This paper is in the following e-collection/theme issue:

Longitudinal and Cohort Studies in Public Health (301) Cardiology (151) Cardiovascular Disease Prevention (131) Asian Health Disparities (4) Social Determinants of Health (28) Diabetes Surveillance and Epidemiology (27) Cardiac Risk and Cardiac Risk Calculators (41) Hypertension Prevention and Treatment (218)

Published on in Vol 10 (2024)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/55261, first published .
Changes in 10-Year Predicted Cardiovascular Disease Risk for a Multiethnic Semirural Population in South East Asia: Prospective Study

Changes in 10-Year Predicted Cardiovascular Disease Risk for a Multiethnic Semirural Population in South East Asia: Prospective Study

Changes in 10-Year Predicted Cardiovascular Disease Risk for a Multiethnic Semirural Population in South East Asia: Prospective Study

Authors of this article:

Hamimatunnisa Johar1, 2 Author Orcid Image ;   Chiew Way Ang1 Author Orcid Image ;   Roshidi Ismail1 Author Orcid Image ;   Zaid Kassim3 Author Orcid Image ;   Tin Tin Su1, 2 Author Orcid Image
Article Authors Cited by Tweetations Metrics

    Original Paper

    1South East Asia Community Observatory (SEACO) & Global Public Health, Jeffrey Cheah School of Medicine and Health Sciences, Monash University Malaysia, Subang Jaya, Malaysia

    2Heidelberg Institute of Global Health, Faculty of Medicine, University of Heidelberg, Heidelberg, Germany

    3Segamat Health Office, Ministry of Health Malaysia, Segamat, Malaysia

    *these authors contributed equally

    Corresponding Author:

    Tin Tin Su, MBBS, MSc, Dr med

    South East Asia Community Observatory (SEACO) & Global Public Health

    Jeffrey Cheah School of Medicine and Health Sciences

    Monash University Malaysia

    Jalan Lagoon Selatan

    Bandar Sunway

    Subang Jaya, 47500

    Malaysia

    Phone: 60 55146000

    Email: TinTin.Su@monash.edu


    Background: Cardiovascular disease (CVD) risk factors tend to cluster and interact multiplicatively and have been incorporated into risk equations such as the Framingham risk score, which can reasonably predict CVD over short- and long-term periods. Beyond risk factor levels at a single time point, recent evidence demonstrated that risk trajectories are differentially related to CVD risk. However, factors associated with suboptimal control or unstable CVD risk trajectories are not yet established.

    Objective: This study aims to examine factors associated with CVD risk trajectories in a semirural, multiethnic community-dwelling population.

    Methods: Data on demographic, socioeconomic, lifestyle, mental health, and cardiovascular factors were measured at baseline (2013) and during follow-up (2018) of the South East Asia Community Observatory cohort. The 10-year CVD risk change transition was computed. The trajectory patterns identified were improved; remained unchanged in low, moderate, or high CVD risk clusters; and worsened CVD risk trajectories. Multivariable regression analyses were used to examine the association between risk factors and changes in Framingham risk score and predicted CVD risk trajectory patterns with adjustments for concurrent risk factors.

    Results: Of the 6599 multiethnic community-dwelling individuals (n=3954, 59.92% female participants and n=2645, 40.08% male participants; mean age 55.3, SD 10.6 years), CVD risk increased over time in 33.37% (n=2202) of the sample population, while 24.38% (n=1609 remained in the high-risk trajectory pattern, which was reflected by the increased prevalence of all major CVD risk factors over the 5-year follow-up. Meanwhile, sex-specific prevalence data indicate that 21.44% (n=567) of male and 41.35% (n=1635) of female participants experienced an increase in CVD risk. However, a stark sex difference was observed in those remaining in the high CVD risk cluster, with 45.1% (n=1193) male participants and 10.52% (n=416) female participants. Regarding specific CVD risk factors, male participants exhibited a higher percentage increase in the prevalence of hypertension, antihypertensive medication use, smoking, and obesity, while female participants showed a higher prevalence of diabetes. Further regression analyses identified that Malay compared to Chinese (P<.001) and Indian (P=.04) ethnicity, nonmarried status (P<.001), full-time employment (P<.001), and depressive symptoms (P=.04) were all significantly associated with increased CVD risk scores. In addition, lower educational levels and frequently having meals from outside were significantly associated to higher odds of both worsening and remaining in high CVD risk trajectories.

    Conclusions: Sociodemographics and mental health were found to be differently associated with CVD risk trajectories, warranting future research to disentangle the role of psychosocial disparities in CVD. Our findings carry public health implications, suggesting that the rise in major risk factors along with psychosocial disparities could potentially elevate CVD risk among individuals in underserved settings. More prevention efforts that continuously monitor CVD risk and consider changes in risk factors among vulnerable populations should be emphasized.

    JMIR Public Health Surveill 2024;10:e55261

    doi:10.2196/55261

    Keywords

    cardiovascular risk trajectory; Framingham risk score; population-based study; low- and middle-income countries 


    Background

    Cardiovascular diseases (CVDs) continue to be the main contributor to deaths globally, with a significant proportion of the mortality occurring in low- and middle-income countries (LMICs) rather than in high-income countries [1]. In Malaysia, CVD is the leading cause of death, as it accounts for an estimated 20% of all noncommunicable disease mortality [2], escalating the health care and economic burden over the past decades [3]. As a middle-income country, rapid urbanization with major demographic and socioeconomic changes and concurrent shifts in diet and activity levels contribute to the epidemiological shift toward much higher rates of CVD.

    Regular assessments of CVD risk factors, as advocated by the Regional Action Framework for Noncommunicable Disease Prevention and Control in the Western Pacific, can inform lifestyle and medical interventions, potentially preventing cardiovascular-related deaths [4]. Multiple CVD risk factors, by contrast, tend to cluster and interact multiplicatively to promote further vascular risk. As a result, CVD risk factors have been incorporated into risk equations such as the Framingham risk score (FRS), which can reasonably predict CVD over short- and long-term periods [5,6]. Furthermore, as CVD risks may change over time, reassessment of CVD risk is important to identify changes, assess the effectiveness of interventions, and ensure adherence to recommended treatments [7]. Previous large epidemiological European studies analyzing these changes in risk scores have been shown to improve risk stratification beyond the single time point risk score classification [8,9].

    Although it has been recognized that adverse levels of risk factors often develop early in life and change over time [10,11], the underlying factors that may influence these changes are not fully understood. Noncommunicable diseases might be preceded by a relatively sudden deterioration in risk factors before disease onset. However, previous research demonstrated that traditional cardiometabolic risk factors do not fully account for or explain the excess burden of CVDs in the population, of which the risk may be additionally explained by mental health factors [12,13]. Moreover, the deterioration in CVD risk factors has also been shown to be amplified by other factors, such as sociodemographic, behavioral, and mental health factors [14]. In LMICs, where research in CVD risk trajectories is still limited, demographic, socioeconomic, behavioral, and mental health factors may be highly prevalent and may contribute significantly to unhealthy CVD risk trajectories. For instance, low socioeconomic conditions are strongly linked to financial stress, psychosocial stress [15], and limited access to health care facilities [16], which could further amplify the deterioration of CVD risk factors such as hypertension and diabetes. Therefore, such insight into these multifaceted factors that interact and influence pathophysiological changes provides indications for optimal preventive actions in diverse global contexts.

    While most prior studies have mainly focused on the impact of risk score trajectories on CVD morbidity [9], studies that examine mental health and behavioral factors associated with CVD risk trajectories remain scarce, particularly in LMICs. Factors that may improve CVD risk include higher educational levels [17], stable employment [18], healthy lifestyle behaviors [19] such as balanced diet and regular exercise, and good mental health conditions [20]. Conversely, factors that may lead to deterioration of CVD risk include lower socioeconomic status (SES) [21]; unhealthy behaviors [19] such as poor diet, sedentary lifestyle, smoking, and harmful alcohol consumption; and chronic psychological stress, depression, and anxiety [20]. A better understanding of the underlying factors that may influence worsened CVD risk trajectories would facilitate targeting at-risk population groups.

    Objective

    This study aims to examine 5-year changes in CVD risk based on 2 assessments of the FRS and to evaluate concurrent demographic, socioeconomic, mental health, and lifestyle factors that may influence the improvement or deterioration of CVD risk trajectories in a representative sample of the multiethnic semirural adult population in Southeast Asia.


    Population and Study Setting

    This study used the secondary data stem from the community health survey of the South East Asia Community Observatory (SEACO) in 2013 and 2018. SEACO is a health and demographic surveillance system (HDSS) established in Segamat, Johor state, Malaysia, by Monash University Malaysia in 2011 [22]. SEACO conducted house-to-house interviews to obtain information on demographic and socioeconomic characteristics (eg, age, education, income, and marital status), self-reported health status (eg, hypertension status and diabetes status), and mental health condition from 5 subdistricts in Segamat, particularly, Bekok, Chaah, Gemereh, Jabi, and Sungai Segamat. All individuals aged ≥5 years in the selected 5 subdistricts who participated in the SEACO censuses were invited to complete the survey.

    In the community health survey, the respondents aged ≥35 years underwent health screening services (anthropometric measurement, blood pressure [BP], and random glucose) and a personal interview conducted by the trained study personnel.

    In 2013, a total of 25,158 individuals aged ≥35 years participated in the Health Survey 2013. Of these, 55% (n=13,828) underwent health screening. A reexamination of the individuals in SEACO HDSS was conducted in 2018. A reexamination of the individuals in SEACO HDSS was conducted in 2018. The SEACO health survey uses a dynamic or open cohort design. However, the study sample was limited to those who participated in both baseline (2013) and follow-up (2018) surveys as well as those without missing data on CVD risk factors at baseline and follow-up, which included 6599 participants, as depicted in Figure 1. In a dropout analysis, excluded participants were more likely to be aged ≥70 years and to have higher FRS scores compared to the study sample (Table S1 in Multimedia Appendix 1). However, no significant sex differences were found between excluded and included participants.

    ‎
    Figure 1. Flowchart of participant selection. At baseline, 13,828 individuals aged ≥35 years were included in the Health Survey 2013 of the South East Asia Community Observatory health and demographic surveillance system, established in Segamat, Johor, Malaysia. A subsample of 7462 individuals was reexamined in the Health Survey 2018. The final study population included 6599 participants who completed both the 2013 and 2018 surveys and provided complete data on cardiovascular disease risk factors.

    Outcome Measurement (Cardiovascular Risk Score)

    The outcome variable was modified FRS proposed by D’Agostino et al [5] in estimating the 10-year predicted CVD risk among the respondents. The 10-year CVD risk score was calculated using nonlaboratory predictors: age (in years), BMI, use of antihypertensive medication, systolic BP, smoking status, and diabetes mellitus status [5]. Participants who reported “currently smoking” were classified as smokers. Anthropometric measurements, including height (in cm) and weight (in kg), were obtained during home-based interviews. BMI was calculated by dividing weight by height in meters squared. BP was measured 3 times using the Omron HEM 7120 E Blood Pressure Monitor M2 Basic Digital Intellisense, following the standard STEPwise guidelines [23]. For data analysis, the second and third BP readings were averaged, as the first reading may overestimate mean BP [24].

    In Malaysia, for primary prevention, the Clinical Practice Guideline committee recommends the use of the FRS general CVD risk score for risk stratification [25]. The FRS score is commonly used as a summary measure in Malaysia and has been shown to predict CVD risk more accurately than other risk scores [26]. Besides that, the modified FRS was used and validated by Su et al [27] in predicting the 10-year CVD risk among the urban population in Malaysia. In this study, the FRS score obtained in 2013 and 2018 were calculated among the study population. The change in CVD risk score over a 5-year period was calculated by subtracting the CVD risk score at follow-up in 2018 from the CVD risk score at baseline in 2013 (CVD risk score at baseline, 2013 – CVD risk score at follow-up, 2018). A negative score indicates an increase in CVD risk (worsened), while a positive score indicates a decrease in CVD risk (improvement).

    Assessment of the CVD Risk Factors and Covariates

    The covariates selected were a priori based on the availability of survey questions and the support from the past studies. Several studies showed that there is an association of demographic and socioeconomic characteristics with 10-year CVD risk [28,29], lifestyle practice [30,31], and mental health status [32,33]. The demographic variables included in this study are age, sex, ethnicity, and marital status. Besides that, education, household income, and occupation were included in this study. Due to the lack of extensive dietary data and considering the influence of salt intake on CVD risks among the study population, the frequency of meals taken outside per week was used as the proxy for healthy eating habit, as described previously [31]. Physical activity was assessed using the Global Physical Activity Questionnaire, and data were transformed to derive estimates of the average number of minutes per week engaged in physical activity in 3 domains (occupational, transport, and leisure) and total physical activity at each time point [34]. We classified participants as insufficiently active if they engaged in <600 metabolic equivalent of task minutes of physical activity per week [34]. Symptoms of depression, anxiety, and stress were assessed using the Depression, Anxiety and Stress Scale-21 items [35]. Higher scores indicate greater levels of mental health symptoms. Individuals were classified as having depressive symptoms with a cutoff score of 10 [35].

    Statistical Analysis

    Participant characteristics, including demographic, socioeconomic, lifestyle, and mental health factors, are presented for the total sample and according to sex at baseline and follow-up. Categorical variables are summarized as proportions and compared between the 2 time points using chi-square tests, whereas continuous variables are summarized as mean (SD) and compared between the 2 time points were using paired 2-tailed t test. Similar analyses were performed to compare the participants’ characteristics and CVD risk groups. To visualize the transition of the predicted 10-year CVD risk change in the study population, we created a river plot using low, moderate, and high-risk CVD groups. To increase the interpretability of the CVD risk trajectories, a stacked bar chart was also created.

    Multivariable linear regression models were fit to assess baseline factors associated with changes in CVD risk scores (CVD risk score at baseline, 2013 – CVD risk score at follow-up, 2018). Model 1 was adjusted for demographic and socioeconomic factors (sex, ethnicity, marital status, educational level, income, and employment status). Model 2 was further adjusted for lifestyle factors (frequency of meals taken from outside and physical activity). Model 3 was additionally adjusted for mental health factors (depression, anxiety, and stress). In a sensitivity analysis, multinomial logistic regression models were fitted to assess the association between baseline demographic, socioeconomic, lifestyle, and mental health factors and changes in the predicted CVD risk groups.

    We performed complete-case analyses and included participants with available CVD risk factors (at baseline and follow-up) and other covariates data. Descriptive and regression analyses were performed in SPSS (version 20; IBM Corp), and the river plot was created by using the web-based SankeyMATIC software by Steve Bogart [36].

    Ethical Considerations

    The surveys were approved by the Monash University Human Research Ethics Committee (MUHREC; MUHREC 3837 [Health Survey 2013, baseline] and MUHREC 13,242 [Health Survey 2018, follow-up]). All participants were given an information sheet and requested to sign a consent form before the data collectors started the interview.


    Participant Characteristics

    A total of 6599 participants (n=3954, 59.92% female participants and n=2645, 40.08% male participants) with a mean age of 55.3 (SD 10.6) years were included in this analysis. Participants’ characteristics of demographic, socioeconomic, lifestyle, clinical, and mental health factors stratified by sex are presented in Table 1. Overall, most participants are categorized in the 50 to 59 years age group (n=2232, 33.83%); of Malay ethnicity (n=4202, 63.68%); were married (n=5551, 84.21%); completed primary school (n=2976, 45.68%); had a monthly household income of <RM 3000 (approximately US $663.35; n=4523, 68.54%); not in full-time employment or self-employment (n=3776, 57.22%); and did not report any symptoms of depression (n=5711, 87.61%), anxiety (n=5425, 82.84%), or stress (n=6253, 95.66%; Table 1). In terms of sex difference, male participants were more likely to be older (men: mean age 56.7, SD 10.7 years; women: mean age 54.4, SD 10.4 years), be married, and currently employed; reported taking more frequent meals from outside; had higher levels of physical activity; and experienced more moderate to severe mental health symptoms than their female counterparts (Table 1). The participants’ baseline demographic (apart from age) and socioeconomic characteristics do not change drastically and are quite stable over the follow-up period (Table S2 in Multimedia Appendix 1).

    Table 1. Baseline demographic, socioeconomic, lifestyle, and mental health characteristics of a subsample of participants from the South East Asia Community Observatory health survey in 2013 by sex (N=6599).
    VariableTotalMale (n=2645)Female (n=3954)Chi-square (df)P valuea
    Age (y), mean (SD)55.29 (10.60)56.68 (10.74)54.36 (10.40)—b—
    Age groups (y), n (%)14.1 (2).001

    		35-39504 (7.63)175 (6.62)329 (8.32)


    		40-491549 (23.47)549 (20.75)1000 (25.29)


    		50-592232 (33.83)808 (30.55)1424 (36.01)


    		60-691665 (25.24)799 (30.21)866 (21.90)


    		≥70649 (9.83)314 (11.87)335 (8.48)

    Ethnicity, n (%)5.8 (4).21

    		Aborigine74 (1.12)33 (1.25)41 (1.04)


    		Chinese1500 (22.73)626 (23.67)874 (22.10)


    		Indian751 (11.38)290 (10.96)461 (11.66)


    		Malay4202 (63.68)1674 (63.29)2528 (63.94)


    		Others72 (1.09)22 (0.83)50 (1.26)

    Marital status, n (%)266.3 (3)<.001

    		Married5551 (84.21)2411 (91.19)3140 (79.52)


    		Never married214 (3.25)113 (4.28)101 (2.59)


    		Separated or divorced115 (1.74)21 (0.79)94 (2.37)


    		Widow, widower, or others712 (10.80)99 (3.74)613 (15.52)

    Education, n (%)76.3 (3)<.001

    		Primary2976 (45.68)1205 (45.85)1771 (45.57)


    		Secondary2898 (44.48)1228 (46.73)1670 (42.96)


    		Tertiary236 (3.62)112 (4.26)124 (3.19)


    		Others405 (6.22)83 (3.16)322 (8.28)

    Household incomec (RM), n (%)9.0 (3).03

    		<10001371 (20.78)585 (22.12)786 (19.87)


    		1000-19991952 (29.58)758 (28.66)1194 (30.20)


    		2000-29991200 (18.18)503 (19.02)697 (17.63)


    		≥30002076 (31.46)799 (30.20)1277 (32.30)

    Occupation, n (%)1975.5 (3)<.001

    		Paid employee1577 (23.90)964 (36.45)613 (15.50)


    		Self-employed1246 (18.88)968 (36.60)278 (7.03)


    		Not working724 (10.97)322 (12.17)402 (10.17)


    		Others3052 (46.25)391 (14.78)2661 (67.30)

    Frequency of meals taken outside per week, n (%)180.0 (3)<.001

    		02474 (38.06)747 (28.79)1727 (44.21)


    		1-52653 (40.81)1153 (44.43)1500 (38.40)


    		6-10643 (9.89)342 (13.18)301 (7.71)


    		≥11731 (11.24)353 (13.60)378 (9.68)

    Level of total physical activity, n (%)74.1 (2)<.001

    		Low5830 (88.35)2256 (85.29)3574 (90.39)


    		Moderate365 (5.53)145 (5.48)220 (5.56)


    		High404 (6.12)244 (9.23)160 (4.05)

    Depression, n (%)2.7 (3).44

    		Normal5711 (87.61)2264 (86.94)3447 (88.07)


    		Mild357 (5.48)145 (5.57)212 (5.42)


    		Moderate340 (5.22)145 (5.57)195 (4.98)


    		Severe and extremely severe110 (1.69)50 (1.92)60 (1.53)

    Anxiety, n (%)3.0 (3).39

    		Normal5425 (82.84)2160 (82.32)3265 (83.18)


    		Mild276 (4.21)104 (3.96)172 (4.38)


    		Moderate619 (9.45)260 (9.91)359 (9.15)


    		Severe and extremely severe229 (3.50)100 (3.81)129 (3.29)

    Stress, n (%)7.5 (3).06

    		Normal6253 (95.66)2494 (95.12)3759 (96.02)


    		Mild156 (2.39)65 (2.48)91 (2.32)


    		Moderate88 (1.35)39 (1.49)49 (1.25)


    		Severe and extremely severe40 (0.60)24 (0.91)16 (0.41)

    aP values for differences between male and female participants were calculated using the χ2 test.

    bNot applicable.

    cUS $1 is equivalent to RM 3.00 (May 13, 2013).

    Table 2 presents the prevalence of CVD risk factors of the study participants at baseline (2013) and follow-up (2018), with all major CVD risk factors showing a significant increase from 2013 to 2018. The prevalence of hypertension increased from 22.47% (n=1483) in 2013 to 38.53% (n=2542) in 2018, as also reflected by the increased antihypertensive medication use (n=980, 14.85% in 2013 to n=1172, 17.76% in 2018). Similarly, an increasing trend was also observed in the prevalence of diabetes (n=795, 12.05% in 2013 to n=1422, 21.55% in 2018), obesity (n=1737, 26.28% in 2013 to n=1871, 28.39% in 2018), and smoking (n=901, 13.65% in 2013 to n=1619, 24.53% in 2018). Similar increasing trends were also observed in systolic BP, BMI, and random blood glucose levels. The mean blood glucose levels increased from 8.0 (SD 3.6) mmol/L in 2013 to 8.3 (SD 4.1) in 2018 (t6597=–4.7; P<.001). Of note, the prevalence of the high CVD risk group significantly increased from 29.78% (n=1965) in 2013 to 44.26% (n=2921) in 2018. Similarly, in comparison to the baseline values, higher mean of total FRS score (2013: mean 11.7, SD 5.5; 2018: mean 14.2, SD 5.4) and overall percentage of predicted CVD risk (2013: mean 14.1%, SD 9.8%; 2018: mean 17.8%, SD 9.9%) at follow-up were also observed (Table S3 in Multimedia Appendix 1).

    Table 2. Cardiovascular disease risk factors in 2013 and 2018 among the South East Asia Community Observatory study participants (prevalence; N=6599).
    Variables2013, n (%)2018, n (%)Chi-square (df)P valuea
    Hypertension status993.7 (1)<.001

    		Yes1483 (22.47)2542 (38.53)


    		No5116 (77.53)4056 (61.47)

    On antihypertensive medication265.7 (1)<.001

    		Yes980 (14.85)1172 (17.76)


    		No5619 (85.15)5427 (82.24)

    Smoking status2408.1 (1)<.001

    		Yes901 (13.65)1619 (24.53)


    		No5698 (86.35)4980 (75.47)

    Diabetes status1392.3 (1)<.001

    		Yes795 (12.05)1422 (21.55)


    		No5804 (87.95)5177 (78.45)

    BMI4900.3 (4)<.001

    		Underweight or normal2327 (35.27)2128 (32.29)


    		Overweight2537 (38.45)2591 (39.32)


    		Obese1734 (26.28)1871 (28.39)

    Predicted cardiovascular risk3101.2 (4)<.001

    		Low1820 (27.58)989 (14.99)


    		Moderate2814 (42.64)2689 (40.75)


    		High1965 (29.78)2921 (44.26)

    aP values for differences between baseline (2013) and follow-up (2018) were calculated using the χ2 test for categorical variables.

    The sex-specific prevalence of CVD risk factors in the study population at baseline and follow-up shows an increase in all risk factors from 2013 to 2018 for both male and female participants, as detailed in Table S4 in Multimedia Appendix 1. In 2018, male participants exhibited the highest prevalence of smoking and high predicted CVD risk, while female participants had the highest prevalence of hypertension, diabetes, and obesity. There was a slightly greater increase in the prevalence of diabetes among female participants (2013: n=493, 12.5%; 2018: n=895, 22.6%) compared to male participants (2013: n=302 11.4%; 2018: n=527, 19.9%). This increase in diabetes prevalence was also reflected in a slight increase in blood glucose levels, from 8.1 (SD 4.0) mmol/L to 8.4 (SD 4.1) mmol/L in female participants (t3952=–3.2; P=.02) and from 8.1 (SD 4.0) mmol/L to 8.3 (SD 4.2) mmol/L in male participants (t2644=–3.3; P=.001). Notably, the prevalence of hypertension, antihypertensive medication use, smoking, and obesity showed a higher percentage increase among male participants from 2013 to 2018 compared to female participants (Table 3).

    Table 3. Sex-stratified cardiovascular disease risk factors in 2013 and 2018 among the South East Asia Community Observatory study participants (prevalence; n=6599).
    VariablesMale (n=2645) Female (n=3954)

    		20132018Chi square (df)P value20132018Chi square (df)P valuea
    Hypertension status, n (%)381.5 (1)<.001
    		608.8 (1)<.001

    		Yes551 (20.83)973 (36.80)

    		932 (23.57)1569 (39.68)


    		No2094 (79.17)1671 (63.20)

    		3022 (76.43)2385 (60.32)

    Antihypertensive medication, n (%)259.1 (1)<.001

    		457.7 (1)<.001

    		Yes366 (13.84)488 (18.45)

    		614 (15.53)684 (17.30)


    		No2279 (86.16)2157 (81.55)

    		3340 (84.47)3270 (82.70)

    Smoking, n (%)528.6 (1)<.001

    		876.1 (1)<.001

    		Yes874 (33.04)1577 (59.62)

    		27 (0.68)42 (1.06)


    		No1771 (66.96)1068 (40.38)

    		3927 (99.32)3912 (98.94)

    Diabetes, n (%)533.0 (1)<.001

    		856.5 (1)<.001

    		Yes302 (11.42)527 (19.92)

    		493 (12.47)895 (22.64)


    		No2343 (88.58)2118 (80.08)

    		3461 (87.53)3059 (77.36)

    BMI1826.8 (4)<.001

    		2986.3 (4).001

    		Underweight or normal1113 (42.08)1046 (39.55)

    		1214 (30.71)1089 (27.54)


    		Overweight1041 (39.36)1054 (39.85)

    		1496 (37.84)1537 (38.87)


    		Obese491 (18.56)545 (20.60)

    		1243 (31.45)1328 (33.59)

    Predicted cardiovascular disease risk825.9 (4)<.001

    		1643.7 (4)<.001

    		Low194 (7.33)140 (5.29)

    		1626 (41.12)849 (21.47)


    		Moderate1019 (38.53)875 (33.08)

    		1795 (45.40)1814 (45.88)


    		High1432 (54.14)1630 (61.63)

    		533 (13.48)1291 (32.65)

    aP values for differences between baseline (2013) and follow-up (2018) by sex were calculated using the χ2 test for categorical variables.

    With regard to sex differences, female participants had a comparable level of diastolic and systolic BP as well as blood glucose levels to their male counterparts. However, female participants recorded a slightly higher mean of BMI of 28.0 (SD 5.4) compared to male participants, who had a mean BMI of 26.3 (SD 4.5). In terms of high predicted CVD risk, males experienced an increase in prevalence from 54.14% (n=1432) in 2013 to 61.63% (n=1630) in 2018, while female participants showed an increase from 13.48% (n=533) in 2013 to 32.65% (n=1291) in 2018. In addition, male participants exhibited an increase in the mean total FRS score (2013: mean 12.7, SD 4.8; 2018: mean 15.7, SD 5.0), while female participants showed an increase from 11.0 (SD 5.8) in 2013 to 15.1 (SD 9.5) in 2018.

    The Transition of CVD Risk Groups Over Time

    Figure 2 presents river plots that illustrate transitions in CVD risk groups (low, moderate, and high) from baseline in 2013 to follow-up in 2018. On the basis of the 10-year CVD risk classification, the prevalence of low, moderate, and high CVD risk was 27.58% (n=1820), 42.64% (n=2814), and 29.78% (n=1965) at baseline and 14.99% (n=989), 40.75% (n=2689), and 44.26% (n=2921) at follow-up (2018), respectively. The percentage of high predicted CVD risk among the total population had increased significantly from 29.78% (n=1965) in 2013 to 44.26% (n=2921) in 2018, whereas the prevalence of low predicted CVD risk decreased from 27.58% (n=1820) in 2013 to 14.99% (n=989) in 2018. Sex-specific river plots show that male participants were the major contributors to high predicted CVD risk, with a prevalence of 54.14% (n=1432) at baseline in 2013 and 61.63% (n=1630) at follow-up in 2018. In contrast, 13.48% (n=533) of female participants presented with high predicted CVD risk at baseline, which increased to 32.65% (n=1291) at follow-up.

    ‎
    Figure 2. River plot describing transition patterns of low, moderate, or high predicted cardiovascular disease (CVD) risk change from baseline in 2013 to follow-up in 2018 among (A) total population, (B) male population, and (C) female population. Note: The light red lines show the transition of high predicted CVD risk cluster from baseline 2013 branching out to high, moderate or low predicted CVD risk in 2018; the yellow lines indicate the moderate predicted CVD risk cluster from baseline 2013 branching out to high, moderate or low predicted CVD risk in 2018; and the blue lines indicate the low predicted CVD risk cluster from baseline 2013 branching out to high, moderate or low predicted CVD risk in 2018. The percentage (%) values indicate the prevalence of low, moderate, and high CVD risk, based on the 10-year CVD risk classification.

    On the basis of the changes from baseline (2013) to follow-up (2018), four distinct trajectory patterns were identified: (1) remained high (n=1609), (2) remained low or moderate (low-low: n=815; moderate-moderate: n=1452), (3) adverse (n=2202), and (4) improved (n=521; Figure S1 in Multimedia Appendix 1).

    Figure 3 present the transition of overall and sex-specified predicted CVD risk trajectories from 2013 to 2018. As can be seen in Figure 3, 33.37% of participants had worsened their CVD risk (from low or moderate in 2013 to high in 2018; Most of the participants remained in the high, moderate, or low CVD risk group at both time points with percentages of 24.38% (n=1609), 22% (n=1452), and 12.35% (n=815), respectively. Only 7.9% (n=521) of the study participants showed improvement in their predicted CVD risks. Sex-stratified analyses revealed that almost half (n=1193, 45.1%) of the male participants remained in the high CVD risk cluster at both time points, 19.4% (n=513) of male participants remained in the moderate-risk group, whereas only 2.08% (n=55) remained in the low-risk category. In contrast, most female participants remained in the moderate (n=939, 23.75%) and low (n=760, 19.22%) CVD risk classification over time. Only 5.16% (n=204) of females had a high CVD risk at both time points. Interestingly, the prevalence of female participants who had improved predicted CVD risk (higher risk in 2013 to lower risk in 2018) was lower than that of males (n=516, 5.16% vs n=317, 11.98%).

    ‎
    Figure 3. Percentages of predicted cardiovascular disease (CVD) risk trajectories by sex from baseline in 2013 to follow-up in 2018. The distinct trajectory patterns identified were improved; remained unchanged in low, moderate, or high CVD risk clusters; and adverse (worsen) CVD risk trajectories. Notes: improved—the predicted CVD risk had shifted from the high-risk cluster in 2013 to the low-risk cluster in 2018, and adverse (worsen)—the predicted CVD risk had shifted from the low-risk cluster in 2013 to the high-risk cluster in 2018.

    Factors Associated With 10-Year CVD Risk Classification

    In the present study, high predicted CVD risk groups from both 2013 and 2018 were predominantly male, of older age (60-69 years), of Malay ethnicity, married, and had a lower SES (completed only up to the primary level of education, did not have a full-time paid employment or self-employment, and earned <RM 3000 (equivalent to US $1000) than the lower CVD risk categories (Table S5 in Multimedia Appendix 1). Frequent eating outside was not significantly different across the CVD risk groups at both time points, although the proportion of participants in the high CVD risk group who did not report having meals outside increased from 37.46% (n=726) in 2013 to 54.78% (n=1600) in 2018. Low physical activity was significantly associated with high CVD risk in 2018 but not in 2013. Regarding mental health, the high CVD risk group was less likely to report mental health conditions (depression, anxiety, or stress).

    Table 4 displays the multivariable linear regression models for the associations of demographic, socioeconomic, lifestyle, and mental health factors with changes in CVD risk from baseline to follow-up (CVD risk score at baseline, 2013 – CVD risk score at follow-up, 2018). A negative value indicates an increase in CVD score (worsening), and a positive value indicates a decrease in CVD score (improvement). Chinese and Indian ethnic groups were associated with a decline in CVD risk scores compared to Malays (Chinese: β=.62, 95% CI .39-.84; Indian: β=.33, 95% CI .44-.62). Being unmarried (single, widowed, or other marital status) was associated with an increase in CVD risk scores compared to being married (β=–.52, 95% CI –.77 to –.26). With regard to SES, low monthly income (<RM 1000) was associated with a decrease in CVD risk score (β=.50, 95% CI .25-.76), whereas full-time self-employment and paid employment were associated with an increase in CVD risk score (self-employment: β=–.64, 95% CI –1.00 to –.29; paid employment [full time, part time, and casual job]: β=–.78, 95% CI –1.12 to –.44). The significant associations remained even after further adjustments for behavioral and mental health factors (model 2 and model 3). In model 3, there was an inverse association between severe depressive symptoms and changes in CVD risk score (–1.25, 95% CI –2.44 to –.06; P=.04). However, there was no significant evidence for associations between educational levels, lifestyle behaviors (frequency of meals taken outside and physical activity), and 2 other mental health conditions (anxiety and stress) and changes in CVD risk scores. ANOVA tests assessed if adding predictors enhances model fit. A low P value suggests that a more complex model (eg, model 3 with added variables) significantly improves fit over a simpler model (eg, model 1 with fewer variables). The ANOVA value indicated that model 3 had the preferred fit, supported by significant ANOVA values that were consistent across all models, even in the most complex models.

    Table 4. β estimates, 95% CIs, and P values of the prospective associations of baseline demographic, socioeconomic, lifestyle, and mental health factors with changes in cardiovascular disease risk score (n=6599)a.
    VariablesFramingham risk score changesb

    		Model 1Model 2Model 3

    		β (95% CI)P valueβ (95% CI)P valueβ (95% CI)P value
    ANOVA2636.4c—d2599.6c—2658.2c—
    Ethnicity

    		Aborigine or others.59 (−.30 to 1.48).19.64 (−.29 to 1.57)0.181.53 (−.21 to 1.69).13

    		Chinese.62 (.39 to .84)<.001.56 (.34 to .79)<.001.56 (.33 to .79)<.001

    		Indian.33 (.04 to .62).03.29 (−.01 to .58)0.06.30 (.00 to .60).04

    		Malay1—1—1—
    Marital status

    		Nonmarried−.52 (−.77 to −.26)<.001−.54 (−.80 to −.28)<.001−.52 (−.78 to −.26)<.001

    		Married1—1—1—
    Education

    		Other.45 (−.17 to 1.06).15.45 (−.17 to 1.07).15.41 (−.21 to 1.03).19

    		Primary.19 (−.31 to .69).46.22 (−.28 to .72).39.20 (.30 to .70).43

    		Secondary−.27 (−.77 to .22).28−.22 (−.71 to .28).40−.21 (−.71 to .29).40

    		Tertiary1—1—1—
    Monthly incomeb (RM)

    		>1000.50 (.25 to .76)<.001.50 (.24 to .76)<.001.50 (.35 to .76)<.001

    		1000-1999−.04 (−.28 to .19).71−.05 (−.28 to .18).68−.07 (−.30 to .16).55

    		2000-2999.12 (−.14 to .39).36.10 (−.17 to .37).46.11 (−.16 to .38).43

    		≥30001—1—1—
    Employment

    		Other.02 (−.30 to .33).92.02 (−.29 to .34).89.00 (−.32 to .32).99

    		Self-employed−.64 (−1.00 to −.29)<.001−.66 (−1.01 to −.30)<.001−.68 (−1.04 to −.31)<.001

    		Paid employee−.78 (−1.12 to −.44)<.001−.75 (−1.10 to −.40)<.001−.73 (−1.08 to −.38)<.001

    		Not working1—1—1—
    Frequent outside meals——−.01 (−.03 to .00).12−.01 (−.03 to .00).13
    Physical activity levels

    		High——.23 (−.16 to .61).24.23 (−.15 to .62).24

    		Moderate——.28 (−.12 to .68).17.27 (−.13 to .67).19

    		Low——1—1—
    Depression

    		Severe————−1.25 (−2.44 to −.06).04

    		Moderate————−.10 (−.68 to .48).74

    		Mild————.10 (−.38 to .58).68

    		No symptom————1—
    Anxiety

    		Severe————−.35 (−1.16 to .46).40

    		Moderate————−.15 (−.57 to .26).47

    		Mild————.07 (−.42 to .55).79

    		No symptom————1—
    Stress

    		Severe————.63 (−.93 to 2.19).43

    		Moderate————.64 (−.53 to 1.80).29

    		Mild————−.11 (−.87 to .65).78

    		No symptom————1—

    aDepicted are β estimates, 95% CIs, and P values estimated by multivariable linear regression models.

    bUS $1 was equivalent to RM 3.00 on May 13, 2013, and RM 3.97 on May 17, 2018.

    cP<.001.

    dNot applicable.

    Sensitivity Analyses

    The transition between CVD risk clusters was categorized as either “improved” (the CVD risk scores became better in 2018), “adverse” (the CVD risk worsened in 2018), or “unchanged (high CVD risk)” (being stable at high CVD risk groups) in comparison to being stable in low or moderate-risk groups at both time points. Therefore, we further analyzed multinomial logistic regression models to examine the association between baseline demographic, socioeconomic, lifestyle, and mental health factors and changes in CVD risk groups (no changes or remained low or moderate CVD risk as the reference category; Table S6 in Multimedia Appendix 1).

    Among all investigated factors associated with CVD risk trajectories, lower educational levels and frequently having meals from outside were significantly associated with increased odds of worsening CVD risk trajectories. Aborigines and other ethnic groups (compared with Malays), income groups of RM 2000 to RM 2999 and <RM 1000 (compared with >RM 3000), and physical activity were significantly associated with decreased odds of being in adverse CVD risk trajectories.

    In addition, lower educational levels and frequently having meals outside were associated with increased odds of remaining in high CVD risk trajectories. Meanwhile, Indian, aborigines, and other ethnicity; being not married; and not being employed for full time (self-employed or others) were associated with decreased odds of remaining in high CVD risk trajectories. However, income levels and physical activity were not associated with unchanged high CVD risk trajectories.

    In terms of improvement in CVD risk, income <RM 3000 (compared to higher income levels) and self-employment or other types of employment (compared to not working) were associated with decreased odds of having improved CVD risk score from 2013 to 2018. Other demographic, socioeconomic, lifestyle, and mental health factors were not significantly associated with improved CVD risk trajectories.

    Of note, mental health factors were not significantly associated with all CVD risk group trajectories.


    Principal Findings

    This study evaluated the changes in CVD risk over a 5-year follow-up period and factors associated with the CVD risk clustering in 6599 multiethnic community-dwelling men and women. Tracking of CVD risk clusters over 5 years of follow-up time revealed that most participants remained in their baseline CVD risk clusters, with 24.38% (n=1609) of the total population remaining in the same high CVD risk profile and 22% (n=1452) in the moderate CVD risk profile, which was dominated by male participants. More alarming, 33.37% (n=2202) of study participants experienced extreme shifts from lower predicted CVD risks in 2013 to higher risks in 2018. This appeared to be reflected by the increased prevalence of all major CVD risk factors over the 5-year follow-up. Of note, our analyses revealed that demographic, SES, depressive symptoms, and dietary habits were important determinants of increasing CVD risk scores and worsening CVD risk trajectories over time.

    Comparison With Prior Work

    In this study, we found that most participants (n=3811, 55.37%) had increased CVD risks or remained in the high CVD risk category, which aligns with the rise in all major CVD risk factors from 2013 to 2018. Previous studies in Malaysia confirmed our findings on the distribution of baseline CVD risk categories using the FRS model, with 20% to 23% at high risk, 29% to 38.5% at moderate risk, and 41% to 48% at low risk [26,27]. However, when compared to other studies, including ours, a recent study found a slightly lower prevalence of the high CVD risk category (16.8%) [37]. Despite recent health care service advancements and CVD preventive measures in Malaysia, our analysis confirmed the increasing trend in individual CVD risk factors, as demonstrated by previous nationwide estimates that reported an increase in the prevalence of diabetes and obesity (diabetes: 13.4% in 2015 to 18.3% in 2019; for obesity [BMI≥27.5 kg/m²]: 17.7% in 2015 to 19.7% in 2019), and hypertension remained high at 30% in both years, all of which are comparable to our study (n=2202, 33.37%) [38]. These data have enormous public health implications, suggesting that the increase in major risk factors in Malaysia may place individuals at higher CVD risk, which is supported by other studies from Asia or multiethnic settings showing an upward trend in CVD risk over time. Li et al [39] showed that the increase in the high CVD risk category was primarily due to an increase in the prevalence of hypertension and diabetes. A prospective population-based study from the Multi-Ethnic Study of Atherosclerosis (N=6800) has assessed temporal changes in individual CVD risk factors and found that almost half (45%) of participants remained in the unfavorable CVD risk category [40]. The Multi-Ethnic Study of Atherosclerosis, like ours, also showed an increase in poor blood glucose control over the follow-up period, suggesting more participants’ glucose control deteriorated or became medication dependent later in life, as observed in this investigation. Similar to our findings, the higher predicted CVD risk score may be attributable to the increased prevalence of individual risk factors, such as obesity [41] and smoking [42], translating into a higher prevalence of diabetes mellitus and hypertension [43]. Thus, this study emphasizes the importance of CVD risk factors management, which could be potential candidates for lifestyle or pharmacological interventions explicitly targeted at improving the cardiovascular health of the high-risk groups.

    However, the observed increase in mean FRS scores between 2013 and 2018 and the higher prevalence of stable high-risk and worsened CVD-risk categories of this investigation is in contrast with the findings from the United States and Europe [44-46]. First, a US nationwide study using the FRS found a reduction in mean 10-year risk of CVD from 9.2% to 8.7% between 1999 and 2010, among those aged 30 to 74 years, with a larger decline in risk score and proportion of change from high to low risk in men than women [44]. Second, the Tromsø study reported that the mean of the NORRISK 2 CVD risk score decreased and distribution in risk categories moved from higher to lower risk in both sexes and all age groups between the first (2007-2008) and second surveys (2015-2016) [46]. Similarly, a study from England reported a decrease in high risk (QRISK2 score>20%) of 2.4% and 6.8% and medium risk (QRISK2 ≥10%) of 3.2% and 5.3%, in 10-year risk of CVD per decade during 1994 to 2013, for women and men, respectively [45]. Moreover, both increases and decreases in CVD risk score trajectories were associated with CVD risk and disease-free life-years, respectively [8,9], suggesting that higher scores were associated with increased CVD risks, while an improvement in CVD risk scores over a 5-year period was linked to reduced risk. This demonstrates the beneficial effect of reducing modifiable risk factors in the general population, resulting in improved CVD risk factors due to effective preventive strategies. In contrast, a population-based study among 3699 individuals (aged 53 years) from Iran indicated that most low-risk individuals remained low risk and high-risk individuals remained high risk over 10-year follow-up period, confirming our findings on the stability of the CVD risk clusters over 5 years [47]. Despite the favorable increases in high-density lipoprotein and decreased smoking and total cholesterol trajectory patterns, Koohi et al [47] found no evidence of a worsening CVD risk trajectory, which could be explained by the unfavorable increases in fasting glucose.

    Of note, previous research demonstrated that traditional cardiometabolic risk factors do not fully account for or explain the excess burden of CVD in the population [13,48]; therefore, assessing other contributing factors of CVD risks is paramount. In this study, differential associations between employment status and changes in CVD risks were observed. First, being employed was associated with a continuous increase in CVD risk score, supporting a previous cross-sectional study, where paid employees experienced the highest prevalence of CVD risk factors compared to the other employment categories [49]. Indeed, employed individuals who are exposed to stress at work, termed as “job strain,” are at an increased risk of coronary heart disease [50,51]. Second, self-employment and others (pensioners, homemakers, and students), which can be linked to both stable and unstable financial situations, were associated with having a high CVD risk at both time points as well as improvement in CVD risk trajectories. Evidence indicates a modest association between job insecurity and incident coronary heart disease, partly attributable to unstable employment status, poorer socioeconomic circumstances, and less favorable risk factor profiles among people with job insecurity [18]. This highlights the potential exposure to job strain in employed individuals and job insecurity in financially unstable types of employment that may influence an individual’s cardiovascular health. In general, having continuous, formal, full-time, and stable jobs provide an income that allows people to meet their basic needs and leads to a better quality of life in adulthood and old age [52,53].

    Although the first line of evidence presents conflicting findings with a lower monthly income associated with a decrease in CVD risk score, results from the sensitivity analysis further revealed that lower educational status, used as a proxy of low SES, was associated with adverse CVD risk trajectories, supporting previous evidence on the role of socioeconomic disparities in substantiating CVD risks. Our findings resonate with previous studies among the low-income urban population in Kuala Lumpur, whereby individuals in the lowest income category (<RM 1000) had the highest prevalence of CVD risk factors, and unstable employment was associated with a higher risk of developing CVD [49]. Moreover, epidemiological studies have demonstrated that lower SES is related to risky health behaviors, such as regular smoking and excessive alcohol consumption, that are significant risk factors for the onset of CVD [54]. Semirural residents considered socioeconomically disadvantaged in the LMIC settings have poorer lifestyle behaviors, have a greater tendency to be unable to purchase health-promoting products and services, and experience more negative life events (such as unemployment, marital conflict, and financial hardship), which would lead to negative mental health [55-57]. Consequently, these prolonged stresses may also trigger compulsive behaviors such as overeating, excessive drinking, and tobacco use [58,59].

    Regarding lifestyle risk factors, our sensitivity analysis reveals a significant association between diet and adverse CVD risk in a semirural setting. Consistent with previous cross-sectional findings, consuming meals frequently from outside was significantly associated with adverse CVD risk trajectory [31]. The high CVD risk among populations is primarily due to the high salt content of food brought from outside; therefore, health promotion programs focusing on salt reduction should be prioritized. Against expectation, our investigation revealed that physical activity was not significantly associated with changes in CVD risk. Although the protective effects of leisure time physical activity were consistent with a lower CVD risk score in previous studies [60], the lack of association between the risk score and physical activity may have been attributable to the low percentage of participants who are physically active and the homogeneity of activity levels among the study participants from this semirural setting, as demonstrated in a previous study [61].

    We additionally found a preliminary association between depressive symptoms and a decline in the CVD score; however, this effect was not observed with CVD risk trajectories as the outcome. One speculation is due to the high prevalence of smoking, which could be associated with lower mental stress levels in the study population. Smoking is highly prevalent in low SES settings in Malaysia and is associated with depression and anxiety among individuals in low-income urban areas [62]. Although longitudinal studies show that smoking cessation (compared with continued smoking) is associated with reduced stress, anxiety, and depression [63], approximately 40% of smokers in England report that they smoke to cope with stress or anxiety [64]. Moreover, according to the stress paradigm, socioeconomic disadvantage is both a source of adversity and a strain on an individual’s coping abilities [65]. Given these circumstances, health-risky behaviors (eg, smoking, overeating, and inactivity) may represent forms of pleasure and relaxation that regulate mood among the population considered disadvantaged. Despite financial instability, maintaining a healthy lifestyle should remain a priority. For instance, even with unstable employment status, a life course free of regular tobacco and alcohol use shows protective effects against CVD in the general population [54]. Therefore, more research on understanding the impact of stress-induced risky behaviors in lower SES groups are warranted.

    Limitations

    This prospective study included a large sample of community-dwelling men and women with a high response rate and strict quality assessment. One of the limitations of this prospective study is that direct cause-and-effect relationships between identified factors at baseline and follow-up cannot be discerned. The adapted FRS did not include several other potential CVD risk factors, family history of CVD, and cholesterol levels. Although we have adjusted for a comprehensive set of confounding variables, we cannot exclude that risk factors not included herein may have biased the results. We also acknowledge the potential selection bias in the study due to the exclusion of participants with missing CVD risk factors. Nevertheless, as demonstrated in the dropout analysis, it is expected that older individuals would have higher FRS scores and were also more likely to be excluded from the study. Moreover, several studies have investigated behavioral risk factors in relation to CVD incidence and suggested that adhering to a combination of healthy behaviors (nonsmoking, moderate alcohol intake, physical activity, and fruit and vegetable consumption) was associated with a lower risk of CVD morbidity and mortality [45,66]. However, in our study, we used 2 additional self-reported behavioral factors (frequent eating out and physical activity) in relation to the objective measure used in previous studies. Furthermore, the definition of diabetes would be more accurate if fasting blood samples or HbA1C were assessed. Finally, data on alcohol consumption in Malaysia are underreported due to social and religious aspects.

    Conclusions

    Our findings highlight the high prevalence of individuals from a semirural multiethnic setting remaining in high CVD risk clusters and worsening CVD trajectories as they are potentially in the progression of developing CVDs. To our knowledge, this study is the first to examine demographic, socioeconomic, behavioral, and mental health factors associated with changes in CVD risk categories, and our results highlight psychosocial disparities that will perpetuate an unacceptable status quo if left unaddressed.

    Both increases and decreases in CVD risk score trajectories were associated with CVD risk and disease-free life-years, respectively. Therefore, interventions targeting an improvement and relatively stable low CVD risk trajectories are favorable, as it indicates a lower risk for the future onset of CVD [67,68]. Recent data from several LMICs in Asia reported that hypertension treatment and control remain challenging in underresourced settings [69]. Therefore, more population-based prevention efforts that focus on CVD risk factors control among populations considered vulnerable should be emphasized. To this end, health care providers should continuously monitor individuals’ CVD risk using updated risk scoring methods and consider any changes in risk factors.

    Acknowledgments

    The authors express their gratitude to all the study participants. Furthermore, the authors thank the staff involved in the planning, organization, and conduct of the South East Asia Community Observatory study.

    Data Availability

    The data sets generated during and analyzed during this study are available from the corresponding author on reasonable request.

    Conflicts of Interest

    None declared.

    Multimedia Appendix 1

    Supplementary material.

    DOCX File , 65 KB
    1. Vardell E. Global Health Observatory Data Repository. Med Ref Serv Q. 2020;39(1):67-74. [FREE Full text] [CrossRef] [Medline]
    2. Statistics on causes of death, Malaysia, 2022. Department of Statistics Malaysia. Oct 27, 2022. URL: https://www.dosm.gov.my/portal-main/release-content/statistics-on-causes-of-death-malaysia-2022 [accessed 2024-09-03]
    3. The direct health-care cost of noncommunicable diseases in Malaysia. Ministry of Health Malaysia. 2022. URL: https:/​/www.​moh.gov.my/​moh/​resources/​Penerbitan/​Rujukan/​NCD/​NCD_Laporan/​HEALTH-COST_of_NCDs-7a-WEB.​pdf [accessed 2024-09-03]
    4. Regional action framework for noncommunicable disease prevention and control in the Western Pacific. World Health Organization. Jun 22, 2023. URL: https://www.who.int/publications/i/item/9789290620044 [accessed 2024-09-11]
    5. D'Agostino RB, Grundy S, Sullivan LM, Wilson P, CHD Risk Prediction Group. Validation of the Framingham coronary heart disease prediction scores: results of a multiple ethnic groups investigation. JAMA. Jul 11, 2001;286(2):180-187. [CrossRef] [Medline]
    6. Brindle P, Beswick A, Fahey T, Ebrahim S. Accuracy and impact of risk assessment in the primary prevention of cardiovascular disease: a systematic review. Heart. Dec 2006;92(12):1752-1759. [FREE Full text] [CrossRef] [Medline]
    7. Piepoli MF, Hoes AW, Agewall S, Albus C, Brotons C, Catapano AL, et al. 2016 European guidelines on cardiovascular disease prevention in clinical practice: the Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European Association for Cardiovascular Prevention and Rehabilitation (EACPR). Eur Heart J. Aug 01, 2016;37(29):2315-2381. [FREE Full text] [CrossRef] [Medline]
    8. Chamnan P, Simmons RK, Sharp SJ, Khaw KT, Wareham NJ, Griffin SJ. Repeat cardiovascular risk assessment after four years: is there improvement in risk prediction? PLoS One. Feb 19, 2016;11(2):e0147417. [FREE Full text] [CrossRef] [Medline]
    9. Lindbohm JV, Sipilä PN, Mars N, Knüppel A, Pentti J, Nyberg ST, et al. Association between change in cardiovascular risk scores and future cardiovascular disease: analyses of data from the Whitehall II longitudinal, prospective cohort study. The Lancet Digital Health. Jul 2021;3(7):e434-e444. [CrossRef]
    10. Kemper HC, Snel J, Verschuur R, Storm-van Essen L. Tracking of health and risk indicators of cardiovascular diseases from teenager to adult: Amsterdam Growth and Health Study. Prev Med. Nov 1990;19(6):642-655. [FREE Full text] [CrossRef] [Medline]
    11. Twisk JW, Kemper HC, van Mechelen W, Post GB. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study. Am J Epidemiol. May 15, 1997;145(10):888-898. [CrossRef] [Medline]
    12. Ladwig KH, Baumert J, Marten-Mittag B, Lukaschek K, Johar H, Fang X, et al. Room for depressed and exhausted mood as a risk predictor for all-cause and cardiovascular mortality beyond the contribution of the classical somatic risk factors in men. Atherosclerosis. Feb 2017;257:224-231. [FREE Full text] [CrossRef] [Medline]
    13. Everson-Rose SA, Lewis TT. Psychosocial factors and cardiovascular diseases. Annu Rev Public Health. 2005;26:469-500. [CrossRef] [Medline]
    14. Havranek EP, Mujahid MS, Barr DA, Blair IV, Cohen MS, Cruz-Flores S, et al. Social determinants of risk and outcomes for cardiovascular disease. Circulation. Sep 2015;132(9):873-898. [CrossRef]
    15. Santosa A, Rosengren A, Ramasundarahettige C, Rangarajan S, Gulec S, Chifamba J, et al. Psychosocial risk factors and cardiovascular disease and death in a population-based cohort from 21 low-, middle-, and high-income countries. JAMA Netw Open. Dec 01, 2021;4(12):e2138920. [FREE Full text] [CrossRef] [Medline]
    16. Powell-Wiley TM, Baumer Y, Baah FO, Baez AS, Farmer N, Mahlobo CT, et al. Social determinants of cardiovascular disease. Circ Res. Mar 04, 2022;130(5):782-799. [CrossRef]
    17. Liu K, Cedres LB, Stamler J, Dyer A, Stamler R, Nanas S, et al. Relationship of education to major risk factors and death from coronary heart disease, cardiovascular diseases and all causes, findings of three Chicago epidemiologic studies. Circulation. Dec 1982;66(6):1308-1314. [CrossRef] [Medline]
    18. Virtanen M, Nyberg ST, Batty GD, Jokela M, Heikkilä K, Fransson EI, et al. Perceived job insecurity as a risk factor for incident coronary heart disease: systematic review and meta-analysis. BMJ. Aug 08, 2013;347(aug08 1):f4746. [FREE Full text] [CrossRef] [Medline]
    19. Colpani V, Baena CP, Jaspers L, van Dijk GM, Farajzadegan Z, Dhana K, et al. Lifestyle factors, cardiovascular disease and all-cause mortality in middle-aged and elderly women: a systematic review and meta-analysis. Eur J Epidemiol. Sep 2018;33(9):831-845. [CrossRef] [Medline]
    20. Hackett RA, Steptoe A. Psychosocial factors in diabetes and cardiovascular risk. Curr Cardiol Rep. Oct 2016;18(10):95. [FREE Full text] [CrossRef] [Medline]
    21. Framke E, Sørensen JK, Andersen P, Svane-Petersen A, Alexanderson K, Bonde J, et al. Contribution of income and job strain to the association between education and cardiovascular disease in 1.6 million Danish employees. Eur Heart J. Mar 14, 2020;41(11):1164-1178. [FREE Full text] [CrossRef] [Medline]
    22. Partap U, Young EH, Allotey P, Soyiri IN, Jahan N, Komahan K, et al. HDSS profile: the South East Asia Community Observatory Health and Demographic Surveillance System (SEACO HDSS). Int J Epidemiol. Oct 01, 2017;46(5):1370-131g. [FREE Full text] [CrossRef] [Medline]
    23. WHO STEPS surveillance manual: the WHO STEPwise approach to chronic disease risk factor surveillance. World Health Organization. 2005. URL: https://iris.who.int/handle/10665/43376 [accessed 2024-09-03]
    24. Lacruz ME, Kluttig A, Kuss O, Tiller D, Medenwald D, Nuding S, et al. Short-term blood pressure variability - variation between arm side, body position and successive measurements: a population-based cohort study. BMC Cardiovasc Disord. Jan 18, 2017;17(1):31. [FREE Full text] [CrossRef] [Medline]
    25. Clinical practice guidelines on primary and secondary prevention of cardiovascular disease 2017. Ministry of Health Malaysia. 2017. URL: https://www.moh.gov.my/moh/resources/Penerbitan/CPG/CARDIOVASCULAR/3.pdf [accessed 2023-07-27]
    26. Selvarajah S, Kaur G, Haniff J, Cheong KC, Hiong TG, van der Graaf Y, et al. Comparison of the Framingham Risk Score, SCORE and WHO/ISH cardiovascular risk prediction models in an Asian population. Int J Cardiol. Sep 2014;176(1):211-218. [FREE Full text] [CrossRef] [Medline]
    27. Su TT, Amiri M, Mohd Hairi F, Thangiah N, Bulgiba A, Majid HA. Prediction of cardiovascular disease risk among low-income urban dwellers in metropolitan Kuala Lumpur, Malaysia. Biomed Res Int. 2015;2015:516984. [FREE Full text] [CrossRef] [Medline]
    28. Rosengren A, Smyth A, Rangarajan S, Ramasundarahettige C, Bangdiwala SI, AlHabib KF, et al. Socioeconomic status and risk of cardiovascular disease in 20 low-income, middle-income, and high-income countries: the Prospective Urban Rural Epidemiologic (PURE) study. The Lancet Global Health. Jun 2019;7(6):e748-e760. [CrossRef]
    29. Schultz WM, Kelli HM, Lisko JC, Varghese T, Shen J, Sandesara P, et al. Socioeconomic status and cardiovascular outcomes: challenges and interventions. Circulation. May 15, 2018;137(20):2166-2178. [FREE Full text] [CrossRef] [Medline]
    30. Barbiellini Amidei C, Trevisan C, Dotto M, Ferroni E, Noale M, Maggi S, et al. Association of physical activity trajectories with major cardiovascular diseases in elderly people. Heart. Mar 14, 2022;108(5):360-366. [CrossRef] [Medline]
    31. Ang CW, Ismail R, Kassim Z, Mohd Ghazali AN, Reidpath D, Su TT. Frequent eating out and 10-year cardiovascular disease risk: evidence from a community observatory in Malaysia. Biomed Res Int. Mar 7, 2022;2022:2748382. [FREE Full text] [CrossRef] [Medline]
    32. Daumit GL, Dalcin AT, Dickerson FB, Miller ER, Evins AE, Cather C, et al. Effect of a comprehensive cardiovascular risk reduction intervention in persons with serious mental illness: a randomized clinical trial. JAMA Netw Open. Jun 01, 2020;3(6):e207247. [FREE Full text] [CrossRef] [Medline]
    33. Nielsen RE, Banner J, Jensen SE. Cardiovascular disease in patients with severe mental illness. Nat Rev Cardiol. Feb 2021;18(2):136-145. [CrossRef] [Medline]
    34. Global physical activity questionnaire (GPAQ) analysis guide. World Health Organization. URL: https://www.who.int/docs/default-source/ncds/ncd-surveillance/gpaq-analysis-guide.pdf [accessed 2024-09-03]
    35. Lovibond SH, Lovibond PF. Depression anxiety stress scales (DASS--21, DASS--42). APA PsycTests. 1995. URL: https://psycnet.apa.org/doiLanding?doi=10.1037%2Ft01004-000 [accessed 2024-08-28]
    36. Bogart S. SankeyMATIC: make beautiful flow diagrams. SankeyMATIC. URL: https://www.sankeymatic.com/ [accessed 2023-06-01]
    37. Kasim SS, Ibrahim N, Malek S, Ibrahim KS, Aziz MF, Song C, et al. Validation of the general Framingham Risk Score (FRS), SCORE2, revised PCE and WHO CVD risk scores in an Asian population. Lancet Reg Health West Pac. Jun 2023;35:100742. [FREE Full text] [CrossRef] [Medline]
    38. National Health and Morbidity Survey 2019: non-communicable diseases, healthcare demand, and health literacy—key findings. National Institutes of Health (NIH), Ministry of Health Malaysia. 2019. URL: https://iptk.moh.gov.my/images/technical_report/2020/4_Infographic_Booklet_NHMS_2019_-_English.pdf [accessed 2024-09-03]
    39. Li Z, Yu S, Han X, Liu J, Yao H. Changes to cardiovascular risk factors over 7 years: a prospective cohort study of in situ urbanised residents in the Chaoyang District of Beijing. BMJ Open. Mar 16, 2020;10(3):e033548. [FREE Full text] [CrossRef] [Medline]
    40. Sheng Q, Ding J, Gao Y, Patel RJ, Post WS, Martin SS. Cardiovascular health trajectories and subsequent cardiovascular disease and mortality: the multi-ethnic study of atherosclerosis (MESA). Am J Prev Cardiol. Mar 2023;13:100448. [FREE Full text] [CrossRef] [Medline]
    41. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999-2004. JAMA. Apr 05, 2006;295(13):1549-1555. [CrossRef] [Medline]
    42. Keto J, Ventola H, Jokelainen J, Linden K, Keinänen-Kiukaanniemi S, Timonen M, et al. Cardiovascular disease risk factors in relation to smoking behaviour and history: a population-based cohort study. Open Heart. Jul 12, 2016;3(2):e000358. [FREE Full text] [CrossRef] [Medline]
    43. Harris MI, Flegal KM, Cowie CC, Eberhardt MS, Goldstein DE, Little RR, et al. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults. The Third National Health and Nutrition Examination Survey, 1988-1994. Diabetes Care. Apr 1998;21(4):518-524. [CrossRef] [Medline]
    44. Ford ES. Trends in predicted 10-year risk of coronary heart disease and cardiovascular disease among U.S. adults from 1999 to 2010. J Am Coll Cardiol. Jun 04, 2013;61(22):2249-2252. [FREE Full text] [CrossRef] [Medline]
    45. Alageel S, Wright AJ, Gulliford MC. Changes in cardiovascular disease risk and behavioural risk factors before the introduction of a health check programme in England. Prev Med. Oct 2016;91:158-163. [FREE Full text] [CrossRef] [Medline]
    46. Nilsen A, Hanssen TA, Lappegård KT, Eggen AE, Løchen ML, Njølstad I, et al. Secular and longitudinal trends in cardiovascular risk in a general population using a national risk model: the Tromsø study. Eur J Prev Cardiol. Nov 2019;26(17):1852-1861. [CrossRef] [Medline]
    47. Koohi F, Ahmadi N, Hadaegh F, Safiee S, Azizi F, Khalili D. Trajectories of cardiovascular disease risk and their association with the incidence of cardiovascular events over 18 years of follow-up: the Tehran Lipid and Glucose study. J Transl Med. Jul 16, 2021;19(1):309. [FREE Full text] [CrossRef] [Medline]
    48. Dar T, Radfar A, Abohashem S, Pitman RK, Tawakol A, Osborne MT. Psychosocial stress and cardiovascular disease. Curr Treat Options Cardiovasc Med. Apr 26, 2019;21(5):23. [FREE Full text] [CrossRef] [Medline]
    49. Amiri M, Majid HA, Hairi F, Thangiah N, Bulgiba A, Su TT. Prevalence and determinants of cardiovascular disease risk factors among the residents of urban community housing projects in Malaysia. BMC Public Health. 2014;14 Suppl 3(Suppl 3):S3. [FREE Full text] [CrossRef] [Medline]
    50. Kivimäki M, Nyberg ST, Batty GD, Fransson EI, Heikkilä K, Alfredsson L, et al. Job strain as a risk factor for coronary heart disease: a collaborative meta-analysis of individual participant data. Lancet. Oct 27, 2012;380(9852):1491-1497. [FREE Full text] [CrossRef] [Medline]
    51. Emeny RT, Zierer A, Lacruz ME, Baumert J, Herder C, Gornitzka G, et al. Job strain-associated inflammatory burden and long-term risk of coronary events: findings from the MONICA/KORA Augsburg case-cohort study. Psychosom Med. Apr 2013;75(3):317-325. [CrossRef] [Medline]
    52. Di Gessa G, Corna L, Price D, Glaser K. Lifetime employment histories and their relationship with 10-year health trajectories in later life: evidence from England. Eur J Public Health. Aug 01, 2020;30(4):793-799. [FREE Full text] [CrossRef] [Medline]
    53. Ponomarenko V. Cumulative disadvantages of non-employment and non-standard work for career patterns and subjective well-being in retirement. Adv Life Course Res. Dec 2016;30:133-148. [FREE Full text] [CrossRef]
    54. Madero-Cabib I, Azar A, Bambs C. Lifetime employment, tobacco use, and alcohol consumption trajectories and cardiovascular diseases in old age. SSM Popul Health. Mar 2021;13:100737. [FREE Full text] [CrossRef] [Medline]
    55. Baum A, Garofalo JP, Yali AM. Socioeconomic status and chronic stress. Does stress account for SES effects on health? Ann N Y Acad Sci. 1999;896:131-144. [CrossRef] [Medline]
    56. Lantz PM, House JS, Mero RP, Williams DR. Stress, life events, and socioeconomic disparities in health: results from the Americans' Changing Lives Study. J Health Soc Behav. Sep 2005;46(3):274-288. [CrossRef] [Medline]
    57. Allen L, Williams J, Townsend N, Mikkelsen B, Roberts N, Foster C, et al. Socioeconomic status and non-communicable disease behavioural risk factors in low-income and lower-middle-income countries: a systematic review. The Lancet Global Health. Mar 2017;5(3):e277-e289. [CrossRef]
    58. Björntorp P. Do stress reactions cause abdominal obesity and comorbidities? Obes Rev. May 07, 2001;2(2):73-86. [CrossRef] [Medline]
    59. Marmot M. The Status Syndrome: How Social Standing Affects Our Health and Longevity. New York, NY. Henry Holt and Company; 2007.
    60. Hu G, Tuomilehto J, Borodulin K, Jousilahti P. The joint associations of occupational, commuting, and leisure-time physical activity, and the Framingham risk score on the 10-year risk of coronary heart disease. Eur Heart J. Feb 2007;28(4):492-498. [CrossRef] [Medline]
    61. LaMonte MJ, Durstine JL, Addy CL, Irwin ML, Ainsworth BE. Physical activity, physical fitness, and Framingham 10-year risk score: the cross-cultural activity participation study. J Cardiopulm Rehabil. 2001;21(2):63-70. [CrossRef] [Medline]
    62. Abd Rashid R, Kanagasundram S, Danaee M, Abdul Majid H, Sulaiman AH, Ahmad Zahari MM, et al. The prevalence of smoking, determinants and chance of psychological problems among smokers in an urban community housing project in Malaysia. Int J Environ Res Public Health. May 18, 2019;16(10):1762. [FREE Full text] [CrossRef] [Medline]
    63. Taylor GM, Lindson N, Farley A, Leinberger-Jabari A, Sawyer K, Te Water Naudé R, et al. Smoking cessation for improving mental health. Cochrane Database Syst Rev. Mar 09, 2021;3(3):CD013522. [FREE Full text] [CrossRef] [Medline]
    64. Perski O, Theodoraki M, Cox S, Kock L, Shahab L, Brown J. Associations between smoking to relieve stress, motivation to stop and quit attempts across the social spectrum: a population survey in England. PLoS One. May 17, 2022;17(5):e0268447. [FREE Full text] [CrossRef] [Medline]
    65. Pearlin LI. The sociological study of stress. J Health Soc Behav. Sep 1989;30(3):241-256. [Medline]
    66. Eriksen A, Tillin T, O'Connor L, Brage S, Hughes A, Mayet J, et al. The impact of health behaviours on incident cardiovascular disease in Europeans and South Asians--a prospective analysis in the UK SABRE study. PLoS One. Mar 2, 2015;10(3):e0117364. [FREE Full text] [CrossRef] [Medline]
    67. Wang Y, Wan EY, Mak IL, Ho MK, Chin WY, Yu EY, et al. The association between trajectories of risk factors and risk of cardiovascular disease or mortality among patients with diabetes or hypertension: a systematic review. PLoS One. 2022;17(1):e0262885. [FREE Full text] [CrossRef] [Medline]
    68. Hagen AN, Ariansen I, Hanssen TA, Lappegård KT, Eggen AE, Løchen ML, et al. Achievements of primary prevention targets in individuals with high risk of cardiovascular disease: an 8-year follow-up of the Tromsø study. Eur Heart J Open. Sep 2022;2(5):oeac061. [FREE Full text] [CrossRef] [Medline]
    69. Geldsetzer P, Tan MM, Dewi FS, Quyen BT, Juvekar S, Hanifi SM, et al. Hypertension care in demographic surveillance sites: a cross-sectional study in Bangladesh, India, Indonesia, Malaysia, Vietnam. Bull World Health Organ. Oct 01, 2022;100(10):601-609. [FREE Full text] [CrossRef] [Medline]

    ‎
    BP: blood pressure
    CVD: cardiovascular disease
    FRS: Framingham risk score
    HDSS: health and demographic surveillance system
    LMIC: low- and middle-income country
    MUHREC: Monash University Human Research Ethics Committee
    SEACO: South East Asia Community Observatory
    SES: socioeconomic status

    Edited by A Mavragani; submitted 08.12.23; peer-reviewed by Y Xu, JW Park; comments to author 08.05.24; revised version received 18.07.24; accepted 09.08.24; published 26.09.24.

    Copyright

    ©Hamimatunnisa Johar, Chiew Way Ang, Roshidi Ismail, Zaid Kassim, Tin Tin Su. Originally published in JMIR Public Health and Surveillance (https://publichealth.jmir.org), 26.09.2024.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Public Health and Surveillance, is properly cited. The complete bibliographic information, a link to the original publication on https://publichealth.jmir.org, as well as this copyright and license information must be included.