Health Inequality Among Older Adults with Percutaneous Coronary Intervention and Universal Health Coverage in Japan

Abstract

This study aimed to comprehensively evaluate whether income affects long-term health outcomes for older patients who underwent percutaneous coronary intervention (PCI) provided by a universal health coverage system. Data were from the Latter Stage Elderly Healthcare Insurance database in Fukuoka Prefecture, Japan. A total of 5625 individuals aged ≥65 years who underwent PCI in 2014–2016 were included. Cox proportional hazards models were used to assess the association between income status and the incidence of health outcomes. With a median follow-up of 1095 days, 554 acute myocardial infarction (AMI) cases, 1075 stroke cases, 1690 repeat revascularization cases, and 1094 deaths were observed. Risk of all-cause mortality decreased significantly with increasing income level in both unadjusted and adjusted Cox regression models. Patients in the low-income level had a significantly higher rate of AMI (log-rank P = 0.003), stroke (log-rank P = 0.039), and all-cause mortality (log-rank P = 0.001) compared with patients in the high-income level. Observed rates for repeat revascularization also were high in the first year after PCI. In the Japanese universal health setting, low-income patients had a comparatively higher mortality risk after PCI. Poor long-term outcomes might be attributed to patients' baseline characteristics rather than treatment processes.

Introduction

Percutaneous coronary intervention (PCI) is one of the most effective methods for treating coronary artery diseases in an aging society,¹ and the use of PCI is increasing worldwide.^2,3 PCI is used widely in Japan. It is estimated that more than 200,000 PCI procedures are performed annually in Japan,⁴ and the proportion of older people who undergo PCI has increased over the past decade.⁵ Patients with PCI are at high risk of myocardial infarction, stroke, and repeat revascularizations.⁶ To minimize the risk of these complications, long-term postoperative care must be provided.

Low income was reported to be associated with higher mortality and poorer quality of life after PCI.^7,8 However, the impact of income on long-term survival outcomes for older adults after PCI has seldom been evaluated. A large cohort study in England found an increased mortality rate after PCI in the lowest socioeconomic status patients compared with the highest socioeconomic status patients.⁹ Although issues related to health care inequity among patients by income status was unheard of under the universal health insurance coverage in Japan, many studies have demonstrated that patients' postoperative outcomes and quality of life after cardiovascular-related surgery, including PCI, varied substantially among the populations with different income levels.^10,11 One possible reason is that sufficient postoperative management might not be provided. Although Japanese universal health insurance set a limit on what patients pay out of pocket for expensive medical care,¹² such as the cost for PCI, the cost for long-term postoperative care following hospital discharge is still a huge burden for patients with low-income status.⁵

Understanding the relationship between income status and clinical outcomes in patients with PCI is important for addressing health care allocation inequality and for increasing access to quality care. Therefore, this retrospective cohort study was conducted that involves older patients with PCI status and uses an insurance claims database in Japan—a country that provides universal health coverage to older adults. This study aimed to clarify whether income has an impact on PCI patients' long-term outcomes after discharge from the hospital.

Methods

Study design and data source

This study was designed as a retrospective cohort study. Data were obtained from the Latter Stage Elderly Healthcare Insurance database of Fukuoka Prefecture with permission from the local government. Latter Stage Elderly Healthcare Insurance is an insurance plan designed specifically for residents aged ≥75 years and for residents aged between 65–74 years who have a specified disability. Each individual's identification number, sex, birth date, income level, diagnostic information, and treatment procedures are available in the claims database. Records were extracted for 5 fiscal years (ie, April 1, 2014, until March 31, 2019). A total of 5666 individuals who were discharged from the hospital between April 1, 2014, and March 31, 2016 were identified. Follow-up covered a 3-year period, starting from the date of hospital discharge.

International Classification of Diseases Tenth Revision (ICD-10) codes were used to capture patients' comorbidity status. Charlson comorbidity index (CCI) was used to measure severity of comorbidity status. To assess the potential impact on treatment prognosis and mortality, weights for CCI scores ranging from 1 to 6 were assigned. CCI scores for each patient were calculated by adding the weights of all comorbidities.¹³ A previous study indicated that dementia and acute myocardial infarction (AMI) had a significant impact on the survival of patients who underwent PCI.¹⁴ Therefore, the present study also extracted the history of dementia and AMI status from the database separately. ICD-10 codes for dementia were adapted from the work of Chartier et al (F00–F04, F05.1, F10.7, F12.7, G30, G31.0, and G31.1).¹⁵ The ICD-10 code for myocardial infarction was defined as I21, and the type of diagnosis procedure combination (DPC) from the claims data was used to distinguish patients' acute conditions because the DPC is a Japanese mixed classification that is used by almost all acute care hospitals.

The claims database was used to identify the use of antiplatelet medications, which included clopidogrel, prasugrel, ticlopidine, beraprost, cilostazol, and aspirin. The database also was used to identify the use of commonly prescribed medications for post-PCI patients, including statins, β-blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, nitrates, warfarin, proton pump inhibitors, and H2 blockers.

Patients without income records were excluded from the study. A total of 5625 patients were identified as study subjects. The selection and exclusion processes are presented in Figure 1. The databases that were used to obtain study data were computer administered. Based on the Japanese government report, the insurance databases must have had a high penetration rate of up to approximately 98.6% until April 2015.¹⁶ The Institutional Review Board of Kyushu University (Clinical Bioethics Committee of the Graduate School of Medical Sciences, Kyushu University) approved this study. Because anonymized claims insurance data were used, there was no requirement to obtain informed consent.

FIG. 1.

Flowchart of subject identification and inclusion and exclusion criteria details. PCI, percutaneous coronary intervention.

Patients' characteristics

This study assessed information on sex, age, income level, comorbidities, and CCI. Age was categorized into the following 4 groups: 65–74 years, 75–79 years, 80–84 years, and ≥85 years. The insurance database records income levels for the older population as low, middle, and high. Therefore, in this study, income levels were classified into low-, middle-, and high-income levels. These categories are applied by insurance associations operated in all prefectures in Japan. Such categories also correspond to the threshold set by the Japanese Ministry of Health, Labour and Welfare based on annual income: low-level, <1,550,000 yen ($13,868.5); middle-level, 1,550,000—3,830,000 yen (($13,868.5—34,268.7); high-level, >3,830,000 yen ($34,268.7). CCI scores were applied to define severity of comorbid conditions as mild (CCI = 0–1), moderate (CCI = 2–3), or severe (CCI ≥4).

End points

The end points of this study were 3 major adverse cardiac events and 3-year all-cause mortality. This study defined major clinical events as AMI, stroke, and repeat revascularization. Therefore, these 4 outcomes (AMI, stroke, repeat revascularization, and death for any cause) were analyzed separately. Repeat revascularization was defined as an event during which a PCI patient received subsequent procedures of PCI or coronary artery bypass grafting (CABG) after being discharged from hospital. Patients were censored at the time of death or at the end of the 3-year follow-up period. Information on these outcomes, including the dates of these outcomes, were extracted from the insurance claims database.

Statistical analysis

Baseline patient characteristics, medications, and clinical outcomes were compared among different income levels using one-way analysis of variance for continuous variables and Pearson's chi-square test for categorical variables. Cox proportional hazards models were used to compute crude and adjusted hazard ratios (AHRs) and 95% confidence intervals (CIs) for the end points to estimate the association between income levels and each outcome. In the adjusted Cox regression models, sex, age group, income level, CCI scores, AMI status, dementia status, and number of medications were adjusted covariates. Proportional hazards assumptions were checked based on Schoenfeld residuals. Cumulative event rates for each outcome were calculated by means of the Kaplan-Meier method. Log-rank tests were performed to assess the significance of income levels for each observed event time. All reported P values were 2-tailed, and the level of significance was set at P < 0.05. All statistical analyses for this study were performed using Stata Statistical Software: Release 15 (StataCorp LLC, College Station, TX, USA).

Results

Descriptive analysis

The study includes 5625 patients who previously had PCI. Table 1 shows the characteristics of the patients, 1960 (34.8%) of whom were at the low-income level, 3201 (56.9%) at the middle-income level, and 464 (8.2%) at the high-income level. Compared with males, more females were in the low-income level group. Patients between ages 75–79 had higher income levels compared with patients in other age groups. Patients with low-income status had a higher proportion of high CCI scores compared with the other patients. Patients with AMI were not significantly different among the different income levels, while patients with dementia showed a significant difference among the income levels.

Table 1.

Baseline Characteristics of Patients Undergoing Percutaneous Coronary Intervention, by Income

	Income level			P value
	Low (n = 1960)	Middle (n = 3201)	High (n = 464)	P value
Sex
Men	971 (50%)	2270 (71%)	375 (81%)	<0.001
Women	989 (50%)	931 (29%)	89 (19%)	<0.001
Age, years
65–74	396 (20%)	511 (16%)	63 (14%)	<0.001
75–79	694 (35%)	1289 (40%)	209 (45%)
80–84	520 (27%)	892 (28%)	129 (28%)
≥85	350 (18%)	509 (16%)	63 (14%)
CCI scores
0–1	964 (49%)	1614 (50%)	220 (47%)	0.007
2–3	651 (33%)	1129 (35%)	156 (34%)
≥4	345 (18%)	458 (14%)	88 (19%)
History of AMI
Yes	738 (38%)	1175 (37%)	148 (32%)	0.068
No	1222 (62%)	2026 (63%)	316 (68%)	0.068
History of dementia
Yes	472 (24%)	609 (19%)	77 (17%)	<0.001
No	1488 (76%)	2592 (81%)	387 (83%)	<0.001

AMI, acute myocardial infarction; CCI, Charlson comorbidity index.

Regarding medications, clopidogrel was the most frequently prescribed antiplatelet drug (Table 2). There also was frequent use of aspirin, nitrates, proton pump inhibitors, and statins. Statins were prescribed more often among patients in the high-income level, while clopidogrel, proton pump inhibitors, and H2 blockers were prescribed more often among patients in the low- and middle-income levels. Median follow-up time was 1095 days. During follow-up, 554 (9.8%) MI, 1075 (19.1%) stroke, 1690 (30.0%) repeat revascularization, and 1094 (19.4%) death events were observed. Proportions of AMI, stroke, and death cases were significantly higher among low-income patients.

Table 2.

Medications and Health Outcomes, by Income

	Income level			P value
	Low (n = 1960)	Middle (n = 3201)	High (n = 464)	P value
Medications
Any antiplatelet drug	1887 (96%)	3108 (97%)	443 (95%)	0.014
Clopidogrel	1570 (80%)	2541 (79%)	346 (75%)	0.029
Prasugrel	555 (28%)	900 (28%)	145 (31%)	0.372
Ticlopidine	179 (9%)	256 (8%)	42 (9%)	0.328
Beraprost	131 (7%)	167 (5%)	17 (4%)	0.014
Aspirin	1854 (95%)	3036 (95%)	431 (93%)	0.219
Cilostazol	254 (13%)	398 (12%)	51 (11%)	0.508
Other medications
Statins	1533 (78%)	2583 (81%)	391 (84%)	0.006
Beta-blockers	798 (41%)	1372 (43%)	206 (44%)	0.196
Calcium channel blockers	1109 (57%)	1775 (55%)	241 (52%)	0.191
ACE-I/ARB	612 (31%)	997 (31%)	130 (28%)	0.369
Nitrates	1905 (97%)	3106 (97%)	449 (97%)	0.873
Warfarin	341 (17%)	479 (15%)	72 (16%)	0.066
Proton pump inhibitors	1600 (82%)	2517 (79%)	360 (78%)	0.018
H2 blockers	660 (34%)	932 (29%)	131 (28%)	0.001
Total number of other medications, median (IQR)	4 (2)	4 (2)	4 (2)	0.208
Outcomes
AMI	235 (12%)	276 (9%)	43 (9%)	<0.001
Stroke	409 (21%)	579 (18%)	87 (19%)	0.047
Repeat revascularization	593 (30%)	964 (30%)	133 (29%)	0.791
Death	427 (22%)	596 (19%)	71 (15%)	0.001

Total number of other medications (excluding antiplatelet and aspirin) was measured at the time of discharge. The number of total medications was selected to be a covariate in the Cox proportional hazard models. Patients received subsequent performance of percutaneous coronary intervention or coronary artery bypass grafting.

ACE-I, angiotensin-converting enzyme inhibitors; ARB, angiotensin receptor blockers; AMI, acute myocardial infarction; IQR, interquartile range.

Clinical outcomes

Table 3 shows hazard ratios for AMI, stroke, repeat revascularization, and death from any cause by income level. Risks of AMI and stroke decreased significantly in middle-income-level patients. Risk of AMI was similar even after the estimate was adjusted for sex, age, CCI scores, AMI status, dementia status, and number of medications. Risk of all-cause mortality decreased significantly with increasing income level in both unadjusted and adjusted Cox regression model (middle: AHR 0.83, 95% CI 0.73–0.95, P = .005; high: AHR 0.65, 95% CI 0.50–0.84, P = .001) (Table 3, Figure 2). The Kaplan-Meier curves for outcomes that compare cumulative rates of patients by different income levels are presented in Figure 3. The Kaplan-Meier estimates showed that patients in the low-income level had a significantly higher AMI rate (log-rank P = 0.003), stroke rate (log-rank P = 0.039), and long-term mortality rate (log-rank P = 0.001) compared with patients in the high-income level.

FIG. 2.

Adjusted hazard ratios among different income levels for AMI rehospitalization, stroke, revascularization, and death from any cause. Covariables in the models include age, sex, Charlson comorbidity index scores, history of AMI, history of dementia, and numbers of medications. AHR, adjusted hazard ratio; AMI, acute myocardial infarction.

FIG. 3.

Kaplan-Meier curves among different income levels for AMI (A), stroke (B), repeat revascularization (C), and death from any cause (D) were compared among different income levels. AMI, acute myocardial infarction.

Table 3.

Crude and Adjusted Hazard Ratios for Acute Myocardial Infarction, Stroke, Revascularization, and Death, by Income

	HR (95% CI)	P value	AHR (95% CI)	P value
AMI (n = 554)
Low (n = 235)	Ref.		Ref.
Middle (n = 276)	0.71 (0.59 − 0.84)	<0.001	0.71 (0.59 − 0.85)	<0.001
High (n = 43)	0.76 (0.55 − 1.05)	0.094	0.73 (0.59 − 1.15)	0.264
Stroke (n = 1075)
Low (n = 409)	Ref.		Ref.
Middle (n = 579)	0.85 (0.75 − 0.96)	0.012	0.89 (0.78 − 1.02)	0.091
High (n = 87)	0.88 (0.70 − 1.12)	0.300	0.91 (0.72 − 1.15)	0.410
Repeatrevascularization (n = 1690)
Low (n = 593)	Ref.		Ref.
Middle (n = 964)	0.99 (0.89 − 1.09)	0.804	0.97 (0.87 − 1.07)	0.527
High (n = 133)	0.93 (0.77 − 1.13)	0.482	0.88 (0.73 − 1.06)	0.185
Death from any cause (n = 1094)
Low (n = 427)	Ref.		Ref.
Middle (n = 596)	0.84 (0.74 − 0.95)	0.005	0.83 (0.73 − 0.95)	0.005
High (n = 71)	0.68 (0.53 − 0.87)	0.002	0.65 (0.50 − 0.84)	0.001

AHRs were estimated by Cox proportional hazard models. Covariables in the models include sex, age, Charlson comorbidity index scores, history of AMI, history of dementia, and numbers of medications.

AHR, adjusted hazard ratio; AMI, acute myocardial infarction; CI, confidence interval; HR, hazard ratio.

Discussion

This retrospective cohort study comprehensively evaluated the association between income levels and clinical outcomes among older patients who underwent PCI in the setting of universal health coverage in Japan. This study found that patients with a high-income level were less likely to have a mortality risk compared with patients with a low-income level in both unadjusted and adjusted analyses (AHR: 0.65). There was a progressive increase in the cumulative rates of death from any cause from the high- to low-income level. Significant effects were not observed for the risk of AMI, stroke, and repeat revascularization among different income levels.

This is the first study to investigate the effect of income status on clinical outcomes after PCI in Japan. A number of studies conducted in western countries showed that a lower income level is associated with a higher all-cause mortality rate following PCI.^7
–9,17,18 The present study is consistent with and extends the findings of previous studies in the setting of a universal health coverage in Japan. A study involving 1346 PCI patients reported that low-income patients who were treated with PCI have more readmission events than high-income patients.⁷ A study involving 13,770 patients who were undergoing PCI reported that the all-cause mortality rate was significantly higher in patients in the lowest quintile socioeconomic status than the highest quintile during a median follow-up of 3.7 years.⁹ However, these studies were focused on valuating a specific adverse event after PCI. The present study concurrently evaluated a number of major adverse cardiac events (AMI, stroke, and repeat revascularization) and death from any cause. This study found that low-income patients faced a higher risk of all-cause mortality rather than other adverse cardiac events. By assessing these major clinical outcomes, this study ascertained the presence of disparities in outcomes among patients with different income status in Japan.

Some previous studies demonstrated several possible reasons why income level might influence PCI patients' long-term mortality rates. One possible reason might be social inequalities in the long-term management of chronic diseases among patients with different income levels.^8,9 Some evidence demonstrated that the measures for long-term management of chronic diseases, including post-PCI medication compliance and cardiac rehabilitation, are reduced in patients with a lower income level,^19,20 which might contribute to increased long-term mortality for these patients.^21,22 Despite some consistency with previous studies, the present study failed to observe a significant difference in any outcome of AMI, stroke, and repeat revascularization among patients with different income status. To the best of the authors' knowledge, this could be explained by the provision of universal health coverage in Japan. Because most out-of-pocket expenses could be subsidized by the government under this scheme,¹² it is possible to have identical medical treatment processes and clinical outcomes for all patients with different economic status. Another possible reason might be different lifestyle and dietary habits among patients with different income levels.^9,23 A previous study reported that patients with low-income status had a considerably worse diet and a lower quality of life,⁷ which also might contribute to increased morbidity and mortality. A past study of 5622 PCI patients reported that low-income patients with PCI had a worse prognosis than high-income patients during a mean follow-up of 3.7 years,⁸ and that the poor outcomes could be explained by differences in patients' baseline characteristics rather than treatment procedures. Therefore, in a universal health coverage setting, patients with different income levels have equitable access to health care services. Thus, the higher mortality rate is related to patients' characteristics rather than the treatment process.

In addition, despite a previous study concluding that lower economic status was associated with higher rates of repeat PCI and CABG,⁹ the present study did not observe a significant difference in the outcome of repeat revascularization among patients with different income levels. The reason might be that the universal health coverage insurance system in Japan ensures that all patients receive access to medical treatment with affordable out-of-pocket expenses.¹² Nevertheless, it was observed that repeat revascularization rates in the first year after PCI were comparatively high in older patients, which is consistent with some previous clinical studies.^24,25 Thus, in the setting of universal health coverage, long-term secondary medical prevention of restenosis seemed necessary.²⁴

These findings have some important implications. To improve the quality of care for older adults who are undergoing PCI, it is important for post-PCI patients to have proper secondary prevention strategies and a healthy lifestyle, which might be substantially influenced by their economic status.^9,26 Policies to address poor post-PCI prognosis issues for older adults must cover secondary community prevention programs. To provide an environment that supports high quality of life, Japan is currently structuring a nationwide program called the Community-based Integrated Care System (CICS). Municipal governments are required to facilitate the construction of this CICS through collaborations with health care, nursing care providers, health promotion, and public health practitioners.^27,28 In order for this system to succeed, it is necessary to identify older people with low incomes and those with high mortality risk.^8,29 Identification of these vulnerable populations in the CICS system is useful in providing intensified secondary prevention and promoting a healthy lifestyle.³⁰

This study also has some limitations. First, the insurance claims data used were not specifically designed for clinical research purposes. Data analyzed did not provide information on disease severity, adherence to preventative therapies, or cardiac rehabilitation services. Researchers should explore these health care interventions, such as cardiac rehabilitation, to overcome the limitations faced by the current study. To compensate for this limitation, the current study provided information regarding medications prescribed after discharge and used the total number of other medications except antiplatelet drugs as a covariate. Second, the database does not contain information on living conditions. After discharge from the hospital, patients might choose to either go back home or to other health care facilities for recuperation. Patients who returned home may live alone or together with family. Third, by performing a power log-rank test to examine whether the sample had sufficient size power to detect the significance of the finding for all-cause mortality as one of the study outcomes, it was found that the number of samples in the high-income group (n = 464) is smaller than the sample size needed in each group. Therefore, study estimates might be statistically less accurate. Furthermore, there is a possibility of coding errors in the database. Upon inspection, it also was observed that a number of records for income status were missing. However, such an issue represents <1% of the total observation; therefore, the authors believe that it had little influence on the statistical inferences.

Despite these limitations, there are some strengths to this study. The main strength is the assurance of wide population coverage and a large sample size. The insurance claim database covered more than 600,000 insured people in Fukuoka Prefecture, and the penetration rate was as high as 98.6%, which was cited in some previous studies that used a similar insurance database.¹⁶ Based on the wide population coverage, the insurance database provides complete information regarding patient-level characteristics, use of medications, and complete follow-up, which could minimize the risk of selection bias.

Conclusions

This study indicates that low-income patients faced a worse survival outcome after PCI even under universal health coverage in Japan. This association persisted despite adjustment by sex. The poor long-term survival outcome seems to be explained by patients' characteristics rather than the acute treatment processes. Intensified secondary prevention of restenosis and healthy lifestyle promotion in the CICS for older patients, especially those with a lower income status, could provide a promising solution to address this health inequality.

Footnotes

Authors' Contributions

Mr. Li conceived the design, acquired the data, performed statistical analysis, and drafted the manuscript. Dr. Babazono supervised the design and interpreted the results. Mr. Jamal performed the analysis and interpreted the results. Drs. Ohmori and Yoshida provided input from the clinical perspective. Ms. Kim, Dr. Fujita, and Ms Liu assisted with acquiring the data and revised the manuscript critically.

Acknowledgments

The authors thank the Wide-area Association of Latter Stage Elderly Healthcare of Fukuoka Prefecture for their provision of a health care claims database. We thank Jodi Smith, PhD, from Edanz Group for editing a draft of this manuscript.

Author Disclosure Statement

The authors declare that there are no conflicts of interest.

Funding Information

No funding was received for this article.

References

Madhavan

, Gersh

, Alexander

, Granger

, Stone

. Coronary artery disease in patients ≥80 years of age. J Am Coll Cardiol 2018 71:2015–2040.

Bhatt

DL.

Percutaneous coronary intervention in 2018. JAMA, 2018; 31:2127–2128.

Al-Lamee

, Thompson

, Dehbi

H-M

, et al. Percutaneous coronary intervention in stable angina (ORBITA): a double-blind, randomised controlled trial. Lancet, 2018; 391:31–40.

Inohara

, Kohsaka

, Yamaji

, et al. Impact of institutional and operator volume on short-term outcomes of percutaneous coronary intervention: a report from the Japanese Nationwide Registry. JACC Cardiovasc Interv, 2017; 10:918–927.

Kumar

, McDaniel

, Samady

, Forouzandeh

. Contemporary revascularization dilemmas in older adults. J Am Heart Assoc, 2020; 9:e014477–e014477.

Serruys

, Morice

, Kappetein

, et al. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med, 2009; 360:961–972.

Denvir

, Lee

, Rysdale

, et al. Influence of socioeconomic status on clinical outcomes and quality of life after percutaneous coronary intervention. J Epidemiol Community Health, 2006; 60:1085–1088.

Jakobsen

, Niemann

, Thorsgaard

, et al. Dimensions of socioeconomic status and clinical outcome after primary percutaneous coronary intervention. Circ Cardiovasc Interv, 2012; 5:641–648.

Daniel

, James

, Krishnaraj

, et al. The impact of socio-economic status on all-cause mortality after percutaneous coronary intervention: an observational cohort study of 13,770 patients. EuroIntervention, 2015; 10:e1–e8.

10.

Iso

Lifestyle and cardiovascular disease in Japan. J Atheroscler Thromb, 2011; 18:83–88.

11.

Tsuji

, Arima

, Ohkubo

, et al. Socioeconomic status and knowledge of cardiovascular risk factors: NIPPON DATA2010. J Epidemiol, 2018; 28 Suppl 3 (Suppl 3):S46–S52.

12.

, Babazono

, Jamal

, Jiang

, Fujita

Cost-sharing effects on hospital service utilization among older people in Fukuoka Prefecture, Japan. Int J Health Policy Manag 2020. doi:10.3417/2/ijhpm.2020.190

13.

Sundararajan

, Henderson

, Perry

, Muggivan

, Quan

, Ghali

. New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. J Clin Epidemiol, 2004; 57:1288–1294.

14.

Cermakova

, Szummer

, Johnell

, et al. Management of acute myocardial infarction in patients with dementia: data from SveDem, the Swedish Dementia Registry. J Am Med Dir Assoc, 2017; 18:19–23.

15.

Chartier

, Bolton

, Mota

, et al. Mental illness among adult manitobans. Winnipeg MB., Manitoba Centre for Health Policy. 2018. http://mchp-appserv.cpe.umanitoba.ca/reference/mh2015_Report_web.pdf Accessed July 2, 2020 .

16.

Jamal

, Babazono

, Li

, Yoshida

, Fujita

. Multilevel analysis of hemodialysis-associated infection among end-stage renal disease patients: results of a retrospective cohort study utilizing the insurance claim data of Fukuoka Prefecture, Japan. Medicine, 2020; 99:e19871.

17.

Biswas

, Andrianopoulos

, Duffy

, et al. Impact of socioeconomic status on clinical outcomes in patients with ST-segment-elevation myocardial infarction. Circ Cardiovasc Qual Outcomes, 2019; 12:e004979.

18.

Chang

W-C

, Kaul

, Westerhout

, Graham

, Armstrong

. Effects of socioeconomic status on mortality after acute myocardial infarction. Am J Med, 2007; 120:33–39.

19.

Alter

, Franklin

, Ko

, et al. Socioeconomic status, functional recovery, and long-term mortality among patients surviving acute myocardial infarction. PLoS One, 2013; 8:e65130–e65130.

20.

Shanmugasegaram

, Oh

, Reid

, McCumber

, Grace

. Cardiac rehabilitation barriers by rurality and socioeconomic status: a cross-sectional study. Int J Equity Health, 2013; 12:72–72.

21.

Beatty

, Doll

, Schopfer

, et al. Cardiac rehabilitation participation and mortality after percutaneous coronary intervention: insights from the veterans affairs clinical assessment, reporting, and tracking program. J Am Heart Assoc, 2018; 7:e010010–e010010.

22.

Beatty

, Bradley

, Maynard

, McCabe

. Referral to cardiac rehabilitation after percutaneous coronary intervention, coronary artery bypass surgery, and valve surgery. Circ Cardiovasc Qual Outcomes, 2017; 10:e003364.

23.

Lowry

, Kann

, Collins

, Kolbe

. The effect of socioeconomic status on chronic disease risk behaviors among US adolescents. JAMA, 1996; 276:792–797.

24.

Stolker

, Cohen

, Kennedy

, et al. Repeat revascularization after contemporary percutaneous coronary intervention. Circ Cardiovasc Interv, 2012; 5:772–782

25.

Parasca

, Head

, Milojevic

, et al. Incidence, characteristics, predictors, and outcomes of repeat revascularization after percutaneous coronary intervention and coronary artery bypass grafting. JACC Cardiovasc Interv, 2016; 9:2493–2507.

26.

Salari

, Rouhi Balasi

, Ashouri

, Moaddab

, Zaersabet

, Nourisaeed

. Medication adherence and its related factors in patients undergoing coronary artery angioplasty. J Caring Sci, 2018; 7:213–218.

27.

Song

, Tang

. The community-based integrated care system in Japan: Health care and nursing care challenges posed by super-aged society. Biosci Trends, 2019; 13:279–281.

28.

Tsutsui

Implementation process and challenges for the community-based integrated care system in Japan. Int J Integr Care, 2014; 14:e002.

29.

Arpey

, Gaglioti

, Rosenbaum

. How socioeconomic status affects patient perceptions of health care: a qualitative study. J Prim Care Community Health, 2017; 8:169–175.

30.

Mandsager

, Harb

, Cremer

, Phelan

, Nissen

, Jaber

. Association of cardiorespiratory fitness with long-term mortality among adults undergoing exercise treadmill testing. JAMA Netw Open, 2018; 1:e183605–e183605.