European Journal of Cardiovascular Medicine

Hypertension remains a major cause of cardiovascular disease and premature mortality worldwide. Although antihypertensive therapies are widely available, global mean blood pressure has declined only marginally, while prevalence continues to rise, particularly in low- and middle-income countries (LMICs). In 2010, an estimated 31.1% of adults—about 1.39 billion people—were hypertensive, with LMICs (31.5%, ≈ 1.04 billion) carrying a greater burden than high-income countries (28.5%, ≈ 349 million) [1]. Urbanisation and lifestyle changes—reduced physical activity, energy-dense diets, psychosocial stress, and shifting social behaviours—are major contributors to this trend [2].

Urban adults frequently exhibit multiple behavioural risks (tobacco, alcohol, poor diet, inactivity) alongside metabolic abnormalities (obesity, dysglycaemia, dyslipidaemia), each independently linked to hypertension [3]. Evidence shows that these factors cluster within individuals, amplifying overall risk. A national survey in China found each additional behavioural risk raised the likelihood of higher blood pressure by 9–20% [4], while South African data demonstrated co-occurrence of poor diet, alcohol use, and hypertension across urban and rural settings [5].

Such clustering is particularly relevant in urban populations, where sedentary work, processed foods, and chronic stress foster central adiposity and insulin resistance, accelerating vascular damage [6]. In Nepal, 68% of adults had at least one cardiometabolic risk factor and 4.7% had all three (obesity, diabetes, hypertension), highlighting clustering as a pervasive problem [7].

In the Indian urban scenario, there is evidence of rising metabolic syndrome prevalence (a prototypical cluster of metabolic risk factors) – one survey in an Iranian urban population reported a metabolic syndrome prevalence of 22.8% and noted that hypertension was one of its components. [8] In a community-based study conducted in an urban slum in Nairobi, Kenya, 26.2% of adults were overweight and 17% were obese, and overweight/obesity remained independently associated with hypertension after multivariable adjustment [9]. Furthermore, behavioural and metabolic risk‐factor studies in Indian and South Asian adult populations highlight the interplay of urbanisation, rising sedentary lifestyles, and diet shift. Collectively, these findings point to an urgent need to examine clustering in urban contexts.

Therefore, this study aims to investigate among urban adults: (1) the prevalence and patterns of clustering of behavioural risk factors (tobacco use, harmful alcohol use, unhealthy diet, physical inactivity) and metabolic risk factors (overweight/obesity, elevated fasting glucose or diabetes, dyslipidaemia, raised waist circumference) and (2) the combined effect of these clusters on the likelihood of hypertension. We hypothesise that individuals with higher counts of clustered behavioural + metabolic risk factors will exhibit significantly greater odds of hypertension compared with those with fewer or no clustered risk factors.

Study design and setting

A cross-sectional analytical study was conducted at Department of General Medicine ,SVS Medical College & Hospital , Mahabubnagar (Telangana, India) from June 2020 to December 2020. The target population comprised adults (aged ≥ 18 years) residing in the urban catchment area of the institution for at least 6 months. The study aimed to assess clustering of behavioural and metabolic risk factors and their combined association with hypertension.

Study population and eligibility criteria

All eligible urban adults attending the outpatient department or community screening camps associated with the hospital during the study period were considered. Inclusion criteria were: age ≥ 18 years; residence in the urban area of Mahbubnagar for ≥6 months; willingness to provide written informed consent and to undergo measurements and biochemical testing. Exclusion criteria included known secondary hypertension (e.g., renal disease, endocrine disorders), pregnancy, and inability or unwillingness to comply with study procedures.

Sample size calculation

Based on previous Indian community studies, a prevalence of hypertension among urban adults of approximately 25% was assumed [10]. Using the formula for estimating a proportion in a single cross-sectional survey:

Where:

• = 1.96 for 95% confidence interval
• = 0.25 (assumed prevalence)
• = allowable absolute error (0.05)

Allowing for a non-response rate of 15%, the sample size was increased:

Therefore, a total sample of 340 individuals was targeted for recruitment.

Sampling procedure

A systematic random sampling technique was employed. From the list of adult individuals attending the outpatient department or community screening on designated days, every kth eligible person was selected (where k = total flow / required sample per day) until the required sample size was achieved. Information on non-respondents (age, sex, reason) was recorded.

Data collection and measurements

Sociodemographic and behavioural data

A structured interviewer-administered questionnaire was used, covering sociodemographic variables (age, sex, education, occupation, socioeconomic status) and behavioural risk factors: tobacco use (current, former, never), alcohol consumption (frequency, quantity), dietary habits (fruit/vegetable intake, high-salt diet, processed foods), and physical activity (using the Global Physical Activity Questionnaire – short form) [11]. Behavioural risk factor definitions were adopted in line with the WHO STEPS framework [12].

Anthropometric and Metabolic Measurements

Height was measured to the nearest 0.1 cm using a stadiometer with the participant standing upright and barefoot. Weight was recorded to the nearest 0.1 kg using a calibrated digital scale with light clothing. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m²). Waist circumference was measured midway between the lowest rib and the iliac crest at the end of normal expiration, and hip circumference at the widest part of the buttocks; waist-hip ratio was computed accordingly [13,14].

BP was measured using a standard mercury sphygmomanometer or a validated automated device, after the participant had been seated quietly for at least five minutes, with the arm supported and the cuff size appropriate to arm circumference. Two readings were taken five minutes apart, and their average recorded. Hypertension was defined as systolic BP ≥ 140 mmHg and/or diastolic BP ≥ 90 mmHg, or current use of antihypertensive medication, following JNC 7 and ACC/AHA 2017 guidelines [15,16]. Fasting venous blood samples were drawn after 8–12 hours of fasting to estimate fasting plasma glucose and serum lipids (total cholesterol, HDL-cholesterol, triglycerides, LDL-cholesterol). Dysglycaemia was defined according to the American Diabetes Association (ADA) criteria—fasting plasma glucose ≥ 126 mg/dL or previously diagnosed diabetes [17]. Dyslipidaemia was defined using NCEP-ATP III cut-offs for abnormal total cholesterol, HDL, LDL, or triglycerides [18].

Definition of risk-factor clustering

Behavioural risk factors considered were current tobacco use, hazardous alcohol consumption, low fruit/vegetable intake (<5 servings/day), high salt intake (>5 g/day), and physical inactivity (less than 150 minutes of moderate activity/week). Metabolic risk factors included overweight/obesity (BMI ≥ 25 kg/m^2 as per Asian criteria), central obesity (waist circumference ≥90 cm men / ≥80 cm women), dysglycaemia, dyslipidaemia, and raised waist-hip ratio. Clustering was defined as presence of two or more risk factors (behavioural and/or metabolic) in the same individual. This operationalisation aligns with previous work on clustering of CVD risk factors. [19][20]

Data management and statistical analysis

Data were entered into EpiData software and cleaned with consistency checks. Analysis was performed using SPSS version 22.0. Descriptive statistics (means, standard deviations for continuous variables; frequencies, percentages for categorical variables) were computed. Patterns of clustering were explored by cross-tabulation of number of risk factors (0, 1, 2, ≥3) by hypertension status. The combined effect of clustered risk factors on hypertension was evaluated using logistic regression analysis: univariate odds ratios (OR) and 95% confidence intervals (CI) were calculated, and then multivariable logistic regression was performed adjusting for age, sex, education and socioeconomic status. A p-value < 0.05 was considered statistically significant.

Ethical considerations

The study was reviewed and approved by the Institutional Ethics Committee of SVS Medical College. Written informed consent was obtained from all participants. Confidentiality of individual data was maintained, and participants found to have elevated blood pressure, dysglycaemia or dyslipidaemia were counselled and referred for appropriate care.

Table 1 presents the sociodemographic profile of the 340 urban adults included in the study. The mean (SD) age of participants was 41.3 (± 12.7) years, reflecting a relatively young-to-middle-aged urban cohort, which typically represents the productive working population vulnerable to lifestyle changes associated with urbanization. Males constituted 52.4 % of the sample, indicating a near-equal sex distribution and minimizing gender bias in exposure estimation. In terms of educational attainment, 43.2 % had secondary-level education, whereas 30.9 % reported primary education or less, highlighting the mixed literacy levels in this urban setting. Approximately 26 % had tertiary education, reflecting a moderate representation of educated professionals. Socio-economic stratification showed that 40 % of participants belonged to the low SES group, 46.8 % to the middle SES, and 13.2 % to the high SES category. This profile is consistent with semi-urban and small-city populations in India, where a majority fall within low-to-middle income brackets.

Table 1: Sociodemographic characteristics

Table 2 summarizes the prevalence of key behavioural and metabolic risk factors observed in the cohort. Among behavioural risks, low fruit and vegetable intake (57.9 %) was the most prevalent, underscoring suboptimal dietary quality—a common feature in rapidly urbanizing environments where consumption of processed foods and refined carbohydrates increases. High salt intake was reported by 43.8 %, aligning with earlier national surveys linking excessive sodium consumption with urban dietary patterns. Physical inactivity was present in 37.4 %, consistent with the mechanization of daily activities and reduced leisure-time exercise. The prevalence of tobacco use (30.3 %) and alcohol consumption (25 %) also reflects the persistence of behavioural risks in urban adults, particularly among men.

Regarding metabolic factors, overweight/obesity (40.3 %) and central obesity (37.9 %) were predominant, confirming the nutritional transition toward positive energy balance. Dyslipidaemia affected 32.9 % of participants, while dysglycaemia was present in 12.4 %, values comparable to national NCD risk-factor surveillance estimates. Overall, hypertension was detected in 61.2 % of individuals—remarkably high for an urban cross-section—suggesting a strong clustering of modifiable risk factors and possibly undiagnosed cases.

Table 2: Prevalence of behavioural and metabolic risk factors

Table 3 provides a detailed view of risk-factor clustering within individuals. For behavioural factors, the modal class was two risk factors (34.1 %), followed by one (29.7 %) and three (19.4 %). Only 7.4 % had no behavioural risk, demonstrating that nearly all adults exhibit at least one unhealthy lifestyle habit. Among metabolic factors, one (39.1 %) and two (25.9 %) risk factors were most common, with only 24.4 % showing none. This indicates a substantial burden of early metabolic derangements even among relatively young adults.

The total risk-factor count, combining both behavioural and metabolic domains, revealed a right-skewed pattern: 25 % had three risk factors, 22.6 % had four, and 12.1 % had five. Only 2.1 % were free from any risk factor, while a small but important subset (≈ 5.6 %) had six or more concurrent risks. The presence of individuals with up to nine risk factors underscores the complexity of multimorbidity in modern urban health contexts.

Table 3: Combined Distribution of Behavioural, Metabolic and Total Risk-Factor Counts

Table 4 summarizes the adjusted odds ratios (aORs) for hypertension after controlling for confounders. Age emerged as a significant independent predictor: for every 10-year increase, the odds of hypertension increased by ~1.42 times (aOR = 1.42; 95 % CI 1.17–1.72; p < 0.001), consistent with the biological progression of vascular stiffness and metabolic decline with age. Male sex was not significantly associated with hypertension (aOR = 0.95; p = 0.84), possibly due to the high prevalence of behavioural risks in both sexes within this cohort. Similarly, low SES showed a non-significant but positive trend (aOR = 1.27; p = 0.32), indicating a potential social gradient that may reach significance in larger samples.

Among clustering variables, neither behavioural cluster ≥ 2 (aOR = 0.78; p = 0.41) nor metabolic cluster ≥ 2 (aOR = 0.91; p = 0.75) individually predicted hypertension after adjustment. This suggests overlapping contributions that lose statistical independence when total clustering is modelled simultaneously. However, participants with ≥ 4 total clustered risk factors exhibited more than double the odds of hypertension (aOR = 2.29; 95 % CI 1.19–4.39; p = 0.012), a strong and statistically significant association.

Table 4: Multivariable Logistic Regression (Hypertension)

Figure 1 illustrates the adjusted odds ratios (aORs) with 95 % confidence intervals for the principal predictors of hypertension derived from the multivariable logistic regression model (Table 4). The x-axis represents the aORs on a logarithmic scale, and the vertical dashed line at 1.0 indicates the null reference (no association). The figure shows a clear upward shift for the total cluster ≥ 4, where the entire confidence interval lies to the right of 1.0, signifying a statistically significant positive association with hypertension. Participants with four or more combined behavioural + metabolic risk factors had approximately twice the odds of being hypertensive compared with those with fewer clustered risks, confirming the cumulative burden effect.

In contrast, the error bars for behavioural cluster ≥ 2, metabolic cluster ≥ 2, male sex, and low socio-economic status (SES) cross the null line, indicating no independent significant associations after adjustment for other covariates. However, the direction of effect for low SES and clustering variables remains on the right side of the reference line, suggesting a potential positive trend that might reach significance in larger samples. Finally, age (per 10 years) consistently appears on the right of the null, reaffirming that increasing age is a robust independent predictor of hypertension in this urban adult population.

In this cross-sectional study of 340 urban adults resident around SVS Medical College, Mahbubnagar, we observed a high prevalence of both behavioural and metabolic risk factors, frequent clustering of these risks, and importantly, a strong association between a high total cluster count (≥ 4 risk-factors) and hypertension. Our findings offer key insights into how multi-domain risk-aggregation operates in urban Indian populations, and have implications for prevention, screening and public‐health policy.

First, the high burden of individual risk factors observed in our sample—low fruit/vegetable intake 57.9 %, high salt intake 43.8 %, physical inactivity 37.4 %, overweight/obesity 40.3 %, central obesity 37.9 % and hypertension 61.2 %—points to a population already well into the non-communicable disease (NCD) transition. Previous Indian studies have similarly documented rising trends of obesity, dyslipidaemia and hypertension in urban settings [6,21]. For example, in a nationally-representative India survey, hypertension prevalence in adults was approximately 27% in 2012-14 [22]. Our elevated hypertension prevalence likely reflects the enriched urban adult sample and high clustering of risks.

Second, as detailed, risk-factor clustering was common: only ~7.4 % had zero behavioural risks, and only ~24.4 % had zero metabolic risks. The modal behavioural count was two risks (34.1 %), while ~22.6 % had four total risks and ~12.1 % had five. These results align with prior evidence showing that behavioural risks rarely exist in isolation, especially in urban environments [3,21]. For example, Ramachandran et al. found significant clustering of CVD risk factors in urban Asian Indians [21]. The observation of clustering across behavioural and metabolic domains emphasises the need to view risk in a multi-dimensional way.

Third, our multivariable model confirms that age remains a robust independent predictor of hypertension (aOR ~1.42 per 10 years). By contrast, when considered singly, behavioural cluster ≥2 and metabolic cluster ≥2 did not reach significance after adjustment (p = 0.405 and 0.749 respectively). However, participants with a total cluster ≥4 had more than double the odds of hypertension (aOR ≈ 2.29; 95% CI 1.19–4.39; p = 0.012). This demonstrates that while individual risk domains matter, it is the aggregate risk‐burden that signals major elevation in hypertension risk—consistent with findings from China that greater number of behavioural risks correlated with higher BP levels [4].

The finding that a threshold of four or more combined behavioural/metabolic risks is associated with a marked increase in hypertension odds suggests a possible tipping-point effect rather than a linear incremental effect. The accumulation of multiple adverse exposures likely overwhelms compensatory physiological mechanisms (vascular endothelium, insulin sensitivity, autonomic regulation) leading to sustained blood-pressure elevation. Behavioural risks (tobacco, high salt, low fruit/vegetable consumption, inactivity) funnel into metabolic dysregulation (elevated BMI, central obesity, dysglycaemia, dyslipidaemia) which in turn drives vascular damage and hypertension. Once multiple such pathways co-exist in a single individual, synergistic damage may accelerate the transition to hypertension.

Our observation that individual behavioural or metabolic clusters did not show independent significance may reflect pathway overlap (i.e., metabolic derangements mediate some of the behavioural effects) and insufficient sample-size to detect moderate effects. However, the presence of a strong effect for high total cluster suggests that interventions should focus on the aggregate risk‐profile rather than isolated risk behaviours.

These findings are consistent with clustering literature. An urban Indian study documented clustering of CVD risk factors in Puducherry and found higher odds among older, male and lower-SES adults [23]. Internationally, Hong et al. showed, in a Chinese survey, that clustering of risk factors produced worse outcomes in both rural and urban populations [24], while Li et al. demonstrated a gradient in BP levels with increasing number of behavioural risks [4]. Our results extend these by combining both behavioural and metabolic domains and linking clustering to hypertension rather than just risk-factor presence.

Public-health implications

The public-health implications of our study are multifold.

Integrated screening and risk stratification: Our results suggest that screening programmes should not focus only on single risk behaviours (e.g., smoking, inactivity) or metabolic factors (e.g., BMI, lipids) in isolation. Instead, programmes should count the number of concurrent risks. Individuals with ≥ 4 risks could be flagged for intensive intervention. This approach aligns with precision‐public‐health strategies focusing resources on high-risk clusters.

Targeted prevention: The behavioural risk pattern (high salt, low fruit/vegetable intake, inactivity) indicates modifiable exposures that are amenable to lifestyle interventions. However, as clustering often precedes metabolic derangement, upstream prevention (e.g., promotion of physical activity, healthy diet, salt reduction) in younger (30–50 years) urban adults may delay or prevent hypertension onset. Given the high prevalence of overweight/obesity in our sample, interventions must integrate weight management and waist-circumference reduction.

Policy and urban design: Urbanisation in India has altered food environments (easy access to processed foods) and reduced physical activity (sedentary work, mechanised transport). Policy initiatives—such as salt-reduction regulation, urban planning for walkability, green spaces, fresh-produce markets—are key to modifying the upstream environment. The clustering phenomenon implies that environmental exposures affect multiple behavioural pathways simultaneously; hence multi-sectoral collaboration is needed.

Clinical practice: For clinicians in urban primary care, these findings emphasise the need to assess not just individual risk factors, but a composite risk‐count in each patient. A patient with two behavioural risks and two metabolic risks (i.e., total cluster = 4) may warrant early BP screening, lifestyle counselling and possibly more intensive follow-up. Also, given that hypertension prevalence in our cohort was 61.2 %—much higher than national averages—clinicians in similar urban settings should adopt proactive screening for undiagnosed hypertension.

Strengths and limitations

Strengths of our study include the assessment of both behavioural and metabolic risk-factors, clustering across domains, and the linkage of clustering to hypertension in an Indian urban adult population—a context with high NCD transition. The use of standard measurement protocols (anthropometry, BP, biochemical data) strengthens internal validity.

However, several limitations must be acknowledged. First, the cross-sectional design means causality cannot be established; reverse causation (hypertension leading to behaviour change) remains a possibility. Second, self-reported behavioural data (tobacco, alcohol, diet, physical activity) may be affected by recall and social-desirability bias. Third, our sample—drawn from hospital’s urban field practice area—may not fully represent the general urban adult population; the high hypertension prevalence suggests a possible selection bias. Fourth, residual confounding may exist (e.g., family history of hypertension, sleep disorders, stress levels, salt sensitivity) that we did not address. Fifth, our threshold of ≥4 risk-factors was determined empirically and may lack external validation; replication in larger and more diverse samples is warranted.

Finally, while our model adjusted for age, sex and SES, other variables such as dietary sodium quantified objectively, physical activity measured by accelerometer, and longitudinal follow-up would further strengthen causal inference.

Future research directions

Going forward, several avenues are indicated:

Longitudinal cohort studies in urban Indian settings that track risk‐factor accumulation, clustering trajectories and incident hypertension and cardiovascular outcomes. This would clarify temporal sequence and allow estimation of population‐attributable risk of clustering.

Intervention studies targeting individuals with high cluster counts (≥4) to ascertain whether multi-component lifestyle interventions (diet, activity, weight, salt) reduce hypertension incidence more effectively than standard single–risk‐factor interventions.

Stratified analyses exploring whether clustering patterns differ by sex, age group, socioeconomic strata or urban neighbourhood (e.g., informal settlements vs formal housing).

Mechanistic research linking clustering to vascular biomarkers (e.g., arterial stiffness, endothelial function, inflammatory markers) to delineate how multiple risks converge to produce hypertension.

Public-health implementation research to test screening algorithms based on risk-count, and to evaluate cost-effectiveness of cluster-based targeting in resource-constrained urban Indian settings.

In sum, our study among urban adults in Mahbubnagar demonstrates that behavioural and metabolic risk-factor clustering is common and that a total risk count of ≥ 4 is significantly associated with more than doubling of hypertension risk. Age remains a strong non-modifiable predictor, but the composite risk burden is a critical modifiable determinant amenable to early prevention. These findings support a shift in hypertension prevention and control—from singular risk-factor focus to a composite risk-cluster paradigm—in urban Indian contexts undergoing rapid epidemiologic transition.

1. Mills KT, Stefanescu A, He J. The global epidemiology of hypertension. Nat Rev Nephrol. 2020 Apr;16(4):223-237. doi: 10.1038/s41581-019-0244-2. Epub 2020 Feb 5. PMID: 32024986; PMCID: PMC7998524.
2. Sarki AM, Nduka CU, Stranges S, Kandala NB, Uthman OA. Prevalence of Hypertension in Low- and Middle-Income Countries: A Systematic Review and Meta-Analysis. Medicine (Baltimore). 2015 Dec;94(50):e1959. doi: 10.1097/MD.0000000000001959. PMID: 26683910; PMCID: PMC5058882.
3. Rajkumar E, Romate J. Behavioural Risk Factors, Hypertension Knowledge, and Hypertension in Rural India. Int J Hypertens. 2020 Mar 9;2020:8108202. doi: 10.1155/2020/8108202. PMID: 32231796; PMCID: PMC7085393.
4. Li Y, Feng X, Zhang M, Zhou M, Wang N, Wang L. Clustering of cardiovascular behavioral risk factors and blood pressure among people diagnosed with hypertension: a nationally representative survey in China. Sci Rep. 2016 Jun 9;6:27627. doi: 10.1038/srep27627. PMID: 27279273; PMCID: PMC4899782.
5. van Zyl S, van Rooyen FC, Joubert G, Kruger WH, Walsh CM. A Comparison of the Socio-Behavioral-Metabolic Risk Profiles and Associated Factors for Chronic Diseases of Lifestyle in Urban and Rural Communities in Central South Africa. Front Public Health. 2020 Oct 16;8:570676. doi: 10.3389/fpubh.2020.570676. PMID: 33178661; PMCID: PMC7596254.
6. Shokeen D, Aeri BT. Prevalence of Cardio-metabolic Risk Factors: A Cross-sectional Study among Employed Adults in Urban Delhi, India. J Clin Diagn Res. 2017 Aug;11(8):LC01-LC04. doi: 10.7860/JCDR/2017/29087.10336. Epub 2017 Aug 1. PMID: 28969161; PMCID: PMC5620802.
7. Gyawali B, Mishra SR, Ghimire S, Hansen MRH, Shah KJ, Subedee KC, Soti PB, Neupane D, Kallestrup P. The burden and correlates of multiple cardiometabolic risk factors in a semi-urban population of Nepal: a community-based cross-sectional study. Sci Rep. 2019 Oct 25;9(1):15382. doi: 10.1038/s41598-019-51454-9. PMID: 31653888; PMCID: PMC6814741.
8. Shahbazian H, Latifi SM, Jalali MT, Shahbazian H, Amani R, Nikhoo A, Aleali AM. Metabolic syndrome and its correlated factors in an urban population in South West of Iran. J Diabetes Metab Disord. 2013 Feb 8;12(1):11. doi: 10.1186/2251-6581-12-11. PMID: 23497506; PMCID: PMC3598198.
9. Olack B, Wabwire-Mangen F, Smeeth L, Montgomery JM, Kiwanuka N, Breiman RF. Risk factors of hypertension among adults aged 35-64 years living in an urban slum Nairobi, Kenya. BMC Public Health. 2015 Dec 17;15:1251. doi: 10.1186/s12889-015-2610-8. PMID: 26679701; PMCID: PMC4683777.
10. Anchala R, Kannuri NK, Pant H, Khan H, Franco OH, Di Angelantonio E, Prabhakaran D. Hypertension in India: a systematic review and meta-analysis of prevalence, awareness, and control of hypertension. J Hypertens. 2014 Jun;32(6):1170-7. doi: 10.1097/HJH.0000000000000146. PMID: 24621804; PMCID: PMC4011565.
11. Armstrong R, Bull F. Development of the World Health Organization Global Physical Activity Questionnaire (GPAQ). J Public Health. 2006;14(2):66-70. doi:10.1007/s10389-006-0024-x.
12. World Health Organization. STEPwise Approach to Noncommunicable Disease Risk Factor Surveillance (STEPS) Manual. Geneva: WHO; 2017
13. World Health Organization. WHO Expert Committee on Physical Status: The Use and Interpretation of Anthropometry. WHO Technical Report Series 854. Geneva: WHO; 1995.
14. Misra A, Vikram NK. Clinical and pathophysiological consequences of abdominal adiposity and abdominal adipose tissue depots. Nutrition. 2003 May;19(5):457-66. doi: 10.1016/s0899-9007(02)01003-1. PMID: 12714101.
15. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 2003 Dec;42(6):1206-52. doi: 10.1161/01.HYP.0000107251.49515.c2. Epub 2003 Dec 1. PMID: 14656957.
16. Whelton PK, Carey RM, Aronow WS, Casey DE Jr, Collins KJ, Dennison Himmelfarb C, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018 Jun;71(6):1269-1324. doi: 10.1161/HYP.0000000000000066. Epub 2017 Nov 13. Erratum in: Hypertension. 2018 Jun;71(6):e136-e139. doi: 10.1161/HYP.0000000000000075. Erratum in: Hypertension. 2018 Sep;72(3):e33. doi: 10.1161/HYP.0000000000000080. PMID: 29133354.
17. American Diabetes Association. Diabetes Diagnosis & Tests. Arlington (VA): ADA; [cited 2019 Aug 25]. Available from: https://diabetes.org/about-diabetes/diagnosis
18. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002 Dec 17;106(25):3143-421. PMID: 12485966.
19. Khanal MK, Mansur Ahmed MSA, Moniruzzaman M, Banik PC, Dhungana RR, Bhandari P, Devkota S, Shayami A. Prevalence and clustering of cardiovascular disease risk factors in rural Nepalese population aged 40-80 years. BMC Public Health. 2018 May 31;18(1):677. doi: 10.1186/s12889-018-5600-9. PMID: 29855293; PMCID: PMC5984400.
20. Thangiah N, Chinna K, Su TT, Jalaludin MY, Al-Sadat N, Majid HA. Clustering and Tracking the Stability of Biological CVD Risk Factors in Adolescents: The Malaysian Health and Adolescents Longitudinal Research Team Study (MyHeARTs). Front Public Health. 2020 Mar 17;8:69. doi: 10.3389/fpubh.2020.00069. PMID: 32257989;
21. Ramachandran A, Snehalatha C, Latha E, Satyavani K, Vijay V. Clustering of cardiovascular risk factors in urban Asian Indians. Diabetes Care. 1998 Jun;21(6):967-71. doi: 10.2337/diacare.21.6.967. PMID: 9614615.
22. Bischops AC, De Neve JW, Awasthi A, Vollmer S, Bärnighausen T, Geldsetzer P. A cross-sectional study of cardiovascular disease risk clustering at different socio-geographic levels in India. Nat Commun. 2020;11:5891. doi:10.1038/s41467-020-19647-3.
23. Selvaraj K, Kar SS, Ramaswamy G, Premarajan KC, Saya GK, Kalidoss V. Clustering of cardiovascular disease risk factors - Syndemic approach: Is sit a time to shift toward integrated noncommunicable disease clinic? Indian J Public Health. 2019 Jul-Sep;63(3):186-193. doi: 10.4103/ijph.IJPH_158_18. PMID: 31552846.
24. Hong X, Ye Q, He J, Wang Z, Yang H, Qi S, et al. Prevalence and clustering of cardiovascular risk factors: a cross-sectional survey among Nanjing adults in China. BMJ Open. 2018;8(6):e020530. doi:10.1136/bmjopen-2017-020530.