The current study analysing data from three large population-based studies involving Chinese, Malay, and Indian adults free of diabetes, hypertension, and CVD showed that higher levels of blood glucose, HbA1c, and BMI were associated with prehypertension in all three ethnic groups. High levels of LDL in Chinese and triglycerides in Malays were associated with prehypertension. These associations were independent of potential confounding factors including age, sex, education, smoking and alcohol consumption and were consistently present when metabolic variables were analyzed as continuous or categorical variables. There are several notable points in our findings. First, to our knowledge this is the first study that examined the contribution of adverse cardiometabolic risk factors in healthy Asian adults to prehypertension, an earlier stage where preventive efforts have been shown to be effective in delaying or preventing the onset of hypertension and cardiovascular outcomes. Second, the main determinants of prehypertension were metabolic factors that are also well-known risk factors for diabetes. Third, these risk factors were not entirely the same ones identified in other Asian populations. For instance, the main determinants of prehypertension in a Japanese population were BMI and hyperlipidemia .
Our study found that BMI was the consistent risk factor associated with prehypertension across the three ethnic groups particularly among Malays. This finding supports past research which showed that being overweight is an established risk factor for prehypertension [18–21]. In the current study, and previous National Health surveys [8, 22], Malays have been found to have the highest BMI among the three races, and have been reported to be the most overweight, followed by Indians and Chinese . Our finding of an association of BMI with prehypertension is consistent with other Asian population: a previous meta-analysis of 24 studies in China where overweight was associated with 62% increased risk of developing hypertension . Inflammation has been postulated to deregulate blood pressure control leading to prehypertension in individuals with high BMI . People who are overweight and obese exhibit central adiposity and have an expanded visceral adipose tissue compartment. The increasing visceral adipose tissue accumulation renders the visceral adipose tissue dysfunctional, resulting in altered adipose tissue secretions that manifest as enhanced systemic inflammation .
In addition to BMI, our study also found HbA1c and blood glucose to be associated with high odds of prehypertension in all three ethnic groups. Studies have found that adults with a high BMI, in parallel with the enhanced systemic inflammation, tend to develop greater insulin resistance [2, 26–29]. Having a large BMI leads to higher fasting serum insulin levels and, in some individuals, a high-normal level of glycosylated haemoglobin . High insulin levels and glucose concentrations above 5.5 mmol/L are associated with enhanced CVD risk  and people with prehypertension have been found to have increased insulin resistance . In the current study, we found that higher levels of LDL cholesterol in Chinese and higher levels of triglycerides to be associated with prehypertension consistent with previous studies that have reported associations between various lipid components and prehypertension [18, 30, 31]. The increased whole body adipose tissue burden has also been found to foster dyslipidemia .
It is interesting to note that the risk factors identified in our study were not entirely the same ones identified in other Asian populations. Specifically, though BMI was a common risk factor among the three ethnicities in our study, BMI was more prominent among the Chinese living in China [32, 33] and the Japanese;  whereas glycemic biomarkers showed a more notable presence in this group of Asians. This alludes to a possible interplay between nature and nurture on variations in risk factor patterns between populations. Our findings also further support our argument of the need to study these associations separately among major Asian ethnic groups in order to gain a better understanding of the contribution of the key cardiometabolic factors to prehypertension in different ethnic groups. Apart from age and gender, primary or below education was found to be associated with prehypertension in Indians. Educational attainment was selected as one of the proxies of SES because it has been reported as a key determinant of health status, particularly cardiovascular conditions . The finding that a lower level of education is associated with prehypertension is not surprising. For instance, in one study, non-Hispanic white men and women with less years of education were found to be at higher risk of developing hypertension compared with their more educated counterparts . It is possible that people with higher education levels hold occupations that are more stable and usually comes with a higher income bracket than people with a lower education who are often in blue-collar roles. The higher income often equates to more security, access to better food and shelter, and optional life indulgences such as going on relaxing holidays.
Our study has several strengths. Our sample size is very large and it is the first study to document and examine the risk factors of prehypertension in a representative multi-ethnic Asian cohort. However, as our study design was cross-sectional rather than longitudinal, causal inferences were not able to be made and it was not possible to determine whether these patients eventually converted to hypertension in the presence of these risk factors. Future research could follow up these patients and investigate the temporal relationship between BMI, glucose, and lipids and prehypertension. In addition, as the data for this study were collected from three cross-sectional studies conducted in different years, it is possible that differences in the prevalence of risk factors for CVD might have influenced our results. However, reports from the Singapore 2004 and 2010 National Health Surveys suggest that except for overweight/obesity, the prevalence of metabolic risk factors did not vary significantly between the two time points (prevalence of hypertension was 24.9% vs. 23.5% in 2004 vs. 2010, high total blood cholesterol was 18.7% vs. 17.4%, diabetes was 8.2% vs. 11.3% and overweight/obesity was 32.5% vs. 40.1%). Therefore, we do not believe that the main study results would be different if they were conducted in the same period [8, 22]. Finally, readers should keep in mind that our results may not be generalizable to Chinese, Indians, and Malays living in countries other than Singapore.