Prevalence And Risk Factors Of Cardiovascular Diseases Worldwide

Global Impact of Cardiovascular Diseases

According to the World Health Organization (WHO), Cardiovascular Diseases are the leading cause of deaths worldwide resulting in the deaths of 17.9 million individuals annually (Alonso et al., 2021). Of the statistic 38% of the deaths were premature for individuals aged below 70 years. These heart diseases constitute a range of disorders that affect both the heart and blood disorders such as coronary heart diseases, rheumatic heart conditions among other (Benjamin et al., 2017). Approximately four out of five heart disease deaths are a result of health attacks and strokes. Notably, the cardiovascular diseases stem from behavioral factors such as unhealthy dietary habits, harmful consumption of alcohol, use of tobacco, and physical inactivity (Brown, & O’Connor, 2010; Ulbinait?, & Radzevi?ius, 2021). Luckily these aspects can be registered medically through increased blood pressure levels, high blood glucose, raised blood lipids, obesity and overweight issues.

Moreover, certain lifestyle changes may mitigate the risk of cardiovascular diseases. For instance, reduced salt intake, regular exercise, consumption of healthier foods such as fruits and vegetables, discontinuing the use of alcohol and tobacco (Merz, 2020). Some of the symptoms of heart attacks are pain in the center of one’s chest or pain in the elbows, jaw, the left shoulder, arms or back. Also, shortness of breath, lightheadedness, nausea and a pale appearance can be a symptom of a heart attack. For stroke the symptoms are similar to those of a heart attack with the inclusion of severe headache, fainting, numbness, confusion, difficult speech and visibility among others. As no symptom of an inherent disease that can be detected by examining blood vessels, it is vital that one seeks medical attention as soon as they observe the stated symptoms. Nonetheless, various researchers have generated predictive models for heart disease to ensure early diagnosis and treatment mitigating premature deaths (Faizal et al., 2021; Kalyanasundaram, Prasanth, Tamizhselvan, & Kumaran, 2017).

In the United States, heart diseases are equally the primary cause of fatalities even though the number has declined from its initial count in the 50s (CDC, 2011). In 2018, the country recorded 164 deaths for every 100000 people from heart disease. With coronary heart disease accounting for 42% of the overall CVD fatalities (Benjamin et al., 2019; Tsao et al., 2022). Notably, a 44% decline in heart fatalities was observed between 1980 and 2000 after people made lifestyle changes (Lee et al., 2022). In Switzerland, coronary heart disease is at the forefront with the highest number of fatalities (Zellweger, Junker, & Bopp, 2019). In fact, the death rate is 47.47% per 100000 people. According to a study done in Switzerland by Faeh, Gutzwiller, and Bopp (2009), individuals residing in high altitudes have low mortality rates caused by coronary heart diseases and stroke. The study findings imply that geographical location of an individual influences the mortality rates due to heart diseases.

On factors influencing prevalence of heart disease, gender was one of them. From literature gender influences early adoption of various health behaviors in children, adolescents and young adults towards drinking, smoking and involvement in physical activities (O’Neil et al., 2018). Gender also plays a role in management of psychological stress with variations observed across the group in controlling hypertension and diabetes mellitus (Peters, Muntner, & Woodward, 2019). Other studies have also indicated the role gender plays on heart disease prevalence (Sharma, Volgman, & Michos, 2020; Peters et al., 2018). Age difference and lipid retention were equally factors that influences CVD diseases (Richardson et al., 2020: Radford et al., 2018).

Behavioral Factors that Contribute to Heart Disease

Consequently, this research sought to compare the findings on the presence of heart disease for patients in Cleveland and those in Switzerland. To achieve the research objective both descriptive and inferential statistics were computed on Minitab. Ideally, the research is going to test one’s proficiency of using Statistical techniques and tools to analyze the relationship between the two datasets.

The Dataset for heart disease was collected from the machine learning database. Heart disease data for both Switzerland and Cleveland were retrieved. From the datasets, we amine the variables age, gender where 1 denoted male and 0 denoted female, chest pain type (cp) which had values 1 to 4 each referencing the following: Value 1: typical angina, Value 2: atypical angina, Value 3: non-anginal pain, and Value 4: asymptomatic. Other variables examined were resting blood pressure in mm Hg on admission to the hospital (trestbps), serum cholesterol in mg/dl (chol), and whether the fasting blood sugar > 120 mg/dl(fbs) with the responses recorded as 1 = true and 0 = false. The other variables used alongside their code values are relayed below:

restecg: resting electrocardiographic results

Value 0: normal

Value 1: having ST-T wave abnormality (T wave inversions and/or ST

elevation or depression of > 0.05 mV)

by Estes’ criteria

1. thalach: maximum heart rate achieved
2. exang: exercise induced angina (1 = yes; 0 = no)
3. oldpeak = ST depression induced by exercise relative to rest
4. slope: the slope of the peak exercise ST segment

Value 1: upsloping

Value 2: flat

Value 3: downsloping

1. 6. ca: number of major vessels (0-3) colored by flourosopy
2. thal: 3 = normal; 6 = fixed defect; 7 = reversable defect – the heart rate(thal)
3. num: diagnosis of heart disease (angiographic disease status)

Value 0: < 50% diameter narrowing

Value 1: > 50% diameter narrowing

Notably of the 76 attribute values collected, only the stated 14 variables will be examined across both countries.

Upon retrieving the data from the ML database, the data was converted to excel format. For the columns with missing values and those that where outcomes were denoted by a question mark, the value was replaced by 0. The data was then entered into the Minitab software for analysis.

The basic descriptive tool was used to obtain the descriptive statistics for the study variables. The frequency distribution tables were also derived

Hypothesis testing was then performed to determine if the number of individuals diagnosed with heart disease was equal in proportion for the two countries. A Chi- square goodness of fit analysis was also performed on the num variable for the two countries. This is done under the stats option, then tables and select chi square goodness of fit option. z A regression analysis was performed to determine how the independent variables (all other variables) influenced diagnosis of heart disease(num). Under the stats option on Minitab, I selected the regression option then proceeded to select fitted line plot, and fitted regression model respectively.

From the descriptive analysis performed on both countries, Table 1 and 2 present a summary of dispersion. The datasets had different sample sizes one was 303(Cleveland) the other was 123(Switzerland). Notably in Table 1 the minimum and maximum cholesterol values were 0. In addition, the minimum age of respondents was 29 and 32 while the maximum age was 77 and 74 for Cleveland and Switzerland respectively. Based on gender majority of the respondents in Switzerland were male (91%, n = 112) while in Cleveland majority of the respondents (68%, n= 206) were equally male. Summary of these findings are relayed in Figure 1 and 2 below.

Lifestyle Changes that Can Reduce Risk

Figure 1. Distribution by Gender

Figure 2. Distribution by Gender

From the frequency distribution analysis done on the num variable from both countries the outputs are relayed in the tables below. In Cleveland majority of the respondents 54% had less than 50 percent diameter narrowing while in Switzerland, majority of the participants recorded a value of 1 implying the have more than 50 percent diameter narrowing.

Table 1. Descriptive statistics of heart disease variables in Switzerland

Variable	N	N*	Mean	SE Mean	StDev	Variance	CoefVar	Minimum	Q1
Age	123	0	55.317	0.814	9.032	81.579	16.33	32.000	51.000
Sex	123	0	0.9106	0.0258	0.2865	0.0821	31.47	0.0000	1.0000
cp	123	0	3.6992	0.0621	0.6887	0.4743	18.62	1.0000	4.0000
trestbps	123	0	128.09	2.51	27.82	773.98	21.72	0.00	115.00
chol	123	0	0.000000	0.000000	0.000000	0.000000	*	0.000000	0.000000
restecg	123	0	0.3577	0.0531	0.5886	0.3464	164.53	0.0000	0.0000
thalach	123	0	120.57	2.53	28.10	789.43	23.30	0.00	103.00
exang	123	0	0.4390	0.0449	0.4983	0.2483	113.50	0.0000	0.0000
oldpeak	123	0	0.6220	0.0937	1.0394	1.0804	167.12	-2.6000	0.0000
slope	123	0	1.5528	0.0768	0.8513	0.7246	54.82	0.0000	1.0000
num	123	0	1.8049	0.0914	1.0135	1.0272	56.15	0.0000	1.0000

Variable	Median	Q3	Maximum	IQR	Mode	N for Mode	Skewness	Kurtosis
Age	56.000	62.000	74.000	11.000	61	9	-0.50	-0.11
Sex	1.0000	1.0000	1.0000	0.0000	1	112	-2.91	6.59
cp	4.0000	4.0000	4.0000	0.0000	4	98	-2.59	6.51
trestbps	125.00	145.00	200.00	30.00	115	14	-1.17	6.24
chol	0.000000	0.000000	0.000000	0.000000	0	123	*	*
restecg	0.0000	1.0000	2.0000	1.0000	0	86	1.43	1.05
thalach	121.00	140.00	182.00	37.00	120	9	-0.61	1.91
exang	0.0000	1.0000	1.0000	1.0000	0	69	0.25	-1.97
oldpeak	0.2000	1.5000	3.7000	1.5000	0	48	0.23	0.44
slope	2.0000	2.0000	3.0000	1.0000	2	61	-0.37	-0.50
num	2.0000	3.0000	4.0000	2.0000	1	48	0.26	-0.77

Table 2. Descriptive statistics of heart disease variables in Cleveland

Descriptive Statistics: Age, Sex, cp, trestbps, chol, … dpeak, slope, num

Variable	N	N*	Mean	SE Mean	StDev	Variance	CoefVar	Minimum	Q1	Median
Age	303	0	54.439	0.519	9.039	81.697	16.60	29.000	48.000	56.000
Sex	303	0	0.6799	0.0268	0.4673	0.2184	68.73	0.0000	0.0000	1.0000
cp	303	0	3.1584	0.0552	0.9601	0.9218	30.40	1.0000	3.0000	3.0000
trestbps	303	0	131.69	1.01	17.60	309.75	13.36	94.00	120.00	130.00
chol	303	0	246.69	2.97	51.78	2680.85	20.99	126.00	211.00	241.00
restecg	303	0	0.9901	0.0572	0.9950	0.9900	100.49	0.0000	0.0000	1.0000
thalach	303	0	149.61	1.31	22.88	523.27	15.29	71.00	133.00	153.00
exang	303	0	0.3267	0.0270	0.4698	0.2207	143.79	0.0000	0.0000	0.0000
oldpeak	303	0	1.0396	0.0667	1.1611	1.3481	111.68	0.0000	0.0000	0.8000
slope	303	0	1.6007	0.0354	0.6162	0.3797	38.50	1.0000	1.0000	2.0000
num	303	0	0.9373	0.0706	1.2285	1.5093	131.07	0.0000	0.0000	0.0000

Variable	Q3	Maximum	IQR	Mode	N for Mode	Skewness	Kurtosis
Age	61.000	77.000	13.000	58	19	-0.21	-0.52
Sex	1.0000	1.0000	1.0000	1	206	-0.77	-1.41
cp	4.0000	4.0000	1.0000	4	144	-0.84	-0.40
trestbps	140.00	200.00	20.00	120	37	0.71	0.88
chol	275.00	564.00	64.00	197, 204, 234	6	1.14	4.49
restecg	2.0000	2.0000	2.0000	0	151	0.02	-2.00
thalach	166.00	202.00	33.00	162	11	-0.54	-0.05
exang	1.0000	1.0000	1.0000	0	204	0.74	-1.46
oldpeak	1.6000	6.2000	1.6000	0	99	1.27	1.58
slope	2.0000	3.0000	1.0000	1	142	0.51	-0.63
num	2.0000	4.0000	2.0000	0	164	1.06	-0.14

A chi square analysis was performed on the predicted variable num to determine the goodness of fit of the observed counts within the variable. For Cleveland’s num variable, the chi square value was 235.135 with a p value of less than .000 while for the Switzerland num variable, the chi square value was 52.4878 with a p value less than .000. Both predicted variables of heart disease have categories that are statistically significant. A summary of these results in relayed in table 3 and 4 below.

Table 3. Chi Square Analysis Switzerland 

Chi-Square Goodness-of-Fit Test for Categorical Variable: num

Observed and Expected Counts

Category	Observed	Test Proportion	Expected	Contribution to Chi-Square
0	8	0.2	24.6	11.2016
1	48	0.2	24.6	22.2585
2	32	0.2	24.6	2.2260
3	30	0.2	24.6	1.1854
4	5	0.2	24.6	15.6163

Chi-Square Test

N	N*	DF	Chi-Sq	P-Value
123	0	4	52.4878	0.000

Chart of Observed and Expected Values

Chart of Contribution to the Chi-Square Value by Category

Table 4. Chi Square Analysis Cleveland 

Chi-Square Goodness-of-Fit Test for Categorical Variable: num

Observed and Expected Counts

Category	Observed	Test Proportion	Expected	Contribution to Chi-Square
0	164	0.2	60.6	176.428
1	55	0.2	60.6	0.517
2	36	0.2	60.6	9.986
3	35	0.2	60.6	10.815
4	13	0.2	60.6	37.389

Chi-Square Test

N	N*	DF	Chi-Sq	P-Value
303	0	4	235.135	0.000

Chart of Observed and Expected Values

Chart of Contribution to the Chi-Square Value by Category

When performing the regression analysis, the variable chol which represented serum cholesterol was omitted from the Switzerland data model as all its observations were recorded as zero. All the other study variables were regressed against num (diagnosis of heart disease). From the analysis, ANOVA tables, model summary, coefficient, and the regression equation.

From the ANOVA tables of both countries, it is evident that most of the predictor variables are statistically insignificant. For each model only oldpeak, cp, thal, and ca were statistically significant that is p <.05. Based on the model summaries, the predictor variables for Model1(Switzerland) only account for 36.36% on the prediction of absence of heart disease in patients. While the number of study variables that are statistically insignificant in the ANOVA test for the Cleveland dataset are many, the model is able to 59% of the variation in the data. The variation predictions are values corresponding to R squared.

Further analysis on the coefficients table for each model, affirms the findings observed in the ANOVA table. For the coefficients that were statistically significant, only coefficients for cp(4), oldpeak, thal(7), slope(2) and ca were significant for the Cleveland model. For the Switzerland model, variables whose coefficients were statistically significant were only thalach, oldpeak and thal(2 and 3). Therefore, in order to make the models more fit in diagnosing heart disease, the insignificant study variables need to be omitted momentarily until the model is fit. However, in this case we will omit all the variables that are statistically insignificant at once. Worth noting, is that none of the variables had a VIF value more than 10 in both models implying that there is n multicollinearity within the predictor variables. First the tables below offer a summary of the first regression analysis (Table 5 and 6).

Heart Disease in the US and Switzerland

Table 5. Regression analysis Switzerland (Simulation 1)

Analysis of Variance

Source	DF	Adj SS	Adj MS	F-Value	P-Value
Regression	21	45.570	2.17000	2.75	0.000
Age	1	1.133	1.13324	1.44	0.234
thalach	1	7.789	7.78887	9.86	0.002
oldpeak	1	8.763	8.76322	11.10	0.001
trestbps	1	0.348	0.34802	0.44	0.508
Sex	1	0.453	0.45287	0.57	0.451
cp	3	6.673	2.22440	2.82	0.043
fbs	1	2.839	2.83888	3.60	0.061
restecg	2	1.835	0.91754	1.16	0.317
exang	1	0.428	0.42797	0.54	0.463
slope	3	1.436	0.47860	0.61	0.613
thal	3	7.652	2.55050	3.23	0.026
ca	3	0.296	0.09852	0.12	0.945
Error	101	79.747	0.78958
Total	122	125.317

S	R-sq	R-sq(adj)	R-sq(pred)
0.888581	36.36%	23.13%	*

Term	Coef	SE Coef	T-Value	P-Value	VIF
Constant	3.24	1.09	2.98	0.004
Age	-0.0136	0.0114	-1.20	0.234	1.63
thalach	-0.01161	0.00370	-3.14	0.002	1.67
oldpeak	0.2990	0.0897	3.33	0.001	1.34
trestbps	-0.00235	0.00354	-0.66	0.508	1.49
Sex
1	0.241	0.318	0.76	0.451	1.28
cp
2	-0.676	0.691	-0.98	0.330	2.34
3	-0.462	0.532	-0.87	0.387	5.25
4	0.163	0.522	0.31	0.755	6.87
fbs
1	0.831	0.438	1.90	0.061	1.17
restecg
1	0.325	0.219	1.49	0.140	1.37
2	0.198	0.419	0.47	0.638	1.47
exang
1	-0.143	0.194	-0.74	0.463	1.44
slope
1	0.313	0.323	0.97	0.335	3.18
2	0.133	0.267	0.50	0.619	2.77
3	-0.047	0.402	-0.12	0.908	2.22
thal
3	0.806	0.337	2.39	0.019	2.31
6	0.968	0.354	2.73	0.007	1.46
7	0.461	0.246	1.87	0.064	2.12
ca
1	0.262	0.685	0.38	0.702	1.17
2	0.123	0.608	0.20	0.839	1.37
8	-0.401	0.919	-0.44	0.663	1.06

num

=

3.24 – 0.0136 Age – 0.01161 thalach + 0.2990 oldpeak – 0.00235 trestbps + 0.0 Sex_0 + 0.241 Sex_1 + 0.0 cp_1 – 0.676 cp_2 – 0.462 cp_3 + 0.163 cp_4 + 0.0 fbs_0 + 0.831 fbs_1 + 0.0 restecg_0 + 0.325 restecg_1 + 0.198 restecg_2 + 0.0 exang_0 – 0.143 exang_1 + 0.0 slope_0 + 0.313 slope_1 + 0.133 slope_2 – 0.047 slope_3 + 0.0 thal_0 + 0.806 thal_3 + 0.968 thal_6 + 0.461 thal_7 + 0.0 ca_0 + 0.262 ca_1 + 0.123 ca_2 – 0.401 ca_8

Table 6. Regression analysis Cleveland (Simulation 1)

The second regression analysis performed on the Cleveland data upon omitting insignificant variables was successful. All the new predictor variables were statistically significant. 

Table 7. Regression analysis Cleveland (Simulation 2)

The second simulation for the Switzerland model was not successful thus a momentary omission of predictor variable approach needs to be attempted. The retained variables did not offer a better prediction of the model.

Table 8. Regression analysis Switzerland (Simulation 2)

To perform momentary elimination on the Switzerland dataset to determine what variables will yield statistically significant coefficients we perform a stepwise regression using backward elimination with an alpha value of 0.05. All predictor values are keyed into the initial regression model, both categorical and non-categorical variables. The process then eliminates the least significant terms for each step until all the regression co efficient retained are statistically significant.  

From the table we observe that thalach, oldpeak, fbs and thal are the variables that statistically predict num in Switzerland. From the ANOVA table all the p values for the variables are statistically significant (p<0.05). The variables explain 25.98% of the variations in num that is diagnosis of heart disease. The VIF values are less than 10 indicating there is no multicollinearity within the predictor variables. 

Backward Elimination of Terms

α to remove = 0.05

Analysis of Variance

Source	DF	Adj SS	Adj MS	F-Value	P-Value
Regression	6	32.561	5.4268	6.79	0.000
oldpeak	1	7.788	7.7879	9.74	0.002
thalach	1	8.884	8.8835	11.11	0.001
fbs	1	3.954	3.9542	4.95	0.028
thal	3	8.649	2.8830	3.61	0.016
Error	116	92.756	0.7996
Total	122	125.317

Model Summary

S	R-sq	R-sq(adj)	R-sq(pred)
0.894217	25.98%	22.15%	18.54%

Coefficients

Term	Coef	SE Coef	T-Value	P-Value	VIF
Constant	2.594	0.431	6.02	0.000
oldpeak	0.2549	0.0817	3.12	0.002	1.10
thalach	-0.01043	0.00313	-3.33	0.001	1.18
fbs
1	0.938	0.422	2.22	0.028	1.07
thal
3	0.434	0.258	1.68	0.096	1.34
6	0.981	0.315	3.11	0.002	1.14
7	0.367	0.203	1.81	0.073	1.42

Regression Equation

fbs	thal
0	0	num	=	2.594 + 0.2549 oldpeak – 0.01043 thalach
0	3	num	=	3.028 + 0.2549 oldpeak – 0.01043 thalach
0	6	num	=	3.574 + 0.2549 oldpeak – 0.01043 thalach
0	7	num	=	2.961 + 0.2549 oldpeak – 0.01043 thalach
1	0	num	=	3.532 + 0.2549 oldpeak – 0.01043 thalach
1	3	num	=	3.967 + 0.2549 oldpeak – 0.01043 thalach
1	7	num	=	3.899 + 0.2549 oldpeak – 0.01043 thalach

References

Ulbinait?, A., & Radzevi?ius, R. (2021). Relationship between healthcare spendings and life expectancy in the USA. In Accounting, audit, analysis: transformation of science and business in new economic reality: international scientific conference, November 25-26, 2021: book of extended abstracts (pp. 111-116). Vilnius University Press.

Faizal, A. S. M., Thevarajah, T. M., Khor, S. M., & Chang, S. W. (2021). A review of risk prediction models in cardiovascular disease: conventional approach vs. artificial intelligent approach. Computer methods and programs in biomedicine, 207, 106190.

Centers for Disease Control and Prevention (CDC. (2011). Prevalence of coronary heart disease–United States, 2006-2010. MMWR. Morbidity and mortality weekly report, 60(40), 1377-1381.

Brown, J. R., & O’Connor, G. T. (2010). Coronary heart disease and prevention in the United States. New England Journal of Medicine, 362(23), 2150-2153.

Richardson, T. G., Sanderson, E., Palmer, T. M., Ala-Korpela, M., Ference, B. A., Davey Smith, G., & Holmes, M. V. (2020). Evaluating the relationship between circulating lipoprotein lipids and apolipoproteins with risk of coronary heart disease: A multivariable Mendelian randomisation analysis. PLoS medicine, 17(3), e1003062.

Radford, N. B., DeFina, L. F., Leonard, D., Barlow, C. E., Willis, B. L., Gibbons, L. W., … & Levine, B. D. (2018). Cardiorespiratory fitness, coronary artery calcium, and cardiovascular disease events in a cohort of generally healthy middle-age men: results from the Cooper Center Longitudinal Study. Circulation, 137(18), 1888-1895.

Peters, S. A., Colantonio, L. D., Zhao, H., Bittner, V., Dai, Y., Farkouh, M. E., … & Woodward, M. (2018). Sex differences in high-intensity statin use following myocardial infarction in the United States. Journal of the American College of Cardiology, 71(16), 1729-1737.

Sharma, G., Volgman, A. S., & Michos, E. D. (2020). Sex differences in mortality from COVID-19 pandemic: are men vulnerable and women protected?. Case Reports, 2(9), 1407-1410.

Peters, S. A., Muntner, P., & Woodward, M. (2019). Sex differences in the prevalence of, and trends in, cardiovascular risk factors, treatment, and control in the United States, 2001 to 2016. Circulation, 139(8), 1025-1035.

O’Neil, A., Scovelle, A. J., Milner, A. J., & Kavanagh, A. (2018). Gender/sex as a social determinant of cardiovascular risk. Circulation, 137(8), 854-864.

Benjamin, E. J., Blaha, M. J., Chiuve, S. E., Cushman, M., Das, S. R., Deo, R., … & Muntner, P. (2017). Heart disease and stroke statistics—2017 update: a report from the American Heart Association. circulation, 135(10), e146-e603.

Ruperti Repilado, F. J., Greutmann, M., Bouchardy, J., Tobler, D., & Schwerzmann, M. (2021). The impact of the coronavirus disease 2019 pandemic among adult congenital heart disease patients in Switzerland. Kardiovaskulare Medizin, 51S-51S.

Zellweger, U., Junker, C., & Bopp, M. (2019). Cause of death coding in Switzerland: evaluation based on a nationwide individual linkage of mortality and hospital in-patient records. Population health metrics, 17(1), 1-15.

Faeh, D., Gutzwiller, F., & Bopp, M. (2009). Lower mortality from coronary heart disease and stroke at higher altitudes in Switzerland. Circulation, 120(6), 495-501.

Kalyanasundaram, R., Prasanth, A., Tamizhselvan, B. R., & Kumaran, U. (2017, May). Calculating the heart disease in switzerland using Pearson’s correlation. In 2017 International Conference on Trends in Electronics and Informatics (ICEI) (pp. 1118-1121). IEEE.

Hahad, O., Kröller-Schön, S., Daiber, A., & Münzel, T. (2019). The cardiovascular effects of noise. Deutsches Ärzteblatt International, 116(14), 245.

Lee, Y. T. H., Fang, J., Schieb, L., Park, S., Casper, M., & Gillespie, C. (2022). Prevalence and Trends of Coronary Heart Disease in the United States, 2011 to 2018. JAMA cardiology.

Merz, L. (2020). E-Cigarette Taxation in the United States: a Comparison and Health Policy Review Study (Doctoral dissertation, University of Zurich).

Benjamin, E. J., Muntner, P., Alonso, A., Bittencourt, M. S., Callaway, C. W., Carson, A. P., … & American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. (2019). Heart disease and stroke statistics—2019 update: a report from the American Heart Association. Circulation, 139(10), e56-e528.

Tsao, C. W., Aday, A. W., Almarzooq, Z. I., Alonso, A., Beaton, A. Z., Bittencourt, M. S., … & American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. (2022). Heart Disease and Stroke Statistics—2022 Update: A Report From the American Heart Association. Circulation, 145(8), e153-e639.

Alonso, A., Aparicio, F. H. J., Benjamin, E. J., Bittencourt, M. S., Callaway, C. W., Carson, F. A. P., … & Lee, C. D. (2021). Heart disease and stroke statistics—2021 update. Circulation, 2021(143), e00-e00.