Factors Associated With Achievement of Blood Pressure, Low-Density Lipoprotein Cholesterol (LDL-C), and Glycemic Targets for Primary Prevention of Cardiovascular Diseases Among High Cardiovascular Risk Malaysians in Primary Care

Abstract

Introduction:

Cardiovascular diseases (CVD) remain the world’s leading cause of death. About half of Malaysian adults have at least 2 risk factors; thus, rigorous primary preventions are crucial to prevent the first cardiovascular (CV) event. This study aimed to determine the achievement of treatment targets and factors associated with it among high CV risk individuals.

Methods:

This cross-sectional study included 390 participants from a primary care clinic in Selangor, Malaysia, between February and June 2022. The inclusion criteria were high-CV risk individuals, that is, Framingham risk score >20%, diabetes without target organ damage, stage 3 kidney disease, and very high levels of low-density lipoprotein cholesterol (LDL-C) >4.9 mmol/L or blood pressure (BP) >180/110 mmHg. Individuals with existing CVD were excluded. The treatment targets were BP <140/90 mmHg (≤135/75 for diabetics), LDL-C <2.6 mmol/L, and HbA1c ≤6.5%. Multiple logistic regressions determined the association between sociodemographic, clinical characteristics, health literacy, and medication adherence with the achievements of each target.

Results:

About 7.2% achieved all treatment targets. Of these, 35.1% reached systolic and diastolic (46.7%) BP targets. About 60.2% and 28.2% achieved optimal LDL-C and HbA1c, respectively. Working participants had lower odds of having optimal systolic (aOR = 0.34, 95% CI: 0.13-0.90) and diastolic (aOR = 0.41, 95% CI: 0.17-0.96) BP. Those who adhered to treatments were more likely to achieve LDL-C and HbA1c targets; (aOR = 1.72, 95% CI: 1.10-2.69) and (aOR = 2.46, 95% CI: 1.25-4.83), respectively.

Conclusions:

The control of risk factors among high CV risk patients in this study was suboptimal. Urgent measures such as improving medication adherence are warranted.

Introduction

Cardiovascular diseases (CVD) remain the leading cause of mortality, accounting for 32% of deaths globally. ¹ Ischemic heart disease is responsible for 13.7% of deaths among Malaysian adults. ² Almost half (43.2%) of Malaysians have a combination of at least 2 cardiovascular (CV) risk factors, such as smoking, obesity, hypercholesterolemia, hypertension, and diabetes. ³

Individuals who have not had CV events but have a clustering of risk factors, such as hypertension, diabetes, and dyslipidemia are considered high CV-risk individuals. These individuals are vulnerable and require intensive primary prevention therapy to prevent their first CV event. Various tools and algorithms, such as the Framingham Risk Score (FRS), Revised Pooled Cohort Equations (RPCE), and Systematic COronary Risk Evaluation 2 (SCORE 2) incorporate CV risk factors to stratify individuals into different risk categories.^{4 -6} Recent literature discovered that FRS and RPCE have good discrimination and are clinically useful in predicting cardiovascular events among Malaysians with clustering of CV risk factors. ⁷ Malaysian Clinical Practice Guidelines on the Management of Primary and Secondary Prevention of Cardiovascular Diseases 2017 recommended utilizing FRS to stratify individuals into CV risk categories. A high-risk individual includes those who have more than 20% of 10-year CV risk calculated using FRS, diabetic patients without target organ damage, patients with stage 3 chronic kidney disease, and individuals with very high levels of an individual risk factor, that is, low-density lipoprotein (LDL-C) >4.9 mmol/L or blood pressure (BP) >180/110 mmHg). ⁸ The risk factors in these patients need to be treated aggressively using pharmacotherapy as well as incorporating therapeutic lifestyle intervention to achieve their treatment targets.

Previous literature has shown various findings regarding achieving treatment targets among patients with high CV risk. Kotseva et al ⁹ reported that about 47% and 46.9% of high CV-risk individuals on treatment achieved BP and LDL-C targets, respectively. The findings were worse in Malaysia, where only 37.4% achieved the BP target, and 18.1% of diabetic patients achieved an HbA1c of less than 6.5%.^10,11 As for LDL-C, only 16.1% of high CV-risk patients achieved their LDL-C target. ¹²

Several factors have been identified to be associated with patients not achieving their treatment targets, including patients’ sociodemographic characteristics, such as age and gender, as well as clinical factors, such as obesity and comorbidities.^{13 -16} In addition, non-adherence to medications and low health literacy were also associated with poor disease control and overall health outcomes.^17,18

In Malaysia, there have been many studies concerning the control of individual risk factors, such as hypertension and diabetes.^10,11,13 However, they were conducted without stratifying individual patients’ CV risk. The literature about the achievements of all treatment targets (BP, glycemic control, and LDL-C) among individuals with high CV risk remains limited. Thus, this study aimed to determine the prevalence of treatment target achievement for BP, glycemic control, and LDL-C, and their associated factors, among high CV risk patients in a primary care setting.

Methods

This cross-sectional study was conducted at a university primary care clinic in Selangor, a central state in Malaysia. This clinic provides acute medical care, chronic disease follow-up, and preventative care to approximately 347,092 population in the Sungai Buloh district in Malaysia. ¹⁹ The sampling frame included patients attending the non-communicable disease (NCD) clinic follow-ups at this primary care clinic between February 2022 and June 2022. Convenience sampling method was used until the target sample size was achieved. This method was chosen due to the absence of an NCD registry, such as diabetes or hypertension.

Inclusion and Exclusion Criteria

The inclusion criteria were patients aged more than 18 years old, attended clinics at least twice in the last 12 months with fasting blood test results (lipid profile, HbA1c) and had high CV risk, as defined by at least one of these criteria: (a) >20% of 10-year CV risk calculated using FRS, (b) individuals with very high levels of individual risk factors (LDL-C >4.9 mmol/L and/or BP >180/110 mmHg), (c) stage 3 chronic kidney disease (eGFR 30-59 mL/min per 1.73 m²), or (d) diabetic patients without documented proteinuria or retinopathy. ⁸ The exclusion criteria were those with existing CVD and target organ damage (coronary artery disease, stroke, peripheral vascular disease, and diabetes with proteinuria), who were pregnant or had cognitive impairment or a mental health disorder which would impair their ability to consent or answer the questionnaire reliably, and those with an acute medical condition, such as a hypertensive emergency.

Variable Definition

Hypertension, diabetes, and dyslipidemia were defined based on their clinician’s diagnosis and/or if patients were taking any antihypertensive, anti-diabetic treatment (oral hypoglycemic agent, insulin), lipid-lowering medications, respectively. A smoker is defined as those who currently smokes any tobacco product. Previous smokers were those who had stopped smoking for more than 5 years, and non-smokers were defined as those who had never smoked. Self-management booklet or mobile application use was defined as participants’ self-reported utilization of EMPOWER-SUSTAIN^® booklet and mobile application as self-empowerment tools to manage their CV risk factors. ²⁰

For this study, the treatment targets for these high-risk participants were defined according to the Malaysian Clinical Practice Guidelines on Management of Primary and Secondary Prevention of Cardiovascular Diseases, as per the following: BP of <140/90 mmHg (≤135/75 mmHg for diabetics), LDL-c target of <2.6 mmol/L, and HbA1c ≤6.5%). ⁸ The outcomes of this study were documented using parameters from their current clinic visit. The best BP on the day, and the latest LDL-C and HbA1c readings from the EMR were used as outcomes for this study.

Study Tool

Health literacy (HL) was assessed using the Short-Form Health Literacy Instrument (HLS-SF-12). ²¹ The HSL-SF-12 has 12 items, and each item was scored using a Likert scale from 1 to 4 (1: very difficult, 2: moderately difficult, 3: fairly difficult, and 4: very easy). The HL score range from 0 to 50 and was determined by the formula: (mean –1) × (50/3), where the mean is the mean of all participating items for each individual. The score determines the HL level: Inadequate (0-25), Problematic (>25-33), Sufficient (>33-42), and Excellent (>42-50). The “inadequate” and “problematic” categories were combined to form a “limited” HL level (score 0-33), while “sufficient” and “excellent” were combined to form an “adequate” HL level (score >33-50). It is valid and reliable for Malaysians with Cronbach’s alpha of .85. ²¹

Medication adherence was assessed using the 12-item Malaysian Medication Adherence Assessment Tool (MyMAAT). ²² This tool has good internal consistency (Cronbach’s α = .91). The items were scored on a 5-point Likert scale, ranging from “strongly disagree” (5) to “strongly agree” (1). The minimum score was 12, and the maximum score was 60. A score from 12 to 53 is considered non-adherence, while a score from 54 to 60 is considered adherence. ²² The original authors have granted permission to use these questionnaires in this study.

Study Conduct: Recruitment and Data Collection Procedures

Patients attending NCD follow-ups were screened for eligibility through face-to-face interviews at the clinic waiting area. Their eligibility was confirmed through an EMR review. They were stratified according to FRS via reviews of electronic medical record, using parameters (blood pressure and lipids profile) from their previous clinic visit. Patients who were eligible and agreed to participate were given a study information sheet, and written consent was obtained. Each patient was allocated a blinded research identification number to preserve confidentiality. Demographic and medical history were gathered through a self-administered questionnaire and the review of EMR. The questionnaire was checked for completeness before the participants left the clinic. Their clinical information, such as blood test results, was retrieved from the EMR.

Anthropometry measurement was performed by trained staff. Height in meters (m) and weight in kilograms (kg) were measured using a standardized stadiometer (seca 787). The body mass index (BMI) was calculated manually. BP measurements were done twice, 2 minutes apart, on the right arm supported at the level of the heart, in a sitting position using Omron automatic digital BP monitor (Omron HBP-1100). The best measurement from the 2 readings was used.

Sample Size Calculation

Based on a study by Kotseva et al, ⁹ among high-risk and treated patients, 47% achieved the BP target, 46.9% achieved the LDL-C target, and 65.2% achieved the HbA1c target. Based on this, using the single proportion formula, taking an α value of .05 with absolute precision of 5%, the minimum calculated sample sizes were 383, 383, and 349, respectively.

Statistical Analysis

The normally distributed descriptive data were expressed as mean with standard deviation (SD), and the non-normally distributed data as median with interquartile range (IQR). Categorical variables were described in numbers and percentages. Simple logistic regression was initially performed to determine the factors associated with achieving the treatment target. Variables with a P-value of <0.25 were included in the multiple logistic regressions to determine the associated factors after adjusting for the confounders. P-values of <0.05 was considered significant. Achievements of BP, LDL-C, or HbA1c targets were the dependent variables. The factors associated with the achievements of these treatment targets were the independent variables. Multiple logistic regression analyses were conducted for each of the dependent variables (BP, LDL-C, and HbA1c).

Results

390 participants were eligible and agreed to enrol in this study. The median (IQR) age was 63 (9) years old. There was almost equal distribution between males (50.8%) and females (49.2%). As for clinical characteristics, more than half were obese (55.6%), had diabetes (78.5%), hypertension (81.5%), and dyslipidemia (96.2%). The median (IQR) HbA1c was 7.0 (1) %, the mean (SD) systolic BP was 141.8 (14.6) mmHg, and the median (IQR) LDL-C was 2.3 (1.0) mmol/L. The sociodemographic and clinical characteristics of participants are shown in Tables 1 and 2.

Table 1.

Socio-demographic Characteristics of Participants, n = 390.

Sociodemographic characteristics
Age (years) [median (IQR)]	63 (9)
Age groups (years) (n, %)
≤50	33 (8.5)
51-60	101 (25.9)
>60	256 (65.6)
Gender (n, %)
Male	198 (50.8)
Female	192 (49.2)
Ethnicity (n, %)
Malay	331 (85.1)
Chinese	24 (6.2)
Indian	28 (7.2)
Others/Indigenous	6 (1.5)
Education attainment (n, %)
No formal education/primary school	26 (6.7)
Secondary school	199 (51)
Tertiary education	165 (42.3)
Marital status (n, %)
Married	357 (91.5)
Single/widowed/divorced	33 (8.5)
Occupation (n, %)
Pensioner/homemaker/unemployed	291 (74.6)
Technician/armed forces/self-employed/elementary	51 (13.1)
Managerial and professional	48 (12.3)

Table 2.

Clinical Characteristics of Participants, n = 390.

Clinical characteristics
Smoking status (n, %)
Non-smoker	293 (75.1)
Previous smoker	67 (17.2)
Current smoker	30 (7.7)
Body mass index (n, %)
Underweight (<18.5 kg/m2)	4 (1.0)
Normal (18.5-22.9 kg/m2)	23 (6.0)
Overweight (23-27.4 kg/m2)	143 (37.3)
Obese (≥27.5 kg/m2)	213 (55.6)
Comorbidities: diabetes (n, %)	306 (78.5)
HbA1c (%) [median (IQR)]	7.0 (1.0)
Treatment with oral hypoglycemic agent (n, %)
No medication	105 (26.9)
1 medication	141 (36.2)
2 medications	112 (28.7)
≥ 3 medications	32 (8.2)
Treatment with metformin (n, %)	280 (71.8)
Treatment with sulfonylureas (n, %)	112 (28.7)
Treatment with dipeptidyl peptidase 4 (DPP-4) inhibitors (n, %)	46 (11.8)
Treatment with sodium-glucose cotransporter-2 (SGLT2) inhibitors (n, %)	27 (6.9)
Treatment with insulin (n, %)	54 (13.8)
Comorbidities: hypertension (n, %)	318 (81.5)
Systolic blood pressure (mmHg) [mean (SD)]	141.8 (14.6)
Diastolic blood pressure (mmHg) [mean (SD)]	78.9 (10.3)
Treatment with antihypertensive (n, %)
No medication	78 (20.0)
1 medication	119 (30.5)
2 medications	134 (34.4)
≥ 3 medications	59 (15.1)
Treatment with angiotensin receptor blockers (ARBs) or angiotensin-converting enzyme (ACE) inhibitors (n, %)	241 (61.8)
Comorbidities: dyslipidemia (n, %)	375 (96.2)
Low-density lipoprotein cholesterol (LDL-C; mmol/L) [median (IQR)]	2.3 (1.0)
Treatment with lipid-lowering medications (n, %)
No medication	25 (6.4)
1 medication	343 (87.9)
2 medications	22 (5.6)
Treatment with statin (n, %)	360 (92.3)
Polypharmacy (n, %)
No (<5 total medications)	281 (72.1)
Yes (≥ 5 total medications)	109 (27.9)
Family history (n, %)
Premature cardiovascular disease (age < 45 years old)	19 (4.9)
Diabetes	244 (62.6)
Hypertension	241 (61.8)
Dyslipidemia	132 (33.8)
Self-management booklet/mobile application use (n, %)	225 (57.7)
Health literacy (n, %)
Limited	133 (34.1)
Adequate	257 (65.9)
Medication adherence (n, %)
Non-adherent	143 (36.7)
Adherent	247 (63.3)

In terms of treatment targets, only 28 participants (7.2%) achieved all treatment targets. Achievements of individual treatment targets are shown in Figure 1.

Figure 1.

Achievements of treatment targets.

After adjusting for confounders, current smokers were more likely to achieve the systolic BP target (aOR = 3.39, 95% CI: 1.30-8.85). Participants taking 1 or 2 antihypertensives also had lower odds of achieving systolic BP targets (Table 3).

Table 3.

Factors Associated With Achieving Systolic Blood Pressure Target ^a .

Variables	Crude OR (95% CI)	P-value	Adjusted OR (95% CI)	P-value
Age groups (years)
≤50	1		1
51–60	0.76 (0.34-1.67)	.487	0.66 (0.25-1.72)	.397
<60	0.58 (0.28-1.20)	.140	0.51 (0.17-1.53)	.233
Gender
Male	1		1
Female	0.43 (0.28-0.66)	<.001	0.61 (0.34-1.09)	.094
Ethnicity
Non-Malay	1
Malay	0.87 (0.49-1.55)	.639	—	—
Education attainment
No formal education/primary school	1		1
Secondary school	1.58 (0.61-4.13)	.350	1.23 (0.45-3.39)	.683
Tertiary education	2.28 (0.87-5.97)	.094	1.49 (0.53-4.20)	.450
Marital status
Single/widowed/divorced	1
Married	1.49 (0.67-3.30)	.326	—	—
Occupation
Pensioner/homemaker/unemployed	1		1
Technician/arm-forces/self-employed/elementary	0.75 (0.39-1.44)	.383	0.34 (0.13-0.90)	.029
Managerial and professional	2.14 (1.16-3.96)	.015	1.55 (0.66-3.62)	.316
Family history premature cardiovascular disease (age < 45 years old)
No	1		—
Yes	0.85 (0.31-2.28)	.740	—
Family history hypertension
No	1		1
Yes	0.63 (0.42-0.97)	.036	0.62 (0.38-1.01)	.54
Smoking status
Non-smoker	1		1
Previous smoker	1.89 (1.10-3.25)	.022	1.50 (0.77-2.95)	.237
Current smoker	4.02 (1.84-8.81)	<.001	3.39 (1.30-8.85)	.013
Body mass index (kg/m2)
Underweight (<18.5)/normal (18.5-22.9)	1		1
Overweight (23-27.4)	2.03 (0.86-4.81)	.107	1.85 (0.72-4.74)	.202
Obese (≥27.5)	1.73 (0.75-4.01)	.201	1.65 (0.66-4.12)	.281
Self-management booklet/mobile application use
No	1		1
Yes	0.72 (0.48-1.10)	.131	0.83 (0.52-1.33)	.437
Treatment with antihypertensive
No medication	1		1
1 medication	0.40 (0.22-0.73)	.002	0.39 (0.20-0.76)	.005
2 medications	0.32 (0.18-0.57)	<.001	0.32 (0.16-0.65)	.002
≥3 medications	0.47 (0.24-0.95)	.035	0.59 (0.22-1.58)	.290
Polypharmacy
No (<5 total medications)	1		1
Yes (≥5 total medications)	0.66 (0.41-1.06)	.086	1.17 (0.58-2.35)	.657
Health literacy
Limited	1
Adequate	0.99 (0.64-1.53)	.95	—	—
Medication adherence
Non-adherent	1
Adherent	1.01 (0.66-1.56)	.96	—	—

1 = Reference group. Emboldened: Significant at P < .05.

^aSystolic Blood Pressure ≤135 mmHg for diabetes, <140 mmHg for non-diabetes.

Variable with P < .25 were included in multiple logistic regression: age, gender, education level, occupation, family history of hypertension, smoking status, body mass index, using self-management app/booklet, and treatment with antihypertensive and polypharmacy. Model fits the data well (Hosmer Lemeshow goodness of fit test P = .178). Cox & Snell R² = 12.8%, Nagelkerke R² = 17.8%. All assumptions (interaction, multicollinearity) were met.

Those who worked as a technician, arm-forces, self-employed, or in an elementary role were less likely to achieve both systolic (aOR = 0.34, 95% CI: 0.13-0.90) and diastolic (aOR = 0.41, 95% CI: 0.17-0.96) BP targets, compared to those who were pensioner, homemaker, or unemployed (Tables 3 and 4). Furthermore, obese participants also had lower odds of achieving their diastolic BP target (aOR = 0.42, 95% CI: 0.19-0.91).

Table 4.

Factors Associated With Achieving Diastolic Blood Pressure Target ^a .

Variables	Crude OR (95% CI)	P-value	Adjusted OR (95% CI)	P-value
Age groups (years)
≤50	1		1
51-60	1.22 (0.52-2.85)	.646	0.92 (0.36-2.35)	.868
< 60	2.65 (1.21-5.79)	.015	1.56 (0.55-4.37)	.401
Gender
Male	1
Female	0.86 (0.58-1.28)	.465	—	—
Ethnicity
Non-Malay	1
Malay	0.93 (0.53-1.63)	.805	—	—
Education attainment
No formal education/primary school	1
Secondary school	0.82 (0.36-1.85)	.625	—	—
Tertiary education	0.65 (0.28-1.49)	.305	—	—
Marital status
Single/widowed/divorced	1
Married	0.92 (0.45-1.89)	.827	—	—
Occupation
Pensioner/homemaker/unemployed	1		1
Technician/arm-forces/self-employed/ elementary	0.26 (0.13-0.52)	<.001	0.41 (0.17-0.96)	.041
Managerial and professional	0.73 (0.40-1.35)	.318	1.22 (0.56-2.66)	.612
Family history: premature cardiovascular disease (age < 45 years old)
No	1
Yes	1.61 (0.63-4.09)	.318	—	—
Family history: hypertension
No	1		1
Yes	0.75 (0.50-1.14)	.177	0.75 (0.48-1.18)	.214
Smoking status
Non-smoker	1
Previous smoker	1.12 (0.66-1.91)	.675	—	—
Current smoker	0.88 (0.41-1.88)	.747	—	—
Body mass index (kg/m2)
Underweight (<18.5)/normal (18.5-22.9)	1		1
Overweight (23-27.4)	0.42 (0.19-0.93)	.032	0.46 (0.20-1.03)	.060
Obese (≥ 27.5)	0.36 (0.17-0.78)	.010	0.42 (0.19-0.91)	.029
Self-management booklet/mobile application use
No	1
Yes	1.14 (0.76-1.70)	.538	—	—
Treatment with antihypertensive
No medication	1		1
1 medication	1.28 (0.72-2.28)	.405	1.16 (0.62-2.18)	.634
2 medications	1.44 (0.82-2.53)	.207	1.42 (0.76-2.66)	.277
≥3 medications	1.21 (0.61-2.40)	.579	1.24 (0.58-2.64)	.585
Polypharmacy
No (<5 total medications)	1
Yes (≥5 total medications)	0.82 (0.53-1.28)	.382	—	—
Health literacy
Limited	1		1
Adequate	0.73 (0.48-1.11)	.138	0.75 (0.48-1.17)	.200
Medication adherence
Non-adherent	1
Adherent	1.08 (0.72-1.63)	.715	—	—

1 = Reference group. Emboldened: Significant at P < .05.

^aDiastolic nlood pressure ≤75 mmHg for diabetes and <90 mmHg for non-diabetes.

Variable with P < .25 were included in multiple logistic regression: age, occupation, family history of hypertension, body mass index, treatment with antihypertensive, and health literacy. Model fits the data well (Hosmer Lemeshow goodness of fit test P = .442). Cox & Snell R² = 7.4%, Nagelkerke R² = 9.9%. All assumptions (interaction, multicollinearity) were met.

Those who were previous smokers had lower odds of achieving LDL of <2.6 mmol/L (aOR = 0.54, 95% CI: 0.31-0.94; Table 5). Participants who adhered to their medications were more likely to achieve their LDL-C and HbA1c targets; (aOR = 1.72, 95% CI: 1.10-2.69) and (aOR = 2.46, 95% CI: 1.25-4.83), respectively (Tables 5 and 6).

Table 5.

Factors Associated With Achieving Low-Density Lipoprotein Cholesterol (LDL-C) <2.6 mmol/L.

Variables	Crude OR (95% CI)	P-value	Adjusted OR (95% CI)	P-value
Age groups (years)
≤ 50	1		1
51-60	1.66 (0.75-3.69)	.215	1.79 (0.73-4.38)	.200
<60	1.86 (0.89-3.89)	.100	1.90 (0.70-5.14)	.205
Gender
Male	1
Female	0.94 (0.63-1.41)	.757	—	—
Ethnicity
Non-Malay	1
Malay	1.07 (0.61-1.89)	.809	—	—
Education attainment
No formal education/primary school	1
Secondary school	0.73 (0.31-1.71)	.461	—	—
Tertiary education	0.87 (0.37-2.07)	.755	—	—
Marital status
Single/widowed/divorced	1		1
Married	1.92 (0.94-3.93)	.075	1.85 (0.87-3.92)	.111
Occupation
Pensioner/homemaker/unemployed	1		1
Technician/arm-forces/self-employed/elementary	0.64 (0.35-1.16)	.139	0.84 (0.37-1.93)	.688
Managerial and professional	1.16 (0.61-2.19)	.652	1.43 (0.64-3.23)	.386
Family history of premature cardiovascular disease (age < 45 years old)
No	1		1
Yes	0.58 (0.23-1.46)	.248	0.49 (0.19-1.30)	.152
Family history of dyslipidemia
No	1
Yes	0.92 (0.60-1.41)	.69	—
Smoking status
Non-smoker	1		1
Previous smoker	0.51 (0.30-0.87)	.014	0.54 (0.31-0.94)	.029
Current smoker	1.13 (0.51-2.51)	.772	1.23 (0.50-3.02)	.649
Body mass index (kg/m2)
Underweight (<18.5)/normal (18.5-22.9)	1
Overweight (23-27.4)	0.92 (0.43-1.95)	.817	—	—
Obese (≥ 27.5)	1.18 (0.56-2.46)	.663	—	—
Self-management booklet/mobile application use
No	1		1
Yes	1.46 (0.97-2.20)	.070	1.50 (0.97-2.30)	.068
Treatment with lipid-lowering medication
No medication	1		1
1 medication	2.95 (1.27-6.87)	.012	1.02 (0.13-7.84)	.983
2 medications	1.96 (0.60-6.39)	.267	0.82 (0.09-7.58)	.858
Treatment with statin
No	1		1
Yes	2.83 (1.31-6.13)	.008	2.76 (0.43-17.81)	.287
Polypharmacy
No (<5 total medications)	1
Yes (≥5 total medications)	1.24 (0.79-1.97)	.351	—	—
Health literacy
Limited	1
Adequate	0.95 (0.62-1.46)	.828	—	—
Medication adherence
Non-adherent	1		1
Adherent	1.77 (1.16-2.69)	.008	1.72 (1.10-2.69)	.017

1 = Reference group. Emboldened: Significant at P < .05.

Variable with P < .25 were included in multiple logistic regression: age, marital status, occupation, family history of premature cardiovascular disease, smoking status, using self-management app/booklet, treatment with lipid-lowering medication, treatment with statin, and medication adherence. Model fits the data well (Hosmer Lemeshow goodness of fit test P = .321). Cox & Snell R² = 7.9%, Nagelkerke R² = 10.7%. All assumptions (interaction, multicollinearity) were met.

Table 6.

Factors Associated With Achieving HbA1c ≤ 6.5% ^a .

Variables	Crude OR (95% CI)	P-value	Adjusted OR (95% CI)	P-value
Age groups (years)
≤50	1		1
51-60	4.88 (1.07-22.13)	.040	6.29 (1.17-34.00)	.033
<60	5.34 (1.22-23.38)	.026	3.69 (0.64-21.40)	.145
Gender
Male	1
Female	1.13 (0.67-1.88)	.652	—	—
Ethnicity
Non-Malay	1
Malay	1.27 (0.62-2.56)	.515	—	—
Education attainment
No formal education/primary school	1
Secondary school	1.16 (0.40-3.35)	.781	—	—
Tertiary education	1.66 (0.57-4.82)	.352	—	—
Marital status
Single/widowed/divorced	1
Married	0.64 (0.28-1.46)	.290	—	—
Occupation
Pensioner/homemaker/unemployed	1		1
Technician/arm-forces/self-employed/elementary	0.45 (0.19-1.07)	.071	0.46 (0.15-1.46)	.187
Managerial and professional	0.63 (0.27-1.45)	.276	0.82 (0.28-2.45)	.725
Family history premature of cardiovascular disease (age < 45 years old)
No	1		1
Yes	0.38 (0.08-1.70)	.205	0.44 (0.08-2.44)	.351
Family history of diabetes
No	1		1
Yes	0.69 (0.40-1.19)	.183	0.81 (0.42-1.56)	.522
Smoking status
Non-smoker	1
Previous smoker	1.48 (0.71-3.07)	.296	—	—
Current smoker	1.36 (0.49-3.78)	.551	—	—
Body mass index (kg/m2)
Underweight (<18.5)/normal (18.5-22.9)	1
Overweight (23-27.4)	1.03 (0.41-2.61)	.948	—	—
Obese (≥ 27.5)	0.79 (0.32-1.93)	.600	—	—
Self-management booklet/mobile application use
No	1
Yes	0.90 (0.54-1.50)	.686	—
Treatment with oral hypoglycemic agent
No medication	1		1
1 medication	0.52 (0.20-1.30)	.161	0.46 (0.15-1.36)	.159
2 medications	0.10 (0.04-0.29)	<.001	0.10 (0.03-0.36)	<.001
≥3 medications	0.08 (0.02-0.34)	<.001	0.09 (0.02-0.48)	.005
Treatment with insulin
No	1		1
Yes	0.04 (0.01-0.27)	.001	0.03 (0.004-0.25)	<.001
Polypharmacy
No (<5 total medications)	1		1
Yes (≥5 total medications)	0.31 (0.17-0.58)	<.001	0.67 (0.32-1.42)	.297
Health literacy
Limited	1		1
Adequate	0.60 (0.36-1.01)	.053	0.57 (0.29-1.12)	.102
Medication adherence
Non-adherent	1		1
Adherent	1.68 (0.98-2.90)	.062	2.46 (1.25-4.83)	.009

1 = Reference group. Emboldened: Significant at P < .05.

Variable with P < .25 were included in multiple logistic regression: age, occupation, family history of premature cardiovascular disease, family history of diabetes, treatment with oral hypoglycemic agents, treatment with insulin, polypharmacy, health literacy, and medication adherence. Model fits the data well (Hosmer Lemeshow goodness of fit test P = .278). Cox & Snell R² = 25.8%, Nagelkerke R² = 37.1%. All assumptions (interaction, multicollinearity) were met.

^aFor diabetes only.

Participants aged between 51 and 60 years old were more likely to achieve HbA1c of ≤6.5% (aOR=6.29, 95% CI: 1.17-34.00), compared to those who were 50 years or younger. Other factors associated with less odds of achieving glycemic target were taking 2 (aOR = 0.10, 95% CI: 0.03-0.36), 3 or more (aOR = 0.09, 95% CI: 0.02-0.48) OHA and being treated with insulin (aOR = 0.03, 95% CI: 0.004-0.25; Table 6).

Discussion

International and local clinical practice guidelines outline strict BP, LDL-C, and glycemic control among patients with high-CV risk based on findings from convincing and strong levels of evidence.^8,23 This study discovered that only a small proportion (7.2%) of these high-CV risk participants from primary care achieved all their targets. Another study in Norway, although it had different cut-off points and study outcomes, discovered slightly better results where 9.8% achieved all targets, which comprised of BP <140/90 (<135/85 for diabetics) mmHg, total cholesterol <5mmol/L, LDL-C <3 mmol/L, and smoking cessation. ²⁴ Other existing literature reported on the achievement of individual treatment targets, such as BP, lipid, and glycemic targets, and findings on the achievements of all these treatment targets remain scarce locally and internationally.^{9 -12,25} Yusuf et al ²⁶ discovered that smoking, hypertension, diabetes, and dyslipidemia attributed to about 75.8% risk of acute myocardial infarction among their study population. Thus, optimizing management of all these risk factors is paramount to prevent patient’s first CV event.

The local Malaysian guideline advocated target BP of <140/90 for most patients and ≤135/75 mmHg for those with diabetes. ⁸ Approximately a third of the participants from this study achieved their systolic BP (35.1%) and diastolic BP (46.7%) targets. Compared to other Asian countries, BP control ranged widely across Asia, from 5.5% in Pakistan to 70% in Taiwan. ²⁷ Focusing on high-CV risk populations, a European study reported that about 63.4% achieved their BP target. ²⁵ Chung et al ²⁸ studied the time spent at BP target among primary prevention patients and discovered that those who maintained their target BP for the duration of 6 to 8.9 months per year had 78% lower odds of myocardial infarction, stroke, and cardiovascular death, compared to those who never reached their targets. These findings confirm the need to treat BP to target to minimize adverse CVD outcomes among patients.

As for factors associated with the achievement of systolic BP targets, this study discovered an association between occupation and the achievement of BP targets. Research from the United States showed that protective service workers, such as police officers, had higher rates of hypertension and were less likely to achieve control ²⁹ ; findings shared with our study showed that the middle-occupation category, including armed forces, had lesser odds of achieving systolic and diastolic BP target. Various hypotheses, including job stress, have been attributed to poor BP control among them. ²⁹ This study also found that current smokers had higher odds of achieving systolic BP targets, parallel with previous findings, which showed current smokers having lower BP. ³⁰ Nevertheless, smoking is a strong risk factor for CVD ⁸ ; therefore, it should not be advocated as a measure to achieve BP control among patients. Another finding from this study was the significant association between obesity and achievement of diastolic BP target. The participants who were obese had lower odds of achieving diastolic BP target. This finding is supported by previous literature which showed various mechanism in which obesity can contribute to hypertension, including compression of the kidney by the surrounding fat, leading to the activation of the renin-angiotensin-aldosterone system, as well as an increased in the activity of sympathetic nervous system. ³¹ Thus, weight management counseling should be actively implemented by the primary care providers as part of the holistic management of hypertension.

This research discovered that 60.2% of its participants attained LDL-C of <2.6 mmol/L, while another Malaysian community-based study reported that a smaller proportions (16.1%) of its high-risk participants achieved the target. ¹² International studies focusing on the LDL-C target achievements among high-risk individuals discovered varying findings, where only 16.8% of high-risk individuals in Hungary achieved LDL-C. ²⁵ In contrast, the EUROSPIRE V study discovered that 46.9% of high-risk participants treated with lipid-lowering medications achieved their LDL-C target. ⁹ Bruckert et al, ³² in their systematic review of European populations, found that approximately 46% of high-risk individuals attained their LDL-C target. The 2019 European Society of Cardiology and European Atherosclerosis Society Guideline recommend an even lower LDL-C target of <1.4 mmol/L for those with very high CV risk and <1.8 mmol/L for high-risk individuals. ²³ Recent literature, however, debated whether individuals with lower LDL-C had better CV outcomes, where some showed continued effectiveness, even with LDL-C <1 mmol/L (40 mg/dL) for patients with high CV risk. ³³ At the same time, Rong et al ³⁴ found the contrary, in which those with LDL-C lower than 1.8 mmol/L (70 mg/dL) had higher odds of developing CVD mortality with a hazard ratio (HR) of 1.60 (95% CI, 1.01-2.54), compared to those with LDL-C of 2.59 mmol/L (100 mg/dL) to 3.36 mmol/L (129.9 mg/dL). In the same study, Rong et al ³⁴ also concluded that those with very high LDL-C of more than 4.91 mmol/L (190 mg/dL) had a higher risk of CVD death. Thus, the Malaysian guideline on dyslipidemia management’s recommendation for an LDL-C target of less than 2.6 mmol/L is appropriate for primary prevention among patients with high CV risk until further research can elucidate the optimal LDL-C levels among the various risk categories. In terms of factors associated with the achievement of the LDL-C target, a positive association between medication adherence and achievement of LDL-C was discovered in this study, consistent with previous literature, which showed that those who adhered to medications had a higher likelihood of reaching their optimal LDL-C. ³⁵ Lastly, previous smokers were also found to have lower odds of achieving LDL-c target, in line with findings from previous literature which showed that former smokers, along with active smokers, were more likely to have dyslipidemia compared to non-smokers. ³⁶ This knowledge further affirms existing evidence on the magnitude of negative effects that smoking has on individuals.

Diabetes has been recognized as a strong CVD risk factor, in which those with diabetes have a high risk of developing CVD. ⁸ The risk becomes more prominent if they develop any signs of target organ damage, such as retinopathy or proteinuria. These individuals will be considered to be in the very high-risk group, even without coronary heart disease or stroke. ⁸ The glycemic control for each individuals with diabetes is set according to his or her comorbidities and risk of developing hypoglycemia. ⁸ A strict HbA1c target of less than 6.5% is appropriate for participants in this study as they have yet to develop any CVD. The glycemic control among the participants with diabetes in this study was poor, where only 28.2% achieved HbA1c of less than 6.5%, and their median (IQR) HbA1c was 7.0 (1.0) %. Like LDL-C, medication adherence was also positively associated with achieving glycemic targets among participants in this study, consistent with previous literature. ³⁷ Furthermore, it was also proven to reduce hospitalization and visits to the emergency department. ³⁸ Thus, improving medication adherence should be prioritized to achieve optimal glycemic control among these populations. Additionally, middle-aged participants (51-60 years old) were more like to have HbA1c of less than 6.5%, compared to those aged 50 years and younger. This finding was consistent with the outcome from a previous study, which concluded that younger age was associated with worse glycemia control. ³⁹ Thus, it is imperative to treat diabetes more aggressively in younger patients to prevent adverse CV outcomes among them. This study also discovered that participants prescribed with 2 or more OHA, or insulin had lower odds of having HbA1c less than 6.5%. The escalation of pharmacotherapy in these patients is needed to achieve optimal glycemic control, in line with the recommendation from clinical practice guidelines to utilize OHA or insulins as a single therapy or in combination to achieve glycemic targets for individual patient. ⁸

Strength, Limitations, and Implications for Clinical Practice and Future Research

The main strength of this research is the discovery of achievement of all treatment targets among high-CV risk individuals in Malaysia, which has important clinical implications. The findings showed that further treatment optimization is required to achieve their BP, LDL-C, and glycemic targets. This alarming finding should alert the primary care providers to treat these risk factors aggressively, using appropriate measures to reach the targets. Efforts should be targeted to factors found to be significant in achieving these targets. For example, high job stress and demand among armed forces may contribute to poor BP control. Thus, providing proactive employer-led onsite check-ups and treatment of BP may assist them in reaching their optimal BP target. Measures to improve medication adherence such as educational programs to improve health awareness, knowledge, and self-efficacy should also be implemented. ⁴⁰ Treatment-related factors, such as simplifying the treatment regime with a single-pill combination, should also be used, as it has been shown to improve treatment adherence. ⁴¹

This study has some limitations. The convenience sampling methods may introduce selection bias. However, this was minimized as each consecutive patient was approached and screened for eligibility on data collection days, which occurred daily during the study period. Also, the study was conducted at 1 center. While the results may still be generalizable to primary care clinics with similar sociodemographic characteristics, the findings may not be generalizable to other healthcare facilities in Malaysia. Nevertheless, this research provides a valuable glimpse into the state of risk factor control among high-CV risk patients in a Malaysian primary care setting. Future research using more robust sampling methods representing broader populations should be conducted to improve the generalizability of the findings and subsequently improve the care provided to these high-CV risk populations. This study did not include some essential variables, such as dietary habits and physical activities, due to limited human and financial resources. The results should thus be interpreted in this context. Future studies should consider these factors in their multiple logistic regression.

Conclusion

Management of CVD risk factors among patients at high CV risk is still suboptimal. Immediate actions, such as measures to improve medication adherence, are needed to improve cardiovascular outcomes among these high-risk individuals.