Abstract
Background: Although reductions in hospitalizations for myocardial infarction and heart failure have been reported during the period of COVID-19 pandemic restrictions, it is unclear how the overall number of hospitalizations for cardiovascular disease (CVD) treatment changed in the early stages of the pandemic.
Methods and Results: We analyzed the records of 574 certified hospitals affiliated with the Japanese Circulation Society and retrieved data from April 2015 to March 2020. Records were obtained from the nationwide Japanese Registry of All Cardiac and Vascular Diseases–Diagnosis Procedure Combination database. A quasi-Poisson regression model was used to estimate the number of hospitalizations for CVD treatment. Between January and March 2020, when the number of COVID-19 cases was relatively low in Japan, the actual/estimated number of hospitalizations for acute CVD was 18,233/21,634 (84.3%), whereas the actual/estimated number of scheduled hospitalizations was 16,921/19,066 (88.7%). The number of hospitalizations for acute heart failure and scheduled hospitalizations for valvular disease and aortic aneurysm were 81.1%, 84.6%, and 83.8% of the estimated values, respectively. A subanalysis that considered only facilities without hospitalization restrictions did not alter the results for these diseases.
Conclusions: The spread of COVID-19 was associated with a decreased number of hospitalizations for CVD in Japan, even in the early stages of the pandemic.
Because of its high transmissibility and associated mortality rate, COVID-19 continues to negatively affect the lives of people worldwide. As a result of a lack of adequate testing systems and personal protective equipment, as well as limited knowledge of the disease, the COVID-19 pandemic, particularly in the early stages, significantly affected the care of patients with non-communicable diseases at medical institutions. The number of patients hospitalized due to acute myocardial infarction and heart failure reportedly decreased during periods of COVID-19 pandemic restrictions.1–4 Without government restrictions, the initial response to the COVID-19 pandemic in Japan was successful, with few cases and deaths. However, as in other countries, the pandemic led to restrictions accessing hospitals and medical treatment, and some scheduled treatments could not be performed because of a lack of personal protective equipment.5 In fact, during the COVID-19 pandemic in Japan, the use of medical facilities for health-related events was reportedly decreased, especially among the elderly.6 Despite the COVID-19 pandemic having affected people’s lifestyles and behaviors in Japan and many other parts of the world, it remains unclear whether these changes affected cardiovascular events. In this study, we explored changes in cardiovascular hospitalizations in the early phase of the COVID-19 pandemic in Japan, with a special focus on acute cardiovascular diseases (CVD) and scheduled treatments.
Methods
The Japanese Registry of All Cardiac and Vascular Diseases–Diagnosis Procedure Combination (JROAD-DPC) is a nationwide claims database comprising data from 1,086 hospitals in Japan collected between April 2012 and March 2020. From the JROAD-DPC database, we selected 574 facilities with all data from April 2015 to March 2020 in the database (Figure 1). For each year, we used International Classification of Diseases, Tenth Revision (ICD-10) codes to extract data for patients aged ≥20 years who were hospitalized for acute CVD (acute myocardial infarction, acute heart failure, ruptured aortic aneurysm, or venous thromboembolism) or had scheduled surgeries or procedures for CVD (ischemic heart disease, valvular heart disease, aortic aneurysm, atrial septal defect, peripheral arterial disease, or venous thromboembolism), scheduled catheter ablation, scheduled permanent pacemaker implantation and exchange, or scheduled left ventricular assist device implantation between January and March. The ICD-10 codes registered as the main, admission-precipitating, or most resource-consuming diagnosis for each disease or procedure are listed in the Supplementary Table.
First, we compared inpatient and background data of the study participants for each CVD before and after (January–March 2020) the onset of the COVID-19 pandemic. Next, we applied a quasi-Poisson regression model to data from January to March in 2015 to 2019 to estimate the weekly number of hospitalizations between January and March from 2015 to 2020.7,8 Unlike the normal Poisson regression model, the quasi-Poisson regression model can handle overdispersed count variables, which are frequently observed, especially under pandemic situations. Our quasi-Poisson regression model consisted of a trend and seasonality expressed as Fourier terms with 2 sine and cosine waves considering 1-year and 0.5-year cycles,8–10 as follows:
var(Yt) = ϕE(Yt)
and
where E is the expectation operator, Yt
is the number of hospitalizations at Week t, α and β are regression parameters, and ϕ is the dispersion parameter. The parameters, including φ, were estimated using the quasi-likelihood approach.
Using the equations above, we calculated the ratio of actual to estimated number of hospitalizations from January to March 2020 as the primary endpoint, and set the point estimate and upper and lower bounds of the 2-sided 95% prediction interval (PI) for the purpose of comparison with hospitalizations before the COVID-19 pandemic as the secondary endpoints.
All data are presented as the mean±SD, number (percentage), or median with interquartile range, as appropriate. Between-group differences were evaluated using the Student’s t-test for normally distributed continuous variables, the Mann-Whitney U test for non-normally distributed continuous variables, and Chi-squared or Fisher’s exact tests for categorical variables. For all tests, 2-sided P<0.05 was considered significant. Data were analyzed using the open source statistical software R (version 4.0.2; R Foundation for Statistical Computing; www.R-project.org).
Subgroup Analysis Using Questionnaire Data
To minimize the effects of factors that influenced hospitals’ decisions to limit hospitalizations for CVD, we performed a subgroup analysis using questionnaire data on the restriction of hospitalizations for CVD. The Japanese Circulation Society regularly performed questionnaire surveys to record departmental experiences and polices during the COVID-19 pandemic.11,12 We used data from the first questionnaire, performed on April 13, 2020. That survey evaluated hospital and departmental policies regarding the restriction of emergency hospitalizations and procedures for CVD from January to March 2020. We merged the questionnaire data with data from the JROAD-DPC dataset. Subgroup analyses were conducted for diseases for which hospitalizations were at least 15% below estimated levels and below the 95% PI per week between January and March 2020.
Results
Acute CVD
Table 1 summarizes the characteristics of patients hospitalized for acute CVD. Compared with those hospitalized in 2015–2019, patients hospitalized for acute CVD in 2020 did not differ in age, but there was a higher proportion of men and patients had a higher body mass index, more comorbidities, lower in-hospital mortality, and a higher Barthel Index score at admission and discharge. The actual/estimated number of hospitalizations for acute CVD between January and March was 16,934/17,026 (99.5%) in 2018, 18,605/19,278 (96.5%) in 2019, and 18,233/21,634 (84.3%) in 2020. During the third week of January 2020, the number of hospitalizations for acute CVD was less than the 95% PI of the estimated value (Figure 2).
Table 1. Inpatient and Background Data of the Study Participants
| |
January–March
2015–2019C |
January–March
2020 |
P value |
| Hospitalization for acute CVDA |
| Total no. hospitalizations |
79,312 |
18,251 |
– |
| Length of stay (days) |
15.56±11.81 |
15.32±11.80 |
0.013 |
| In-hospital mortality (%) |
12.5 |
11.9 |
0.022 |
| Within 24 h |
4.9 |
4.2 |
<0.001 |
| Within 7 days |
7.6 |
7.2 |
0.058 |
| Barthel Index score at discharge |
80.0±31.4 |
80.7±30.85 |
0.009 |
| Hospitalization costs (JPY) |
857,000 [519,000–1,537,000] |
897,000 [568,000–1,534,000] |
<0.001 |
| Age (years) |
76.61±13.40 |
76.70±13.58 |
0.377 |
| Male (%) |
57.7 |
58.8 |
0.008 |
| BMI (kg/m2) |
22.89±4.98 |
23.08±5.19 |
<0.001 |
| Barthel Index score at admission |
46.6±41.62 |
48.45±41.7 |
<0.001 |
| Charlson comorbidity index |
2.18±1.39 |
2.28±1.46 |
<0.001 |
| Scheduled surgeries or procedures for CVDB |
| Total no. hospitalizations |
68,819 |
16,952 |
– |
| Length of stay (days) |
6.96±9.14 |
7.02±9.29 |
0.479 |
| In-hospital mortality (%) |
0.53 |
0.50 |
0.567 |
| Hospitalization costs (JPY) |
1,105,000 [871,000–765,000] |
1,051,000 [819,000–2,036,000] |
<0.001 |
| Age (years) |
70.81±11.57 |
71.74±11.77 |
<0.001 |
| Male (%) |
74.3 |
73.8 |
0.211 |
| BMI (kg/m2) |
23.86±4.51 |
23.84±4.18 |
0.751 |
| Barthel Index score at admission |
95.1±16.1 |
94.95±16.35 |
0.2 |
| Charlson comorbidity index |
1.70±1.41 |
1.76±1.45 |
<0.001 |
| Scheduled catheter ablation |
| Total no. hospitalizations |
23,142 |
7,393 |
– |
| Length of stay (days) |
3.96±3.20 |
3.71±2.99 |
<0.001 |
| In-hospital mortality (%) |
0.03 |
0.02 |
>0.999 |
| Hospitalization costs (JPY) |
2,059,000 [1,560,000–2,325,000] |
2,064,000 [1,727,000–2,285,000] |
0.007 |
| Age (years) |
62.91±13.94 |
64.39±13.66 |
<0.001 |
| Male (%) |
66.6 |
66.6 |
>0.999 |
| BMI (kg/m2) |
24.21±6.97 |
24.56±9.72 |
0.001 |
| Barthel Index score at admission |
98.9±7.65 |
98.95±7.85 |
0.624 |
| Charlson comorbidity index |
0.72±0.89 |
0.79±0.94 |
<0.001 |
| Scheduled permanent pacemaker implantation |
| Total no. hospitalizations |
5,133 |
1,351 |
– |
| Length of stay (days) |
12.01±8.66 |
11.47±8.08 |
0.038 |
| In-hospital mortality (%) |
0.20 |
0.07 |
0.556 |
| Hospitalization costs (JPY) |
1,729,000 [1,555,000–3,354,000] |
1,498,000 [1,393,000–3,777,000] |
<0.001 |
| Age (years) |
73.49±13.21 |
74.49±12.95 |
0.013 |
| Male (%) |
56.8 |
58.8 |
0.196 |
| BMI (kg/m2) |
23.26±3.87 |
23.22±3.82 |
0.73 |
| Barthel Index score at admission |
93.0±19.1 |
93.25±18.8 |
0.64 |
| Charlson comorbidity index |
1.20±1.21 |
1.29±1.26 |
0.011 |
| Scheduled pacemaker generator exchange |
| Total no. hospitalizations |
4,846 |
1,233 |
– |
| Length of stay (days) |
7.06±4.85 |
6.37±4.09 |
<0.001 |
| In-hospital mortality (%) |
0.17 |
0.08 |
0.787 |
| Hospitalization costs (JPY) |
1,171,000 [1,031,000–1,419,000] |
1120000 [994,000–1,298,000] |
<0.001 |
| Age (years) |
75.95±14.98 |
76.99±14.81 |
0.03 |
| Male (%) |
50.3 |
49.6 |
0.659 |
| BMI (kg/m2) |
23.15±4.76 |
23.01±4.15 |
0.347 |
| Barthel Index score at admission |
88.1±25.45 |
88.15±25.8 |
0.938 |
| Charlson comorbidity index |
0.97±1.07 |
1.15±1.10 |
<0.001 |
| Scheduled LVAD implantation |
| Total no. hospitalizations |
350 |
118 |
– |
| Length of stay (days) |
20.52±12.07 |
19.50±12.20 |
0.429 |
| In-hospital mortality (%) |
1.43 |
1.69 |
>0.999 |
| Hospitalization costs (JPY) |
5,303,000 [4,419,000–6,037,000] |
5,473,000 [4,533,000–5,944,000] |
0.648 |
| Age (years) |
62.04±16.49 |
64.44±14.70 |
0.162 |
| Male (%) |
62.3 |
68.6 |
0.257 |
| BMI (kg/m2) |
22.65±3.73 |
23.50±3.17 |
0.027 |
| Barthel Index score at admission |
95.1±15.1 |
92.8±22.7 |
0.216 |
| Charlson comorbidity index |
1.83±1.04 |
1.75±1.06 |
0.505 |
Unless indicated otherwise, data are given as the mean±SD or median [interquartile range]. AAcute myocardial infarction, acute heart failure, ruptured aortic aneurysm, and VTE. BIschemic heart disease, valvular heart disease, aortic aneurysm, atrial septal defect, peripheral arterial disease, and VTE. CData were collected between January and March for the period 2015–2019. BMI, body mass index; CVD, cardiovascular disease; JPY, Japanese Yen; LVAD, left ventricular assist device; VTE, venous thromboembolism.
Among patients hospitalized for acute CVD, the number of hospitalizations for acute myocardial infarction, acute heart failure, ruptured aortic aneurysm, and venous thromboembolism were 90.1%, 81.1%, 88.8, and 96.0% of the estimated values, respectively. During the third week of January 2020, the number of hospitalizations for acute heart failure was less than the 95% PI of the estimated value (Figure 3). The characteristics of patients hospitalized for acute myocardial infarction, acute heart failure, ruptured aortic aneurysm, and venous thromboembolism are summarized in Table 2. There were no significant changes in the age or in-hospital mortality of patients hospitalized for acute heart failure. However, the proportion of patients with New York Heart Association Class III or IV at admission was significantly higher in 2020 (75%) than in 2015–2019 (66%).
Table 2. Characteristics of Patients Hospitalized for Acute Cardiovascular Diseases
| |
January–March
2015–2019A |
January–March
2020 |
P value |
| Hospitalization for AMI |
| Total no. hospitalizations |
20,986 |
5,031 |
– |
| Length of stay (days) |
12.29±10.12 |
11.94±10.01 |
0.026 |
| In-hospital mortality (%) |
15.1 |
12.9 |
<0.001 |
| Within 24 h |
8.6 |
6.6 |
<0.001 |
| Within 7 days |
11.7 |
10.1 |
0.001 |
| Barthel Index score at discharge |
89.7±25.4 |
89.75±25.3 |
0.916 |
| Hospitalization costs (JPY) |
1,723,000 [1,226,000–2,357,000] |
1,623,000 [1,164,000–2,157,000] |
<0.001 |
| Age (years) |
70.51±13.34 |
70.73±13.28 |
0.304 |
| Male (%) |
72.0 |
73.0 |
0.188 |
| BMI (kg/m2) |
23.65±4.53 |
23.75±4.33 |
0.207 |
| Barthel Index score at admission |
36.95±43.4 |
39.95±43.65 |
<0.001 |
| Killip classification III or IV (%) |
22.2 |
21.2 |
0.14 |
| Charlson comorbidity index |
2.10±1.19 |
2.16±1.27 |
0.002 |
| Hospitalization for acute HF |
| Total no. hospitalizations |
52,352 |
11,867 |
– |
| Length of stay (days) |
17.14±12.10 |
16.99±12.13 |
0.224 |
| In-hospital mortality (%) |
10.5 |
10.6 |
0.822 |
| Within 24 h |
2.3 |
2.3 |
0.958 |
| Within 7 days |
4.9 |
5.1 |
0.378 |
| Barthel Index score at discharge |
76.25±32.45 |
77.0±31.95 |
0.03 |
| Hospitalization costs (JPY) |
727,000 [492,000–1,076,000] |
766,000 [530,000–1,117,000] |
<0.001 |
| Age (years) |
79.58±12.15 |
79.79±12.34 |
0.089 |
| Male (%) |
53.4 |
54.2 |
0.107 |
| BMI (kg/m2) |
22.56±4.89 |
22.74±5.10 |
0.001 |
| Barthel Index score at admission |
50.05±40.15 |
51.7±40.15 |
<0.001 |
| NYHA Class III or IV (%) |
66.0 |
75.3 |
<0.001 |
| Charlson comorbidity index |
2.31±1.39 |
2.44±1.47 |
<0.001 |
| Hospitalization for ruptured aortic aneurysms |
| Total no. hospitalizations |
1,436 |
326 |
– |
| Length of stay (days) |
9.20±13.76 |
10.84±15.05 |
0.057 |
| In-hospital mortality (%) |
62.4 |
59.2 |
0.313 |
| Within 24 h |
48.9 |
43.3 |
0.076 |
| Within 7 days |
57.1 |
52.1 |
0.117 |
| Barthel Index score at discharge |
68.05±39.55 |
65.05±38.7 |
0.412 |
| Hospitalization costs (JPY) |
378,000 [156,000–3,681,000] |
662,000 [173,000–4,247,000] |
0.009 |
| Age (years) |
79.40±10.31 |
80.04±11.40 |
0.325 |
| Male (%) |
66.9 |
65.0 |
0.573 |
| BMI (kg/m2) |
21.90±5.25 |
21.88±4.73 |
0.969 |
| Barthel Index score at admission |
17.5±34.65 |
16.2±47.7 |
0.563 |
| Charlson comorbidity index |
1.64±1.10 |
1.77±1.27 |
0.052 |
| Hospitalization for VTE |
| Total no. hospitalizations |
4,731 |
1,082 |
– |
| Length of stay (days) |
14.78±11.22 |
14.24±10.76 |
0.154 |
| In-hospital mortality (%) |
7.8 |
6.6 |
0.193 |
| Within 24 h |
2.8 |
2.2 |
0.309 |
| Within 7 days |
4.8 |
4.0 |
0.267 |
| Barthel Index score at discharge |
81.45±32.15 |
82.1±31.85 |
0.559 |
| Hospitalization costs (JPY) |
680,000 [432,000–1,049,000] |
675,000 [466,000–987,000] |
0.735 |
| Age (years) |
69.95±15.88 |
69.58±16.26 |
0.497 |
| Male (%) |
39.3 |
42.1 |
0.497 |
| BMI (kg/m2) |
23.55±6.92 |
24.20±8.44 |
0.011 |
| Barthel Index score at admission |
58.25±40.75 |
60.2±41.2 |
0.165 |
| Charlson comorbidity index |
1.27±1.77 |
1.31±1.71 |
0.484 |
Unless indicated otherwise, data are given as the mean±SD or median [interquartile range]. AData were collected between January and March for the period 2015–2019. AMI, acute myocardial infarction; BMI, body mass index; HF, heart failure; JPY, Japanese Yen; NYHA, New York Heart Association; VTE, venous thromboembolism.
Table 3 summarizes the characteristics of patients hospitalized for scheduled treatments or procedures. Patients hospitalized for scheduled surgeries or procedures for CVD in 2020 were older and had more comorbidities than those hospitalized in 2015–2019, but there were no differences in sex, body mass index, in-hospital mortality, and Barthel Index score at admission. Except for patients who underwent left ventricular assist device implantation, the mean age of patients hospitalized in 2020 for scheduled surgery or procedures was higher than of those hospitalized in 2015–2019, but the in-hospital mortality or Barthel Index score at admission did not differ. The actual/estimated number of hospitalizations for scheduled CVD-related surgeries or procedures between January and March was 14,024/14,673 (95.6%) in 2018, 16,720/17,016 (98.3%) in 2019, and 16,921/19,066 (88.7%) in 2020. The number of hospitalizations for a scheduled surgery or procedure did not deviate from the 95% PI of the estimated value in any week. The number of hospitalizations for a scheduled catheter ablation, permanent pacemaker implantation, pacemaker generator exchange, and left ventricular assist device implantation were 91.6%, 95.5%, 97.2, and 90.1% of the estimated values, respectively. The number of hospitalizations for scheduled catheter ablation declined between January and March 2020, but the estimated 95% PI for hospitalizations for scheduled catheter ablation was exceeded in the fifth week of March 2020 (Supplementary Figure 1).
Table 3. Characteristics of Patients Hospitalized for Scheduled Surgeries or Procedures for Cardiovascular Diseases
| |
January–March
2015–2019A |
January–March
2020 |
P value |
| Scheduled surgery or procedures for IHD |
| Total no. hospitalizations |
46,142 |
10,649 |
– |
| Length of stay (days) |
4.97±7.11 |
4.84±7.30 |
0.102 |
| In-hospital mortality (%) |
0.31 |
0.28 |
0.759 |
| Hospitalization costs (JPY) |
1,043,000 [868,000–1,354,000] |
948,000 [790,000–1,223,000] |
<0.001 |
| Age (years) |
70.29±10.15 |
70.93±10.28 |
<0.001 |
| Male (%) |
77.6 |
77.7 |
0.828 |
| BMI (kg/m2) |
24.39±4.54 |
24.46±4.22 |
0.152 |
| Barthel Index score at admission |
97.3±11.55 |
97.3±11.55 |
0.962 |
| Charlson comorbidity index |
1.45±1.31 |
1.50±1.37 |
0.002 |
| Scheduled surgeries or procedures for VHD |
| Total no. hospitalizations |
4,895 |
1,617 |
– |
| Length of stay (days) |
21.70±11.08 |
19.25±11.22 |
<0.001 |
| In-hospital mortality (%) |
1.82 |
1.86 |
>0.999 |
| Hospitalization costs (JPY) |
5,430,000 [4,698,000–6,109,000] |
5,557,000 [4,914,000–6,096,000] |
<0.001 |
| Age (years) |
73.45±13.40 |
75.46±13.26 |
<0.001 |
| Male (%) |
49.2 |
50.0 |
0.608 |
| BMI (kg/m2) |
22.90±3.80 |
22.86±3.74 |
0.73 |
| Barthel Index score at admission |
93.7±16.9 |
94.0±16.4 |
0.568 |
| Charlson comorbidity index |
1.48±1.21 |
1.57±1.28 |
0.006 |
| Scheduled surgeries or procedures for aortic aneurysm |
| Total no. hospitalizations |
5,605 |
1,646 |
– |
| Length of stay (days) |
15.07±10.44 |
14.28±9.98 |
0.007 |
| In-hospital mortality (%) |
1.30 |
1.09 |
0.587 |
| Hospitalization costs (JPY) |
3,627,000 [2,927,000–5,025,000] |
3,789,000 [2,948,000–5,186,000] |
0.076 |
| Age (years) |
73.40±10.65 |
74.06±10.52 |
0.029 |
| Male (%) |
78.9 |
79.0 |
0.963 |
| BMI (kg/m2) |
23.58±4.66 |
23.55±3.57 |
0.77 |
| Barthel Index score at admission |
95.45±15.1 |
95.75±14.8 |
0.463 |
| Charlson comorbidity index |
2.00±1.21 |
1.98±1.15 |
0.584 |
| Scheduled surgeries or procedures for ASD |
| Total no. hospitalizations |
774 |
208 |
– |
| Length of stay (days) |
8.75±6.57 |
8.25±6.20 |
0.328 |
| In-hospital mortality (%) |
0.13 |
0 |
>0.999 |
| Hospitalization costs (JPY) |
1,790,000 [1,526,000–2,473,000] |
1,727,000 [1,507,000–2,379,000] |
0.122 |
| Age (years) |
31.31±26.32 |
32.43±27.08 |
0.592 |
| Male (%) |
40.2 |
36.1 |
0.317 |
| BMI (kg/m2) |
19.95±4.68 |
19.94±4.78 |
0.982 |
| Barthel Index score at admission |
94.05±19.1 |
95.2±17.75 |
0.486 |
| Charlson comorbidity index |
0.65±0.77 |
0.76±0.85 |
0.076 |
| Scheduled surgeries or procedures for PAD |
| Total no. hospitalizations |
12,721 |
3,272 |
– |
| Length of stay (days) |
6.39±9.15 |
6.21±9.66 |
0.336 |
| In-hospital mortality (%) |
0.90 |
0.70 |
0.316 |
| Hospitalization costs (JPY) |
938,000 [738,000–1,286,000] |
976,000 [766,000–1,328,000] |
<0.001 |
| Age (years) |
73.20±9.59 |
74.13±9.33 |
<0.001 |
| Male (%) |
72.1 |
72.9 |
0.409 |
| BMI (kg/m2) |
22.56±4.01 |
22.60±3.92 |
0.617 |
| Barthel Index score at admission |
87.45±25.45 |
87.2±25.3 |
0.64 |
| Charlson comorbidity index |
2.65±1.45 |
2.68±1.49 |
0.215 |
| Scheduled surgeries or procedures for VTE |
| Total no. hospitalizations |
244 |
46 |
– |
| Length of stay (days) |
21.47±16.05 |
19.26±15.28 |
0.389 |
| In-hospital mortality (%) |
2.46 |
0 |
0.61 |
| Hospitalization costs (JPY) |
1,688,000 [876,000–2,288,000] |
1,568,000 [836,000–2,167,000] |
0.584 |
| Age (years) |
65.77±14.67 |
66.87±13.84 |
0.643 |
| Male (%) |
39.3 |
30.4 |
0.329 |
| BMI (kg/m2) |
23.30±4.12 |
23.26±3.76 |
0.949 |
| Barthel Index score at admission |
86.2±27.65 |
81.15±35.4 |
0.288 |
| Charlson comorbidity index |
2.45±2.63 |
2.63±2.60 |
0.671 |
Unless indicated otherwise, data are given as the mean±SD or median [interquartile range]. AData were collected between January and March for the period 2015–2019. ASD, atrial septal defects; BMI, body mass index; IHD, ischemic heart disease; JPY, Japanese Yen; PAD, peripheral arterial disease; VHD, valvular heart disease; VTE, venous thromboembolism.
Among patients hospitalized for scheduled CVD-related surgeries or procedures, the number of hospitalizations for valvular heart disease (84.6% of the estimated value) and aortic aneurysm (83.8% of the estimated value) was less than the 95% PI of the estimated value during the third week of January 2020 (valvular heart disease) and the first week of March 2020 (valvular heart disease and aortic aneurysm; Supplementary Figure 2). Except for those for venous thromboembolism (117.9% of the estimated value), hospitalizations for scheduled surgeries or procedures were lower than estimated. As indicated in Table 3, there was no significant change in the in-hospital mortality for any scheduled hospitalization.
Subgroup Analysis
We performed a subgroup analysis for acute heart failure, valvular heart disease, and aortic aneurysm from the primary and secondary endpoints. The results of the subanalyses of hospitalizations for these 3 diseases did not change, even when only facilities without the restriction of emergency hospitalizations and procedures for CVD were considered. Among the 243 facilities (42% of facilities participating in the main analysis) that had no restrictions on emergency care, the actual number of hospitalizations for acute heart failure was 72.4% of the estimated value, with hospitalizations below the 95% PI during the third and fourth weeks of January 2020 (Figure 4). Among the 239 facilities (42% of facilities participating in the main analysis) that had no restrictions on hospitalization for scheduled treatments, the actual number of hospitalizations for valvular heart disease was 75.0% of the estimated value, with hospitalizations below the 95% PI during the third week of January 2020 and first week of March 2020. Similarly, the actual number of hospitalizations for aortic aneurysm was 83.3% of the estimated value, with hospitalizations below the 95% PI during the first week of March 2020.
Discussion
Figure 5 summarizes the findings of the present study. The major findings are that: (1) the number of hospitalizations for acute CVD between January and March 2020 was low (84% of the expected value), with hospitalizations for acute heart failure being the lowest among all conditions (81% of the expected value) and (2) the number of hospitalizations for a scheduled treatment was 89% of the expected value, with hospitalizations for the scheduled treatment of aortic aneurysms and valvular disease being 84% and 85% of the expected values, respectively. There were periods when the number of hospitalizations per week for acute heart failure and scheduled hospitalizations for aortic aneurysms and valvular disease were clearly lower than expected.
Hospitalizations for CVD in Japan are increasing each year as the population ages, and it has become a major public health concern.13 An increase in the number of hospitalizations was estimated by the model used in the present study. However, during the period January–March 2020, which corresponds to the beginning of the COVID-19 pandemic, hospitalizations for acute CVD and the number of scheduled treatments for CVD were both lower than expected. There was no increase in the in-hospital mortality associated with any disease or treatment, as reported in previous studies,1,14,15 and no decrease in the Barthel Index score at admission.
The decrease in the number of hospitalizations for CVD observed in the present study can be attributed to environmental factors (e.g., the infectious disease pandemic), hospital-related factors (e.g., restrictions of hospitalizations for CVD), and patient-related factors (e.g., avoiding using medical facilities). The first case of COVID-19 in Japan was confirmed on January 15, 2020. At the end of March 2020, the number of patients diagnosed with COVID-19 per day (1.2 per million people) and the total number of patients with COVID-19 (18 per million people) in Japan were very low, making it unlikely that the COVID-19 pandemic was directly related to the hospitalization of patients with CVD. Even in hospitals without restrictions on emergency hospitalizations and procedures for CVD, the number of hospitalizations for acute heart failure and scheduled hospitalizations for aortic aneurysms and valvular disease decreased, suggesting that patient-related factors were the main cause of the decrease in the number of hospitalizations. These findings are consistent with research reports indicating that patients tended to avoid medical facilities during the COVID-19 pandemic in Japan.6 Increases were observed in the proportion of critically ill (New York Heart Association Class III or IV) patients hospitalized for acute heart failure and in the number of scheduled hospitalizations for venous thromboembolism. However, it is difficult to determine from our results whether patients’ reluctance to visit doctors or changes in their lifestyles to reduce the risk of developing CVD contributed to the decrease in the number of hospitalizations for CVD. A reduction in the number of hospitalizations does not necessarily indicate improved patient outcomes; rather, it is important to consider that patients may have missed out on necessary treatment.
This study has several limitations that need to be discussed. First, this study is not based on exhaustive data; therefore, the results should be interpreted with caution. Because this study included most facilities participating in the JROAD-DPC and an analysis of data collected through a questionnaire survey regarding medical restrictions, we believe that our results are adequate. Second, the subanalysis of hospitals without medical restrictions revealed lower rates of hospitalization for acute heart failure (72.4%) and scheduled hospitalizations for valvular heart disease (75.0%) than the main analysis (81.1% and 84.6%, respectively). It is possible that hospitalizations were concentrated in those hospitals where restrictions were imposed; however, because we did not evaluate the uneven distribution of hospitalizations among the different facilities, additional research is required. Third, this study does not sufficiently explain why the number of scheduled hospitalizations for venous thromboembolism increased. Factors related to the COVID-19 pandemic may have been responsible for an increase in the incidence of venous thromboembolism, because individuals may have refrained from habitual exercise. Finally, this study only examined the early phase of the COVID-19 pandemic, and longer-term studies are required to determine the effects of COVID-19 on cardiovascular care.
Conclusions
The spread of COVID-19 was associated with a decrease in the number of hospitalizations for CVD in Japan, even in the early stages of the COVID-19 pandemic. The possibility of a decline in hospitalizations for CVD should be considered in the event of an emerging infectious disease pandemic.
Acknowledgment
The authors thank Editage (www.editage.com) for English language editing.
Sources of Funding
This work was supported by the Japanese Ministry of Health, Labour and Welfare Policy Research Grants, Special Research Program (Grant no. 21CA2002).
Disclosures
None of the authors has any conflicts of interest related to this study. However, K.N. and Y. Miyamoto are members of the Circulation Reports’ Editorial Team.
IRB Information
The ethics committees of the Japanese Circulation Society and Mie University (U2021-034) approved the study protocol and waived the requirement for obtaining informed consent from participants because no personal identifying information was used in the study.
Supplementary Files
Please find supplementary file(s);
http://dx.doi.org/10.1253/circrep.CR-22-0058
References
- 1.
Cannatà A, Bromage DI, Rind IA, Gregorio C, Bannister C, Albarjas M, et al. Temporal trends in decompensated heart failure and outcomes during COVID-19: A multisite report from heart failure referral centres in London. Eur J Heart Fail 2020; 22: 2219–2224.
- 2.
Bodilsen J, Nielsen PB, Søgaard M, Dalager-Pedersen M, Speiser LOZ, Yndigegn T, et al. Hospital admission and mortality rates for non-covid diseases in Denmark during covid-19 pandemic: Nationwide population based cohort study. BMJ 2021; 373: n1135.
- 3.
Fairweather SM, Chang CL, Mansell CJ, Shafuddin E, Hancox RJ. Impact of COVID-19 pandemic restrictions on the cardio-respiratory health of New Zealanders. Respirology 2021; 26: 1041–1048.
- 4.
Morishita T, Takada D, Shin JH, Higuchi T, Kunisawa S, Fushimi K, et al. Effects of the COVID-19 pandemic on heart failure hospitalizations in Japan: Interrupted time series analysis. ESC Heart Fail 2022; 9: 31–38.
- 5.
Sugimoto T, Mizuno A, Kishi T, Ito N, Matsumoto C, Fukuda M, et al. Coronavirus disease 2019 (COVID-19) information for cardiologists: Systematic literature review and additional analysis. Circ J 2020; 84: 1039–1043.
- 6.
Aoki T, Matsushima M. The ecology of medical care during the COVID-19 pandemic in Japan: A nationwide survey. J Gen Intern Med 2022; 37: 1211–1217.
- 7.
Nomura S, Kawashima T, Yoneoka D, Tanoue Y, Eguchi A, Gilmour S, et al. Trends in suicide in Japan by gender during the COVID-19 pandemic, up to September 2020. Psychiatry Res 2021; 295: 113622.
- 8.
Vestergaard LS, Nielsen J, Richter L, Schmid D, Bustos N, Braeye T, et al. Excess all-cause mortality during the COVID-19 pandemic in Europe: Preliminary pooled estimates from the EuroMOMO network, March to April 2020. Euro Surveill 2020; 25: 2001214.
- 9.
Nielsen J, Krause TG, Mølbak K. Influenza-associated mortality determined from all-cause mortality, Denmark 2010/11–2016/17: The FluMOMO model. Influenza Other Respir Viruses 2018; 12: 591–604.
- 10.
Nielsen J, Mazick A, Andrews N, Detsis M, Fenech TM, Flores VM, et al. Pooling European all-cause mortality: Methodology and findings for the seasons 2008/2009 to 2010/2011. Epidemiol Infect 2013; 141: 1996–2010.
- 11.
Mizuno A, Matsumoto C, Kishi T, Ishida M, Sanada S, Fukuda M, et al. Cardiology department policy in Japan after coronavirus disease-2019 (COVID-19): Descriptive summary of 2nd nationwide survey by the Japanese Circulation Society. Circ Rep 2021; 3: 100–104.
- 12.
Mizuno A, Matsumoto C, Yoneoka D, Kishi T, Ishida M, Sanada S, et al. Cardiology department practices in the first wave of the coronavirus disease pandemic: A nationwide survey in Japan by the Japanese Circulation Society. Circ Rep 2021; 3: 137–141.
- 13.
Shimokawa H, Miura M, Nochioka K, Sakata Y. Heart failure as a general pandemic in Asia. Eur J Heart Fail 2015; 17: 884–892.
- 14.
Xiang D, Xiang X, Zhang W, Yi S, Zhang J, Gu X, et al. Management and outcomes of patients with STEMI during the COVID-19 pandemic in China. J Am Coll Cardiol 2020; 76: 1318–1324.
- 15.
De Luca G, Verdoia M, Cercek M, Jensen LO, Vavlukis M, Calmac L, et al. Impact of COVID-19 pandemic on mechanical reperfusion for patients with STEMI. J Am Coll Cardiol 2020; 76: 2321–2330.
