Prevalence of high 10-year cardiovascular risk among the general population in Malaysia and the associated factors: a nationwide community-based study in 2023

Abstract

Background

Cardiovascular disease (CVD) is a leading cause of death in Malaysia. Knowing the current distribution of 10-year CVD risk among the general population is essential for health planning and targeted interventions. Hence, we aim to determine the prevalence of high 10-year CVD risk among the general adult population in Malaysia and the associated factors.

Methods

We conducted a secondary data analysis of a dataset from a nationwide cross-sectional community-based study with stratified random sampling. Individuals aged ≥ 30 years were included. The 10-year CVD risk was estimated using the Framingham General Cardiovascular Risk Score, which was derived from points depending on sex, age groups, total cholesterol, HDL cholesterol, untreated systolic blood pressure, treated systolic blood pressure, smoking, and diabetes status. The total points were assigned corresponding risks, and the low, moderate, and high-risk categories were defined as having < 10%, 10–20%, and > 20% 10-year CVD risk, respectively. Complex sample multiple logistic regression was conducted to determine the factors associated with the high-risk group.

Results

The number of eligible participants was 827. The prevalence of high 10-year CVD risk was 21.4% (95%: 17.9–25.3), projecting to 3.6 million people. People with secondary education (aOR: 0.38) and tertiary education (aOR: 0.31) had lower odds of a high 10-year risk than those without formal or primary education. In contrast, inadequate sleep (aOR: 1.49), physical inactivity (aOR: 2.01), moderate physical activity (aOR: 1.79), and fatty liver (aOR: 2.23) were positively associated with a high 10-year CVD risk.

Conclusions

The prevalence of high 10-year CVD risk among the general adult population in Malaysia is high. Lower education level, sleep inadequacy, physical inactivity, and fatty liver are independent factors associated with a high 10-year CVD risk. Health policymakers, programme managers, and clinicians should screen, identify, and treat people with a high 10-year CVD risk to prevent cardiovascular complications.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-23748-3.

Background

Cardiovascular diseases (CVD) pose an enormous disease burden globally. CVD is the leading cause of worldwide mortality, with 20.5 million deaths in 2021 [1], and they are projected to cause 35.6 million deaths by 2050 [2]. Premature CVD, particularly ischaemic heart disease and stroke, are now causing substantial burdens of morbidity and mortality globally [3]. Furthermore, a systematic review revealed that in most low- and middle-income countries, the annual cost of care and the cost of an acute cardiovascular event are many times higher than the per capita total expenditure on health [4].

Malaysia is an upper-middle-income country in Southeast Asia. Here, ischaemic heart disease and stroke are consistently the leading causes of death [5]. According to the Malaysian Burden of Disease Study, CVD contributed to the highest premature mortality rates in both males and females in the country [6]. It was estimated that the Malaysian government spent Ringgit Malaysia (RM) 5.5 billion (RM 1 is approximately 0.22 US dollars), or 12.0% of the total government healthcare expenditure in 2021, on CVD alone [7]. Meanwhile, premature deaths, absenteeism, and presenteeism caused RM 19.5 billion of indirect costs to be lost, causing substantial economic losses to the country [7].

CVD risk scores or prediction models are essential tools for the primary prevention of CVD [8]. These risk scores use established CVD risk factors, such as age, sex, smoking, blood pressure, cholesterol, and diabetes, to predict five- or ten-year CVD risks in a defined population [8]. Estimating future CVD risks and identifying high-risk populations for preventive measures are the cornerstones of clinical practice guidelines for the primary prevention of CVD [8, 9]. A local study using the Globorisk risk score estimated that 14.0% of the general population in 2015 had a high 10-year CVD risk [10]. Another study, using the 2019 World Health Organisation (WHO) Cardiovascular Disease Risk laboratory-based charts for Southeast Asia, reported that only 4.9% of the general population in 2019 was at high risk of CVD [11]. While these studies revealed important information, the estimates were limited by the tools, which had not been validated specifically for use in the Malaysian setting [10, 11]. Moreover, the current 10-year CVD risk among the general population in Malaysia is unknown, posing a knowledge gap. Hence, we aim to determine the prevalence of high 10-year CVD risk among the general population in Malaysia and the associated factors using a previously validated CVD risk score.

Methods

Source of data

This study analysed a secondary dataset from a national survey in 2023, and the study protocol detailing the methods has been published [12]. It was a cross-sectional and population-based study across all states and federal territories in Malaysia. Stratified random sampling was utilised, and the number of enumeration blocks and living quarters was proportionate to the population size [12]. In addition to face-to-face interviews and clinical assessment, fasting venous blood samples were collected [12].

Sample size

The single proportion formula for prevalence was used to calculate the sample size [13]. Based on the 14.0% prevalence of high 10-year CVD risk, 95% confidence interval, 5% margin of error, 2.0 design effect, and finite population correction to the 2023 population in Malaysia, the sample size was 367 [10, 14]. We also estimated the sample size by factoring in the odds ratio of some independent variables at https://www.openepi.com/SampleSize/SSCohort.htm [15]. Based on a previous nationwide local study, compared to tertiary education, the odds ratio for a high 10-year CVD risk among those with non-formal, primary, and secondary education were 2.36, 3.28, and 1.18, respectively [11]. Using a 95% confidence interval and 80% of power, the estimated samples with continuity correction for these respective odds ratios are 696, 334, and 22,896. The 696 is a more optimum sample size; too little, the study would be underpowered, disallowing analysis of associated factors, while a large sample of over 22,000 would be financially expensive and logistically challenging for a community-based nationwide survey [12]. The secondary dataset had 1,035 respondents, and after excluding 208 of them who were younger than 30 years, and four of them due to missing values required for the computation of 10-year CVD risk, the number of eligible respondents was 823, which was above 696. Hence, our sample size should be optimally powered for this study. We included only respondents aged 30 years and above because the Framingham Risk Study General Cardiovascular Risk Score used in this study was for this specific age category [9, 16]. The four missing values comprised three missing values for systolic blood pressure and one missing value for HDL and total cholesterol.

Study definitions

There are many CVD risk scores and prediction models, such as the Framingham General Cardiovascular Risk Score [16], WHO Cardiovascular Disease Risk [17], Globorisk [18], and Systematic Coronary Risk Evaluation (SCORE) [19]. Prior validation of CVD risk scores is critical while selecting the appropriate one for specific populations [8]. According to current clinical practice guidelines in Malaysia, the Framingham General Cardiovascular Risk Score is the recommended CVD risk score [9]. Two local studies validated the risk score and reported that it was better at discriminating CVD risk in a multiethnic population [20, 21]. Hence, the Framingham General Cardiovascular Risk Score was used in this study.

The Framingham General Cardiovascular Risk Score could assess the general CVD risk and the risk of individual CVD events, namely coronary heart disease, cerebrovascular disease, peripheral arterial disease, and heart failure [16]. CVD points were given for seven variables based on sex, age groups, HDL cholesterol, total cholesterol, untreated systolic blood pressure or treated systolic blood pressure, current smoking, and diabetes status [16]. Additional File 1 shows the CVD points for each variable and the CVD risk for total points stratified by sex. Overall, the points for these variables ranged from −3 to 12 for women and −2 to 15 for men, respectively [16]. The points were then summed up for each respondent, and the total points were assigned the corresponding percentage of risks, which were stratified by sex [16]. The low, moderate, and high 10-year CVD risk was defined as having < 10%, 10% to 20%, and > 20% risks, respectively [9].

In the original article, HDL- and total cholesterol were in mg/dL, whereas the Malaysian Clinical Practice Guidelines adopted the tables but converted the units to mmol/L [9]. The HDL- and total cholesterol assessments in this study were based on fasting venous blood samples analysed in a centralised laboratory to prevent measurement bias [12]. The WHO STEPwise structured questionnaire was used to obtain the medical history for diabetes, hypertension, and hypercholesterolemia [22]. Diabetes was based on self-reported diabetes among physician-diagnosed persons or having fasting venous glucose ≥ 7.0 mmol/L or HbA1c ≥ 6.3% among respondents not known to have diabetes [23]. Hypertension and hypercholesterolemia were based on self-reported physician-diagnosed cases. Among non-cases, hypertension was defined as having the average second and third systolic BP readings of ≥ 140 mmHg or diastolic BP ≥ 90 mmHg [24]. Meanwhile, hypercholesterolaemia was defined as having total cholesterol > 5.2 mmol/L among non-cases [25].

Socio-demographic information was collected, including sex, ethnic groups, age categories, education level, household income categories, and employment status. The three major ethnic groups in Malaysia are Malay, Chinese, and Indian ethnicities. Bumiputera Sabah and Sarawak groups are indigenous people who originated from East Malaysia on the island of Borneo. The highest education levels were categorised into no formal, primary, secondary, or tertiary education. The household income was classified into the Top 20%, Middle 40%, and Bottom 40% based on the Department of Statistics Malaysia [26]. Current drinkers were those who drank any kind of alcoholic beverages in the past 12 months, while current smokers were those who were currently smoking any form of tobacco product, either daily or occasionally [27]. An average sleep duration of less than seven hours was considered inadequate [28, 29]. Obesity, overweight, normal weight, and underweight were defined as body mass index (BMI) ≥ 27.5, 23.0–27.4, 18.5–22.9, and < 18.5 kg/m², respectively, following the local guidelines [30]. Physical activity was measured using the Global Physical Activity Questionnaire (GPAQ), and respondents were categorised into three levels: (1) highly active, achieving at least 3,000 metabolic equivalents of task (MET) minutes per week; (2) moderately active, achieving between 600 and 2,999 MET minutes per week; and (3) physically inactive, achieving less than 600 MET minutes per week [31].

A fatty liver was diagnosed based on a Fatty Liver Index (FLI) of ≥ 60, which was derived from body mass index, waist circumference, gamma-glutamyl transferase, and triglycerides [32]. We did not use metabolic dysfunction-associated fatty liver disease (MAFLD) because its computation included metabolic risk abnormalities such as blood pressure and HDL cholesterol, which were also components of the 10-year CVD risk [33]. This ensured that fatty liver, not MAFLD, was independent of the dependent variable for data analysis.

The risk of advanced liver fibrosis was estimated using the Fibrosis-4 index, which was derived from age, aspartate aminotransferase, alanine aminotransferase, and platelet count [33]. Low fibrosis risk was defined as having a Fibrosis-4 index of < 1.3 for respondents aged < 65 years and < 2.0 for those ≥ 65 years. A score of 1.3 to 2.67 and 2.0 to 2.67 indicated moderate risk in individuals < 65 and ≥ 65 years, respectively. A score of > 2.67 defined a high risk of advanced liver fibrosis [33]. Since the Fibrosis-4 index contained age, it was not included in the multiple logistic regression analysis. Nevertheless, the description of advanced liver fibrosis risk with 10-year CVD risk is still important because local data on this aspect is lacking.

Statistical analysis

This study used complex sample analyses, and the weights were calculated from the inverse probability of selecting a living quarter multiplied by non-response rates and further multiplied by post-stratification factors by sex, age groups, and ethnic groups, consistent with the method used in annual National Health and Morbidity Surveys in Malaysia [27]. The frequency counts and weighted percentages were reported for the baseline characteristics. The prevalence of high 10-year CVD risk was presented in percentages with 95% confidence intervals, and the estimated populations with it were also reported. Weighted proportions of people with high 10-year CVD risk were compared in the bivariate Pearson Chi-Square analyses, and P values were reported. We also reported unadjusted odds ratios and 95% confidence intervals using univariate logistic regression. Independent variables with P values < 0.25 were included in the multivariable model to control for confounding factors using multiple binary logistic regression to determine the factors associated with high 10-year CVD risk [34]. The variables used to compute the 10-year CVD risk, such as sex, age groups, diabetes, and smoking status, were not included in the multivariable model because the dependent variable was derived from these factors. We checked for interaction and multicollinearity and reported the model’s performance in terms of area under the receiving operating characteristics (AUROC), the classification table, and the coefficient of determination (R²).

Ethical approval

The Medical Research and Ethics Committee of the Ministry of Health Malaysia (NMRR-ID-22–02845-GUT) has approved the study. It was conducted according to the principles of the Declaration of Helsinki and the Malaysian Good Clinical Practice Guidelines. All study participants gave written informed consent to participate. The anonymised dataset used in this study did not contain any participant identifiers.

Results

Table 1 shows the characteristics of participants. Among 823 participants, there were more males (52.1%), Malay ethnicity (50.6%), aged 30–39 years (33.9%), having secondary education (63.7%), and in the bottom 40% income category (49.5%). Current drinkers and current smokers were observed in 14.0% and 17.7%, respectively. Meanwhile, 40.8% of participants had inadequate sleep, while 44.6% had high physical activity levels. About 37.7% and 37.6% of participants were overweight or obese.

Table 1.

Baseline characteristics of respondents ≥ 30 years in Malaysia, n = 823

	Characteristics	Total, n	Weighted %
Sex	Male	354	52.1
	Female	469	47.9
Ethnicity	Malay	484	50.6
	Chinese	161	23.8
	Indian	69	6.5
	Bumiputera	74	10.7
	Others	35	8.4
Age groups	30—39	182	33.9
	40—49	182	26.0
	50—59	200	18.4
	60 and above	259	21.7
Education level	No formal education	64	7.6
	Primary education	119	12.2
	Secondary education	527	63.7
	Tertiary education	113	16.5
Household income, n = 815	Bottom 40%	416	49.5
	Middle 40%	240	30.7
	Top 20%	159	19.8
Current drinker, n = 821	Yes	77	14.0
Current smoker, n = 821	Yes	118	17.7
Inadequate sleep, n = 822	Yes	345	40.8
Physical activity level, n = 823	Highly active	364	44.6
	Moderately active	219	27.0
	Inactive	240	28.4
Body mass index, n = 808	Normal	169	22.0
	Underweight	24	2.7
	Overweight	301	37.7
	Obese	314	37.6

The prevalence of high 10-year CVD risk among the general adult population in Malaysia was 21.4%, and the population projection was around 3.6 million people (Table 2). The mean 10-year CVD risk was 27.2% in the high-risk group.

Table 2.

Prevalence and estimated population with low, moderate, and high 10-year CVD risk, n = 823

10-year CVD risk categories	n	Estimated population (‘000)	Prevalence (%)	95% CI		Mean risk (%)	95% CI
Low-risk	429	10,454	61.8	57.2	66.1	4.11	3.80	4.42
Moderate-risk	169	2,856	16.9	14.1	20.1	13.98	13.47	14.48
High-risk	225	3,617	21.4	17.9	25.3	27.19	26.55	27.84

The prevalence of high 10-year CVD risk was significantly higher statistically among males (31.7%), 60 years and above (60.0%), a primary education level (34.3%), current smokers (33.0%), inadequate sleep (24.7%), physically inactive level (27.2%), diabetes (58.8%), hypertension (43.5%), hypercholesterolemia (25.6%), fatty liver (31.2%), and high liver fibrosis risk (54.3%) (Table 3).

Table 3.

Prevalence of high 10-year CVD risk in Malaysia, stratified by sociodemographic factors and comorbidities, n = 823

Characteristics		Low or moderate 10-year risk (n = 598)	High 10-year risk (n = 225)	Population with a high 10-year risk (‘000)	Weighted % for high 10-year risk	95% CI		P values
Sex	Male	195	159	2,796	31.7	24.9	39.3	< 0.001
	Female	403	66	821	10.1	7.4	13.8
Age groups	30—39	181	1	38	0.7	0.1	4.3	< 0.001
	40—49	164	18	427	9.7	5.8	15.8
	50—59	143	57	932	30.0	23.7	37.1
	60 and above	110	149	2,220	60.0	53.5	66.1
Ethnicity	Malay	338	146	2,003	23.4	18.7	28.8	0.387
	Chinese	125	36	688	17.1	12.6	22.8
	Indian	47	22	247	20.3	13.3	34.9
	Bumiputera	56	18	461	25.4	18.8	33.4
	Others	32	3	218	15.3	4.8	39.4
Education level	No formal education	39	25	429	33.4	20.6	49.4	0.015
	Primary education	78	41	710	34.3	24.7	45.4
	Secondary education	388	139	2,030	18.8	15.1	23.1
	Tertiary education	93	20	448	16.1	8.1	29.4
Household income (n = 815)	Bottom 40%	289	127	2,025	24.4	19.0	30.6	0.098
	Middle 40%	176	64	1,019	19.8	15.5	24.9
	Top 20%	128	31	531	16.0	10.5	23.5
Current smoker (n = 821)	Current smoker	64	54	983	33.0	23.9	43.5	0.012
	Non-current smoker	533	170	2,622	18.9	14.8	23.7
Current drinker (n = 821)	Current drinker	59	18	480	20.4	11.1	34.3	0.837
	Non-current drinker	537	207	3,136	21.6	18.0	25.8
Inadequate sleep	Yes	244	102	1,704	24.7	19.1	31.3	0.062
	No	354	123	1,913	19.1	15.8	23.0
Physical activity	Highly active	296	68	1,236	16.4	13.0	20.5	0.007
	Moderately active	152	67	1,073	23.4	16.5	32.2
	Inactive	150	90	1,308	27.2	22.1	32.9
Body mass index (n = 808)	Underweight	18	6	78	17.4	7.2	36.2	0.033
	Normal	137	32	448	12.2	7.8	18.6
	Overweight	205	96	1,602	25.5	19.2	33.0
	Obese	230	84	1,396	22.2	16.1	29.8
Diabetes	Yes	84	127	1,839	58.8	51.9	65.3	< 0.001
	No	514	98	1,778	12.9	9.7	16.9
Hypertension	Yes	194	172	2,828	43.5	37.0	50.2	< 0.001
	No	404	53	789	7.6	5.8	9.9
Hypercholesterolaemia	Yes	405	179	2,864	25.6	21.5	30.2	0.002
	No	193	46	753	13.1	8.4	19.9
Fatty liver	Yes	166	95	1,637	31.2	24.6	38.8	< 0.001
(n= 804)	No	420	123	1,887	16.6	13.0	20.9
Liver fibrosis	Low risk	501	163	2,697	18.7	15.6	22.1	< 0.001
(n= 821)	Intermediate risk	85	50	739	34.9	23.7	48.1
	High risk	10	12	180	54.3	36.6	70.9

In the multivariable model, four variables were statistically significant and independently associated with high 10-year CVD risk: education level, sleep status, physical activity level, and fatty liver (Table 4). People with secondary education (aOR: 0.38) and tertiary education (aOR: 0.31) were less likely to have a high 10-year CVD risk than those without formal or primary education. In contrast, inadequate sleep (aOR: 1.49), physically inactive (aOR: 2.01) and moderately active (aOR: 1.79), and fatty liver (aOR: 2.23) were positively associated with a high 10-year CVD risk.

Table 4.

Factors associated with high 10-year CVD risk, n = 796

Variables	Crude OR	95% CI		P values	Adjusted OR	95% CI		P values
Ethnicity
Malay	1.00				1.00
Chinese	0.68	0.44	1.05	0.078	0.70	0.39	1.26	0.223
Indian	0.94	0.48	1.84	0.850	1.02	0.44	2.39	0.958
Bumiputera	1.12	0.70	1.78	0.637	1.24	0.67	2.33	0.482
Others	0.59	0.17	2.10	0.406	0.48	0.14	1.66	0.237
Education level
No formal/primary education	1.00				1.00
Secondary education	0.45	0.29	0.71	< 0.001	0.38	0.24	0.60	< 0.001
Tertiary education	0.37	0.16	0.88	0.026	0.31	0.13	0.74	0.011
Household income
Bottom 40%	1.00				1.00
Middle 40%	0.76	0.49	1.20	0.232	0.84	0.52	1.34	0.838
Top 20%	0.59	0.36	0.98	0.043	0.68	0.39	1.21	0.684
Current drinker—Yes	0.93	0.44	1.96	0.927
Inadequate sleep—Yes	1.39	0.99	1.96	0.060	1.49	1.02	2.18	0.041
Physical activity level
Highly active	1.00				1.00
Moderately active	1.56	1.03	2.38	0.038	1.79	1.09	2.92	0.022
Inactive	1.91	1.35	2.70	< 0.001	2.01	1.35	2.99	0.001
Body mass index
Obese	1.36	0.48	3.81	0.386
Overweight	1.63	0.40	4.35	0.644
Normal	0.66	0.18	2.41	0.511
Underweight	1.00
Fatty liver—Yes	2.28	1.51	3.46	< 0.001	2.23	1.42	3.52	0.001

The model was adjusted for ethnicity and household income because the P values in the univariate analysis were < 0.25. The results for these variables were not statistically significant in the multivariable model. Variables used to compute the 10-year CVD risk,such as age, sex, smoking, and diabetes status, were omitted as the dependent variable was derived from these variables. Liver fibrosis risk was not included because the fibrosis-4index contained the age variable; its inclusion would violate the principle of independence between dependent and independent variables. The model was valid as the Hosmer and Lemeshow test was insignificant, P= 0.599. There was no multicollinearity or interactions between variables. The overall classification table’s correct percentage was 77.9%, and the coefficient of determination (R²) was 12.1%. The area under the receiving operating characteristics (AUROC) was 67.5% (95% CI: 63.4–71.6), P< 0.001

Discussion

In 2023, we found out from our study that about 21.4% of adults aged 30 years and above in Malaysia had a high 10-year risk of developing CVD. We also extrapolate that this could mean about 3.6 million people are at higher risk of developing a CVD. This projection reflects the significant public health burden due to CVD in this country. It is consistent with the statistics that ischaemic heart disease and cerebrovascular disease were consistently the top causes of mortality in Malaysia for many years [5]. While the figure is alarming, the findings are not unexpected as CVD risk factors, such as diabetes, hypertension, hypercholesterolaemia, and obesity, are highly prevalent in Malaysia [27].

The prevalence of high 10-year CVD risk in our study is similar to the 20.5% and 23% reported in two previous Malaysian studies using the Framingham Risk Score [21, 35]. A study using the 2006 National Health and Morbidity Survey dataset noted that 23% of the general population had a high 10-year CVD risk then [21]. In other words, our study suggested that the prevalence of high 10-year CVD risk has remained stagnant for almost two decades. This problem of high 10-year CVD risk in Malaysia can be partly attributable to the consistently elevated prevalence of CVD risk factors like diabetes, hypertension, and hypercholesterolaemia [27].

Two other studies have used more recent nationwide datasets, namely the 2015 and 2019 National Health and Morbidity Surveys, to estimate the 10-year CVD risk in Malaysia, but with different CVD risk scores [10, 11]. While these studies revealed essential information, the results were limited by the tools, namely the Globorisk risk score and the 2019 WHO Cardiovascular Disease Risk laboratory-based charts for Southeast Asia, which had not been previously validated for use in the Malaysian setting [10, 11].

When the prevalence of high 10-year CVD risk was stratified by sociodemographic and comorbidities, the highest prevalence in our study was observed among people with diabetes, whereby three out of every five were at high risk of developing cardiovascular events. The local clinical practice guidelines on type 2 diabetes noted that 60% of patients with the disease will eventually die from CVDs [23]. Our findings further add evidence for regular CVD risk assessments and multifactorial cardiovascular risk factor control for the prevention of CVD in patients with diabetes [9, 23].

The literature supports the association between lower educational levels and CVD. For example, a recent study in the United States reported that lower educational attainment was linked to a higher 10-year risk of atherosclerotic CVD [36]. In a meta-analysis, the risk of CVD incidence was found to be 44% higher in the low-education group than in the high-education group [37]. In China, a ten-year prospective cohort study reported that education was the leading factor for all-cause mortality, including CVD mortality [38]. Individuals with lower educational attainment may have unstable income and reduced access to healthcare, including treatment for CVD and the risk factors [37, 39]. Moreover, they may have lower health literacy and give lower priority to health and health-related behaviour [37, 39].

Chronic sleep deprivation has been linked to increased CVD risk, and with our modern lifestyle, inadequate sleep is becoming more common [29, 40]. Our study further supports that inadequate sleep is significantly associated with increased CVD risk. A meta-analysis of prospective studies shows that short sleep duration is associated with an increased risk of CVD and higher morbidity of coronary artery disease [41]. Another study in the United Kingdom reported that poor sleepers lost CVD-free life expectancy; males lost 2.31 years, while females lost 1.80 years [42]. The underlying pathophysiology of short sleep duration with CVD is complex and includes association with common CVD risk factors, chronic inflammation, metabolic dysregulation, hormonal disturbance, and immune dysfunction [29, 41].

The protective effects of physical activity on CVD risks are expected. A recent meta-analysis of 103 studies found that a high level of physical activity was linked to a lower risk of overall CVD, coronary heart disease, and stroke [43]. Physical activity improves endothelial function, prevents arterial stiffness, and reduces oxidative stress and inflammatory markers associated with atherosclerosis [43]. Furthermore, physical activity has been consistently shown in meta-analyses to benefit many CVD risk factors, such as high blood pressure, low HDL cholesterol, obesity, and diabetes [43–46].

In our study, the strongest association with a high 10-year CVD risk was seen in those with fatty liver. Chronic excess in energy intake over expenditure increases adiposity, lipo-toxicity, and low-grade systemic inflammation, increasing the risk of atherosclerotic cardiovascular disease [47]. Excess fat accumulation in the liver is one of the manifestations of the associated metabolic dysfunction, and this has a bidirectional relationship with other components of the metabolic syndrome. In the liver, excess free fatty acids, mainly the result of influx from adipose tissue and de novo lipogenesis, are either (1) converted and stored as fat, (2) packed and exported as very low-density lipoprotein, thus giving rise to an atherogenic lipid profile, or (3) undergo beta-oxidation. The beta-oxidation process can be overwhelmed, leading to oxidative stress, liver cell injury and death, and a cascade of inflammatory responses [48]. In the face of chronic and continuous insult, this leads to the development and progression of liver fibrosis that can lead to cirrhosis and an increased risk of hepatocellular carcinoma. The identification of fatty liver represents an opportunity for lifestyle intervention that can alter the course of metabolic dysfunction. Weight loss through lifestyle intervention has been shown to resolve fatty liver, inflammation, and fibrosis, improve glycaemic and lipid profiles, and lower cardiovascular disease risk [49–51].

This study has several limitations. As the temporal effect is unclear in a cross-sectional design, the results only imply associations and do not ascertain causality. The 3.6 million people with a high 10-year CVD risk are just an extrapolation from the analysis, and it is not absolute. While this study analysed many cardiovascular risk factors, including the newer ones like inadequate sleep, some important variables were not captured, such as the family history of cardiovascular disease, diet quality, and psychosocial factors [52]. However, this is unavoidable as the current study analysed secondary data meant primarily for other research objectives.

The main strength of our study was the use of a nationally representative dataset, allowing the results to be externally generalised to the general population in Malaysia. We used the Framingham General Cardiovascular Risk Score, the recommended CVD risk score by the Malaysian Clinical Practice Guidelines, to estimate the 10-year CVD risk [9]. Laboratory-based assessment of venous blood samples was considered the ‘gold standard’ rather than point-of-care testing (POCT) in the field using capillary samples, which are subjected to operational factors and field conditions [22, 53]. For example, it was reported that total cholesterol measured by a device listed in the WHO Stepwise manual had a sensitivity value of only 57.1% compared to laboratory venous blood measurement [53]. Besides that, all our venous blood samples were analysed in a single accredited laboratory to reduce measurement bias [12]. Thus, our study findings were internally valid for estimating the 10-year CVD risk burdens in Malaysia.

An important implication of this study is the projection of high CVD burdens affecting many adults in Malaysia in the coming ten years. Since the majority of patients with cardiovascular risk factors like diabetes, hypertension, and dyslipidaemia receive their treatment in heavily subsidised public healthcare facilities, the government is expected to bear much of the brunt of CVD in terms of healthcare costs [27]. Thus, it is essential to tackle CVD and its risk factors holistically, not just from the clinical perspective, but also for Malaysia to adopt the whole-of-system and health-in-all policies to address the social determinants of health [7, 27]. For example, we found that people with a higher education level and those highly active were less likely to have a high 10-year CVD risk. Accordingly, health policymakers can approach stakeholders within and outside the government to include health topics in the education system and support a safe, health-promoting built environment that encourages physical activities in cities, housing areas, and neighbourhoods [54]. Meanwhile, public health promotion is essential to increase awareness and knowledge of the importance of sufficient sleep, sleep hygiene, and its association with health risks.

Besides that, we recommend that policymakers and programme managers consider including 10-year CVD risk scoring in existing community-based screening programmes in Malaysia. The effort can help stratify those with high CVD risks for referral to health facilities for early diagnosis and management to prevent cardiovascular complications. We also recommend that primary care facilities strengthen routine CVD risk assessments for the general population to prevent, detect, and manage cardiovascular disease and its risk factors, as recommended by the World Health Organization [17, 55]. Intensive treatment based on 10-year CVD risk would require training, financial resources, access to newer Guideline-Directed Medical Therapy (GDMT), and quality assurance to ensure good care quality [56, 57]. Clinicians should also screen their patients with fatty liver for a high 10-year CVD risk and treat them accordingly per guideline recommendations to prevent adverse cardiovascular outcomes [33].

Conclusion

In 2023, this study shows that about 21.4% of adults aged 30 years and above in Malaysia had a high 10-year risk of developing CVD. A lower education level, sleep inadequacy, a lower physical activity level, and fatty liver are independent factors associated with a high 10-year CVD risk. Health policymakers, programme managers, and clinicians should screen and treat people with a high 10-year CVD risk among the general population to prevent complications. Primary prevention programmes and interventions can be targeted at high-risk subpopulations.

Authors’ contributions

KSW, MFMY, HMR, WKC, MM, NMN, PFW, LLL, FM, NAA: Study concept and design. KSW, MFMY, HMR, NA, SWT, MABS, TGRL, TAS, NAA: Acquisition of data. KSW, MFMY, HMR, NAA: Analysis and interpretation of data. KSW: Drafting of the manuscript. KSW, MFMY, HMR, WKC, NA, SWT, MABS, TGRL, TAS, PFW, MM, NMN, LLL, FM, NAA: Critical revision of the manuscript for important intellectual content. KSW, MFMY, HMR, NAA: Obtained funding. NAA: Study supervision. All authors approved the final manuscript.

Funding

This work was supported by the Ministry of Health Malaysia, grant number 91000050.

Data availability

The data analysed in this study is not available publicly due to the local ethics regulations and could be obtained upon reasonable request via written permission from the Director General of Health, Malaysia. The corresponding author can be contacted for further information to retrieve the dataset analysed in this study.

Declarations

Ethics approval and consent to participate

The study was approved by the Medical Research and Ethics Committee of the Ministry of Health, Malaysia (NMRR-ID-22–02845-GUT). It was conducted according to the principles of the Declaration of Helsinki and the Malaysian Good Clinical Practice Guidelines. All respondents provided written informed consent to participate in this research.

Consent for publications

Not applicable.

Competing interests

The authors declare no competing interests.