Anthropometric Indices Associated With Elevated Estimated 10-year Risk of Cardiovascular Disease Among Adults in the Kumasi Metropolis of Ghana: A Cross-Sectional Study
Ref. Link: <a href="https://onlinelibrary.wiley.com/doi/10.1002/hsr2.72849">https://onlinelibrary.wiley.com/doi/10.1002/hsr2.72849<a/></br>
ABSTRACT
Background and Aim
Cardiovascular diseases remain a major cause of morbidity and mortality in Ghana. This study aimed to determine the association between anthropometric indices and elevated estimated 10-year cardiovascular disease (CVD) risk among adults in the Kumasi Metropolis.
Methods
This cross-sectional study involved 480 adults aged 40–79 years. Anthropometric indices and lipid profile were determined using standard methods. Blood pressure was measured using an automated sphygmomanometer. The World Health Organization/International Society of Hypertension laboratory-based risk assessment charts were used to estimate 10-year Atherosclerotic Cardiovascular Disease (ASCVD) risk. Estimated CVD risk ≥ 10% was considered elevated. Anthropometric indices were examined for association with elevated risk using chi-square, Fisher's exact and Wilcoxon rank-sum tests. Multivariable modified Poisson regression analysis was used to determine factors associated with elevated estimated CVD risk.
Results
An elevated estimated10-year ASCVD risk was observed in 56.9% (103/181) of males and 38.8% (116/299) of females. Lower education level was associated with elevated CVD risk. Waist-to-hip ratio (WHR) (aPR = 1.36, 95% CI: 1.11–1.68), body shape index (ABSI) (aPR = 1.35, 95% CI: 1.08–1.69), hip circumference (HC) (aPR = 0.98, 95% CI: 0.97–0.99), and mid-arm circumference (MAC) (aPR = 0.95, 95% CI: 0.92–0.98) were associated with elevated estimated CVD risk. After stratifying by sex, WHR remained independently associated with elevated estimated CVD risk in females (aPR = 1.36, 95% CI: 1.01–1.84); as did ABSI (aPR = 1.45, 95% CI: 1.05–2.01) and MAC (aPR = 0.93, 95% CI: 0.88–0.99) in males.
Conclusion
WHR, ABSI and MAC assessments should be integrated into CVD risk evaluation in Ghana, especially in deprived areas where lipid profile is not routinely assessed.
1 Introduction
Cardiovascular diseases (CVDs) are a group of diseases that affect the normal function of the heart and blood vessels, including stroke, coronary heart disease, and heart failure [1]. According to the World Health Organization (WHO), they remain the leading cause of mortality globally and accounted for 32.0% of mortality in 2022. Stroke and heart attack, for example, are currently responsible for 85.0% of all CVD deaths, with the burden severe in low- and middle- income countries (LMICs) [1]. LMICs, including Ghana, account for approximately 75.0% of CVD-related mortality globally [1].
In Ghana, the burden of CVDs is currently rising, and this is driven mostly by stroke and coronary artery disease [2]. Indeed, a recent systematic review and meta-analysis of observational studies reported a stroke prevalence of 7.96% in the country [3]. Another systematic review and meta-analysis that included 16 studies with 58,912 participants from 1954 to 2022 reported that the pooled prevalence of CVD nationally is 10.34% (95% CI: [8.48, 12.20]) [4]. However, with rising risk factors for CVDs in Ghana, including hypertension, diabetes, old age and increased body mass index (BMI), the prevalence of CVDs is expected to increase significantly over the next decade [4]. Nationally, two key systematic reviews and meta-analyses of observational studies have reported a hypertension prevalence of 27.0% (95% CI 24.0%–30.0%) and 30.3% (95% CI 26.1–34.8%) respectively [5, 6]. An analysis of data on hospital-based mortality records indicated that hypertension is currently the leading cause of death among chronic diseases in Ghana [7].
Although efforts to prevent and mitigate the CVDs in Ghana have largely focused on promoting lower-sodium diets and increased physical activity, assessing Atherosclerotic Cardiovascular Disease (ASCVD) risk is also an important strategy for identifying individuals at risk [8]. Several tools have been reported as feasible for assessing CVD risk, including mobile-based applications for Pooled Cohort Equation (PCE), laboratory-based Framingham Risk Score (FRS), Globorisk and laboratory-based World Health Organization/International Society of Hypertension (WHO/ISH) risk charts [9, 10]. However, these tools have shown varying degrees of agreement when applied in the Ghanaian setting [11]. Currently, the WHO/ISH risk assessment tool is recommended by the CVD treatment guideline for Ghana for assessing and managing CVD risk to ensure uniformity [11, 12]. The WHO/ISH risk score classified 82.0%, 9.4%, and 8.6% of Ghanaians as low-, intermediate-, and high-risk, respectively based on data from the Ghana Heart Study [11]. The WHO/ISH risk assessment tool utilizes parameters including the lipid profile and presence or absence of diabetes, with other factors for the determination of CVD risk. However, these may not always be possible, especially in a low-resource setting like Ghana. Anthropometric indices are non-invasive, low-cost, and can also be used to determine their association with CVD risk.
The use of anthropometric measures as proxies or indicators of CVD risk factors is well established [13, 14]. Consequently, measures of central obesity, including waist circumference (WC), hip circumference (HC), waist-to-hip ratio (WHR), conicity index (CI), waist-to-height ratio (WHtR), abdominal volume index (AVI), body roundness index (BRI), a body shape index (ABSI), and waist-to-height-to-the-power-of-0.5 ratio (WHT.5 R), have been identified as reliable indicators of CVD risk [14-19]. Anthropometric measures, however, differ by gender and geographic location, as well as by the degree of association with CVD risk, as demonstrated by the varying outcomes of previous studies [14-18]. This makes it prudent to investigate anthropometric indices and their association with elevated estimated risk of CVD in specified local populations. Currently, there is a paucity of data on the association between anthropometric indices and CVD risk amongst Ghanaians. This study sought to determine the association between anthropometric indices and elevated estimated 10-year CVD risk among adults in the Kumasi metropolis of Ghana.
2 Methods and Materials
2.1 Study Design and Sampling Technique
The study design was cross-sectional, and our sampling frame consisted of all registered patients at the Centre for Ageing and Elderly Care who met the inclusion criteria. To achieve the required sample size of 457 participants, a simple random sample of 500 eligible patients was selected from the sampling frame using the sample command with the count option in Stata/SE version 17.0 [20]. A random seed (set seed) was specified to ensure reproducibility of the sampling procedure. The 500 eligible participants were randomly selected to ensure adequate statistical power and precision of the study estimates. Selected individuals were contacted and invited to participate in the study. Participants who were unable to report to the hospital but expressed willingness to participate were subsequently visited in their homes, where interviews and study assessments were conducted. A total of 480 participants completed the study and were included in the final analysis (Figure 1). An ethical approval was sought and obtained from the Kwame Nkrumah University of Science and Technology Committee for Human Research and Publication Ethics (CHRPE/AP/035/22) prior to the start of the work. Data were collected over a 9-month period, from 15 February 2022 to 21 November 2022. Participants provided a written informed consent. The study was conducted in compliance with the guidelines stipulated in the Declaration of Helsinki. Informed consent was obtained from all participants who participated in the study.
Details are in the caption following the image
Figure 1
Open in figure viewer
PowerPoint
Flow diagram showing the selection of study participants.
2.2 Inclusion and Exclusion Criteria
Participants were eligible if they were aged ≥ 40 years, were regular outpatient department attendees, and were receiving standard care during routine follow-up visits. Individuals with a history of major cardiovascular events, such as stroke or myocardial infarction, were excluded.
2.3 Study Site
The study was conducted at the Centre for Ageing and Elderly Care at Metro Health Hospital in Kumasi. The Centre for Ageing and Elderly Care at Metro Health Hospital in Kumasi is a private medical facility providing specialist medical and geriatric care as well as services for healthy ageing. Its catchment area is the Kumasi Metropolis which is the capital city of the Ashanti Region of Ghana.
2.4 Sample Size Determination
The sample size required for the study was calculated using the Cochran formula [21].
𝑛=
(𝑍𝛼/2)2𝑝(1−𝑝)
𝑑2
Where n is the minimum sample size required for the study; 𝑍𝛼/2
is the abscissa of the normal curve that cut-off an area at the tail (1.96); p is the estimated proportion of an attribute (elevated 10-year ASCVD risk) that is present in the study population, which is 45.4% [22], d is the desired level of precision (5%). An additional 20% participants were added to allow for withdrawals and missing data. Therefore, the required sample size was 457.
2.5 Data Collection and Study Procedures
Structured questionnaires were administered to participants to collect data on sociodemographic characteristics, history of hypertension or diabetes diagnosis, current medications, including use of statins and acetylsalicylic acid and lifestyle factors such as smoking history.
2.6 Anthropometric, Blood Pressure, and Biochemical Measurements
A trained health care provider measured anthropometric data, including weight, height, WC, HC and blood pressure (BP), using standard protocols. Body weight was measured with participants' excess clothes removed and barefooted using Omron's body composition analyzer (Omron, BF511, Japan). Measurement was recorded to the nearest 0.5 kg. Height was measured using a stadiometer's vertical scale (Seca, Hamburg, Germany) with the participants standing in an erect position against the scale with their heels and buttocks in contact with the scale. This was measured to the closest 0.5 cm. Waist circumference was measured in centimeters (cm) around the smallest area below the rib cage and above the belly button. Hip circumference was measured in cm at the largest circumference between the waist and the knee. Mid-arm circumference (MAC) was also measured in cm at the midpoint between the acromion and olecranon processes on the non-dominant arm, with the arm relaxed. Measurements were done with a non-stretchable and accurately calibrated measuring tape.
Blood pressure was determined using an attended automated sphygmomanometer (Omron M7 series) with properly fitted cuff. Participants were seated with their back supported and both feet flat on the ground. Three BP readings were taken for each participant, with a 3-min rest interval between successive measurements. The mean of the three readings were used for the final analysis. Derived anthropometric indices were calculated using the formulae as described in previous studies [16-19].
2.7 Unit Standardization for Anthropometric Indices
To ensure accuracy, consistency, and comparability with previously published definitions [16-19], all anthropometric measurements were standardized before the computation of derived indices. Directly measured variables were retained in their original units for descriptive analysis, with weight recorded in kilograms (kg), height, WC, HC, and MAC in centimeters (cm), BP in millimeters of mercury (mmHg).
For derived anthropometric indices, unit conversions were performed where required prior to calculation. Specifically, WC and height were converted from centimeters to meters (m) when necessary to align with the original formulations of indices such as ABSI and BRI that are sensitive to unit specification. Indices that incorporated BMI were computed using BMI expressed in kg/m2, with height in meters and weight in kg. This standardization ensured internal consistency across all derived indices and minimized potential computational bias arising from unit misalignment.
WHR=
WC(cm)
HC(cm)
WHtR=
WC(cm)
Height(cm)
BMI=
Weight(kg)
Height(m)2
CI=
WC(m)
0.109×√
Weight(kg)
Height(m)
AVI=
2×WC(cm)2+0.7×[WC(cm)−HC(cm)]2
1000
ABSI=
WC(m)
BMI
2
3
×Height(m)
1
2
BRI=364.2−365.5×
√
√
√
√
√
⎷
1−(
(WC(m)/(2π))2
(0.5×Height(m))2
)
WHT.5R=
WC(cm)
√Height(cm)
Information on fasting lipid profiles was extracted from participants' medical records. The most recent fasting lipid profile result, obtained within 1 month prior to data collection, was used for the analysis.
The estimated 10-year ASCVD risk was calculated using the WHO/ISH laboratory-based risk charts. The laboratory model is based on six variables: diabetes mellitus status, age, sex, smoking status, systolic BP (SBP), and total cholesterol. It stratifies risk as < 5% (green), 5% to < 10% (yellow), 10% to < 20% (orange), 20% to < 30% (red), and ≥ 30% (dark red), corresponding to very low, low, moderate, high, and very high risk, respectively [23]. For the purpose of this study, a cut off of ≥ 10% was indicative of elevated risk [16]. The ≥ 10% threshold was used because WHO cardiovascular risk guidelines commonly categorize individuals with a 10-year ASCVD risk of ≥ 10% as having at least moderate cardiovascular risk that may warrant closer clinical attention and preventive interventions, especially in LMICs, including Ghana [23].
2.8 Data Analysis
Data collected from the study participants were entered into Microsoft Excel and exported to Stata/SE version 17.0 [20] for analysis. Continuous variables were tested for normality using the Shapiro-Wilk test, supported by visual assessment using Q-Q plots or histograms. The characteristics of study participants were summarized using frequencies and percentages for categorical variables, and medians with interquartile ranges for skewed continuous variables. Anthropometric indices were examined for association with elevated estimated ASCVD risk using chi-square or Fisher's exact test where an expected cell frequency was less than five, and Wilcoxon rank-sum test was applied for all skewed continuous variables.
Separate modified Poisson regression models with robust standard errors were fitted for each anthropometric index to assess associations with elevated estimated ASCVD risk. Anthropometric indices were evaluated one-at-a-time because of the strong intercorrelations among adiposity measures. All models [both overall (Figure 1) and sex-stratified (Table 3, Model II)] were adjusted for the same prespecified sociodemographic and clinical covariates, selected for clinical relevance and prior literature [10, 11]. The adjusted models controlled for marital status, diabetes mellitus, hypertension, BP medication, statin initiation, aspirin initiation, and smoking status. Multicollinearity was assessed using the Variance Inflation Factor (VIF), with variables exhibiting VIF values < 5 retained for inclusion in the models (Supplementary File 1). Sensitivity analyses were additionally conducted using alternative thresholds for elevated estimated ASCVD risk (≥ 20% and ≥ 30%) to assess the robustness of observed associations. Separate modified Poisson regression models with robust standard errors were repeated using these alternative categorizations. There were no missing values in the dataset; therefore, all observations were included in the final analyses. Statistical significance was determined at a two-sided p-value < 0.05 or where the 95% CI for the aPR excluded 1.
3 Results
3.1 Socio-Demographic and Clinical Baseline Characteristics of the Study Participants
Median age (IQR) of males (n = 181) was 59 years (50, 67) and females (n = 299) was 59 years (49, 68). An elevated estimated 10-year ASCVD risk was observed in 56.9% (103/181) of males and 38.8% (116/299) of females. The study found differences in baseline and clinical characteristics between participants with estimated ASCVD risk less than 10% and those with a risk greater than or equal to 10%, stratified by gender. Among females, marital status showed an association with elevated estimated ASCVD risk, with a higher proportion (54.31%, 63/116) of unmarried women falling into the elevated estimated risk category. In both male and female groups, individuals with elevated estimated ASCVD risk exhibited higher age, lower level of education, presence of diabetes, higher systolic BP, presence of hypertension, use of BP medication and statin and aspirin initiation. In males, HDL cholesterol levels were lower in the elevated estimated risk group, while in females, total and LDL cholesterol levels were higher among those with elevated estimated risk (Table 1).
Table 1. Socio-demographic and clinical characteristics of study participants stratified by gender.
Characteristics Male n (%) Female n (%)
Total n = 181 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value Total n = 299 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value
Age (years), Median (IQR) 59 (50, 67) 49 (45, 55) 66 (62, 72) < 0.001 59 (49, 68) 52 (46, 59) 69 (64, 74) < 0.001
Marital status 0.801 < 0.001
Married 166 (91.71) 72 (92.31) 94 (91.26) 180 (60.20) 127 (69.40) 53 (45.69)
Not married 15 (8.29) 6 (7.69) 9 (8.74) 119 (39.80) 56 (30.60) 63 (54.31)
Educational status 0.027 0.001
No formal education 2 (1.10) 0 (0.0) 2 (1.94) 29 (9.70) 11 (6.01) 18 (15.52)
Primary 49 (27.07) 14 (17.95) 35 (33.98) 111 (37.12) 59 (32.24) 52 (44.83)
Secondary 29 (16.02) 12 (15.38) 17 (16.50) 30 (10.03) 22 (12.02) 8 (6.90)
Tertiary 101 (55.80) 52 (66.67) 49 (47.57) 129 (43.14) 91 (49.73) 38 (32.76)
Smoking status 0.636 0.150
Current 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00)
Never 165 (91.16) 72 (92.31) 93 (90.29) 297 (99.33) 183 (100.00) 114 (98.28)
Former 16 (8.84) 6 (7.69) 10 (9.71) 2 (0.67) 0 (0.00) 2 (1.72)
Diabetes mellitus 0.002 < 0.001
No 146 (80.66) 71 (91.03) 75 (72.82) 228 (76.25) 164 (89.62) 64 (55.17)
Yes 35 (19.34) 7 (8.97) 28 (27.18) 71 (23.75) 19 (10.38) 52 (44.83)
Systolic BP (mmHg), Median (IQR) 133 (121,148) 128 (117, 138) 138 (126, 154) < 0.001 131 (121, 143) 124 (117, 136) 139 (132, 150) < 0.001
Diastolic BP (mmHg), Median (IQR) 80 (74, 88) 80 (73, 88) 81 (75, 89) 0.287 81 (74, 87) 81 (74, 87) 81 (74, 87) 0.921
Hypertension < 0.001 < 0.001
No 68 (37.57) 51 (65.38) 17 (16.50) 103 (34.45) 86 (46.99) 17 (14.66)
Yes 113 (62.43) 27 (34.62) 86 (83.50) 196 (65.55) 97 (53.01) 99 (85.34)
BP medication < 0.001 < 0.001
No 68 (37.57) 51 (65.38) 17 (16.50) 103 (34.45) 86 (46.99) 17 (14.66)
Yes 113 (62.43) 27 (34.62) 86 (83.50) 196 (65.55) 97 (53.01) 99 (85.34)
Statin initiation 0.001 < 0.001
No 133 (73.48) 67 (85.90) 66 (64.08) 211 (70.57) 148 (80.87) 63 (54.31)
Yes 48 (26.52) 11 (14.10) 37 (35.92) 88 (29.43) 35 (19.13) 53 (45.69)
Aspirin initiation < 0.001 < 0.001
No 133 (73.48) 67 (85.90) 66 (64.08) 244 (81.61) 166 (90.71) 78 (67.24)
Yes 48 (26.52) 11 (14.10) 37 (35.92) 55 (18.39) 17 (9.29) 38 (32.76)
Total cholesterol (mmol/L), Median (IQR) 4.70 (4.08, 5.40) 4.90 (4.20, 5.50) 4.60 (4.00, 5.40) 0.255 4.90 (4.30, 5.90) 4.80 (4.06, 5.50) 5.41 (4.60, 6.15) < 0.001
HDL (mmol/L), Median (IQR) 1.45 (1.14, 1.90) 1.52 (1.31, 1.97) 1.37 (1.05, 1.90) 0.012 1.62 (1.26, 1.93) 1.63 (1.30, 1.93) 1.56 (1.19, 1.94) 0.280
LDL (mmol/L), Median (IQR) 2.74 (2.08, 3.31) 2.62 (1.90, 3.20) 2.80 (2.08, 3.40) 0.311 2.77 (2.12, 3.62) 2.64 (2.00, 3.42) 3.02 (2.27, 4.06) < 0.001
Abbreviations: ASCVD, atherosclerotic cardiovascular disease; BP, blood pressure; HDL, high-density lipoprotein; IQR, interquartile range; LDL, low-density lipoprotein.
3.2 Baseline Anthropometric Measurements of the Study Participants
Several anthropometric indices showed differences between participants with low and elevated estimated 10-year ASCVD risk, stratified by gender. MAC, WHR, CI, and ABSI were associated with elevated estimated risk in both males and females. However, body BMI showed no association with elevated estimated 10-year ASCVD risk in both males and females. WC, WHtR, AVI, BRI, and WHT.5 R were higher among males with elevated estimated risk, but not among females. HC, on the other hand, was lower among females with elevated estimated risk (Table 2).
Table 2. Anthropometric indices of study participants stratified by gender.
Characteristics Male n (%) Female n (%)
Total n = 181 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value Total n = 299 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value
Weight (kg), Median (IQR) 74.50 (65.70, 81.80) 75.55 (66.00, 83.70) 74.40 (65.50, 81.40) 0.407 77.10 (66.90, 85.90) 78.70 (68.40, 88.20) 74.40 (64.55, 83.65) 0.024
Height (cm), Median (IQR) 170.05 (164.90, 175.30) 171.85 (166.40, 177.20) 169.00 (164.30, 174.00) 0.113 159.80 (155.10, 163.70) 160.20 (155.80, 163.90) 158.65 (154.60, 162.90) 0.076
MAC (cm), Median (IQR) 31.80 (29.50, 33.70) 32.50 (30.50, 34.50) 31.00 (29.00, 33.00) < 0.001 34.50 (32.00, 37.50) 35.00 (32.50, 38.00) 34.00 (31.00, 37.00) 0.043
WC (cm), Median (IQR) 95.30 (88.50, 101.50) 91.75 (86.50, 101.00) 96.00 (90.00, 102.50) 0.042 103.00 (95.20, 110.50) 102.50 (95.00, 109.00) 104.00 (96.00, 113.00) 0.221
HC (cm), Median (IQR) 101.50 (96.30, 106.00) 102.50 (97.00, 107.50) 101.00 (96.00, 106.00) 0.250 111.00 (103.50, 118.00) 112.00 (105.00, 118.50) 109.50 (102.00, 117.00) 0.041
WHR, Median (IQR) 0.94 (0.90, 0.97) 0.91 (0.87, 0.95) 0.95 (0.92, 0.98) < 0.001 0.93 (0.88, 0.98) 0.91 (0.87, 0.96) 0.95 (0.91, 0.99) < 0.001
WHtR, Median (IQR) 0.56 (0.52, 0.60) 0.55 (0.50, 0.59) 0.57 (0.53, 0.60) 0.011 0.65 (0.60, 0.69) 0.65 (0.59, 0.69) 0.65 (0.61, 0.71) 0.075
BMI (kg/m2), Median (IQR) 25.60 (23.40, 27.90) 25.75 (22.50, 28.20) 25.60 (23.60, 27.90) 0.900 30.30 (26.80, 33.70) 30.80 (27.30, 33.80) 29.65 (26.15, 33.40) 0.113
CI, Median (IQR) 1.32 (1.27, 1.37) 1.30 (1.23, 1.34) 1.34 (1.31, 1.38) < 0.001 1.36 (1.30, 1.42) 1.34 (1.29, 1.39) 1.40 (1.35, 1.44) < 0.001
AVI, Median (IQR) 18.44 (16.08, 20.81) 17.13 (15.23, 20.60) 18.64 (16.97, 21.12) 0.046 21.37 (18.59, 24.49) 21.06 (18.54, 24.20) 21.66 (18.70, 25.70) 0.239
ABSI, Median (IQR) 0.08 (0.08, 0.09) 0.08 (0.08, 0.08) 0.09 (0.08, 0.09) < 0.001 0.08 (0.08, 0.09) 0.08 (0.08, 0.09) 0.09 (0.08, 0.09) < 0.001
BRI, Median (IQR) 4.64 (3.82, 5.34) 4.27 (3.43, 5.23) 4.79 (4.04, 5.43) 0.011 6.57 (5.38, 7.74) 6.51 (5.27, 7.53) 6.71 (5.64, 8.29) 0.075
WHT.5 R, Median (IQR) 0.73 (0.68, 0.77) 0.71 (0.66, 0.77) 0.74 (0.70, 0.78) 0.019 0.82 (0.76, 0.87) 0.81 (0.76, 0.86) 0.83 (0.77, 0.90) 0.126
Abbreviations: ABSI, a body shape index; ASCVD, atherosclerotic cardiovascular disease; AVI, abdominal volume index; BMI, body mass index; BRI, body roundness index; CI, conicity index; HC, hip circumference; IQR, interquartile range; MAC, mid arm circumference; WC, waist circumference; WHR, waist-to-hip ratio; WHt0.5R, waist-to-height^0.5 ratio; WHtR, waist-to-height ratio.
3.3 Association Between Anthropometric Indices and Elevated Estimated 10-year Risk of ASCVD
After adjustment for potential confounders, WHR (aPR = 1.36, 95% CI: 1.11–1.68) and ABSI (aPR = 1.35, 95% CI: 1.08–1.69) remained positively associated with elevated estimated ASCVD risk. In contrast, HC (aPR = 0.98, 95% CI: 0.97–0.99) and MAC (aPR = 0.95, 95% CI: 0.92–0.98) were independently associated with a lower prevalence of elevated estimated ASCVD risk (Figure 2).
Details are in the caption following the image
Figure 2
Open in figure viewer
PowerPoint
Forest plot of multivariable modified Poisson regression analysis showing anthropometric indices associated with elevated estimated 10-year ASCVD risk among total participants based on the WHO/ISH Risk Assessment Tool. Blue circles represent adjusted prevalence ratios, and horizontal lines indicate 95% confidence intervals. The dashed vertical line represents the null value (adjusted prevalence ratio = 1).
In the sex-stratified analyses, the crude models (Model I) showed that WHR, CI, and ABSI were positively associated with elevated estimated ASCVD risk in both males and females. Specifically, WHR was associated with elevated estimated ASCVD risk among males (PR = 1.52, 95% CI: 1.15–2.00) and females (PR = 1.73, 95% CI: 1.31–2.29). Similarly, CI was associated with elevated estimated ASCVD risk among males (PR = 1.41, 95% CI: 1.11–1.79) and females (PR = 1.58, 95% CI: 1.27–1.96), while ABSI was associated with elevated estimated ASCVD risk among males (PR = 1.56, 95% CI: 1.12–2.17) and females (PR = 1.75, 95% CI: 1.30–2.37). After adjustment for sociodemographic and clinical covariates in Model II, WHR remained independently associated with elevated estimated ASCVD risk among females (aPR = 1.36, 95% CI: 1.01–1.84), while ABSI remained independently associated with elevated estimated ASCVD risk among males (aPR = 1.45, 95% CI: 1.05–2.01). MAC also remained independently associated with a lower prevalence of elevated estimated ASCVD risk among males in both the crude and adjusted models (Model I: PR = 0.93, 95% CI: 0.87–0.98; Model II: aPR = 0.93, 95% CI: 0.88–0.99) (Table 3).
Table 3. Prevalence ratios (PRs) and 95% confidence intervals for associations between anthropometric indices and elevated estimated 10-year ASCVD risk (≥ 10%), stratified by sex.
Variables Model I Model II
Male Female Male Female
MAC 0.93 (0.87, 0.98) 0.97 (0.93, 1.01) 0.93 (0.88, 0.99) 0.98 (0.94, 1.02)
WC 1.01 (0.99, 1.03) 1.01 (0.99, 1.02) 1.00 (0.99, 1.02) 1.00 (0.98, 1.01)
HC 0.99 (0.96, 1.02) 0.98 (0.96, 0.99) 0.98 (0.96, 1.01) 0.99 (0.97, 1.00)
WHR 1.52 (1.15, 2.00) 1.73 (1.31, 2.29) 1.33 (0.99, 1.80) 1.36 (1.01, 1.84)
WHtR 1.09 (0.89, 1.34) 1.16 (0.91, 1.47) 0.99 (0.79, 1.24) 1.02 (0.80, 1.30)
BMI 1.00 (1.00, 1.00) 0.97 (0.94, 1.01) 1.00 (1.00, 1.00) 0.98 (0.94, 1.01)
CI 1.41 (1.11, 1.79) 1.58 (1.27, 1.96) 1.26 (0.99, 1.60) 1.24 (0.98, 1.58)
AVI 1.03 (0.98, 1.07) 1.01 (0.98, 1.05) 1.00 (0.96, 1.05) 0.99 (0.96, 1.03)
ABSI 1.56 (1.12, 2.17) 1.75 (1.30, 2.37) 1.45 (1.05, 2.01) 1.31 (0.95, 1.81)
BRI 1.02 (0.94, 1.10) 1.06 (0.96, 1.16) 0.99 (0.90, 1.08) 0.95 (0.89, 1.01)
WHT.5 R 1.14 (0.92, 1.40) 1.10 (0.91, 1.34) 1.02 (0.82, 1.27) 1.00 (0.82, 1.22)
Note: Model I: Crude Prevalence Ratio (PR).
Model II: Adjusted Prevalence Ratio (aPR). (Fully adjusted for marital status, diabetes mellitus, hypertension, BP medication, statin initiation, aspirin initiation, and smoking status).
Abbreviations: ABSI, a body shape index; AVI, abdominal volume index; BMI, body mass index; BRI, body roundness index; CI, conicity index; HC, hip circumference; MAC, mid arm circumference; WC, waist circumference; WHR, waist-to-hip ratio; WHt0.5R, waist-to-height0.5 ratio; WHtR, waist-to-height ratio.
Sensitivity analyses using alternative thresholds for elevated estimated 10-year ASCVD risk (≥ 20% and ≥ 30%) demonstrated generally consistent patterns of association across models (Supplementary Tables 1 and 2). In both males and females, anthropometric indices such as WHR and ABSI generally retained positive associations with elevated estimated ASCVD risk across thresholds, although several associations became attenuated and lost statistical significance at the ≥ 30% threshold because of reduced precision resulting from smaller numbers of participants in the highest-risk category. Nevertheless, the overall direction and pattern of associations remained largely consistent across thresholds, supporting the robustness of the primary findings.
4 Discussion
The study adds to the existing literature by focusing on an urban Ghanaian population, assessing composite cardiovascular risk using guideline-recommended, sub-region specific WHO/ISH laboratory-based risk assessment charts rather than individual CVD risk factors or other risk scores. This study demonstrates that anthropometric measures including WHR, ABSI, HC and MAC in adults from the Kumasi metropolis are independently associated with elevated estimated cardiovascular risk than the frequently measured BMI and other anthropometric indices.
The clinical variables associated with elevated estimated CVD risk including age, gender, diabetes, hypertension- particularly systolic blood pressure, smoking history and total cholesterol are recognized as components of the risk score and as such are not going to be discussed individually [23]. Associations of antihypertensive medication use, statins and aspirin initiation with elevated estimated risk is related to interventions by physicians to ameliorate this pre-existing risk [8]. In the present study, social determinants of health associated with elevated estimated 10-year risk of ASCVD were level of education in the total population and marital status in females only. Several studies have demonstrated that, irrespective of gender and geographic boundaries, a lower level of education is associated with a worse risk for developing and dying from CVD [24, 25]. Policies on improving the general level of education in the population may be associated with long term CVD risk profile benefits. Two systematic reviews and meta-analyses involving over seven and two million participants each showed that being unmarried was associated with CVD risk in both males and females, even more significantly in males [18, 26]. Gender stratified analysis may have reduced the present study's power to detect this risk in males; however, unmarried females were more likely to have elevated estimated CVD risk.
Anthropometric indices such as BMI, WC, HC, WHR, AVI and CI have been used as measures of obesity and visceral fat; and are shown to be associated with CVD risk factors such as diabetes, hypertension, dyslipidemia and metabolic syndrome in several previous studies in Ghana [27-30]. This study, however, uses these, along with other novel anthropometric indices, and examines their association with elevated estimated CVD risk. Although MAC, WHR, CI, and ABSI were independently associated with elevated estimated risk in both males and females, BMI showed no significant association. This finding contrasts with other studies that found a positive association between BMI and CVD risk [15-18, 27, 29]. This suggests that BMI alone may not be suitable for elevated CVD risk screening and intervention targeting high risk individuals in the Kumasi metropolis. BMI, a simple ratio of weight (Kg) to height (cm) squared, increases proportionally with body weight. Body weight, however, may increase due to increases in muscle mass, adiposity, or bone density. BMI thus fails to discriminate fat accumulation, leading to potential misclassification. Simple measures of central obesity or central adiposity like HC, WC or WHR may thus be preferred as they may be more representative of visceral fat [14, 15, 28, 30-32]. WC, WHtR, AVI, BRI, and WHT.5 R were higher among males with elevated estimated risk relative to females. This may reflect the android and gynecoid obesity phenotypes, with males having more fat deposition in the mid-section, while females have fat deposition at the buttocks, hips and thighs, respectively [33]. HC was inversely associated with elevated estimated CVD risk in the present study and particularly among females. This is corroborated by previous studies showing that HC correlates directly with WC and inversely with cardiometabolic risk factors, CVD risk, and all-cause mortality, and that the effect of central obesity on mortality risk is seriously underestimated without adjustment for hip circumference [31].
Among the anthropometric indices evaluated, WHR and ABSI showed the strongest associations with elevated estimated 10-year ASCVD risk. In the sex-stratified analyses, WHR remained independently associated with elevated estimated ASCVD risk among females, while ABSI remained independently associated with elevated estimated ASCVD risk among males. Similar findings have been reported among Ghanaian migrants, where WHR demonstrated stronger associations with diabetes burden compared with WC and BMI [28]. Although the associations between anthropometric indices and CVD risk may vary across populations, WHR has shown stronger associations with obesity-related cardiovascular risk than BMI among adults in the Eastern Caribbean, Iran, and Southeast Asia [15-17]. Although specific WHR cutoffs were not applied in the present study, the WHO recommends WHR thresholds of 0.90 for men and 0.85 for women as indicators of abdominal obesity [34]. Specific Ghanaian population WHR thresholds for abdominal obesity and CVD risk should be investigated among a nationally representative sample. ABSI, a relatively newer anthropometric index, was derived from WC, BMI and height using data from the National Health and Nutrition Examination Survey (NHANES) 1999–2004 by Krakauer et al. (2012) [35]. It captures both the linear association between WC and BMI, and the nonlinear relationship between WC and height, making it a more nuanced and reliable indicator of central adiposity and body fat distribution in both males and females. A higher ABSI value indicates greater WC relative to expected values for a given height and weight, reflecting increased abdominal fat deposition [19, 35]. Consistent with our findings, ABSI has been reported to be strongly associated with cardiovascular risk and mortality across different sexes, age groups, and ethnic populations in previous studies [16, 18, 19, 35-38]. Findings from Wang et al. (2018) among a Chinese adult population align closely with our study, identifying ABSI as the best anthropometric indicator for assessing coronary heart disease risk in men, with a proposed cut-off value of 0.078 [18]. Since BMI and WC are routinely measured in our clinical setting in Ghana, calculating ABSI from these parameters will be easy to implement. Also, further studies are warranted to assess population-specific cut-off ranges.
MAC also showed an association with elevated estimated 10-year ASCVD risk in the overall study population and in the sex-stratified analysis among males. The mid-arm circumference has been widely used to evaluate nutritional status in children. However, it has been shown to be inversely associated with long-term all-cause and CVD mortality among adults. Individuals in lower MAC quartiles tended to have higher mortality risk, particularly in males, the elderly and non-overweight individuals [39]. Our study reflects this inverse association. Lower MAC may be linked to reduced muscle mass, impaired lipid metabolism or nutritional deficiencies, which may increase CVD risk [39, 40]. The median MAC of a US population was 32.5 cm; however, this was higher than what was observed in a Chinese cohort [39, 40]. It is therefore necessary to set local reference values for CVD risk as MAC differs across populations.
4.1 Strengths and Limitations
The larger sample size and the adherence to protocols for participants' outcome assessment in this study are strengths. That notwithstanding, because the present study is cross-sectional, causal inferences cannot be drawn from the observed associations. The findings of our study may not be generalizable to the Ghanaian population; however, the Kumasi metropolis reflects a typical Ghanaian urban population. Also, one limitation of this study is that anthropometric indices were modeled as continuous variables using their original measurement scales, which may limit the immediate clinical interpretability of some effect estimates. However, this approach was retained to preserve statistical power, avoid information loss associated with categorization, and maintain comparability with previous epidemiological studies evaluating anthropometric measures and cardiovascular risk. Another limitation of this study is the lack of additional serum-based cardiovascular biomarkers beyond those incorporated in the WHO/ISH laboratory-based risk charts. Consequently, potentially important biochemical markers such as inflammatory markers and detailed lipid subfractions could not be evaluated alongside anthropometric indices in cardiovascular risk stratification.
5 Conclusion
This study found that several anthropometric indices, particularly WHR and ABSI, were independently associated with elevated estimated 10-year ASCVD risk among adults in the Kumasi Metropolis. In the sex-stratified analyses, WHR remained independently associated with elevated estimated ASCVD risk among females, while ABSI remained independently associated with elevated estimated ASCVD risk among males. MAC and HC were inversely associated with elevated estimated ASCVD risk, although these findings should be interpreted cautiously, as they may reflect underlying differences in age, body composition, frailty, sarcopenia, or other chronic disease-related factors rather than direct protective effects.
The findings of the present study suggest that anthropometric indices beyond BMI may provide additional insights into cardiovascular risk stratification in this population. However, given the cross-sectional design and the use of estimated ASCVD risk categories rather than incident cardiovascular events, these indices should not yet be considered stand-alone screening or diagnostic tools. Further prospective studies are needed to validate these associations against observed cardiovascular outcomes, evaluate their performance relative to established cardiovascular risk prediction tools, and determine appropriate population-specific cut-offs for clinical application.
Author Contributions
Phyllis Tawiah: conceptualization, methodology, supervision, validation, writing – original draft, writing – review and editing. Kwadwo F. Gyan: conceptualization, methodology, supervision, validation, writing – original draft, writing – review and editing. Isaac Amoah: conceptualization, methodology, supervision, validation, writing – original draft, writing – review and editing. Ibok N. Oduro: supervision, validation, writing – review and editing. Julius K. Karikari: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Douglas A. Opoku: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Emmanuel Konadu: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Ebenezer O. A. Ansah: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Solomon Gyabaah: supervision, validation, writing – review and editing. Godfred K. Twumasi: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Nana K. Ayisi-Boateng: supervision, validation, writing – review and editing.
6 Acknowledgments
The authors would like to thank the staff of the Metro Health Hospital for their cooperation during this study.
Funding
The authors have nothing to report.
Ethics Statement
Ethical approval for this study was sought and obtained from the Kwame Nkrumah University of Science and Technology Committee for Human Research and Publication Ethics prior to the start of the work (Approval number: CHRPE/AP/035/22). Written informed consent was obtained from all participants before they participated in the study.
Conflicts of Interest
The authors declare no conflicts of interest.
Transparency Statement
Phyllis Tawiah and Isaac Amoah the corresponding authors of this work affirm that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Data is available from the corresponding author upon reasonable request.
ABSTRACT
Background and Aim
Cardiovascular diseases remain a major cause of morbidity and mortality in Ghana. This study aimed to determine the association between anthropometric indices and elevated estimated 10-year cardiovascular disease (CVD) risk among adults in the Kumasi Metropolis.
Methods
This cross-sectional study involved 480 adults aged 40–79 years. Anthropometric indices and lipid profile were determined using standard methods. Blood pressure was measured using an automated sphygmomanometer. The World Health Organization/International Society of Hypertension laboratory-based risk assessment charts were used to estimate 10-year Atherosclerotic Cardiovascular Disease (ASCVD) risk. Estimated CVD risk ≥ 10% was considered elevated. Anthropometric indices were examined for association with elevated risk using chi-square, Fisher's exact and Wilcoxon rank-sum tests. Multivariable modified Poisson regression analysis was used to determine factors associated with elevated estimated CVD risk.
Results
An elevated estimated10-year ASCVD risk was observed in 56.9% (103/181) of males and 38.8% (116/299) of females. Lower education level was associated with elevated CVD risk. Waist-to-hip ratio (WHR) (aPR = 1.36, 95% CI: 1.11–1.68), body shape index (ABSI) (aPR = 1.35, 95% CI: 1.08–1.69), hip circumference (HC) (aPR = 0.98, 95% CI: 0.97–0.99), and mid-arm circumference (MAC) (aPR = 0.95, 95% CI: 0.92–0.98) were associated with elevated estimated CVD risk. After stratifying by sex, WHR remained independently associated with elevated estimated CVD risk in females (aPR = 1.36, 95% CI: 1.01–1.84); as did ABSI (aPR = 1.45, 95% CI: 1.05–2.01) and MAC (aPR = 0.93, 95% CI: 0.88–0.99) in males.
Conclusion
WHR, ABSI and MAC assessments should be integrated into CVD risk evaluation in Ghana, especially in deprived areas where lipid profile is not routinely assessed.
1 Introduction
Cardiovascular diseases (CVDs) are a group of diseases that affect the normal function of the heart and blood vessels, including stroke, coronary heart disease, and heart failure [1]. According to the World Health Organization (WHO), they remain the leading cause of mortality globally and accounted for 32.0% of mortality in 2022. Stroke and heart attack, for example, are currently responsible for 85.0% of all CVD deaths, with the burden severe in low- and middle- income countries (LMICs) [1]. LMICs, including Ghana, account for approximately 75.0% of CVD-related mortality globally [1].
In Ghana, the burden of CVDs is currently rising, and this is driven mostly by stroke and coronary artery disease [2]. Indeed, a recent systematic review and meta-analysis of observational studies reported a stroke prevalence of 7.96% in the country [3]. Another systematic review and meta-analysis that included 16 studies with 58,912 participants from 1954 to 2022 reported that the pooled prevalence of CVD nationally is 10.34% (95% CI: [8.48, 12.20]) [4]. However, with rising risk factors for CVDs in Ghana, including hypertension, diabetes, old age and increased body mass index (BMI), the prevalence of CVDs is expected to increase significantly over the next decade [4]. Nationally, two key systematic reviews and meta-analyses of observational studies have reported a hypertension prevalence of 27.0% (95% CI 24.0%–30.0%) and 30.3% (95% CI 26.1–34.8%) respectively [5, 6]. An analysis of data on hospital-based mortality records indicated that hypertension is currently the leading cause of death among chronic diseases in Ghana [7].
Although efforts to prevent and mitigate the CVDs in Ghana have largely focused on promoting lower-sodium diets and increased physical activity, assessing Atherosclerotic Cardiovascular Disease (ASCVD) risk is also an important strategy for identifying individuals at risk [8]. Several tools have been reported as feasible for assessing CVD risk, including mobile-based applications for Pooled Cohort Equation (PCE), laboratory-based Framingham Risk Score (FRS), Globorisk and laboratory-based World Health Organization/International Society of Hypertension (WHO/ISH) risk charts [9, 10]. However, these tools have shown varying degrees of agreement when applied in the Ghanaian setting [11]. Currently, the WHO/ISH risk assessment tool is recommended by the CVD treatment guideline for Ghana for assessing and managing CVD risk to ensure uniformity [11, 12]. The WHO/ISH risk score classified 82.0%, 9.4%, and 8.6% of Ghanaians as low-, intermediate-, and high-risk, respectively based on data from the Ghana Heart Study [11]. The WHO/ISH risk assessment tool utilizes parameters including the lipid profile and presence or absence of diabetes, with other factors for the determination of CVD risk. However, these may not always be possible, especially in a low-resource setting like Ghana. Anthropometric indices are non-invasive, low-cost, and can also be used to determine their association with CVD risk.
The use of anthropometric measures as proxies or indicators of CVD risk factors is well established [13, 14]. Consequently, measures of central obesity, including waist circumference (WC), hip circumference (HC), waist-to-hip ratio (WHR), conicity index (CI), waist-to-height ratio (WHtR), abdominal volume index (AVI), body roundness index (BRI), a body shape index (ABSI), and waist-to-height-to-the-power-of-0.5 ratio (WHT.5 R), have been identified as reliable indicators of CVD risk [14-19]. Anthropometric measures, however, differ by gender and geographic location, as well as by the degree of association with CVD risk, as demonstrated by the varying outcomes of previous studies [14-18]. This makes it prudent to investigate anthropometric indices and their association with elevated estimated risk of CVD in specified local populations. Currently, there is a paucity of data on the association between anthropometric indices and CVD risk amongst Ghanaians. This study sought to determine the association between anthropometric indices and elevated estimated 10-year CVD risk among adults in the Kumasi metropolis of Ghana.
2 Methods and Materials
2.1 Study Design and Sampling Technique
The study design was cross-sectional, and our sampling frame consisted of all registered patients at the Centre for Ageing and Elderly Care who met the inclusion criteria. To achieve the required sample size of 457 participants, a simple random sample of 500 eligible patients was selected from the sampling frame using the sample command with the count option in Stata/SE version 17.0 [20]. A random seed (set seed) was specified to ensure reproducibility of the sampling procedure. The 500 eligible participants were randomly selected to ensure adequate statistical power and precision of the study estimates. Selected individuals were contacted and invited to participate in the study. Participants who were unable to report to the hospital but expressed willingness to participate were subsequently visited in their homes, where interviews and study assessments were conducted. A total of 480 participants completed the study and were included in the final analysis (Figure 1). An ethical approval was sought and obtained from the Kwame Nkrumah University of Science and Technology Committee for Human Research and Publication Ethics (CHRPE/AP/035/22) prior to the start of the work. Data were collected over a 9-month period, from 15 February 2022 to 21 November 2022. Participants provided a written informed consent. The study was conducted in compliance with the guidelines stipulated in the Declaration of Helsinki. Informed consent was obtained from all participants who participated in the study.
Details are in the caption following the image
Figure 1
Open in figure viewer
PowerPoint
Flow diagram showing the selection of study participants.
2.2 Inclusion and Exclusion Criteria
Participants were eligible if they were aged ≥ 40 years, were regular outpatient department attendees, and were receiving standard care during routine follow-up visits. Individuals with a history of major cardiovascular events, such as stroke or myocardial infarction, were excluded.
2.3 Study Site
The study was conducted at the Centre for Ageing and Elderly Care at Metro Health Hospital in Kumasi. The Centre for Ageing and Elderly Care at Metro Health Hospital in Kumasi is a private medical facility providing specialist medical and geriatric care as well as services for healthy ageing. Its catchment area is the Kumasi Metropolis which is the capital city of the Ashanti Region of Ghana.
2.4 Sample Size Determination
The sample size required for the study was calculated using the Cochran formula [21].
𝑛=
(𝑍𝛼/2)2𝑝(1−𝑝)
𝑑2
Where n is the minimum sample size required for the study; 𝑍𝛼/2
is the abscissa of the normal curve that cut-off an area at the tail (1.96); p is the estimated proportion of an attribute (elevated 10-year ASCVD risk) that is present in the study population, which is 45.4% [22], d is the desired level of precision (5%). An additional 20% participants were added to allow for withdrawals and missing data. Therefore, the required sample size was 457.
2.5 Data Collection and Study Procedures
Structured questionnaires were administered to participants to collect data on sociodemographic characteristics, history of hypertension or diabetes diagnosis, current medications, including use of statins and acetylsalicylic acid and lifestyle factors such as smoking history.
2.6 Anthropometric, Blood Pressure, and Biochemical Measurements
A trained health care provider measured anthropometric data, including weight, height, WC, HC and blood pressure (BP), using standard protocols. Body weight was measured with participants' excess clothes removed and barefooted using Omron's body composition analyzer (Omron, BF511, Japan). Measurement was recorded to the nearest 0.5 kg. Height was measured using a stadiometer's vertical scale (Seca, Hamburg, Germany) with the participants standing in an erect position against the scale with their heels and buttocks in contact with the scale. This was measured to the closest 0.5 cm. Waist circumference was measured in centimeters (cm) around the smallest area below the rib cage and above the belly button. Hip circumference was measured in cm at the largest circumference between the waist and the knee. Mid-arm circumference (MAC) was also measured in cm at the midpoint between the acromion and olecranon processes on the non-dominant arm, with the arm relaxed. Measurements were done with a non-stretchable and accurately calibrated measuring tape.
Blood pressure was determined using an attended automated sphygmomanometer (Omron M7 series) with properly fitted cuff. Participants were seated with their back supported and both feet flat on the ground. Three BP readings were taken for each participant, with a 3-min rest interval between successive measurements. The mean of the three readings were used for the final analysis. Derived anthropometric indices were calculated using the formulae as described in previous studies [16-19].
2.7 Unit Standardization for Anthropometric Indices
To ensure accuracy, consistency, and comparability with previously published definitions [16-19], all anthropometric measurements were standardized before the computation of derived indices. Directly measured variables were retained in their original units for descriptive analysis, with weight recorded in kilograms (kg), height, WC, HC, and MAC in centimeters (cm), BP in millimeters of mercury (mmHg).
For derived anthropometric indices, unit conversions were performed where required prior to calculation. Specifically, WC and height were converted from centimeters to meters (m) when necessary to align with the original formulations of indices such as ABSI and BRI that are sensitive to unit specification. Indices that incorporated BMI were computed using BMI expressed in kg/m2, with height in meters and weight in kg. This standardization ensured internal consistency across all derived indices and minimized potential computational bias arising from unit misalignment.
WHR=
WC(cm)
HC(cm)
WHtR=
WC(cm)
Height(cm)
BMI=
Weight(kg)
Height(m)2
CI=
WC(m)
0.109×√
Weight(kg)
Height(m)
AVI=
2×WC(cm)2+0.7×[WC(cm)−HC(cm)]2
1000
ABSI=
WC(m)
BMI
2
3
×Height(m)
1
2
BRI=364.2−365.5×
√
√
√
√
√
⎷
1−(
(WC(m)/(2π))2
(0.5×Height(m))2
)
WHT.5R=
WC(cm)
√Height(cm)
Information on fasting lipid profiles was extracted from participants' medical records. The most recent fasting lipid profile result, obtained within 1 month prior to data collection, was used for the analysis.
The estimated 10-year ASCVD risk was calculated using the WHO/ISH laboratory-based risk charts. The laboratory model is based on six variables: diabetes mellitus status, age, sex, smoking status, systolic BP (SBP), and total cholesterol. It stratifies risk as < 5% (green), 5% to < 10% (yellow), 10% to < 20% (orange), 20% to < 30% (red), and ≥ 30% (dark red), corresponding to very low, low, moderate, high, and very high risk, respectively [23]. For the purpose of this study, a cut off of ≥ 10% was indicative of elevated risk [16]. The ≥ 10% threshold was used because WHO cardiovascular risk guidelines commonly categorize individuals with a 10-year ASCVD risk of ≥ 10% as having at least moderate cardiovascular risk that may warrant closer clinical attention and preventive interventions, especially in LMICs, including Ghana [23].
2.8 Data Analysis
Data collected from the study participants were entered into Microsoft Excel and exported to Stata/SE version 17.0 [20] for analysis. Continuous variables were tested for normality using the Shapiro-Wilk test, supported by visual assessment using Q-Q plots or histograms. The characteristics of study participants were summarized using frequencies and percentages for categorical variables, and medians with interquartile ranges for skewed continuous variables. Anthropometric indices were examined for association with elevated estimated ASCVD risk using chi-square or Fisher's exact test where an expected cell frequency was less than five, and Wilcoxon rank-sum test was applied for all skewed continuous variables.
Separate modified Poisson regression models with robust standard errors were fitted for each anthropometric index to assess associations with elevated estimated ASCVD risk. Anthropometric indices were evaluated one-at-a-time because of the strong intercorrelations among adiposity measures. All models [both overall (Figure 1) and sex-stratified (Table 3, Model II)] were adjusted for the same prespecified sociodemographic and clinical covariates, selected for clinical relevance and prior literature [10, 11]. The adjusted models controlled for marital status, diabetes mellitus, hypertension, BP medication, statin initiation, aspirin initiation, and smoking status. Multicollinearity was assessed using the Variance Inflation Factor (VIF), with variables exhibiting VIF values < 5 retained for inclusion in the models (Supplementary File 1). Sensitivity analyses were additionally conducted using alternative thresholds for elevated estimated ASCVD risk (≥ 20% and ≥ 30%) to assess the robustness of observed associations. Separate modified Poisson regression models with robust standard errors were repeated using these alternative categorizations. There were no missing values in the dataset; therefore, all observations were included in the final analyses. Statistical significance was determined at a two-sided p-value < 0.05 or where the 95% CI for the aPR excluded 1.
3 Results
3.1 Socio-Demographic and Clinical Baseline Characteristics of the Study Participants
Median age (IQR) of males (n = 181) was 59 years (50, 67) and females (n = 299) was 59 years (49, 68). An elevated estimated 10-year ASCVD risk was observed in 56.9% (103/181) of males and 38.8% (116/299) of females. The study found differences in baseline and clinical characteristics between participants with estimated ASCVD risk less than 10% and those with a risk greater than or equal to 10%, stratified by gender. Among females, marital status showed an association with elevated estimated ASCVD risk, with a higher proportion (54.31%, 63/116) of unmarried women falling into the elevated estimated risk category. In both male and female groups, individuals with elevated estimated ASCVD risk exhibited higher age, lower level of education, presence of diabetes, higher systolic BP, presence of hypertension, use of BP medication and statin and aspirin initiation. In males, HDL cholesterol levels were lower in the elevated estimated risk group, while in females, total and LDL cholesterol levels were higher among those with elevated estimated risk (Table 1).
Table 1. Socio-demographic and clinical characteristics of study participants stratified by gender.
Characteristics Male n (%) Female n (%)
Total n = 181 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value Total n = 299 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value
Age (years), Median (IQR) 59 (50, 67) 49 (45, 55) 66 (62, 72) < 0.001 59 (49, 68) 52 (46, 59) 69 (64, 74) < 0.001
Marital status 0.801 < 0.001
Married 166 (91.71) 72 (92.31) 94 (91.26) 180 (60.20) 127 (69.40) 53 (45.69)
Not married 15 (8.29) 6 (7.69) 9 (8.74) 119 (39.80) 56 (30.60) 63 (54.31)
Educational status 0.027 0.001
No formal education 2 (1.10) 0 (0.0) 2 (1.94) 29 (9.70) 11 (6.01) 18 (15.52)
Primary 49 (27.07) 14 (17.95) 35 (33.98) 111 (37.12) 59 (32.24) 52 (44.83)
Secondary 29 (16.02) 12 (15.38) 17 (16.50) 30 (10.03) 22 (12.02) 8 (6.90)
Tertiary 101 (55.80) 52 (66.67) 49 (47.57) 129 (43.14) 91 (49.73) 38 (32.76)
Smoking status 0.636 0.150
Current 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00)
Never 165 (91.16) 72 (92.31) 93 (90.29) 297 (99.33) 183 (100.00) 114 (98.28)
Former 16 (8.84) 6 (7.69) 10 (9.71) 2 (0.67) 0 (0.00) 2 (1.72)
Diabetes mellitus 0.002 < 0.001
No 146 (80.66) 71 (91.03) 75 (72.82) 228 (76.25) 164 (89.62) 64 (55.17)
Yes 35 (19.34) 7 (8.97) 28 (27.18) 71 (23.75) 19 (10.38) 52 (44.83)
Systolic BP (mmHg), Median (IQR) 133 (121,148) 128 (117, 138) 138 (126, 154) < 0.001 131 (121, 143) 124 (117, 136) 139 (132, 150) < 0.001
Diastolic BP (mmHg), Median (IQR) 80 (74, 88) 80 (73, 88) 81 (75, 89) 0.287 81 (74, 87) 81 (74, 87) 81 (74, 87) 0.921
Hypertension < 0.001 < 0.001
No 68 (37.57) 51 (65.38) 17 (16.50) 103 (34.45) 86 (46.99) 17 (14.66)
Yes 113 (62.43) 27 (34.62) 86 (83.50) 196 (65.55) 97 (53.01) 99 (85.34)
BP medication < 0.001 < 0.001
No 68 (37.57) 51 (65.38) 17 (16.50) 103 (34.45) 86 (46.99) 17 (14.66)
Yes 113 (62.43) 27 (34.62) 86 (83.50) 196 (65.55) 97 (53.01) 99 (85.34)
Statin initiation 0.001 < 0.001
No 133 (73.48) 67 (85.90) 66 (64.08) 211 (70.57) 148 (80.87) 63 (54.31)
Yes 48 (26.52) 11 (14.10) 37 (35.92) 88 (29.43) 35 (19.13) 53 (45.69)
Aspirin initiation < 0.001 < 0.001
No 133 (73.48) 67 (85.90) 66 (64.08) 244 (81.61) 166 (90.71) 78 (67.24)
Yes 48 (26.52) 11 (14.10) 37 (35.92) 55 (18.39) 17 (9.29) 38 (32.76)
Total cholesterol (mmol/L), Median (IQR) 4.70 (4.08, 5.40) 4.90 (4.20, 5.50) 4.60 (4.00, 5.40) 0.255 4.90 (4.30, 5.90) 4.80 (4.06, 5.50) 5.41 (4.60, 6.15) < 0.001
HDL (mmol/L), Median (IQR) 1.45 (1.14, 1.90) 1.52 (1.31, 1.97) 1.37 (1.05, 1.90) 0.012 1.62 (1.26, 1.93) 1.63 (1.30, 1.93) 1.56 (1.19, 1.94) 0.280
LDL (mmol/L), Median (IQR) 2.74 (2.08, 3.31) 2.62 (1.90, 3.20) 2.80 (2.08, 3.40) 0.311 2.77 (2.12, 3.62) 2.64 (2.00, 3.42) 3.02 (2.27, 4.06) < 0.001
Abbreviations: ASCVD, atherosclerotic cardiovascular disease; BP, blood pressure; HDL, high-density lipoprotein; IQR, interquartile range; LDL, low-density lipoprotein.
3.2 Baseline Anthropometric Measurements of the Study Participants
Several anthropometric indices showed differences between participants with low and elevated estimated 10-year ASCVD risk, stratified by gender. MAC, WHR, CI, and ABSI were associated with elevated estimated risk in both males and females. However, body BMI showed no association with elevated estimated 10-year ASCVD risk in both males and females. WC, WHtR, AVI, BRI, and WHT.5 R were higher among males with elevated estimated risk, but not among females. HC, on the other hand, was lower among females with elevated estimated risk (Table 2).
Table 2. Anthropometric indices of study participants stratified by gender.
Characteristics Male n (%) Female n (%)
Total n = 181 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value Total n = 299 10-year ASCVD risk < 10 10-year ASCVD risk ≥ 10 p value
Weight (kg), Median (IQR) 74.50 (65.70, 81.80) 75.55 (66.00, 83.70) 74.40 (65.50, 81.40) 0.407 77.10 (66.90, 85.90) 78.70 (68.40, 88.20) 74.40 (64.55, 83.65) 0.024
Height (cm), Median (IQR) 170.05 (164.90, 175.30) 171.85 (166.40, 177.20) 169.00 (164.30, 174.00) 0.113 159.80 (155.10, 163.70) 160.20 (155.80, 163.90) 158.65 (154.60, 162.90) 0.076
MAC (cm), Median (IQR) 31.80 (29.50, 33.70) 32.50 (30.50, 34.50) 31.00 (29.00, 33.00) < 0.001 34.50 (32.00, 37.50) 35.00 (32.50, 38.00) 34.00 (31.00, 37.00) 0.043
WC (cm), Median (IQR) 95.30 (88.50, 101.50) 91.75 (86.50, 101.00) 96.00 (90.00, 102.50) 0.042 103.00 (95.20, 110.50) 102.50 (95.00, 109.00) 104.00 (96.00, 113.00) 0.221
HC (cm), Median (IQR) 101.50 (96.30, 106.00) 102.50 (97.00, 107.50) 101.00 (96.00, 106.00) 0.250 111.00 (103.50, 118.00) 112.00 (105.00, 118.50) 109.50 (102.00, 117.00) 0.041
WHR, Median (IQR) 0.94 (0.90, 0.97) 0.91 (0.87, 0.95) 0.95 (0.92, 0.98) < 0.001 0.93 (0.88, 0.98) 0.91 (0.87, 0.96) 0.95 (0.91, 0.99) < 0.001
WHtR, Median (IQR) 0.56 (0.52, 0.60) 0.55 (0.50, 0.59) 0.57 (0.53, 0.60) 0.011 0.65 (0.60, 0.69) 0.65 (0.59, 0.69) 0.65 (0.61, 0.71) 0.075
BMI (kg/m2), Median (IQR) 25.60 (23.40, 27.90) 25.75 (22.50, 28.20) 25.60 (23.60, 27.90) 0.900 30.30 (26.80, 33.70) 30.80 (27.30, 33.80) 29.65 (26.15, 33.40) 0.113
CI, Median (IQR) 1.32 (1.27, 1.37) 1.30 (1.23, 1.34) 1.34 (1.31, 1.38) < 0.001 1.36 (1.30, 1.42) 1.34 (1.29, 1.39) 1.40 (1.35, 1.44) < 0.001
AVI, Median (IQR) 18.44 (16.08, 20.81) 17.13 (15.23, 20.60) 18.64 (16.97, 21.12) 0.046 21.37 (18.59, 24.49) 21.06 (18.54, 24.20) 21.66 (18.70, 25.70) 0.239
ABSI, Median (IQR) 0.08 (0.08, 0.09) 0.08 (0.08, 0.08) 0.09 (0.08, 0.09) < 0.001 0.08 (0.08, 0.09) 0.08 (0.08, 0.09) 0.09 (0.08, 0.09) < 0.001
BRI, Median (IQR) 4.64 (3.82, 5.34) 4.27 (3.43, 5.23) 4.79 (4.04, 5.43) 0.011 6.57 (5.38, 7.74) 6.51 (5.27, 7.53) 6.71 (5.64, 8.29) 0.075
WHT.5 R, Median (IQR) 0.73 (0.68, 0.77) 0.71 (0.66, 0.77) 0.74 (0.70, 0.78) 0.019 0.82 (0.76, 0.87) 0.81 (0.76, 0.86) 0.83 (0.77, 0.90) 0.126
Abbreviations: ABSI, a body shape index; ASCVD, atherosclerotic cardiovascular disease; AVI, abdominal volume index; BMI, body mass index; BRI, body roundness index; CI, conicity index; HC, hip circumference; IQR, interquartile range; MAC, mid arm circumference; WC, waist circumference; WHR, waist-to-hip ratio; WHt0.5R, waist-to-height^0.5 ratio; WHtR, waist-to-height ratio.
3.3 Association Between Anthropometric Indices and Elevated Estimated 10-year Risk of ASCVD
After adjustment for potential confounders, WHR (aPR = 1.36, 95% CI: 1.11–1.68) and ABSI (aPR = 1.35, 95% CI: 1.08–1.69) remained positively associated with elevated estimated ASCVD risk. In contrast, HC (aPR = 0.98, 95% CI: 0.97–0.99) and MAC (aPR = 0.95, 95% CI: 0.92–0.98) were independently associated with a lower prevalence of elevated estimated ASCVD risk (Figure 2).
Details are in the caption following the image
Figure 2
Open in figure viewer
PowerPoint
Forest plot of multivariable modified Poisson regression analysis showing anthropometric indices associated with elevated estimated 10-year ASCVD risk among total participants based on the WHO/ISH Risk Assessment Tool. Blue circles represent adjusted prevalence ratios, and horizontal lines indicate 95% confidence intervals. The dashed vertical line represents the null value (adjusted prevalence ratio = 1).
In the sex-stratified analyses, the crude models (Model I) showed that WHR, CI, and ABSI were positively associated with elevated estimated ASCVD risk in both males and females. Specifically, WHR was associated with elevated estimated ASCVD risk among males (PR = 1.52, 95% CI: 1.15–2.00) and females (PR = 1.73, 95% CI: 1.31–2.29). Similarly, CI was associated with elevated estimated ASCVD risk among males (PR = 1.41, 95% CI: 1.11–1.79) and females (PR = 1.58, 95% CI: 1.27–1.96), while ABSI was associated with elevated estimated ASCVD risk among males (PR = 1.56, 95% CI: 1.12–2.17) and females (PR = 1.75, 95% CI: 1.30–2.37). After adjustment for sociodemographic and clinical covariates in Model II, WHR remained independently associated with elevated estimated ASCVD risk among females (aPR = 1.36, 95% CI: 1.01–1.84), while ABSI remained independently associated with elevated estimated ASCVD risk among males (aPR = 1.45, 95% CI: 1.05–2.01). MAC also remained independently associated with a lower prevalence of elevated estimated ASCVD risk among males in both the crude and adjusted models (Model I: PR = 0.93, 95% CI: 0.87–0.98; Model II: aPR = 0.93, 95% CI: 0.88–0.99) (Table 3).
Table 3. Prevalence ratios (PRs) and 95% confidence intervals for associations between anthropometric indices and elevated estimated 10-year ASCVD risk (≥ 10%), stratified by sex.
Variables Model I Model II
Male Female Male Female
MAC 0.93 (0.87, 0.98) 0.97 (0.93, 1.01) 0.93 (0.88, 0.99) 0.98 (0.94, 1.02)
WC 1.01 (0.99, 1.03) 1.01 (0.99, 1.02) 1.00 (0.99, 1.02) 1.00 (0.98, 1.01)
HC 0.99 (0.96, 1.02) 0.98 (0.96, 0.99) 0.98 (0.96, 1.01) 0.99 (0.97, 1.00)
WHR 1.52 (1.15, 2.00) 1.73 (1.31, 2.29) 1.33 (0.99, 1.80) 1.36 (1.01, 1.84)
WHtR 1.09 (0.89, 1.34) 1.16 (0.91, 1.47) 0.99 (0.79, 1.24) 1.02 (0.80, 1.30)
BMI 1.00 (1.00, 1.00) 0.97 (0.94, 1.01) 1.00 (1.00, 1.00) 0.98 (0.94, 1.01)
CI 1.41 (1.11, 1.79) 1.58 (1.27, 1.96) 1.26 (0.99, 1.60) 1.24 (0.98, 1.58)
AVI 1.03 (0.98, 1.07) 1.01 (0.98, 1.05) 1.00 (0.96, 1.05) 0.99 (0.96, 1.03)
ABSI 1.56 (1.12, 2.17) 1.75 (1.30, 2.37) 1.45 (1.05, 2.01) 1.31 (0.95, 1.81)
BRI 1.02 (0.94, 1.10) 1.06 (0.96, 1.16) 0.99 (0.90, 1.08) 0.95 (0.89, 1.01)
WHT.5 R 1.14 (0.92, 1.40) 1.10 (0.91, 1.34) 1.02 (0.82, 1.27) 1.00 (0.82, 1.22)
Note: Model I: Crude Prevalence Ratio (PR).
Model II: Adjusted Prevalence Ratio (aPR). (Fully adjusted for marital status, diabetes mellitus, hypertension, BP medication, statin initiation, aspirin initiation, and smoking status).
Abbreviations: ABSI, a body shape index; AVI, abdominal volume index; BMI, body mass index; BRI, body roundness index; CI, conicity index; HC, hip circumference; MAC, mid arm circumference; WC, waist circumference; WHR, waist-to-hip ratio; WHt0.5R, waist-to-height0.5 ratio; WHtR, waist-to-height ratio.
Sensitivity analyses using alternative thresholds for elevated estimated 10-year ASCVD risk (≥ 20% and ≥ 30%) demonstrated generally consistent patterns of association across models (Supplementary Tables 1 and 2). In both males and females, anthropometric indices such as WHR and ABSI generally retained positive associations with elevated estimated ASCVD risk across thresholds, although several associations became attenuated and lost statistical significance at the ≥ 30% threshold because of reduced precision resulting from smaller numbers of participants in the highest-risk category. Nevertheless, the overall direction and pattern of associations remained largely consistent across thresholds, supporting the robustness of the primary findings.
4 Discussion
The study adds to the existing literature by focusing on an urban Ghanaian population, assessing composite cardiovascular risk using guideline-recommended, sub-region specific WHO/ISH laboratory-based risk assessment charts rather than individual CVD risk factors or other risk scores. This study demonstrates that anthropometric measures including WHR, ABSI, HC and MAC in adults from the Kumasi metropolis are independently associated with elevated estimated cardiovascular risk than the frequently measured BMI and other anthropometric indices.
The clinical variables associated with elevated estimated CVD risk including age, gender, diabetes, hypertension- particularly systolic blood pressure, smoking history and total cholesterol are recognized as components of the risk score and as such are not going to be discussed individually [23]. Associations of antihypertensive medication use, statins and aspirin initiation with elevated estimated risk is related to interventions by physicians to ameliorate this pre-existing risk [8]. In the present study, social determinants of health associated with elevated estimated 10-year risk of ASCVD were level of education in the total population and marital status in females only. Several studies have demonstrated that, irrespective of gender and geographic boundaries, a lower level of education is associated with a worse risk for developing and dying from CVD [24, 25]. Policies on improving the general level of education in the population may be associated with long term CVD risk profile benefits. Two systematic reviews and meta-analyses involving over seven and two million participants each showed that being unmarried was associated with CVD risk in both males and females, even more significantly in males [18, 26]. Gender stratified analysis may have reduced the present study's power to detect this risk in males; however, unmarried females were more likely to have elevated estimated CVD risk.
Anthropometric indices such as BMI, WC, HC, WHR, AVI and CI have been used as measures of obesity and visceral fat; and are shown to be associated with CVD risk factors such as diabetes, hypertension, dyslipidemia and metabolic syndrome in several previous studies in Ghana [27-30]. This study, however, uses these, along with other novel anthropometric indices, and examines their association with elevated estimated CVD risk. Although MAC, WHR, CI, and ABSI were independently associated with elevated estimated risk in both males and females, BMI showed no significant association. This finding contrasts with other studies that found a positive association between BMI and CVD risk [15-18, 27, 29]. This suggests that BMI alone may not be suitable for elevated CVD risk screening and intervention targeting high risk individuals in the Kumasi metropolis. BMI, a simple ratio of weight (Kg) to height (cm) squared, increases proportionally with body weight. Body weight, however, may increase due to increases in muscle mass, adiposity, or bone density. BMI thus fails to discriminate fat accumulation, leading to potential misclassification. Simple measures of central obesity or central adiposity like HC, WC or WHR may thus be preferred as they may be more representative of visceral fat [14, 15, 28, 30-32]. WC, WHtR, AVI, BRI, and WHT.5 R were higher among males with elevated estimated risk relative to females. This may reflect the android and gynecoid obesity phenotypes, with males having more fat deposition in the mid-section, while females have fat deposition at the buttocks, hips and thighs, respectively [33]. HC was inversely associated with elevated estimated CVD risk in the present study and particularly among females. This is corroborated by previous studies showing that HC correlates directly with WC and inversely with cardiometabolic risk factors, CVD risk, and all-cause mortality, and that the effect of central obesity on mortality risk is seriously underestimated without adjustment for hip circumference [31].
Among the anthropometric indices evaluated, WHR and ABSI showed the strongest associations with elevated estimated 10-year ASCVD risk. In the sex-stratified analyses, WHR remained independently associated with elevated estimated ASCVD risk among females, while ABSI remained independently associated with elevated estimated ASCVD risk among males. Similar findings have been reported among Ghanaian migrants, where WHR demonstrated stronger associations with diabetes burden compared with WC and BMI [28]. Although the associations between anthropometric indices and CVD risk may vary across populations, WHR has shown stronger associations with obesity-related cardiovascular risk than BMI among adults in the Eastern Caribbean, Iran, and Southeast Asia [15-17]. Although specific WHR cutoffs were not applied in the present study, the WHO recommends WHR thresholds of 0.90 for men and 0.85 for women as indicators of abdominal obesity [34]. Specific Ghanaian population WHR thresholds for abdominal obesity and CVD risk should be investigated among a nationally representative sample. ABSI, a relatively newer anthropometric index, was derived from WC, BMI and height using data from the National Health and Nutrition Examination Survey (NHANES) 1999–2004 by Krakauer et al. (2012) [35]. It captures both the linear association between WC and BMI, and the nonlinear relationship between WC and height, making it a more nuanced and reliable indicator of central adiposity and body fat distribution in both males and females. A higher ABSI value indicates greater WC relative to expected values for a given height and weight, reflecting increased abdominal fat deposition [19, 35]. Consistent with our findings, ABSI has been reported to be strongly associated with cardiovascular risk and mortality across different sexes, age groups, and ethnic populations in previous studies [16, 18, 19, 35-38]. Findings from Wang et al. (2018) among a Chinese adult population align closely with our study, identifying ABSI as the best anthropometric indicator for assessing coronary heart disease risk in men, with a proposed cut-off value of 0.078 [18]. Since BMI and WC are routinely measured in our clinical setting in Ghana, calculating ABSI from these parameters will be easy to implement. Also, further studies are warranted to assess population-specific cut-off ranges.
MAC also showed an association with elevated estimated 10-year ASCVD risk in the overall study population and in the sex-stratified analysis among males. The mid-arm circumference has been widely used to evaluate nutritional status in children. However, it has been shown to be inversely associated with long-term all-cause and CVD mortality among adults. Individuals in lower MAC quartiles tended to have higher mortality risk, particularly in males, the elderly and non-overweight individuals [39]. Our study reflects this inverse association. Lower MAC may be linked to reduced muscle mass, impaired lipid metabolism or nutritional deficiencies, which may increase CVD risk [39, 40]. The median MAC of a US population was 32.5 cm; however, this was higher than what was observed in a Chinese cohort [39, 40]. It is therefore necessary to set local reference values for CVD risk as MAC differs across populations.
4.1 Strengths and Limitations
The larger sample size and the adherence to protocols for participants' outcome assessment in this study are strengths. That notwithstanding, because the present study is cross-sectional, causal inferences cannot be drawn from the observed associations. The findings of our study may not be generalizable to the Ghanaian population; however, the Kumasi metropolis reflects a typical Ghanaian urban population. Also, one limitation of this study is that anthropometric indices were modeled as continuous variables using their original measurement scales, which may limit the immediate clinical interpretability of some effect estimates. However, this approach was retained to preserve statistical power, avoid information loss associated with categorization, and maintain comparability with previous epidemiological studies evaluating anthropometric measures and cardiovascular risk. Another limitation of this study is the lack of additional serum-based cardiovascular biomarkers beyond those incorporated in the WHO/ISH laboratory-based risk charts. Consequently, potentially important biochemical markers such as inflammatory markers and detailed lipid subfractions could not be evaluated alongside anthropometric indices in cardiovascular risk stratification.
5 Conclusion
This study found that several anthropometric indices, particularly WHR and ABSI, were independently associated with elevated estimated 10-year ASCVD risk among adults in the Kumasi Metropolis. In the sex-stratified analyses, WHR remained independently associated with elevated estimated ASCVD risk among females, while ABSI remained independently associated with elevated estimated ASCVD risk among males. MAC and HC were inversely associated with elevated estimated ASCVD risk, although these findings should be interpreted cautiously, as they may reflect underlying differences in age, body composition, frailty, sarcopenia, or other chronic disease-related factors rather than direct protective effects.
The findings of the present study suggest that anthropometric indices beyond BMI may provide additional insights into cardiovascular risk stratification in this population. However, given the cross-sectional design and the use of estimated ASCVD risk categories rather than incident cardiovascular events, these indices should not yet be considered stand-alone screening or diagnostic tools. Further prospective studies are needed to validate these associations against observed cardiovascular outcomes, evaluate their performance relative to established cardiovascular risk prediction tools, and determine appropriate population-specific cut-offs for clinical application.
Author Contributions
Phyllis Tawiah: conceptualization, methodology, supervision, validation, writing – original draft, writing – review and editing. Kwadwo F. Gyan: conceptualization, methodology, supervision, validation, writing – original draft, writing – review and editing. Isaac Amoah: conceptualization, methodology, supervision, validation, writing – original draft, writing – review and editing. Ibok N. Oduro: supervision, validation, writing – review and editing. Julius K. Karikari: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Douglas A. Opoku: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Emmanuel Konadu: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Ebenezer O. A. Ansah: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Solomon Gyabaah: supervision, validation, writing – review and editing. Godfred K. Twumasi: methodology, formal analysis, validation, writing – original draft, writing – review and editing. Nana K. Ayisi-Boateng: supervision, validation, writing – review and editing.
6 Acknowledgments
The authors would like to thank the staff of the Metro Health Hospital for their cooperation during this study.
Funding
The authors have nothing to report.
Ethics Statement
Ethical approval for this study was sought and obtained from the Kwame Nkrumah University of Science and Technology Committee for Human Research and Publication Ethics prior to the start of the work (Approval number: CHRPE/AP/035/22). Written informed consent was obtained from all participants before they participated in the study.
Conflicts of Interest
The authors declare no conflicts of interest.
Transparency Statement
Phyllis Tawiah and Isaac Amoah the corresponding authors of this work affirm that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Data is available from the corresponding author upon reasonable request.
