Abstract
Background:
The appropriate number of board-certified cardiologists (BCC) for the treatment of acute myocardial infarction (AMI) has not been thoroughly examined in Japan. This study investigated whether the number of BCC/50 cardiovascular beds affects acute outcome in AMI treatment.
Methods and Results:
Data on 751 board-certified teaching hospitals and 63,603 patients with AMI were obtained from the Japanese Registry Of All cardiac and vascular Diseases (JROAD) and JROAD Diagnosis Procedure Combination (JROAD-DPC) databases between 1 April 2012 and 31 March 2014. The hospitals were categorized into 3 groups based on the median number of BCC/50 cardiovascular beds: first tertile, 5.0 (IQR, 4.0–5.7); second, 8.3 (IQR, 7.4–9.8); third, 15.3 (IQR, 12.5–22.7), and the patients with AMI admitted to the categorized hospitals were compared (first tertile, 12,002 patients; second, 23,930; third, 27,671). On hierarchical logistic modeling, the adjusted OR for 30-day mortality were 0.86 (95% CI: 0.74–1.00) for the second tertile and 0.75 (95% CI: 0.65–0.88) for the third tertile.
Conclusions:
Patients with AMI admitted to hospitals with a large number of BCC/50 cardiovascular beds had a lower 30-day mortality rate. This tendency was independent of patient and hospital characteristics. This is the first study to provide new information on the association between the number of BCC and in-hospital AMI-related mortality in Japan.
Reduction of mortality in acute myocardial infarction (AMI) has been an important issue for cardiologists. The age-standardized mortality rate of AMI has been gradually decreasing over the past 3 decades in several countries,1,2
which is probably due to the development of emergency medical network systems, and the carrying out of reperfusion therapy and optimal medical therapy.3,4
AMI, however, remains one of the most common causes of death.1,5
In 2016, 68,907 patients in Japan were hospitalized because of AMI, and the in-hospital mortality rate was 8.2%, which has not improved in recent years despite the advancements in medical technology.6,7
Thus, to further improve the acute outcome of AMI treatment, structural and clinical properties of the hospital, including the quality and quantity of medical staff, require more attention.
The aim of this study was to determine whether or not the number of cardiovascular specialists affects the acute mortality of AMI. The Japanese Circulation Society (JCS) launched the board of cardiovascular specialist system in 1989.8
There were 6,568 JCS board-certified cardiologists (BCC) working full time in 1,612 hospitals in 2012, which included almost all cardiovascular beds in Japan. All hospitals with cardiovascular beds were divided into different classes: class A hospitals have >2 BCC and >30 cardiovascular beds; class B hospitals have >1 BCC and >15 cardiovascular beds, and class C hospitals do not have BCC or they have <15 cardiovascular beds. The number of BCC per hospital in classes A and B widely varied from 1 to 87, and whether this current status is appropriate or not for the treatment of AMI remains unknown, especially given that available related studies in Japan are limited.
The aim of this study was therefore to clarify the associations between the number of BCC/50 cardiovascular beds and 30-day AMI mortality.
Methods
Patients
This was a cross-sectional study, and we used the Japanese Registry Of All cardiac and vascular Diseases (JROAD) database and Diagnosis Procedure Combination (DPC) discharge database from 2012 to 2014. JROAD is conducted by JCS, and the data were collected from almost all teaching hospitals with cardiovascular beds, except for stroke.9,10
Of 1,613 hospitals included in the JROAD, 806 provided DPC data to JCS.
We obtained the following data on JCS-certified class A and B teaching hospitals: number of hospital beds and cardiologists; presence or absence of cardiac surgery division and coronary care unit; hospital teaching status (teaching hospitals are defined as JCS-certified cardiovascular specialist training facilities with >2 BCC and >30 cardiovascular beds); and annual number of percutaneous coronary interventions (PCI) and coronary artery bypass grafts (CABG). Consequently, 751 hospitals were enrolled in this study (Figure 1A).
Subsequently, we collected data for 72,216 patients from the 751 hospitals based on the following inclusion criteria: (1) hospitalized because of AMI between 1 April 2012 and 31 March 2014; and (2) diagnosis of AMI based on the International Classification of Diseases (ICD-10) codes for AMI (I21.0, I21.1, I21.2, I21.3, I21.4, and I21.9). We excluded 19 patients <20 years old, 420 patients discharged on the day of hospitalization, and 8,174 patients with insufficient data. Consequently, we analyzed 63,603 patients in 751 hospitals (Figure 1B).
We categorized the 751 hospitals into 3 groups according to the number of BCC/50 cardiovascular beds and compared the groups of patients admitted to the categorized hospitals. In the subanalysis, we divided the 63,603 patients with AMI into 2 groups according to Killip classification on admission: not severe (Killip I and II) or severe (Killip III and IV).
Ethics
The ethics committees at both JCS and Nara Medical University, which waived the requirement for individual informed consent because no information specifying individuals was included, approved the study protocol. The original DPC data were anonymized using the code change equations, and were then sent to the Ministry of Health, Labor, and Welfare.
Statistical Analysis
Data are presented as mean±SD for normally distributed data, and as median (IQR) for asymmetrically distributed data, or absolute number (proportion) for categorical data. The differences between tertiles were compared using analysis of variance for continuous variables and the chi-squared test with Bonferroni correction for non-continuous and categorical variables. The main outcome measure was in-hospital or 30-day mortality. The tendency for tertile 30-day mortality was analyzed using the Cochran-Armitage trend test. The association between the number of BCC/50 cardiovascular beds and 30-day mortality was analyzed using multilevel mixed-effect logistics regression, with hospital characteristics at the first level and patient characteristics at the second. The number of BCC/50 cardiovascular beds was analyzed as a continuous variable and as a categorical variable (first, second, and third tertile). The models were adjusted for the following covariates based on previous reports: age; gender; Charlson comorbidity index; comorbidity (hypertension, dyslipidemia, diabetes mellitus, chronic kidney disease, and atrial fibrillation); presence of cardiac surgery division and coronary care unit; hospital teaching status; use of ambulance; and Killip classification. Stepwise analysis was performed to determine whether any combination of clinical findings was associated with 30-day mortality. In addition, propensity score analysis was conducted to evaluate the robustness of the results. We performed one-to-one propensity score matching between the 2 groups by the number of cardiologists. The cut-off for the number of cardiologists was 10.0, which was determined on receiver operating characteristic analysis. Propensity scores were estimated using a logistic regression model with independent variables selected on stepwise analysis. A caliper of 0.001-fold the SD was used.
In the sub-analysis, we performed stratified analysis according to AMI severity to ascertain whether the effect of the number of BCC/50 cardiovascular beds on 30-day mortality is dependent on AMI severity. Biological interaction between the number of BCC/50 cardiovascular beds and AMI severity was evaluated by calculating the relative excess risk.11–13
STATA (version 14, Stata Corp., College Station, TX, USA) was used for the analysis.
Results
Hospital Characteristics
The median number of BCC/50 cardiovascular beds was 5.0 (IQR, 4.0–5.7) in the first tertile hospitals, 8.3 (IQR, 7.4–9.8) in the second, and 15.3 (IQR, 12.5–22.7) in the third. As the number of BCC/50 cardiovascular beds increased, the following variables also increased: number of hospital beds and cardiologists; presence of cardiac surgery division and coronary care unit; hospital teaching status; and annual number of PCI and CABG (Table 1).
Table 1.
Hospital Characteristics vs. No. JCS Board-Certified Cardiologists
| |
Total |
First
tertile |
Second
tertile |
Third
tertile |
P-value |
| JCS BCC/50 CV beds (range) |
0.63–75.58 |
0.63–6.58 |
6.63–10.98 |
11.05–75.58 |
|
| No. hospitals§ |
751 |
238 |
282 |
231 |
|
| Hospital beds |
444 (300–558) |
317 (230–400) |
406 (307–500) |
621 (441–735) |
<0.001†
|
| CV beds |
38 (30–46) |
39 (30–48) |
37 (30–42) |
39 (28–46) |
0.005†
|
| JCS BCC/50 CV beds |
8.3 (6.3–13.2) |
5.0 (4.0–5.7) |
8.3 (7.4–9.8) |
15.3 (12.5–22.7) |
<0.001†
|
| JCS BCC/hospital |
5.4 (3–6) |
2.8 (2–3) |
4.2 (3–5) |
9.4 (5–12) |
<0.001†
|
| JCS BCC and non-JCS BCC/hospital |
8.3 (4–9) |
3.7 (3–5) |
6.2 (4–7) |
15.8 (8–19) |
<0.001†
|
| Cardiac intensive care units |
85 |
71 |
87 |
96 |
<0.001‡
|
| Coronary angiography/year |
444 (182–595) |
247 (107–326) |
434 (193–547) |
636 (383–807) |
<0.001†
|
| Emergency PCI for AMI/year |
49 (20–69) |
29 (10–40) |
49 (23–64) |
68 (38–94) |
<0.001†
|
| Cardiac surgery |
61 |
42 |
57 |
87 |
<0.001‡
|
| No. CABG/year |
31 (13–40) |
16 (2–23) |
27 (8–38) |
41 (19–54) |
<0.001†
|
| Hospital teaching status |
94 |
82 |
95 |
99 |
<0.001‡
|
Data given as median (IQR), n§
or %. †Kruskal-Wallis test; ‡χ2 test with Bonferroni correction. AMI, acute myocardial infarction; BCC, board-certified cardiologists; CABG, coronary artery bypass graft; CV, cardiovascular; JCS, Japanese Circulation Society; PCI, percutaneous coronary intervention.
Approximately 82% of the patients were admitted to the second or third tertile hospitals. Sex, age, Charlson comorbidity index, Killip classification, comorbidity, ambulance use, and medication were similar between the 3 groups, whereas the use of circulation devices, such as intra-aortic balloon pump (IABP) and percutaneous cardiopulmonary support system (PCPS), increased and the length of hospital stay decreased as the number of BCC/50 cardiovascular beds increased (Table 2).
Table 2.
Patient Characteristics vs. No. JCS Board-Certified Cardiologists
| |
Total |
First
tertile |
Second
tertile |
Third
tertile |
P-value |
| JCS BCC/50 CV beds (range) |
0.63–75.58 |
0.63–6.58 |
6.63–10.98 |
11.05–75.58 |
|
| Patients |
63,603 (100) |
12,002 (19) |
23,930 (38) |
27,671 (44) |
|
| Age (years) |
69±13 |
70±13 |
69±13 |
69±13 |
<0.001* |
| Male |
73 |
71 |
73 |
74 |
<0.001‡
|
| Charlson comorbidity index |
2 (1–3) |
2 (1–3) |
2 (1–3) |
2 (1–3) |
<0.001†
|
| Killip classification |
| I |
30,336 (48) |
5,028 (42) |
11,658 (49) |
13,650 (49) |
<0.001‡
|
| II |
18,521 (29) |
3,910 (33) |
6,514 (27) |
8,097 (29) |
|
| III |
5,561 (9) |
1,197 (10) |
2,098 (9) |
2,266 (8) |
|
| IV |
9,185 (14) |
1,867 (16) |
3,660 (15) |
3,658 (13) |
|
| Comorbidities |
| Hypertension |
62 |
61 |
61 |
63 |
<0.001‡
|
| Diabetes mellitus |
29 |
29 |
28 |
29 |
<0.001‡
|
| Dyslipidemia |
56 |
55 |
55 |
58 |
<0.001‡
|
| CKD |
4.2 |
3.7 |
4 |
4.5 |
<0.001‡
|
| Atrial fibrillation |
5.0 |
5.6 |
5.2 |
4.7 |
<0.001‡
|
| Ambulance use |
69 |
59 |
64 |
66 |
<0.001‡
|
| Medication |
| ACEI or ARB |
88 |
86 |
87 |
89 |
<0.001‡
|
| β-blockers |
58 |
61 |
60 |
56 |
<0.001‡
|
| Antiplatelets |
98 |
98 |
98 |
99 |
<0.001‡
|
| CCB |
63 |
66 |
64 |
60 |
<0.001‡
|
| Statins |
91 |
89 |
90 |
92 |
<0.001‡
|
| Oral anti-diabetic drugs |
64 |
67 |
66 |
62 |
<0.001‡
|
| Device |
| IABP |
15 |
12 |
15 |
17 |
<0.001‡
|
| PCPS |
2.3 |
1.5 |
2 |
2.8 |
<0.001‡
|
| LVAD |
0.01 |
0.00 |
0.02 |
0.01 |
0.36‡
|
| CRRT |
1.1 |
0.9 |
0.9 |
1.3 |
<0.001‡
|
| Respirator |
18 |
17 |
17 |
18 |
<0.001‡
|
| LOHS (days) |
17 (9–20) |
18 (10–22) |
17 (10–20) |
16 (9–19) |
<0.001†
|
| 30-day mortality |
9.6 |
11.5 |
10.0 |
8.3 |
<0.001* |
Data given as mean±SD, median (IQR), n (%) or %. *ANOVA; †Kruskal-Wallis rank sum test; ‡χ2 test with Bonferroni correction. ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; CCB, calcium channel blockers; CKD, chronic kidney disease; CRRT, continuous renal replacement therapy; IABP, intra-aortic balloon pumping; LOHS, length of hospital stay; LVAD, left ventricular assist device; PCPS, percutaneous cardiopulmonary support. Other abbreviations as in Table 1.
Crude in-hospital mortality was 11.5% for the first tertile, 10.0% for the second, and 8.3% for the third, which all decreased as the number of BCC/50 cardiovascular beds increased.
No. BCC/50 Cardiovascular Beds and 30-Day Mortality
On multilevel mixed-effect logistics regression analysis of the number of BCC/50 cardiovascular beds as a continuous variable, the OR for 30-day mortality was 0.987 (95% CI: 0.982–0.994, P<0.01), and the adjusted OR (AOR) for 30-day mortality was 0.988 (95% CI: 0.982–0.995, P<0.01).
On tertile analysis of the number of BCC/50 cardiovascular beds, unadjusted OR for 30-day mortality was significantly lower in the second tertile (OR, 0.81; 95% CI: 0.71–0.92, P<0.001) and third tertile (OR, 0.66; 95% CI: 0.58–0.76, P<0.001) than in the first tertile. Moreover, inverse trends in 30-day mortality were observed between the tertiles (trend test P<0.01). The AOR for 30-day mortality was 0.88 (95% CI: 0.75–1.03, P=0.10) for the second tertile and 0.77 (95% CI: 0.65–0.92, P<0.001) for the third tertile. The number of BCC remained an independent risk factor after stepwise analysis. On stepwise analysis, the AOR for 30-day mortality was 0.86 (95% CI: 0.74–1.00, P=0.05) in the second tertile and 0.75 (95% CI: 0.65–0.88, P<0.01) in the third tertile (Table 3). The OR for the large BCC group vs. the small BCC group was 0.87 (95% CI: 0.78–0.98, P=0.019) after propensity score matching (Table S1).
Table 3.
No. JCS BCC/50 CV Beds and 30-Day Mortality
| |
No. JCS BCC/
50 CV beds
(range) |
n |
Unadjusted analysis |
Multivariate analysis |
Stepwise analysis |
| OR |
95% CI |
P-value |
AOR |
95% CI |
P-value |
AOR |
95% CI |
P-value |
| First tertile |
0.63–6.58 |
12,002 |
Ref. |
|
|
Ref. |
|
|
Ref. |
|
|
| Second tertile |
6.63–10.98 |
23,930 |
0.81 |
0.71–0.92 |
<0.001 |
0.88 |
0.75–1.03 |
0.10 |
0.86 |
0.74–1.00 |
0.05 |
| Third tertile |
11.05–75.58 |
27,671 |
0.66 |
0.58–0.76 |
<0.001 |
0.77 |
0.65–0.92 |
<0.001 |
0.75 |
0.65–0.88 |
<0.001 |
The model includes age (per 10 years), gender, Charlson comorbidity index, Killip classification, hypertension, hyperlipidemia, diabetes, CKD, atrial fibrillation, cardiac surgery, ambulance use, coronary care unit, hospital teaching status, and JCS BCC/50 CV beds. Stepwise analysis includes age (per 10 years), gender, Charlson comorbidity index, Killip classification, hypertension, hyperlipidemia, diabetes, CKD, atrial fibrillation, and ambulance use. AOR, adjusted OR. Other abbreviations as in Tables 1,2.
To investigate whether AMI severity affects the association between the number of BCC/50 cardiovascular beds and 30-day mortality, we performed a stratified analysis according to AMI severity. Similar tendencies were observed in both the not-severe and severe groups. The number of patients and the use of circulation devices increased as the number of BCC/50 cardiovascular beds increased, whereas patient characteristics and medication remained almost unchanged. The length of hospital stay and the crude 30-day mortality decreased as the number of BCC/50 cardiovascular beds increased (Table 4).
Table 4.
Patient Characteristics vs. AMI Severity and Cardiologist Tertile
| |
Not-severe |
Severe |
| Total |
First
tertile |
Second
tertile |
Third
tertile |
P-value |
Total |
First
tertile |
Second
tertile |
Third
tertile |
P-value |
| JCS BCC/50 CV beds (range) |
0.63–
75.58 |
0.63–
6.58 |
6.63–
10.98 |
11.05–
75.58 |
|
0.63–
75.58 |
0.63–
6.58 |
6.63–
10.98 |
11.05–
75.58 |
|
| Patients |
48,857
(100) |
8,938
(18) |
18,172
(37) |
21,747
(45) |
|
14,746
(100) |
3,064
(21) |
5,758
(39) |
5,924
(40) |
|
| Age (years) |
68±13 |
69±13 |
68±13 |
67±13 |
<0.001* |
72±13 |
74±13 |
73±13 |
72±13 |
<0.001* |
| Male |
75 |
73 |
74 |
75 |
<0.001‡ |
68 |
66 |
68 |
70 |
<0.001‡ |
| Charlson comorbidity index |
2 (1–3) |
2 (1–3) |
2 (1–2) |
2 (1–3) |
<0.001† |
2 (1–3) |
2 (1–3) |
2 (1–3) |
2 (1–3) |
<0.001† |
| Comorbidities |
| Hypertension |
68 |
67 |
67 |
69 |
<0.001‡ |
43 |
42 |
43 |
45 |
<0.001‡ |
| Diabetes mellitus |
29 |
30 |
29 |
30 |
<0.001‡ |
26 |
28 |
25 |
27 |
<0.001‡ |
| Dyslipidemia |
63 |
62 |
61 |
68 |
<0.001‡ |
34 |
33 |
34 |
35 |
0.077‡ |
| CKD |
3.5 |
2.9 |
3.4 |
3.7 |
<0.001‡ |
6.5 |
5.7 |
6 |
7.4 |
<0.001‡ |
| Atrial fibrillation |
4.8 |
5.5 |
5 |
4.5 |
<0.001‡ |
5.7 |
5.9 |
5.8 |
5.4 |
<0.001‡ |
| Ambulance use |
61 |
56 |
61 |
62 |
<0.001‡ |
75 |
70 |
74 |
78 |
<0.001‡ |
| Medication |
| ACEI or ARB |
87 |
86 |
87 |
88 |
<0.001‡ |
88 |
87 |
88 |
89 |
0.04‡ |
| β-blockers |
56 |
58 |
57 |
54 |
<0.001‡ |
68 |
68 |
69 |
66 |
<0.001‡ |
| Antiplatelets |
99 |
99 |
99 |
99 |
<0.001‡ |
97 |
96 |
97 |
98 |
<0.001‡ |
| CCB |
60 |
64 |
62 |
56 |
<0.001‡ |
72 |
73 |
73 |
71 |
0.08‡ |
| Statins |
91 |
90 |
91 |
92 |
<0.001‡ |
89 |
87 |
89 |
90 |
<0.001‡ |
| Oral anti-diabetic drugs |
61 |
64 |
62 |
59 |
<0.001‡ |
76 |
75 |
77 |
76 |
<0.001‡ |
| Device |
| IABP |
8.9 |
6.6 |
9.2 |
9.6 |
<0.001‡ |
37 |
29 |
34 |
44 |
<0.001‡ |
| PCPS |
0.4 |
0.2 |
0.4 |
0.6 |
<0.001‡ |
8.3 |
5.2 |
7.1 |
11 |
<0.001‡ |
| LVAD |
0 |
0 |
0 |
0 |
<0.001‡ |
0.05 |
0 |
0.07 |
0.05 |
0.36‡ |
| CRRT |
0.7 |
0.6 |
0.6 |
0.8 |
0.05‡ |
2.6 |
1.9 |
2 |
3.6 |
<0.001‡ |
| Respirator |
7.7 |
7.2 |
7.5 |
8 |
0.05‡ |
50 |
47 |
47 |
54 |
<0.001‡ |
| LOHS (days) |
16
(10–19) |
17
(11–21) |
16
(10–19) |
15
(9–17) |
<0.001‡ |
20
(5–26) |
22
(3–26) |
19
(4–25) |
21
(7–26) |
<0.001‡ |
Data given as mean±SD, median (IQR), n (%) or (%). *ANOVA; †Kruskal-Wallis rank sum test; ‡χ2 test with Bonferroni correction. Abbreviations as in Tables 1,2.
In the not-severe group, the AOR for 30-day mortality was 0.87 (95% CI: 0.70–1.05, P=0.15) in the second tertile and 0.81 (95% CI: 0.67–0.97, P=0.02) in the third tertile, and an inverse trend for 30-day mortality was observed between the tertiles (trend test P<0.01). In the severe group, the AOR was 0.86 (95% CI: 0.71–1.03, P=0.10) in the second tertile and 0.71 (95% CI: 0.59–0.85, P<0.001) in the third tertile, and an inverse trend for 30-day mortality was observed between the tertiles (trend test P<0.01). No interaction effect between the number of BCC/50 cardiovascular beds and AMI severity was seen on interaction analysis (Figure 2; Table 5).
Table 5.
AOR and 95% CI for 30-Day Mortality vs. AMI Severity and Cardiologist Tertile
| |
No. JCS BCC/
50 CV beds (range) |
Not severe |
Severe |
| n |
AOR |
95% CI |
P-value |
n |
AOR |
95% CI |
P-value |
| First tertile |
0.63–6.58 |
8,938 |
Ref. |
|
|
3,064 |
Ref. |
|
|
| Second tertile |
6.63–10.98 |
18,172 |
0.87 |
0.70–1.05 |
0.15 |
5,758 |
0.86 |
0.71–1.03 |
0.10 |
| Third tertile |
11.05–75.58 |
21,747 |
0.81 |
0.67–0.97 |
0.02 |
5,924 |
0.71 |
0.59–0.85 |
<0.001 |
The model includes age (per 10 years), gender, Charlson comorbidity index, Killip classification, hypertension, hyperlipidemia, diabetes, CKD, atrial fibrillation, and ambulance use. Abbreviations as in Tables 1–3.
Discussion
In this study, we demonstrated that the number of BCC/50 cardiovascular beds is associated with 30-day in-hospital AMI mortality. In addition, after adjustment for patient and hospital baseline characteristics, we showed that the hospital 30-day mortality rate in the third tertile hospitals is significantly lower than that in the first tertile hospitals. Moreover, this tendency was independent of AMI severity.
The present results are generally consistent with previous studies. O’Neill et al compared 30-day mortality in acute coronary syndrome (ACS) patients between those admitted to non-cardiology services and those admitted to cardiology services. Patients with ACS who were admitted to cardiology services more commonly underwent cardiac catheterization and evidence-based pharmacotherapy and had a significantly lower 30-day mortality than those admitted to non-cardiology services.14
Badheka et al found that a large operator volume is associated with reduced mortality in patients undergoing PCI.15
Their study included angina pectoris and AMI,15
whereas the present study focused on patients with AMI and investigated the number of cardiologists/50 cardiovascular beds, which in turn is considered new.
We infer that the decreased 30-day mortality with the increase in the number of BCC/50 cardiovascular beds could be attributed to the following: (1) increased annual number of coronary angiography, emergency PCI, and CABG per hospital; moreover, the use of circulation devices, such as IABP and PCPS, increased as the number of BCC/50 cardiovascular beds increased, which in turn suggests that the treatment options for AMI are wide and varied and AMI could be treated quickly; and (2) the total number of hospital beds was greater in hospitals with a large number of BCC/50 cardiovascular beds than in those with a low number of BCC/50 cardiovascular beds, with the former having several medical staff working and sufficient facilities, and thus cardiologists would more likely receive the needed assistance and cooperation from other departments and medical staff when they encounter problems.
Improvement in the prognosis of AMI requires adherence to treatment guidelines.16
Differences in the adherence to treatment guidelines for AMI between cardiologists and non-cardiologists exist.17,18
In the present study, significant differences in the use of medications, such as angiotensin-converting enzyme inhibitors, angiotensin II receptor blocker, β-blockers, and antiplatelets, between tertiles were not found.
New approaches to further improve the acute outcome of AMI treatment are warranted. This study suggests that one of the methods to improve outcome in Japan is to transport patients with AMI to third tertile hospitals. Furthermore, increasing physician volume may be needed to improve treatment outcome.
Study Limitations
This study had several limitations. First, this was an observational study using the JROAD-DPC database, which included approximately 50% of JCS-certified hospitals and 29% of all hospital beds in Japan.11
Second, the accuracy of the diagnoses and procedures in the DPC database is unclear. Third, the present study could not include some important confounding factors such as door-to-balloon time. Finally, we could not calculate the appropriate number of cardiologists to ensure the best outcome.
Conclusions
Patients with AMI who are admitted to hospitals with a large number of cardiologists/50 cardiovascular beds have a lower 30-day mortality rate. This tendency was independent of patient and hospital characteristics and AMI severity. This study is the first to provide new information about the relationship between the number of BCC and in-hospital AMI-related mortality in Japan.
Acknowledgment
We would like to express our gratitude to Ms Sumita for preparing the data of the JROAD and JROAD-DPC study.
Disclosures
The authors declare no conflicts of interest.
Supplementary Files
Supplementary File 1
Table S1.
No. JCS BCC/50 CV beds and 30-day AMI mortality with PSM
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-18-0487
References
- 1.
Hartley A, Marshall DC, Salciccioli JD, Sikkel MB, Maruthappu M, Shalhoub J. Trends in mortality from ischemic heart disease and cerebrovascular disease in Europe: 1980 to 2009. Circulation 2016; 133: 1916–1926.
- 2.
McGovern PG, Jacobs DR Jr, Shahar E, Arnett DK, Folsom AR, Blackburn H, et al. Trends in acute coronary heart disease mortality, morbidity, and medical care from 1985 through 1997: The Minnesota heart survey. Circulation 2001; 104: 19–24.
- 3.
Ministry of Health, Labour and Welfare. Statistics 2015. http://www.mhlw.go.jp/toukei_hakusho/toukei/index.html (accessed January 1, 2018).
- 4.
Puymirat E, Simon T, Steg PG, Schiele F, Guéret P, Blanchard D, et al. Association of changes in clinical characteristics and management with improvement in survival among patients with ST-elevation myocardial infarction. JAMA 2012; 308: 998–1006.
- 5.
Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, et al. Heart disease and stroke statistics–2015 update: A report from the American Heart Association. Circulation 2015; 131: e29–e322.
- 6.
Japanese Registry Of All cardiac and vascular Diseases (JROAD). Survey on medical cardiovascular disease 2016. http://www.j-circ.or.jp/jittai_chosa/jittai_chosa2016web (accessed February 2, 2018).
- 7.
Cui Y, Hao K, Takahashi J, Miyata S, Shindo T, Nishimiya K, et al. Age-specific trends in the incidence and in-hospital mortality of acute myocardial infarction over 30 years in Japan: Report from the Miyagi AMI Registry Study. Circ J 2017; 81: 520–528.
- 8.
The Japanese Circulation Society. Cardiovascular specialist and membership directry. http://www.j-circ.or.jp/ (accessed February 2, 2018).
- 9.
Tomoike H, Yokoyama H, Sumita Y, Hanai S, Kada A, Okamura T, et al. Nationwide distribution of cardiovascular practice in Japan: Results of Japanese Circulation Society 2010 annual survey. Circ J 2015; 79: 1058–1067.
- 10.
Yasuda S, Nakao K, Nishimura K, Miyamoto Y, Sumita Y, Shishido T, et al. The current status of cardiovascular medicine in Japan: Analysis of a large number of health records from a nationwide claim-based database, JROAD-DPC. Circ J 2016; 80: 2327–2335.
- 11.
Rothman K. Epidemiology: An introduction. New York: Oxford University Press, 2002.
- 12.
Andersson T, Alfredsson L, Kallberg H, Zdravkovic S, Ahlbom A. Calculating measures of biological interaction. Eur J Epidemiol 2005; 20: 575–579.
- 13.
VanderWeele TJ, Tchetgen Tchetgen EJ. Attributing effects to interactions. Epidemiology 2014; 25: 711–722.
- 14.
O’Neill DE, Southern DA, Norris CM, O’Neill BJ, Curran HJ, Graham MM. Acute coronary syndrome patients admitted to a cardiology vs non-cardiology service: Variations in treatment and outcome. BMC Health Serv Res 2017; 17: 354.
- 15.
Badheka AO, Patel NJ, Grover P, Singh V, Patel N, Arora S, et al. Impact of annual operator and institutional volume on percutaneous coronary intervention outcomes: A 5-year United States experience (2005–2009). Circulation 2014; 130: 1392–1406.
- 16.
Drozda J Jr, Messer JV, Spertus J, Abramowitz B, Alexander K, Beam CT, et al. ACCF/AHA/AMA-PCPI 2011 performance measures for adults with coronary artery disease and hypertension: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Performance Measures and the American Medical Association-Physician Consortium for Performance Improvement. J Am Coll Cardiol 2011; 58: 316–336.
- 17.
Ayanian JZ, Hauptman PJ, Guadagnoli E, Antman EM, Pashos CL, McNeil BJ. Knowledge and practices of generalist and specialist physicians regarding drug therapy for acute myocardial infarction. N Engl J Med 1994; 331: 1136–1142.
- 18.
Hickson RP, Robinson JG, Annis IE, Killeya-Jones LA, Korhonen MJ, Cole AL, et al. Changes in statin adherence following an acute myocardial infarction among older adults: Patient predictors and the association with follow-up with primary care providers and/or cardiologists. J Am Heart Assoc 2017; 6: e007106.
