomes of infected patients among primary, secondary, and tertiary level care facilities with the aim to examine whether the level of care they received was appropriate.
2. Material and methods
2.1. Study design and data sources
This retrospective observational study utilized data from the COVIREGI-JP, a Japanese registry of more than 70,000 patients with COVID-19, hospitalized in 695 facilities [1]. Additional details of the registry including patient inclusion criteria and data collection method have been described elsewhere [1,2]. Each research collaborator completed a report form that was developed specifically for the registry. Datasets were obtained from reports filed as of May 31, 2022.
2.2. Healthcare level of facility
Participating facilities in the registry were classified as primary, secondary, or tertiary care hospitals. To determine the level of care, we referred to the emergency care category of the facility, which was designated based on the Medical Care Act. Japan has three categories of emergency hospitals: a “primary” emergency center, dealing with patients who can be managed as outpatients; a “secondary” emergency center, dealing with patients who can be managed as inpatients on a general medical floor; and a “tertiary” emergency center, dealing with patients who require specialized or intensive care. Here in our study, a facility is classified as a primary, secondary, and tertiary care hospital if it is designated as a primary, secondary, and tertiary emergency center, respectively. Smaller facilities that did not participate in emergency care (e.g. psychiatric hospitals or clinics with minimal beds) were also included in the primary care category. Patients treated at primary, secondary, and tertiary care hospitals were defined as primary, secondary, and tertiary care groups, respectively.
2.3. Baseline characteristics
Data on patients' demographics, including age, sex, body mass index (BMI), smoking and alcohol habits, and comorbidities were collected. Comorbidities were pre-specified in the report form and included myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic obstructive pulmonary disease (COPD), chronic lung disease, bronchial asthma, liver disease, peptic ulcer, hypertension, hyperlipidemia, diabetes with and without complications, obesity, kidney disease, hemodialysis, solid tumor, leukemia, lymphoma, immunosuppression, and collagen disease. The vital signs (temperature, heart rate, respiratory rate, and blood pressures), COVID-19 severity, oxygen support and route, and findings on imaging (X-ray and/or CT) on admission were also recorded. COVID-19 severity on admission was defined as either severe or non-severe, based on previous studies [1,3]. Severe cases were those that met at least one of the following criteria: (1) requiring invasive or non-invasive mechanical ventilation, (2) requiring supplemental oxygen, (3) SpO2 ≤ 94% in room air, or (4) tachypnea with a respiratory rate ≥24 breaths per minute. Those who did not meet any of these criteria were classified as non-severe.
2.4. Clinical course and complications
The clinical course for each group was evaluated by intensive care unit (ICU) admissions and the use of supportive care, including oxygen support, prone positioning, tracheotomy, nitric oxide (NO) inhalation, vasopressor/inotropic administration, renal replacement therapy/hemodialysis, and blood transfusion. Complications specified in the report form were collected, which included pneumonia coinfections, respiratory conditions (pneumothorax, pleural effusion, bloody sputum, acute respiratory distress syndrome [ARDS]), bacteremia, cardiovascular conditions (myocarditis/pericarditis, arrhythmia, myocardial ischemia, deep vein thrombosis [DVT]), pulmonary thromboembolism [PE]), neurological conditions (meningitis, seizure, stroke), and gastrointestinal bleeding.
2.5. Outcomes
We collected data on the outcome (death or discharge) and the length of hospitalization. Discharge destination was classified as one of the following: discharge to home, transfer to a non-medical facility, transfer to a rehabilitation/long-term care facility, transfer to a medical facility for intensive care, death, and others. Mortality rates by complications were also calculated to determine which complications had a fatal course and whether they differed at the institutional level.
2.6. Statistical analysis
Continuous variables are expressed as medians and interquartile ranges and categorical variables as number of cases and percentages. Continuous variables are compared between the groups using the Kruskal-Wallis test and categorical variables using the chi-square test. All statistical analysis were performed using R version 4.1.3 (R Foundation for Statistical Computing, Vienna, Austria).
2.7. Ethics
This study was approved by the National Center for Global Health and Medicine Ethics Review Board, (NCGM-G-004037-02) which waived the requirement for informed consent due to the retrospective nature of the study. The protocol was conducted in accordance with the Declaration of Helsinki.
3. Results
Data of 59,707 cases from 585 facilities were included in the analysis, comprising 2,004, 41,420, and 16,283 patients in the primary, secondary, and tertiary care groups, respectively. The healthcare levels of the included facilities were categorized as 20, 421, and 144 primary, secondary, and tertiary care hospitals, respectively.
3.1. Patient characteristics and condition on admission
The patient demographics and comorbidities are summarized in Table 1 . The median age of the patients was 54, 57, and 58 years in the primary, secondary, and tertiary care groups, respectively. The sex and BMI were similar in the primary and tertiary care groups, with 60% males and a median BMI of 23.6. The secondary care group consisted of 55.6% males and had a lower median BMI of 23.2. Regarding alcohol consumption and drinking habits, patients in higher-level facilities were more likely to consume alcohol daily and have a history of smoking. Those who had at least one comorbidity constituted 51.8%, 55.5%, and 60.6% of the patients in the primary, secondary, and tertiary care groups, respectively. The tertiary care group had the highest prevalence for all listed comorbidities, with the only exception of dementia (9.5%, 8.6%, and 6.8% in the primary, secondary, and tertiary care groups, respectively).
Table 1.
Baseline characteristics of patients with COVID-19 by care level of the admitted hospital. Data are presented as n (%) or median [IQR]. Numbers do not always add up due to missing values.
| Total (n = 59,707) | Primary (n = 2004) | Secondary (n = 41,420) | Tertiary (n = 16,283) | p-value | ||
|---|---|---|---|---|---|---|
| Age (years) | 57 [39, 74] | 54 [37, 74] | 57 [39, 74] | 58 [41, 74] | <0.001 | |
| Sex | Male | 33990 (56.9) | 1194 (59.6) | 23026 (55.6) | 9770 (60.0) | <0.001 |
| Female | 25698 (43.0) | 807 (40.3) | 18381 (44.4) | 6510 (40.0) | ||
| BMI (kg/m^2) | 23.4 [20.7, 26.5] | 23.6 [20.8, 26.9] | 23.2 [20.6, 26.3] | 23.6 [20.9, 26.9] | <0.001 | |
| Smoking history | Current smoking | 8936 (15.0) | 347 (17.3) | 6142 (14.8) | 2447 (15.0) | <0.001 |
| Past smoking | 12391 (20.8) | 380 (19.0) | 8136 (19.6) | 3875 (23.8) | ||
| Never | 29023 (48.6) | 1028 (51.3) | 20533 (49.6) | 7462 (45.8) | ||
| Unknown | 9250 (15.5) | 244 (12.2) | 6554 (15.8) | 2452 (15.1) | ||
| Alcohol intake | Daily | 3309 (5.5) | 77 (3.8) | 2201 (5.3) | 1031 (6.3) | <0.001 |
| Occasional | 18185 (30.5) | 372 (18.6) | 12623 (30.5) | 5190 (31.9) | ||
| Never | 23365 (39.1) | 580 (28.9) | 16742 (40.4) | 6043 (37.1) | ||
| Unknown | 14151 (23.7) | 965 (48.2) | 9397 (22.7) | 3789 (23.3) | ||
| Comorbidities | Any comorbidity | 33892 (56.8) | 1039 (51.8) | 22988 (55.5) | 9865 (60.6) | <0.001 |
| Myocardial infarction | 1175 (2.0) | 42 (2.1) | 733 (1.8) | 400 (2.5) | <0.001 | |
| Congestive heart failure | 1979 (3.3) | 48 (2.4) | 1344 (3.2) | 587 (3.6) | 0.008 | |
| Peripheral vascular disease | 1002 (1.7) | 20 (1.0) | 626 (1.5) | 356 (2.2) | <0.001 | |
| Cerebrovascular disease | 4057 (6.8) | 103 (5.1) | 2827 (6.8) | 1127 (6.9) | 0.009 | |
| Dementia | 4851 (8.1) | 190 (9.5) | 3548 (8.6) | 1113 (6.8) | <0.001 | |
| COPD | 1480 (2.5) | 29 (1.4) | 958 (2.3) | 493 (3.0) | <0.001 | |
| Chronic lung disease (excluding COPD) | 940 (1.6) | 34 (1.7) | 602 (1.5) | 304 (1.9) | 0.003 | |
| Bronchial asthma | 3277 (5.5) | 97 (4.8) | 2284 (5.5) | 896 (5.5) | 0.46 | |
| Mild liver disease | 1418 (2.4) | 31 (1.5) | 773 (1.9) | 614 (3.8) | <0.001 | |
| Moderate to severe liver disease | 228 (0.4) | 7 (0.3) | 150 (0.4) | 71 (0.4) | 0.42 | |
| Peptic ulcer | 494 (0.8) | 12 (0.6) | 317 (0.8) | 165 (1.0) | 0.011 | |
| Hypertension | 18259 (30.6) | 508 (25.3) | 12422 (30.0) | 5329 (32.7) | <0.001 | |
| Hyperlipidemia | 8679 (14.5) | 191 (9.5) | 5822 (14.1) | 2666 (16.4) | <0.001 | |
| Diabetes without complications | 8886 (14.9) | 243 (12.1) | 6035 (14.6) | 2608 (16.0) | <0.001 | |
| Diabetes with complications | 1311 (2.2) | 24 (1.2) | 737 (1.8) | 550 (3.4) | <0.001 | |
| Obesity | 4190 (7.0) | 91 (4.5) | 2770 (6.7) | 1329 (8.2) | <0.001 | |
| Moderate to severe kidney disease | 1080 (1.8) | 20 (1.0) | 603 (1.5) | 457 (2.8) | <0.001 | |
| Hemodialysis | 611 (1.0) | 6 (0.3) | 351 (0.8) | 254 (1.6) | <0.001 | |
| Solid tumor | 2148 (3.6) | 70 (3.5) | 1401 (3.4) | 677 (4.2) | <0.001 | |
| Metastatic solid tumor | 555 (0.9) | 10 (0.5) | 375 (0.9) | 170 (1.0) | 0.03 | |
| Leukemia | 176 (0.3) | 5 (0.2) | 109 (0.3) | 62 (0.4) | 0.06 | |
| Lymphoma | 342 (0.6) | 8 (0.4) | 237 (0.6) | 97 (0.6) | 0.52 | |
| Immunosuppression | 1508 (2.5) | 48 (2.4) | 957 (2.3) | 503 (3.1) | <0.001 | |
| Collagen disease | 817 (1.4) | 17 (0.8) | 554 (1.3) | 246 (1.5) | 0.02 |
BMI, body mass index; COPD, chronic obstructive pulmonary disease.
Table 2 shows the patients' medical conditions on hospital admission. The percentage of patients with severe COVID-19 was 26.2%, 29.3%, and 44.9% in the primary, secondary, and tertiary care groups, respectively. Advanced care facilities treated more patients with oxygen demand on admission (10.4% in primary. vs. 13.4% in secondary vs. 13.4% in tertiary care group). Regarding the administration route for patients on oxygen support, 76.1% of patients with oxygen support in the primary care group and 71.3% in the secondary care group required only nasal cannula. In contrast, in the tertiary care group, nasal cannula was sufficient for only 46.3% of those requiring oxygen support, while the remainder of patients required more intensive respiratory support. The percentage of patients with pneumonia findings on imaging in the primary and secondary care groups was similar at 52.7% for X-ray and 71% for CT, while it reached 63.3% for X-ray and 79.8% for CT in the tertiary care group.
Table 2.
Initial condition of patients with COVID-19 by care level of the admitted hospital. Data are presented as n (%) or median [IQR]. Numbers do not always add up due to missing values.
| Total (n = 59,707) | Primary (n = 2004) | Secondary (n = 41,420) | Tertiary (n = 16,283) | p-value | |||
|---|---|---|---|---|---|---|---|
| Vital signs | Temperature (°C) | 37 [36.6, 37.8] | 37 [36.6, 37.7] | 37 [36.6, 37.7] | 37.1 [36.6, 37.8] | <0.001 | |
| Heart rate (beats/minute) | 86 [76, 98] | 87 [76, 98] | 86 [75, 98] | 87 [76, 99] | <0.001 | ||
| Respiratory rate (breaths/minute) | 18 [16,21] | 18 [16,20] | 18 [16,21] | 19 [16,23] | <0.001 | ||
| Systolic blood pressure (mmHg) | 127 [114, 142] | 125 [113, 138] | 127 [114, 142] | 127 [114, 142] | <0.001 | ||
| Diastolic blood pressure (mmHg) | 79 [70, 88] | 80 [70, 88] | 79 [70, 88] | 78 [69, 88] | <0.001 | ||
| Severity of COVID-19 | Severe | 19975 (33.5) | 525 (26.2) | 12142 (29.3) | 7308 (44.9) | <0.001 | |
| Non-severe | 39732 (66.5) | 1479 (73.8) | 29278 (70.7) | 8975 (55.1) | |||
| Oxygen support | Yes | 10401 (17.4) | 209 (10.4) | 5541 (13.4) | 4651 (28.6) | <0.001 | |
| Route of oxygen support | Nasal cannula | 6265 (60.2) | 159 (76.1) | 3953 (71.3) | 2153 (46.3) | ||
| Face mask | 1885 (18.1) | 33 (15.8) | 863 (15.6) | 989 (21.3) | |||
| Reservoir mask | 1326 (12.7) | 16 (7.7) | 502 (9.1) | 808 (17.4) | |||
| High-flow oxygen device | 252 (2.4) | 0 (0) | 108 (1.9) | 144 (3.1) | |||
| Non-invasive ventilation | 46 (0.4) | 0 (0) | 15 (0.3) | 31 (0.7) | |||
| Invasive mechanical ventilation | 623 (6.0) | 1 (0.5) | 100 (1.8) | 522 (11.2) | |||
| ECMO | 4 (0) | 0 (0) | 0 (0) | 4 (0.1) | |||
| None (room air) | 48974 (82) | 1795 (89.6) | 35657 (86.1) | 11522 (70.8) | |||
| Imaging findings | |||||||
| Chest X-ray | Data available | 0 | 0 | 0 | 0 | ||
| No abnormality | 13244 (42.2) | 397 (46.4) | 9320 (45.6) | 3527 (35.1) | <0.001 | ||
| Pneumonia | 17596 (56.1) | 451 (52.7) | 10781 (52.7) | 6364 (63.3) | |||
| Abnormality (excluding pneumonia) | 520 (1.7) | 7 (0.8) | 354 (1.7) | 159 (1.6) | |||
| CT | Data available | 0 | 0 | 0 | 0 | ||
| No abnormality | 10128 (23.8) | 437 (25.8) | 7769 (26.1) | 1922 (17.3) | <0.001 | ||
| Pneumonia | 31244 (73.4) | 1212 (71.5) | 21166 (71.1) | 8866 (79.8) | |||
| Abnormality (excluding pneumonia) | 1214 (2.9) | 46 (2.7) | 844 (2.8) | 324 (2.9) | |||
ECMO, extracorporeal membrane oxygenation; CT, computed tomography.
3.2. Clinical course, complications, and outcomes
Table 3 shows the clinical indicators of the in-hospital course for each group. ICU admission occurred in 0.4%, 3.8%, and 18.1% of patients in the primary, secondary, and tertiary care groups, respectively. Various types of intensive treatment (e.g. oxygen support, vasopressor/inotropic administration, and renal replacement therapy), were consistently used more frequently in higher care facilities. With the exception of nasal cannula and high-flow oxygen device, such intensive treatment was used in less than ten patients in the primary care group. ECMO was used in a total of 208 patients, which consisted of 0, 32, and 176 patients in the primary, secondary, and tertiary care groups, respectively.
Table 3.
In-hospital clinical course of patients with COVID-19 by care level of the admitted hospital. Data are presented as n (%). Numbers do not always add up due to missing values.
| Total (n = 59,707) | Primary (n = 2004) | Secondary (n = 41,420) | Tertiary (n = 16,283) | p-value | ||
|---|---|---|---|---|---|---|
| ICU admission | 4554 (7.6) | 9 (0.4) | 1593 (3.8) | 2952 (18.1) | <0.001 | |
| Oxygen support | 22785 (38.2) | 631 (31.5) | 14174 (34.2) | 7980 (49.0) | <0.001 | |
| Route of oxygen support | Nasal cannula or face/resorvoir mask | 22526 (37.7) | 629 (31.4) | 14069 (34.0) | 7828 (48.1) | |
| High-flow oxygen device | 3085 (5.2) | 52 (2.6) | 1515 (3.7) | 1518 (9.3) | ||
| Non-invasive ventilation | 804 (1.3) | 4 (0.2) | 252 (0.6) | 548 (3.4) | ||
| Invasive mechanical ventilation | 2664 (4.5) | 8 (0.4) | 945 (2.3) | 1711 (10.5) | ||
| ECMO | 208 (0.3) | 0 (0) | 32 (0.1) | 176 (1.1) | ||
| Prone positioning | 2896 (4.9) | 2 (0.1) | 1201 (2.9) | 1693 (10.4) | <0.001 | |
| Tracheotomy | 500 (0.8) | 2 (0.1) | 120 (0.3) | 378 (2.3) | <0.001 | |
| Nitric oxide inhalation | 42 (0.1) | 0 (0) | 13 (0) | 29 (0.2) | <0.001 | |
| Vasopressor/Inotropic administration | 1520 (2.5) | 1 (0) | 507 (1.2) | 1012 (6.2) | <0.001 | |
| RRT or hemodialysis | 806 (1.3) | 7 (0.3) | 352 (0.8) | 447 (2.7) | <0.001 | |
| Blood transfusion | 1126 (1.9) | 7 (0.3) | 410 (1.0) | 709 (4.4) | <0.001 |
ICU, intensive care unit; ECMO, extracorporeal membrane oxygenation; RRT; renal replacement therapy.
As indicated in Table 4 , complication rates tended to be higher in the tertiary care group. In hospitals providing a higher level of care, non-COVID-19 pneumonia coinfection and respiratory-related complications occurred more frequently than in hospitals that provided a lower level of care. Non-COVID-19 pneumonia coinfection cases were infrequent in the primary care group; there were only 5 cases of viral pneumonia, 67 cases of bacterial pneumonia, 3 cases of methicillin-resistant Staphylococcus aureus (MRSA) pneumonia, and 2 cases of Pseudomonas aeruginosa pneumonia. Patients developed ARDS in 1.0%, 3.2%, and 8.5% of the primary, secondary, and tertiary care groups, respectively. The incidence rates for almost all the other complications were lowest in the primary care group and highest in the tertiary care group.
Table 4.
Complications in patients with COVID-19 by care level of the admitted hospital. Data are presented as n (%).
| Total (n = 59,707) | Primary (n = 2004) | Secondary (n = 41,420) | Tertiary (n = 16,283) | p-value | |
|---|---|---|---|---|---|
| Pneumonia coinfection | |||||
| Viral pneumonia (excluding COVID-19) | 511 (0.9) | 5 (0.2) | 308 (0.7) | 198 (1.2) | <0.001 |
| Bacterial pneumonia (including HAP/VAP) | 3226 (5.4) | 67 (3.3) | 1703 (4.1) | 1456 (8.9) | <0.001 |
| MRSA pneumonia | 281 (0.5) | 3 (0.1) | 119 (0.3) | 159 (1.0) | <0.001 |
| Pseudomonas aeruginosa pneumonia | 190 (0.3) | 2 (0.1) | 85 (0.2) | 103 (0.6) | <0.001 |
| MDRP pneumonia | 30 (0.1) | 0 (0) | 8 (0) | 22 (0.1) | <0.001 |
| Respiratory | |||||
| Pneumothorax | 249 (0.4) | 0 (0) | 95 (0.2) | 154 (0.9) | <0.001 |
| Pleural effusion | 1751 (2.9) | 31 (1.5) | 957 (2.3) | 763 (4.7) | <0.001 |
| Bloody sputum | 459 (0.8) | 13 (0.6) | 236 (0.6) | 210 (1.3) | <0.001 |
| ARDS | 2747 (4.6) | 20 (1.0) | 1335 (3.2) | 1392 (8.5) | <0.001 |
| Infectious | |||||
| Bacteremia | 595 (1.0) | 5 (0.2) | 249 (0.6) | 341 (2.1) | <0.001 |
| Cardiovascular | |||||
| Endocarditis | 18 (0) | 0 (0) | 11 (0) | 7 (0) | <0.001 |
| Myocarditis/Pericarditis | 58 (0.1) | 0 (0) | 27 (0.1) | 31 (0.2) | <0.001 |
| Arrhythmia | 262 (0.4) | 2 (0.1) | 150 (0.4) | 110 (0.7) | <0.001 |
| Cardiac ischemia | 104 (0.2) | 7 (0.3) | 47 (0.1) | 50 (0.3) | <0.001 |
| Deep vein thrombosis | 450 (0.8) | 4 (0.2) | 248 (0.6) | 198 (1.2) | <0.001 |
| Pulmonary thromboembolism | 218 (0.4) | 1 (0) | 134 (0.3) | 83 (0.5) | <0.001 |
| Neurological | |||||
| Meningitis | 36 (0.1) | 0 (0) | 21 (0.1) | 15 (0.1) | <0.001 |
| Seizure | 152 (0.3) | 1 (0) | 80 (0.2) | 71 (0.4) | <0.001 |
| Stroke | 212 (0.4) | 2 (0.1) | 119 (0.3) | 91 (0.6) | <0.001 |
| Gastrointestinal | |||||
| Gastrointestinal bleeding | 363 (0.6) | 4 (0.2) | 184 (0.4) | 175 (1.1) | <0.001 |
HAP, hospital-acquired (or nosocomial) pneumonia; VAP, ventilator-associated pneumonia; MRSA, Methicillin-resistant Staphylococcus aureus; MDRP, Multidrug-resistant Pseudomonas aeruginosa; ARDS, acute respiratory distress syndrome.
The outcome and length of hospital stay among the three groups are shown in Table 5 . More deaths and fewer home discharges were observed in patients treated at higher care facilities. The median length of stay was 11 days in the tertiary care group and 10 days in the primary and secondary care groups.
Table 5.
Outcomes of patients with COVID-19 by care level of the admitted hospital. Data are presented as n (%). Numbers do not always add up due to missing values.
| Total (n = 59,707) | Primary (n = 2004) | Secondary (n = 41,420) | Tertiary (n = 16,283) | p-value | |||
|---|---|---|---|---|---|---|---|
| Outcome | Death | 2816 (4.7) | 52 (2.6) | 1672 (4.0) | 1092 (6.7) | <0.001 | |
| Discharge | 56865 (95.2) | 1952 (97.4) | 39723 (95.9) | 15190 (93.3) | |||
| Discharge destination | Home | 44580 (78.4) | 1627 (83.4) | 31347 (78.9) | 11606 (76.4) | ||
| Non-medical facility | 2714 (4.8) | 36 (1.8) | 2023 (5.1) | 655 (4.3) | |||
| Rehabilitation/long-term care facility | 9353 (16.4) | 276 (14.1) | 6162 (15.5) | 2915 (19.2) | |||
| Medical facility for intensive care | 199 (0.3) | 13 (0.7) | 178 (0.4) | 8 (0.1) | |||
| Others | 19 (0) | 0 (0) | 13 (0) | 6 (0) | |||
| Length of hospital stay (days) | 10 [8,15] | 10 [7,14] | 10 [8,14] | 11 [8,17] | <0.001 |
3.3. Outcomes for each complication
Table 6 shows the analysis of in-hospital mortality by complication. The complications with the highest mortality rates overall were MRSA pneumonia (50.2%), pneumothorax (47.8%), gastrointestinal bleeding (44.6%), bacteremia (44.2%), Pseudomonas aeruginosa pneumonia (42.1%, and 43.3% for multiple drug resistant strains), and myocardial ischemia (37.5%).
Table 6.
In-hospital mortality of patients with COVID-19 with various complications by care level of the admitted hospital. Data are presented as n (%). “-” in primary care group indicates that there were no reported cases of the complication.
| Total (n = 59,707) |
Primary (n = 2004) |
Secondary (n = 41,420) |
Tertiary (n = 16,283) |
p-value (primary vs secondary) | p-value (secondary vs tertiary) | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Cases | Deaths (mortality rate) | Cases | Deaths | Cases | Deaths | Cases | Deaths | |||
| Pneumonia coinfection | ||||||||||
| Viral pneumonia (excluding COVID-19) | 511 | 91 (17.8) | 5 | 0 (0) | 308 | 44 (14.3) | 198 | 47 (23.7) | 0.36 | 0.006 |
| Bacterial pneumonia (including HAP/VAP) | 3226 | 913 (28.3) | 67 | 22 (32.8) | 1703 | 448 (26.3) | 1456 | 443 (30.4) | 0.23 | 0.01 |
| MRSA pneumonia | 281 | 141 (50.2) | 3 | 2 (66.7) | 119 | 61 (51.3) | 159 | 78 (49.1) | 0.59 | 0.71 |
| Pseudomonas aeruginosa pneumonia | 190 | 80 (42.1) | 2 | 1 (50.0) | 85 | 34 (40.0) | 103 | 45 (43.7) | 0.77 | 0.61 |
| MDRP pneumonia | 30 | 13 (43.3) | 0 | 0 (−) | 8 | 3 (37.5) | 22 | 10 (45.5) | N/A | 0.69 |
| Respiratory | ||||||||||
| Pneumothorax | 249 | 119 (47.8) | 0 | 0 (−) | 95 | 44 (46.3) | 154 | 75 (48.7) | N/A | 0.71 |
| Pleural effusion | 1751 | 544 (31.1) | 31 | 13 (41.9) | 957 | 316 (33.0) | 763 | 215 (28.2) | 0.29 | 0.03 |
| Bloody sputum | 459 | 130 (28.3) | 13 | 0 (0) | 236 | 68 (28.8) | 210 | 62 (29.5) | 0.02 | 0.86 |
| ARDS | 2747 | 864 (31.5) | 20 | 10 (50) | 1335 | 419 (31.4) | 1392 | 435 (31.3) | 0.07 | 0.93 |
| Infectious | ||||||||||
| Bacteremia | 595 | 263 (44.2) | 5 | 1 (20.0) | 249 | 119 (47.8) | 341 | 143 (41.9) | 0.21 | 0.15 |
| Cardiovascular | ||||||||||
| Endocarditis | 18 | 1 (5.6) | 0 | 0 (−) | 11 | 1 (9.1) | 7 | 0 (0) | N/A | 0.41 |
| Myocarditis/Pericarditis | 58 | 13 (22.4) | 0 | 0 (−) | 27 | 5 (18.5) | 31 | 8 (25.8) | N/A | 0.5 |
| Arrhythmia | 262 | 96 (36.6) | 2 | 1 (50.0) | 150 | 62 (41.3) | 110 | 33 (30.0) | 0.8 | 0.06 |
| Myocardial ischemia | 104 | 39 (37.5) | 7 | 6 (85.7) | 47 | 17 (36.2) | 50 | 16 (32.0) | 0.01 | 0.66 |
| Deep vein thrombosis | 450 | 87 (19.3) | 4 | 0 (0) | 248 | 39 (15.7) | 198 | 48 (24.2) | 0.38 | 0.02 |
| Pulmonary thromboembolism | 218 | 34 (15.6) | 1 | 0 (0) | 134 | 14 (10.4) | 83 | 20 (24.1) | 0.73 | 0.007 |
| Neurological | ||||||||||
| Meningitis | 36 | 10 (27.8) | 0 | 0 (−) | 21 | 7 (33.3) | 15 | 3 (20.0) | N/A | 0.37 |
| Seizure | 152 | 37 (24.3) | 1 | 1 (100.0) | 80 | 19 (23.8) | 71 | 17 (23.9) | 0.07 | 0.97 |
| Stroke | 212 | 70 (33.0) | 2 | 0 (0) | 119 | 34 (28.6) | 91 | 36 (39.6) | 0.37 | 0.09 |
| Gastrointestinal | ||||||||||
| Gastrointestinal bleeding | 363 | 162 (44.6) | 4 | 1 (25.0) | 184 | 85 (46.2) | 175 | 76 (43.4) | 0.39 | 0.59 |
HAP, hospital-acquired (or nosocomial) pneumonia; VAP, ventilator-associated pneumonia; MRSA, Methicillin-resistant Staphylococcus aureus; MDRP, Multidrug-resistant Pseudomonas aeruginosa; ARDS, acute respiratory distress syndrome.
Mortality rates varied substantially by the type of complication. ARDS and non-COVID-19 mixed pneumonia were rare in primary care facilities, but when they did occur, the mortality rate exceeded 30%, which was much higher than those in secondary and tertiary care settings. Myocardial ischemia also had a high fatality rate of 85.7% in the primary care group, which was significantly higher than the 36.2% and 32.0% in the secondary and tertiary care groups, respectively.
Compared between the primary group and the secondary care groups, mortality rate of myocardial ischemia was significantly lower in the secondary group (36.2% vs. 85.7%, p = 0.01), whereas for bloody sputum it was higher in secondary group (28.8% vs. 0%, p = 0.02). The fatality rate was higher in the tertiary care group than in the secondary care group when complicated by viral pneumonia other than COVID-19 (23.7% vs. 14.3%, p = 0.006), bacterial pneumonia (30.4% vs. 26.3%, p = 0.01), deep vein thrombosis (24.2% vs. 15.7%, p = 0.02), and pulmonary embolism (24.1% vs. 10.4%, p = 0.007). Conversely, the fatality rate was higher in the secondary care group than that in the tertiary care group when complicated by pleural effusion (33.0% vs. 28.2%, p = 0.03), with a similar trend observed for arrhythmia (41.3% vs. 30.0%, p = 0.06).
4. Discussion and conclusions
This is the first report in Japan to compare the characteristics and clinical course of patients with COVID-19 according to the health care level of the admitted hospital. Previous reports from Japan have focused on analyses classified according to patient factors; however, to improve actual practice, assessments based on factors regarding healthcare providers are equally important. In Japan, during the stable phases of the pandemic, there have been no major restrictions on inpatient or outpatient practice. However, even 3 years after the start of the pandemic, hospitals remain short of beds during surges of infections. Determining the appropriate level of care for patients with COVID-19 remains unresolved. Here, we revealed that patients treated at higher-level medical facilities had a higher risk of severe disease, more comorbidities, and a poorer clinical course.
Differences in outcomes among healthcare facilities have previously been reported outside Japan. In a study in the USA, during the first 6 months of the pandemic, the mortality rates were strongly associated with the prevalence of COVID-19 in the hospital's surrounding communities, but neither the number of intensive care unit beds nor academic status was relevant [4]. Another USA report suggested that treatment at a hospital dedicated to COVID-19 care performed better in terms of the survival rates of patients with severe COVID-19 [5]. In that study, mortality rates were higher in specialized hospitals, but multivariate adjustment verified that their outcomes were better when patients' severity was accounted for. Socioeconomic determinants, such as poor prognosis in hospitals with poor finances in disadvantaged areas, have also been shown to influence outcomes [6].
We showed that patients at higher risk of severe disease were treated at higher care institutions in Japan. Age, smoking, and the presence of comorbidities are established risk factors for severe COVID-19, and these characteristics are more prevalent in patients in higher care facilities [1,7]. Surprisingly, the opposite trend was observed for dementia. Dementia and end-of-life care are major issues in Japan's aging society [8]. Patients with COVID-19 with dementia may have been sent to a lower-level care hospital because they had previously decided to avoid intensive treatment. Male sex and high BMI are other risk factors for severe disease but were less common in the secondary care group. Japan is a hyper-aged society, with a large proportion of frail elderly women [9], and frailty is also associated with poor prognosis in patients with COVID-19 [10]. A subgroup of elderly women with low BMI may have been treated at higher-care-level facilities, resulting in a unique distribution in terms of sex and BMI among the three groups.
Despite there being no clear difference among the three groups in terms of vital signs alone, severe cases, as defined by respiratory status, were transported to higher care facilities. This was supported by imaging findings and oxygen administration status on admission. Unsurprisingly, patients admitted to higher-level facilities had a poor clinical course after admission. Intensive interventions were most common in tertiary care hospitals, and the tertiary group experienced more complications and had poor outcomes. Conversely, patients treated in primary care settings rarely required intensive care or experienced complications during their course. Although the present study was not adjusted with multivariate analysis, the less favorable clinical course in higher-level care facilities is reasonable, given the percentage of high-risk patients with COVID-19.
The fatality rates classified by each complication provide clues for better management of COVID-19. First, our data also revealed that when complicated by viral pneumonia, bacterial pneumonia, deep vein thrombosis, and pulmonary embolism, tertiary groups had lower rates of survival. This probably reflects the fact that the majority of very severe cases of COVID-19 were treated in tertiary care hospitals. Indeed, severe COVID-19 infection often triggers a process known as immunothrombosis, resulting in DVT and/or PE. Second, for some complications, lower-level care facilities performed worse despite the patients having less severe pneumonia. Specifically, the mortality rate of cardiac ischemia was significantly higher in the primary group than the secondary group, and the mortality rate of pleural effusion and arrhythmia was higher in the secondary group than the tertiary group. This may be attributed to the fact that treatments for these complications are usually procedure-based, whereas other complications in the list are often followed up with pharmacologic or conservative treatment. It is plausible that in patients at higher-level care facilities, who are better prepared by specialists in each field, undergo such procedures quickly and effectively, resulting in better outcomes. Now that the in-hospital survival rate has plateaued because of the development of novel COVID-19 medications [11], further improvement in outcomes depends on the handling of complicated cases. The development of methods for predicting the occurrence of each complication remains a subject for future research.
We also revealed data concerning the healthcare delivery system in Japan. The mortality rate in this study was 2.6%, 4%, and 6.7% in primary, secondary, and tertiary care facilities, respectively, all of which were lower than those reported overseas [[12], [13], [14], [15], [16], [17], [18]]. It is possible that the majority of hospitalized patients in Japan have exceptionally mild disease. Indeed, only 33.5% of the patients in our study were classified as having severe COVID-19 on admission, which is lower than the proportion of severely ill patients reported in several international studies. In a USA study that analyzed 466,677 admissions from April 2020 to April 2021, more than 50% of hospitalized patients presented with acute respiratory failure throughout the study period [19]. Especially with the development of effective oral antiviral medications, clinicians may set a higher bar for admitting non-severe COVID-19 patients. Furthermore, in the present study, the median length of hospital stay was 10–11 days, which is much longer than the 7 days reported in the USA [20]. In fact, a considerable proportion of hospital stay in Japan is due to social reasons (i.e. no caregiver at home) rather than medical indications. The COVID-19 pandemic has had a significant negative impact on the management of other diseases [21,22]. It is crucial to develop a healthcare resource allocation strategy that takes a view of the entire picture as opposed to focusing on COVID-19. At the same time, the government needs to make more efforts to promote smooth transition from acute care centers to recovery/long-term beds.
This study has several limitations. First, selection bias may have occurred due to the nature of this registry study, as described previously [23]. Also, if a patient had a history of multiple COVID-19 hospitalizations or had been transferred, the data may have been duplicated. Furthermore, only hospitalized patients were included in this study. Now that outpatient treatment has become mainstream for mild cases [24], both inpatient and outpatient cases need to be examined to establish a rational healthcare policy.
In conclusion, our study provides evidence that more severely ill patients with COVID-19 with poorer prognoses are treated at higher care facilities in Japan. In addition to the severity of pneumonia, attending to possible complications may assist with selecting an appropriate medical facility for treatment. Furthermore, our findings serve as a reminder for healthcare providers that they may be allocating too many resources toward COVID-19 care at the cost of managing other diseases.
Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Authorship statement
DT and YH contributed to the concept, data interpretation, and writing of the original manuscript. YA was responsible for the data acquisition and analysis. KH, SK, MT, WS, and NO provided supervision and were responsible for the organization and management of the trial. All authors have approved the final version of the manuscript for publication.
Declaration of competing interest
None.
References
- 1.Matsunaga N., Hayakawa K., Terada M., et al. Clinical epidemiology of hospitalized patients with coronavirus disease 2019 (COVID-19) in Japan: report of the COVID-19 registry Japan. Clin Infect Dis. 2021;73 doi: 10.1093/cid/ciaa1470. e3677–89–e3689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Terada M., Ohtsu H., Saito S., Hayakawa K., Tsuzuki S., Asai Y., et al. Risk factors for severity on admission and the disease progression during hospitalisation in a large cohort of patients with COVID-19 in Japan. BMJ Open. 2021;11 doi: 10.1136/bmjopen-2020-047007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Beigel J.H., Tomashek K.M., Dodd L.E., Mehta A.K., Zingman B.S., Kalil A.C., et al. Remdesivir for the treatment of COVID-19 – final. Report. N Engl J Med. 2020;383:1813–1826. doi: 10.1056/NEJMoa2007764. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Asch D.A., Sheils N.E., Islam M.N., Chen Y., Werner R.M., Buresh J., et al. Variation in US hospital mortality rates for patients admitted with COVID-19 during the first 6 months of the pandemic. JAMA Intern Med. 2021;181:471–478. doi: 10.1001/jamainternmed.2020.8193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bergman Z.R., Usher M., Olson A., Chipman J.G., Brunsvold M.E., Beilman G., et al. Comparison of outcomes and process of care for patients treated at hospitals dedicated for COVID-19 care vs other hospitals. JAMA Netw Open. 2022;5 doi: 10.1001/jamanetworkopen.2022.0873. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Asch D.A., Islam M.N., Sheils N.E., Chen Y., Doshi J.A., Buresh J., et al. Patient and hospital factors associated with differences in mortality rates among black and white US Medicare beneficiaries hospitalized with COVID-19 Infection. JAMA Netw Open. 2021;4 doi: 10.1001/jamanetworkopen.2021.12842. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Clift A.K., von Ende A., Tan P.S., Sallis H.M., Lindson N., Coupland C.A.C., et al. Smoking and COVID-19 outcomes: an observational and Mendelian randomisation study using the UK biobank cohort. Thorax. 2022;77:65–73. doi: 10.1136/thoraxjnl-2021-217080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Nakanishi M., Honda T. Processes of decision making and end-of-life care for patients with dementia in group homes in Japan. Arch Gerontol Geriatr. 2009;48:296–299. doi: 10.1016/j.archger.2008.02.009. [DOI] [PubMed] [Google Scholar]
- 9.Kojima G., Iliffe S., Taniguchi Y., Shimada H., Rakugi H., Walters K. Prevalence of frailty in Japan: a systematic review and meta-analysis. J Epidemiol. 2017;27:347–353. doi: 10.1016/j.je.2016.09.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhang X.M., Jiao J., Cao J., Huo X.P., Zhu C., Wu X.J., et al. Frailty as a predictor of mortality among patients with COVID-19: a systematic review and meta-analysis. BMC Geriatr. 2021;21:186. doi: 10.1186/s12877-021-02138-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Moon R.C., Mackey R.H., Cao Z., Emot S., Schott L.L., Gayle J.A., et al. Is coronavirus disease 2019 (COVID-19) less deadly now? Trends in in-hospital mortality among hospitalized COVID-19 patients in the United States. Clin Infect Dis. 2022;74:2238–2242. doi: 10.1093/cid/ciab830. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mozaffari E., Chandak A., Zhang Z., Liang S., Gayle J., Thrun M., et al. Clinical management of hospitalized coronavirus disease 2019 patients in the United States. Open Forum Infect Dis. 2022;9 doi: 10.1093/ofid/ofab498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Finelli L., Gupta V., Petigara T., Yu K., Bauer K.A., Puzniak L.A. Mortality among US patients hospitalized with SARS-CoV-2 infection in 2020. JAMA Netw Open. 2021;4 doi: 10.1001/jamanetworkopen.2021.6556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nguyen N.T., Chinn J., Kirby K., Hohmann S.F., Amin A. Outcomes of COVID-19 adults managed in an outpatient versus hospital setting. PLoS One. 2022;17 doi: 10.1371/journal.pone.0263813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Macedo A., Gonçalves N., Febra C. COVID-19 fatality rates in hospitalized patients: systematic review and meta-analysis. Ann Epidemiol. 2021;57:14–21. doi: 10.1016/j.annepidem.2021.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Roso-Llorach A., Serra-Picamal X., Cos F.X., Pallejà-Millán M., Mateu L., Rosell A., et al. Evolving mortality and clinical outcomes of hospitalized subjects during successive COVID-19 waves in Catalonia, Spain. Glob Epidemiol. 2022;4 doi: 10.1016/j.gloepi.2022.100071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mexico. 2020 - April 2022. Comparative epidemiology of four waves of COVID-19. [DOI] [Google Scholar]
- 18.Drake T.M., Riad A.M., Fairfield C.J., Egan C., Knight S.R., Pius R., et al. Characterisation of in-hospital complications associated with COVID-19 using the ISARIC WHO Clinical Characterisation Protocol UK: a prospective, multicentre cohort study. Lancet. 2021;398:223–237. doi: 10.1016/S0140-6736(21)00799-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Whitfield G.P., Harris A.M., Kadri S.S., Warner S., Bamrah Morris S., Giovanni J.E., et al. vol. 9. Open Forum Infect Dis; 2022. (Trends in clinical severity of hospitalized patients with coronavirus disease 2019 – premier hospital dataset, April 2020 – April 2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mozaffari E., Chandak A., Zhang Z., Liang S., Gayle J., Thrun M., et al. Clinical management of hospitalized coronavirus disease 2019 patients in the United States. Open Forum Infect Dis. 2022;9 doi: 10.1093/ofid/ofab498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yamaguchi S., Okada A., Sunaga S., Ikeda Kurakawa K., Yamauchi T., Nangaku M., et al. Impact of COVID-19 pandemic on healthcare service use for non-COVID-19 patients in Japan: retrospective cohort study. BMJ Open. 2022;12 doi: 10.1136/bmjopen-2021-060390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Dang A., Thakker R., Li S., Hommel E., Mehta H.B., Goodwin J.S. Hospitalizations and mortality from non–SARS-CoV-2 causes among Medicare beneficiaries at US hospitals during the SARS-CoV-2 pandemic. JAMA Netw Open. 2022;5 doi: 10.1001/jamanetworkopen.2022.1754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hayakawa K., Asai Y., Matsunaga N., Tsuzuki S., Terada M., Suzuki S., et al. Evaluation of the representativeness of data in the COVID-19 Registry Japan during the first six waves of the epidemic. Glob Health Med. 2022;4:2022. doi: 10.35772/ghm.2022.01033. 20201033:204–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Perico N., Cortinovis M., Suter F., Remuzzi G. Home as the new frontier for the treatment of COVID-19: the case for anti-inflammatory agents. Lancet Infect Dis. 2022 doi: 10.1016/S1473-3099(22)00433-9. [DOI] [PMC free article] [PubMed] [Google Scholar]