High long-term mortality in ischaemic heart disease accentuated among ethnic minorities in Eastern Europe: findings from a prospective all-comers percutaneous coronary intervention registry in Romania
========================================================================================================================================================================================================

* Paul-Adrian Călburean
* Paul Grebenișan
* Ioana-Andreea Nistor
* Ioana Paula Șulea
* Anda-Cristina Scurtu
* Klara Brinzaniuc
* Horatiu Suciu
* Marius Harpa
* Dan Dobreanu
* Laszlo Hadadi

## Abstract

**Background** Long-term outcomes in cardiovascular diseases are historically under-reported in Eastern Europe. Our aim was to report long-term survival and to identify survival predictors in a prospective Romanian percutaneous coronary intervention (PCI) registry, with an emphasis on important under-resourced minorities, such as Hungarian and Roma ethnicities.

**Methods** An all-comers patient population treated by PCI in a tertiary cardiovascular centre that has been included prospectively in the local registry since January 2016 was analysed. Cardiovascular cause and all-cause mortality data were available as of December 2023.

**Results** A total of 6867 patients with 8442 PCI procedures were included. Romanian group consisted of 5095 (74.2%) patients, the Hungarian group consisted of 1417 (20.6%) patients and the Roma group consisted of 355 (5.1%) patients. During a median follow-up of 3.60 (1.35–5.75) years, a total of 1064 cardiovascular-cause and 1374 all-cause events occurred. Romanian, Hungarian and Roma patients suffered 5.12, 5.89 and 7.71 all-cause deaths per 100 patient-years, respectively. Romanian, Hungarian and Roma patients suffered 3.94, 4.63 and 6.22 cardiovascular-cause deaths per 100 patient-years, respectively. Both Hungarian and Roma patients presented significantly higher all-cause mortality than Romanian patients (adjusted HR (aHR)=1.20 (1.05–1.36), p=0.005 and aHR=1.51 (1.21–1.88), p=0.0001). Similarly, Hungarian and Roma patients presented significantly higher cardiovascular cause mortality than Romanian patients (aHR=1.22 (1.05–1.41), p=0.006 and aHR=1.51 (1.18–1.92), p=0.0008).

**Conclusions** High long-term cardiovascular and all-cause mortality was observed for the entire included population. Long-term survival was significantly lower in ethnic minorities, such as the Hungarian and Roma minority than in the Romanian population.

*   EPIDEMIOLOGY
*   CARDIOVASCULAR DISEASES
*   ETHNIC GROUPS

#### WHAT IS ALREADY KNOWN ON THIS TOPIC

*   Long-term mortality in ischaemic heart disease in Eastern Europe is estimated to be high, however, reports rely on aggregated country-specific data and not individual patient-level data.

#### WHAT THIS STUDY ADDS

*   In an all-comers prospective percutaneous coronary interventions registry from Romania, all-cause deaths occurred in 5.40 cases per 100 patient-years, while cardiovascular-cause deaths occurred in 4.19 cases per 100 patient-years. The crude incidence rate is more than double that reported in Western registries.

*   Mortality was higher in ethnic minorities such as Roma or Hungarian minority and ethnicity was an independent predictor of events, even after adjusting for potential confounders.

#### HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

*   Our results should encourage clinical practice and policymakers to increase primary prevention efforts since such minorities had a higher burden of risk factors.

## Introduction

Ischaemic heart disease is the leading cause of death and premature mortality, both worldwide and in Europe.1 However, data regarding long-term outcomes of not only ischaemic heart disease, but of any cardiovascular disease is typically scarce in Eastern Europe. Reports on morbidity and mortality in cardiovascular diseases from Eastern Europe usually rely on aggregated country-specific data and not individual patient-level data.1 2 Moreover, individual patient-level reports regarding sex‐based, race‐based and ethnicity‐based differences or long-term outcomes in cardiovascular disease are practically absent in Eastern Europe.

Regarding social and economic differences, there are important disparities between Eastern Europe and Western countries. Romania, one of the main Eastern European countries, had 34% of the population at risk of poverty or social exclusion, which is the highest in the European Union (EU) and approximately 70% higher than in Western Europe.3 According to the 2021 census, the population of Romania consists of 9.3% minorities, of which 6.6% is the Hungarian minority and 2.46% is the Roma minority.4 While data on the Roma minority in Romania and across the EU is scarce, the Fundamental Rights Report 2019 revealed an increase in segregation and persistence of anti-gypsyism in the EU.5 Similarly, data related to health disparities in such minorities is lacking. The county where our tertiary hospital is located had the highest percentage of Roma minority in Romania as reported by the 2021 census.4 Moreover, the affiliated counties for the acute coronary syndrome (ACS) treating network that our tertiary hospital is responsible for have the highest percentage of Hungarian minorities in Romania.4 Thus, a minority-focused individual patient report would be of interest, given our location in the most ethnically heterogenous region of Romania.

The aim of this study was to report long-term survival outcomes in an all-comers prospective percutaneous coronary intervention (PCI) population in Romania. Differences among ethnic minorities, in terms of comorbidities, clinical presentation, coronary artery disease (CAD) complexity, PCI procedural characteristics and survival rates were also investigated.

## Materials and methods

### Patient inclusion

The present all-comers prospective PCI registry was previously described.6 Briefly, all patients older than 18 years and treated by PCI in the Emergency Institute for Cardiovascular Diseases and Transplantation of Targu Mures have been included prospectively at hospital discharge in the local PCI registry of the institute since January 2016. The registry is accessible online at the website [http://pci.cardio.ro/](http://pci.cardio.ro/) and is based on the criteria of Cardiology Audit and Registration Data Standards developed by the Department of Health and Children, European Society of Cardiology, Irish Cardiac Society and the European Commission.7 All the information available regarding all the variables proposed in that document was collected for each included patient, at every PCI. All patients available in the registry as of December 2023 were included in the present analysis.

### Research ethics

All patients (or their legal representatives) signed the informed consent regarding the PCI procedure and their participation in the study. The protocol was carried out in accordance with the ethical principles for medical research involving human subjects established by the Declaration of Helsinki, protecting the confidentiality of personal information of the patients.

### Follow-up and clinical outcomes

The clinical endpoint of this study was the incidence of cardiovascular cause and all-cause mortality. In-hospital mortality data was available from the PCI registry. Romanian National Health Insurance Service (NHIS) database supplied mortality rates as of 31 December 2023 for all the patients. For patients who had died during follow-up, Regional Statistics Office of the Romanian National Institute of Statistics supplied the exact date and cause of death according to the 10th revision of the International Classification of Diseases. If the cause of death belonged to diseases of the circulatory system, then death was considered to be of cardiovascular cause.

### Ethnicity background

Included patients were categorised according to the three most common and important ethnic groups that comprise over 97% of Romania’s population—Romanian, Hungarian and Roma (colloquially referred to as Gypsy) ethnicities.4 Categorisation was performed by an algorithmic analysis of the origin of the last name, since ethnic background can be particularly saliently derived from anthroponyms in Romania.8 

### Statistical analysis

A significance level α of 0.05 and a 95% CI were considered. All tests were two-sided. Continuous variables were evaluated for normal distribution using the Shapiro-Wilk test. Continuous variables with parametric distributions were reported as mean±SD and compared using non-paired Student’s t-test, while continuous variables with non-parametric distributions and discrete variables were reported as median (IQR) and compared using Mann-Whitney U test. Categorical variables were reported as absolute and relative frequencies and were compared using the χ2 or Fisher’s exact test as appropriate. The cumulative hazard of all-cause and cardiovascular cause mortality was assessed using the Kaplan-Meier method and compared using the log-rank test. To examine the impact of different predictors on survival, univariable and multivariable Cox proportional hazard models were used to predict the association in the form of HR between observed survival and single or multiple independent variables, respectively. The number of predictors in the multivariable models was chosen so that there would be at least 15 events per covariate.9 Multicollinearity among independent variables in multivariable models was assessed and a strong correlation between variables was considered present if the variance inflation factor was above 2.5. Multivariable models were constructed in a stepwise fashion and overfitting was evaluated using Akaike’s information criterion (AIC). AIC was calculated at each step to assess if an overfitting problem was incurred by adding more variables to the model. Given that the final prediction model and model performance were assessed using the same data set, the model performance statistics might be optimistic. To correct this optimism, all 95% CI were bootstrapped with 1000 samples and then reported in order to ensure internal validation. Statistical analysis was performed using R V.4.1.1 (Foundation for Statistical Computing, Vienna, Austria, [https://www.R-project.org/](https://www.R-project.org/)) and RStudio V.1.4.17 (PBC, Boston, Massachusetts, USA, [http://www.rstudio.com/](http://www.rstudio.com/)). The geographical origin of the treated patients was plotted using GeoPandas and matplotlib libraries implemented in Python V.3.9.13 (Python Software Foundation, [https://www.python.org/](https://www.python.org/)).

## Results

A total of 6867 patients with 8442 PCI procedures were included in the present study. Romanian group consisted of 5095 (74.2%) patients with 6238 (73.9%) PCI procedures, the Hungarian group consisted of 1417 (20.6%) patients with 1780 (21.1%) PCI procedures and the Roma group consisted of 355 (5.1%) patients with 424 (5.0%) PCI procedures. Clinical characteristics of the included population in a per-patient and per-PCI analysis are reported in table 1. The geographical distribution of the included patients is illustrated in figure 1.

![Figure 1](http://jech.bmj.com/https://jech.bmj.com/content/jech/79/4/272/F1.medium.gif)

[Figure 1](http://jech.bmj.com/content/79/4/272/F1)

Figure 1 
Geographical origin of treated patients in Romania.

View this table:
[Table 1](http://jech.bmj.com/content/79/4/272/T1)

Table 1 
Clinical characteristics of the studied population

During a median follow-up of 3.60 (1.35–5.75) years, a total of 1064 cardiovascular-cause and 1374 all-cause events occurred. All-cause deaths occurred in 5.40 cases per 100 patient-years, while cardiovascular-cause deaths occurred in 4.19 cases per 100 patient-years. In patients with ACS, all-cause and cardiovascular-cause deaths occurred in 6.56 cases and 5.26 cases per 100 patient-years, respectively. In patients with chronic coronary syndrome (CCS), all-cause and cardiovascular-cause deaths occurred in 4.01 cases and 2.92 cases per 100 patient-years, respectively. Cumulative hazard curves for cardiovascular and all-cause mortality are reported in online supplemental figure 1. In-hospital mortality occurred in 191 (3.75%) patients for the Romanian group, in 77 (5.43%) patients for the Hungarian group and in 20 (5.63%) patients for the Roma group (p=0.007). Cumulative hazard curves for cardiovascular-cause and all-cause mortality in a per-patient analysis are reported in figure 2 and in a per-PCI analysis are reported in online supplemental figure 2. Cumulative hazard curves for acute and chronic coronary syndromes are reported in online supplemental figure 3-5. There was a significant increase of in-hospital, 1-year and 2 years mortality in 2020 and 2021, consistent with the recent coronavirus pandemic (online supplemental figure 6), however hospitalisations for CCS decreased during that time frame and a relative increase in hospitalisations for ACS was observed (online supplemental figure 7). 18 (0.02%) patients had an active coronavirus infection at the time of PCI, with no differences among ethnic groups. Yearly mortality trends are illustrated in online supplemental figure 7.

### Supplementary data

[[jech-2024-222845supp001.pdf]](pending:yes) 

![Figure 2](http://jech.bmj.com/https://jech.bmj.com/content/jech/79/4/272/F2.medium.gif)

[Figure 2](http://jech.bmj.com/content/79/4/272/F2)

Figure 2 
Cumulative hazard and numbers at risk table for all-cause and cardiovascular cause mortality among investigated groups.

Roma patients were significantly younger, had higher rates of risk factors and comorbidities and presented more often with ACS and with lower left ventricular ejection fraction (LVEF) when compared with Romanian patients (online supplemental table 1 and online supplemental table 2). While the complexity of the CAD was similar among the treated groups, the PCI procedure was less complex (lower complete revascularisation rates, fewer segments treated and fewer implanted stents) when compared with Romanian patients (online supplemental table 1 and online supplemental table 2). However, thrombolysis in myocardial infarction (TIMI) flow post-PCI did not differ among treated groups. Hungarian patients were less frequently males, and had similar rates of risk factors and comorbidities, but presented more often with ACS when compared with Romanian patients (online supplemental table 1 and online supplemental table 2). Both the complexity of the CAD and that of the PCI procedure were similar in Hungarian patients when compared with Romanian patients (online supplemental table 1 and online supplemental table 2).

Prescription of guidelines recommended treatment was present in 1905 (27.74%) of patients for ticagrelor as a potent P2Y12 receptor inhibitor, in 376 (19.24%) and 229 (11.72%) of patients with LVEF≤40% for angiotensin receptor-neprilysin inhibitor (ARNI) and sodium-glucose cotransporter-2 inhibitor (SGLT2) inhibitors, respectively, and in 432 (51.67%) of patients with an anticoagulation indication (836 (12.1%)) for direct oral anticoagulant (DOAC), and there was a significant increasing trend in their prescription (online supplemental figure 8). All 836 (12.1%) patients with an anticoagulation indication also received aspirin and clopidogrel as part of the triple antitrombotic therapy. The increase in prescription trends was consistent with nationwide price-volume agreements and NHIS coverage. There was no difference in the prescription of ARNI, SGLT2 inhibitors or DOAC among the treated groups. However, the 32.2% prescription rate of potent P2Y12 receptor inhibitor in Roma patients was higher than the 27.7% prescription rate in the rest of the population (p=0.04; online supplemental figure 8). 30 (9.35%) Roma patients were uninsured, which is significantly higher than Hungarian patients (12 (0.86%)) and Romanian patients (24 (0.47%); p<0.0001). High prescription rates were observed for other treatment classes, such as aspirin (in 6509 (94.8%) patients), beta-blockers (in 6214 (90.5%) patients), ACE inhibitors or angiotensin receptor blockers (in 6269 (91.3%) patients) and statin (in 6674 (97.2%) patients), with no heterogeneity among ethnic groups.

PCI via radial vascular access was performed in 5452 (79.39%) patients and a vascular closure device was used in 439 (6.39%) patients. Drug-eluting and biodegradable stents were implanted in 6373 (92.8%), and 409 (5.96%) patients, respectively. 85 (1.2%) patients received balloon-only angioplasty. Regarding bleeding risk, 15 (0.002%) patients suffered major clinical overt bleeding, 66 (0.1%) patients had a haemoglobin drop ≥50 g/L after PCI and 189 (2.0%) patients had a haemoglobin drop ≥15% after PCI. No periprocedural stroke occurred. No in-hospital death was attributed to bleeding events. Extensive patient characteristics and procedural details are reported in online supplemental table 3.

When adjusting for potential confounders in a Cox multivariable regression, the presence of guidelines recommended treatment (potent P2Y12 receptor inhibitor, ARNI, SGLT2 inhibitor and DOAC), higher pre-PCI TIMI flow, higher post-PCI TIMI flow or higher LVEF were protective against cardiovascular cause and all-cause mortality (figure 3). Advanced age, higher creatinine, higher Killip class, presence of proximal segment culprit or number of diseased coronary arteries were among risk factors for cardiovascular cause and all-cause mortality (figure 3). Both Hungarian and Roma patients presented significantly higher all-cause mortality than Romanian patients (adjusted HR=1.20 (1.05–1.36), p=0.005 and adjusted HR=1.51 (1.21–1.88), p=0.0001, respectively). Similarly, Hungarian and Roma patients presented significantly higher cardiovascular cause mortality than Romanian patients (adjusted HR=1.22 (1.05–1.41), p=0.006 and adjusted HR=1.51 (1.18–1.92), p=0.0008, respectively). Ethnicity was an independent predictor of events, including in subgroups such as ACS, CCS, LVEF or atrial fibrillation (figure 4). As an exception, Hungarian ethnicity in the ACS subgroup did not have a higher adjusted HR when compared with Romanian ethnicity (figure 4).

![Figure 3](http://jech.bmj.com/https://jech.bmj.com/content/jech/79/4/272/F3.medium.gif)

[Figure 3](http://jech.bmj.com/content/79/4/272/F3)

Figure 3 
Multivariable Cox proportional hazards for mortality prediction. All variables were one-hot encoded, including DOAC, SGLT2I, ARNI and ticagrelor. ACS, acute coronary syndrome; ARNI, angiotensin receptor-neprilysin inhibitor; COPD, chronic obstructive pulmonary disease; DOAC, direct oral anticoagulant; LVEF, left ventricular ejection fraction; SGLT2I, sodium-glucose cotransporter-2 inhibitor; TIMI, thrombolysis in myocardial infarction flow.

![Figure 4](http://jech.bmj.com/https://jech.bmj.com/content/jech/79/4/272/F4.medium.gif)

[Figure 4](http://jech.bmj.com/content/79/4/272/F4)

Figure 4 
Adjusted HR among entire and subgroup populations. *Adjusted for the following variables: age, sex, atrial fibrillation, chronic obstructive pulmonary disease, diabetes mellitus, dilated cardiomyopathy, number of diseased vessels, proximal segment angioplasty, Killip class, thrombolysis in myocardial infarction flow pre-PCI and post-PCI, creatinine, left ventricular ejection fraction (LVEF), acute coronary syndrome (ACS), ticagrelor, angiotensin receptor-neprilysin inhibitor, sodium-glucose cotransporter 2 inhibitors and direct oral anticoagulant administration. CCS, chronic coronary syndrome; PCI, percutaneous coronary intervention.

## Discussions

Our study findings could be summarised as follows: (1) High long-term cardiovascular and all-cause mortality was observed for the entire included population. Long-term survival was significantly lower in ethnic minorities, such as the Hungarian and Roma minorities. (2) Roma patients were significantly younger, had higher rates of risk factors and comorbidities and presented more often with ACS. While the complexity of the CAD was similar, the PCI procedure was less complex, although TIMI flow post-PCI did not differ in Roma patients when compared with Romanian patients. (3) Hungarian patients were less frequently males, had similar rates of risk factors and comorbidities, but presented more often with ACS. Both the complexity of the CAD and that of the PCI procedure were similar in Hungarian patients when compared with Romanian patients.

High mortality was observed in our study when compared with contemporary PCI cohorts. Overall, the 7-year all-cause mortality in the CCS subgroup in our study was 32%, which is higher than the 7-year all-cause mortality of 12.4% in the PCI-treated arm of the ISCHEMIA trial.10 Similarly, the 7-year cardiovascular-cause mortality in the CCS subgroup in our study was 23%, which is higher than the 7-year cardiovascular-cause mortality of 6.4% in the ISCHEMIA trial.10 The incidence rate of all-cause mortality in patients with ACS in our study was 6.56 cases per 100 patient-years, more than twice that of 2.4 cases per 100 patient-years reported in a large PCI-treated ACS Western population.11 Higher Killip class, hospitalisation for ACS, lower pre-PCI TIMI flow, diabetes mellitus and advanced age were among the independent predictors of mortality (figure 3). Our findings are in line with previous studies that revealed higher mortality when cardiogenic shock is associated with diabetes mellitus or low pre-PCI TIMI flow.12 Moreover, bioresorbable stents were associated with improved survival (online supplemental table 3), similar to other reports.13 Regarding periprocedural safety, no stroke (ischaemic or haemorrhagic) occurred in our study. A previous large PCI registry reported a periprocedural stroke rate of approximately 0.02%.14 

Significantly higher mortality rates were observed in our study for Roma and Hungarian minorities. A relatively small study from Central Europe, in which 75% of patients had angiographically significant CAD, reported higher mortality among the Roma minority when compared with the general population.15 That difference was significant even after adjusting for socioeconomic factors. However, the fact that belonging to an ethnic minority is an independent risk factor for higher all-cause and cardiovascular cause mortality needs attention. Possible explanations could be genetic, socioeconomic and even cultural differences in the studied populations.

Our study confirms a higher prevalence of cardiovascular risk factors among Roma patients. Even though the complexity of CAD as reflected by the SYNTAX score and the number of diseased coronary arteries was similar among ethnic groups, Roma patients were younger, had more risk factors and had more comorbidities than their counterparts. A higher risk for a 10-year development of cardiovascular diseases as quantified by Framingham score and a reduction in specific high-density lipoprotein cholesterol subfraction in Roma patients than the general population were also previously reported.16 17 Interestingly, genetic polymorphisms of apolipoprotein E, such as the ε2 allele, are associated with obesity phenotype in Roma patients.18 Moreover, certain cardiovascular risk factors, such as smoking which is a strong part of Roma culture, can reach extremely high rates. While in our study 60% of Roma patients were smokers, other studies report smoking rates of up to 72%.19 Certain attempts towards smoking cessation in Roma patients have ultimately proven to be ineffective.20 While there is a paucity of reliable data regarding the Roma population, it is clear that comprehensive lifestyle modification interventions along with maximising guidelines recommended treatment should be pursued.

Although prescription trends were increasing, average prescription numbers for guidelines-directed pharmacological treatment were relatively low, especially for drug classes whose benefit was established in randomised trials before our PCI registry such as ticagrelor, ARNI or DOAC. Ticagrelor prescription rate was 19% in our study, much lower than that of a recent registry where ticagrelor usage was 63%.21 While other registries reported that ticagrelor was not superior to clopidogrel in reducing adverse events,22 in our population ticagrelor was among the most important clinical parameters that improved patient survival. Low ARNI prescription rates, in 19% of patients with LVEF≤40% in our study, were also observed in previous studies, where despite the strong evidence, prescription rates only rose to 26%.23 24 Similarly, an SGLT2 inhibitor was used in 11% of patients with LVEF≤40% which is less than 37% of SGLT2 inhibitor use in other Western registries.25 DOAC usage instead of vitamin K antagonists for patients with an anticoagulation indication was 51% in our study, while in other PCI registries rose up to 87%.26 Prescription rate was also observed to increase following the introduction of price-volume agreements in other studies. For instance, affordability and cost-sharing are major factors in increasing prescription rates of quadruple therapy for heart failure.27 Not surprisingly, lower affordability of cardiovascular drugs was associated with a higher rate of adverse events and mortality.28 Interestingly, in our study Roma patients had a higher prescription rate of ticagrelor, which may be due to the fact that clopidogrel price is not fully covered in Romania, while ticagrelor price is fully covered and, thus, cheaper. Moreover, prasugrel was not prescribed since it lacks a national cost-volume agreement and is commercially unavailable, emphasising the relationship between affordability and prescription.

An increased mortality was observed during the recent coronavirus pandemic, although the pandemic did not have a significant impact on mortality per se in multivariable analysis, it was due to a decrease in CCS hospitalisations and a relative increase in ACS hospitalisations. A recent registry-based meta-analysis reported higher mortality in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-positive patients.29 However, in the present study, a small number of SARS-CoV-2-positive patients were included, which may explain why hospitalisation during the coronavirus pandemic did not impact mortality in the multivariate analysis.

### Study limitations

Although our tertiary centre is located in the most ethnically heterogeneous region of Romania and the percentages of minorities in the present study were higher than nationwide percentages, the included patients in each minority were still relatively few (20% for Hungarian patients and 5% for Roma patients). Although ethnical background can be easily found by anthroponomic analysis in Romania,8 a self-affirmed and phenotypic-derived ethnical background or genetic ancestry could have led to more precise results and should constitute further research directions.

## Conclusions

High long-term cardiovascular and all-cause mortality was observed for the entire included population and mortality was higher among ethnic minorities than Romanian population. Roma minority had a significantly higher incidence of associated comorbidities, and it had significantly lower survival than the Romanian population, even after adjusting for potential confounders. Moreover, the Hungarian minority had significantly lower survival, even though it shared the same clinical characteristics as the Romanian population.

## Data availability statement

Data are available upon reasonable request. The data underlying this article will be shared on reasonable request to the corresponding author.

## Ethics statements

### Patient consent for publication

Consent obtained directly from patient(s).

### Ethics approval

The study was approved by the ethical committee of our institution (decision number 8646 / 22 December 2015). Participants gave informed consent to participate in the study before taking part.

## Acknowledgments

The authors would like to thank all the resident colleagues who continuously introduced the data regarding percutaneous coronary interventions performed in the Emergency Institute for Cardiovascular Diseases and Transplantation since 2016. We also would like to thank Mr Ioan Matei for his help in achieving long-term mortality data.

## Footnotes

*   Contributors LH, PC and DD supervised and coordinated the data collection in the registry. LH, PC, PG, I-AN, ACS, KB, HS, IPS and MH were involved in investigation and data curation. PC and LH performed the formal analysis and visualisation. PC and LH were involved in the writing of the original draft. PC, LH and DD were involved in the revision of the final revised draft. All authors approved the manuscript. PC is the guarantor.

*   Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

*   Map disclaimer The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.

*   Competing interests None declared.

*   Provenance and peer review Not commissioned; externally peer reviewed.

*   Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

[http://creativecommons.org/licenses/by-nc/4.0/](http://creativecommons.org/licenses/by-nc/4.0/)
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: [http://creativecommons.org/licenses/by-nc/4.0/](http://creativecommons.org/licenses/by-nc/4.0/).

## References

1.   Luengo-Fernandez R , Little M , Gray A , et al . Cardiovascular disease burden due to productivity losses in European Society of Cardiology countries. Eur Heart J Qual Care Clin Outcomes 2024;10:36–44. [doi:10.1093/ehjqcco/qcad031](http://dx.doi.org/10.1093/ehjqcco/qcad031) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1093/ehjqcco/qcad031&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=37280170&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

2.   Luengo-Fernandez R , Walli-Attaei M , Gray A , et al . Economic burden of cardiovascular diseases in the European Union: a population-based cost study. Eur Heart J 2023;44:4752–67. [doi:10.1093/eurheartj/ehad583](http://dx.doi.org/10.1093/eurheartj/ehad583) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1093/eurheartj/ehad583&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=37632363&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

3.  Eurostat. Persons at risk of poverty or social exclusion by age and sex. 2024 Available: [https://doi.org/10.2908/ILC\_PEPS01N](https://doi.org/10.2908/ILC_PEPS01N) 
    
    

4.  Rezultate definitive: Caracteristici etno-culturale demografice – Recensamantul Populatiei si Locuintelor. 2021. Available: [https://www.recensamantromania.ro/rezultate-rpl-2021/rezultate-definitive-caracteristici-etno-culturale-demografice](https://www.recensamantromania.ro/rezultate-rpl-2021/rezultate-definitive-caracteristici-etno-culturale-demografice) 
    
    

5.  European Union Agency for Fundamental Rights. The plight of roma sees little change. European Union Agency for Fundamental Rights; 2019. Available: [http://fra.europa.eu/en/publications-and-resources/infographics/plight-roma-sees-little-change](http://fra.europa.eu/en/publications-and-resources/infographics/plight-roma-sees-little-change) 
    
    

6.   Călburean PA , Grebenișan P , Nistor I-A , et al . Prediction of 3-year all-cause and cardiovascular cause mortality in a prospective percutaneous coronary intervention registry: Machine learning model outperforms conventional clinical risk scores. Atherosclerosis 2022;350:33–40. [doi:10.1016/j.atherosclerosis.2022.03.028](http://dx.doi.org/10.1016/j.atherosclerosis.2022.03.028) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1016/j.atherosclerosis.2022.03.028&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=35483116&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

7.   Flynn MR , Barrett C , Cosío FG , et al . The Cardiology Audit and Registration Data Standards (CARDS), European data standards for clinical cardiology practice. Eur Heart J 2005;26:308–13. [doi:10.1093/eurheartj/ehi079](http://dx.doi.org/10.1093/eurheartj/ehi079) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1093/eurheartj/ehi079&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=15618029&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 
    
    [Web of Science](http://jech.bmj.com/lookup/external-ref?access_num=000228892200016&link_type=ISI) 

8.  1.  
    2.   Felecan O , 
    3.   Bugheșiu A  
    
     Felecan O . Transylvania—an anthroponymic perspective. In:  Felecan O ,  Bugheșiu A  , eds. Names and naming: multicultural aspects. Cham: Springer International Publishing, 2021: 143–63.
    
    

9.   Harrell FE . Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis. Cham: Springer International Publishing, 2015.
    
    

10.  Hochman JS , Anthopolos R , Reynolds HR , et al . Survival After Invasive or Conservative Management of Stable Coronary Disease. Circulation 2023;147:8–19. [doi:10.1161/CIRCULATIONAHA.122.062714](http://dx.doi.org/10.1161/CIRCULATIONAHA.122.062714) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1161/CIRCULATIONAHA.122.062714&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=36335918&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

11.  Marquis-Gravel G , Dalgaard F , Jones AD , et al . Post-Discharge Bleeding and Mortality Following Acute Coronary Syndromes With or Without PCI. J Am Coll Cardiol 2020;76:162–71. [doi:10.1016/j.jacc.2020.05.031](http://dx.doi.org/10.1016/j.jacc.2020.05.031) 
    
    [FREE Full Text](http://jech.bmj.com/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6MzoiUERGIjtzOjExOiJqb3VybmFsQ29kZSI7czo0OiJhY2NqIjtzOjU6InJlc2lkIjtzOjg6Ijc2LzIvMTYyIjtzOjQ6ImF0b20iO3M6MTk6Ii9qZWNoLzc5LzQvMjcyLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

12.  Tokarek T , Dziewierz A , Plens K , et al . Radial approach reduces mortality in patients with ST-segment elevation myocardial infarction and cardiogenic shock. Pol Arch Intern Med 2021;131:421–8. [doi:10.20452/pamw.15886](http://dx.doi.org/10.20452/pamw.15886) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=33739779&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

13.  Rzeszutko Ł , Siudak Z , Tokarek T , et al . Twelve months clinical outcome after bioresorbable vascular scaffold implantation in patients with stable angina and acute coronary syndrome. Data from the Polish National Registry. Postepy Kardiol Interwencyjnej 2016;12:108–15. [doi:10.5114/aic.2016.59360](http://dx.doi.org/10.5114/aic.2016.59360) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=27279869&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

14.  Dziewierz A , Siudak Z , Tokarek T , et al . Determinants of stroke following percutaneous coronary intervention in acute myocardial infarction (from ORPKI Polish National Registry). Int J Cardiol 2016;223:236–8. [doi:10.1016/j.ijcard.2016.08.214](http://dx.doi.org/10.1016/j.ijcard.2016.08.214) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1016/j.ijcard.2016.08.214&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=27544594&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

15.  Sudzinova A , Nagyova I , Rosenberger J , et al . Seven years’ mortality in Roma and non-Roma patients after coronary angiography. Eur J Public Health 2015;25:765–9. [doi:10.1093/eurpub/ckv057](http://dx.doi.org/10.1093/eurpub/ckv057) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1093/eurpub/ckv057&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=25829503&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

16.  Piko P , Kosa Z , Sandor J , et al . Comparative risk assessment for the development of cardiovascular diseases in the Hungarian general and Roma population. Sci Rep 2021;11:3085. [doi:10.1038/s41598-021-82689-0](http://dx.doi.org/10.1038/s41598-021-82689-0) 
    
    

17.  Piko P , Kosa Z , Sandor J , et al . The profile of HDL-C subfractions and their association with cardiovascular risk in the Hungarian general and Roma populations. Sci Rep 2022;12:10915. [doi:10.1038/s41598-022-15192-9](http://dx.doi.org/10.1038/s41598-022-15192-9) 
    
    

18.  Zeljko HM , Škarić-Jurić T , Narančić NS , et al . E2 allele of the apolipoprotein E gene polymorphism is predictive for obesity status in Roma minority population of Croatia. Lipids Health Dis 2011;10:9. [doi:10.1186/1476-511X-10-9](http://dx.doi.org/10.1186/1476-511X-10-9) 
    
    

19.  Skarić-Jurić T , Klarić IM , Narancić NS , et al . Trapped between tradition and transition--anthropological and epidemiological cross-sectional study of Bayash Roma in Croatia. Croat Med J 2007;48:708–19.
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=17948957&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 
    
    [Web of Science](http://jech.bmj.com/lookup/external-ref?access_num=000251130000013&link_type=ISI) 

20.  Petek D , Rotar Pavlic D , Svab I , et al . Attitudes of Roma toward smoking: qualitative study in Slovenia. Croat Med J 2006;47:344–7.
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=16625703&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 
    
    [Web of Science](http://jech.bmj.com/lookup/external-ref?access_num=000237317700019&link_type=ISI) 

21.  Wiens EJ , Leon SJ , Whitlock R , et al . Ticagrelor as Compared to Clopidogrel Following Percutaneous Coronary Intervention for Acute Coronary Syndrome. Am J Cardiol 2023;198:26–32. [doi:10.1016/j.amjcard.2023.03.012](http://dx.doi.org/10.1016/j.amjcard.2023.03.012) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1016/j.amjcard.2023.03.012&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=37196530&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

22.  Völz S , Petursson P , Odenstedt J , et al . Ticagrelor is Not Superior to Clopidogrel in Patients With Acute Coronary Syndromes Undergoing PCI: A Report from Swedish Coronary Angiography and Angioplasty Registry. J Am Heart Assoc 2020;9:e015990. [doi:10.1161/JAHA.119.015990](http://dx.doi.org/10.1161/JAHA.119.015990) 
    
    

23.  Tran JS , Loveland MG , Alamer A , et al . Clinical and Socioeconomic Determinants of Angiotensin Receptor-Neprilysin Inhibitor Prescription at Hospital Discharge in Patients With Heart Failure With Reduced Ejection Fraction. Circ Heart Fail 2022;15:e009395. [doi:10.1161/CIRCHEARTFAILURE.121.009395](http://dx.doi.org/10.1161/CIRCHEARTFAILURE.121.009395) 
    
    

24.  Mohanty AF , Levitan EB , King JB , et al . Sacubitril/Valsartan Initiation Among Veterans Who Are Renin-Angiotensin-Aldosterone System Inhibitor Naïve With Heart Failure and Reduced Ejection Fraction. J Am Heart Assoc 2021;10:e020474. [doi:10.1161/JAHA.120.020474](http://dx.doi.org/10.1161/JAHA.120.020474) 
    
    

25.  Stolfo D , Lund LH , Benson L , et al . Real-world use of sodium-glucose cotransporter 2 inhibitors in patients with heart failure and reduced ejection fraction: Data from the Swedish Heart Failure Registry. Eur J Heart Fail 2023;25:1648–58. [doi:10.1002/ejhf.2971](http://dx.doi.org/10.1002/ejhf.2971) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1002/ejhf.2971&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=37419495&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

26.  Uziębło-Życzkowska B , Krzesiński P , Maciorowska M , et al . Antithrombotic therapy in patients with atrial fibrillation undergoing percutaneous coronary intervention, including compliance with current guidelines-data from the POLish Atrial Fibrillation (POL-AF) Registry. Cardiovasc Diagn Ther 2021;11:14–27. [doi:10.21037/cdt-20-839](http://dx.doi.org/10.21037/cdt-20-839) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.21037/cdt-20-839&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=33708474&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

27.  Faridi KF , Dayoub EJ , Ross JS , et al . Medicare Coverage and Out-of-Pocket Costs of Quadruple Drug Therapy for Heart Failure. J Am Coll Cardiol 2022;79:2516–25. [doi:10.1016/j.jacc.2022.04.031](http://dx.doi.org/10.1016/j.jacc.2022.04.031) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1016/j.jacc.2022.04.031&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=35738713&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom) 

28.  Chow CK , Nguyen TN , Marschner S , et al . Availability and affordability of medicines and cardiovascular outcomes in 21 high-income, middle-income and low-income countries. BMJ Glob Health 2020;5:e002640. [doi:10.1136/bmjgh-2020-002640](http://dx.doi.org/10.1136/bmjgh-2020-002640) 
    
    

29.  De Luca G , Silverio A , Verdoia M , et al . Angiographic and clinical outcome of SARS-CoV-2 positive patients with ST-segment elevation myocardial infarction undergoing primary angioplasty: A collaborative, individual patient data meta-analysis of six registry-based studies. Eur J Intern Med 2022;105:69–76. [doi:10.1016/j.ejim.2022.08.021](http://dx.doi.org/10.1016/j.ejim.2022.08.021) 
    
    [CrossRef](http://jech.bmj.com/lookup/external-ref?access_num=10.1016/j.ejim.2022.08.021&link_type=DOI) 
    
    [PubMed](http://jech.bmj.com/lookup/external-ref?access_num=35999094&link_type=MED&atom=%2Fjech%2F79%2F4%2F272.atom)