The prevalence of hypertension, obesity and dyslipidemia in individuals of over 30 years of age belonging to minorities from the pasture area of Xinjiang

Nearly all the Kazaks of the province of Xinjiang live in the pasture areas which are mostly located in the north of the region. We used a stratified sampling method to select a representative sample of the general population of Kazaks of this area. Three counties (Hefeng, Fuhai and Xinyuan County) were chosen and, based on the government record of registered residence, one participant was randomly selected from each household.

In this way, a total of 2492 participants, 1051 men and 1441 women, were randomly selected from 28 villages of the three counties and were invited to participate. Those who were younger than 30 years old, 84 males and 156 females, were excluded from the analysis, which is motivated by the higher risk of secondary hypertension in individuals younger than 30 years. Furthermore, one 68-year-old female dwarf with height 93 cm was excluded as well. Finally, 2251 individuals (90.33%) were analyzed, 967 men (42.96%) and 1284 women (57.04%). The proportion of participants from the three counties were: Fuhai (n = 566, 25.14%), Hefeng (n = 1030, 45.76%) and Xinyuan (n = 655, 29.10%). The ethnic distribution of the subjects in the present study were: 1604 Kazaks (71.26%), 418 Uygurs (18.57%) and 81 (3.6%) Mongolians. Additionally, 148 Han immigrants, 6.57% of the total sample, were included in this study. As their original provinces or cultural and geographical regions were unknown, all Han immigrants were considered as one group for the purpose of analysis.

Blood pressure was measured three times by the same physician from the hospital's hypertension department. After resting for 5 to 10 minutes in a quiet and warm room, blood pressure was measured with a table sphygmomanometer. For the purposes of analysis, the mean of the three measured values was considered. A wider cuff was used if the participant's upper arm circumference was above 32 cm. New hypertension was defined as systolic blood pressure greater than 140 mmHg and/or diastolic blood pressure greater than 90 mmHg. Known hypertension was defined as hypertension or intake of antihypertensive drugs as self-reported in the questionnaire. Hypertension was in any case defined as either new or known according to the standard guidelines [10]. The awareness of hypertension was defined as having been informed of the hypertensive status by a health professional in the previous year.

Body mass index (BMI) was calculated as weight (kg) divided by height squared (m²). Obesity was defined as body mass index (BMI) ≥ 30 kg/m² and overweight as 25 kg/m² ≤ BMI < 30 kg/m².

Blood lipids and lipoproteins were measured on samples obtained after an overnight fast (12 h at least). Total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-c), and low-density lipoprotein cholesterol (LDL-c) were measured in EDTA plasma. Hypercholesterolemia was defined as TC >5.7 mmol/L, increased LDL-c as LDL-c >3.6 mmol/L, hypertriglyceridemia as TGs >1.7 mmol/L, and reduced HDL-c as HDL-c <0.9 mmol/L. Dyslipidemia was defined by the presence of one or more than one abnormal serum lipid concentration.

Smoking habits were coded as current daily smoker, former smoker or lifetime non-smoker. Similarly, the alcohol intake was recorded as current alcohol-intake, former alcohol-intake or no history of alcohol-intake. Furthermore, for participants with current alcohol-intake, the questionnaire had questions concerning present intake of beer, wine, spirits and others. Average daily intake of alcohol was calculated and three categories of drinking groups were defined as follows: no (0 g alcohol/day), <30 g/day and ≥ 30 g/day alcohol intake.

Differences in categorical data among Kazak, Uygur, Mongolian and Han groups were examined by χ² test and differences in quantitative data by ANOVA, followed by Post-hoc test (Tukey). Logistic regression analysis was performed for exposure dose and risk factors of hypertension. All statistical procedures were performed with SAS 9.1.3 (SAS Institute Inc., Cary, NC, USA). All tests were two tailed and the significance level was set at α = 0.05.

Table 1 shows the main blood pressure related variables, metabolic and lifestyle characteristics respectively, of all participants from the pasture area of Xinjiang.

The prevalence and awareness of hypertension in all referred participants were 51.93% and 15.28%, respectively. No significant difference between men and women was detected (53.98% vs. 50.39%, χ² = 2.85, P = 0.09) in the prevalence of hypertension, as well as in the awareness of hypertension (14.89% vs. 15.58%, χ² = 0.20, P = 0.66). The prevalence and awareness of hypertension in the four ethnic groups are shown in Figure 1. The prevalence of hypertension was highest in Uygur, with 54.55%, and significantly lower in Mongolian, comprising 39.51%. In contrast, the awareness of hypertension was not statistically different between the four ethnic groups (χ² = 1.48, P = 0.69) and it was quite low in all of them. The prevalence of overweight and obesity in the whole population was 47.96%. There was no difference between men and women (48.5% vs. 47.5%, χ² = 0.211 P = 0.65) while the prevalence of dyslipidemia was significantly higher in men (52.4% vs. 46.8%, χ² = 6.41, P = 0.011). The prevalence of overweight and obesity in Mongolians was 62.5%, which was significantly higher than that in Uygurs (62.5% vs. 48.9%, P = 0.003) and that found in Kazaks (62.5% vs. 50.3%, P = 0.005). There seemed to be an even greater difference between Mongolians and the Han population (62.5% vs. 47.3%, P > 0.05) but this was not statistically significant due to the rather small sample size of Han. Similarly, the prevalence of dyslipidemia was 50.79% in the whole study population and varied greatly among the four ethnic groups (F = 15.53, P < 0.0001): it was 53.5% in Kazaks, 49.3% in Mongolians, 47.3% in Han and 34.8% in Uygurs. Pair-wise comparisons of the prevalence of dyslipidemia between the four ethnic groups revealed that only Uygurs had significantly smaller prevalence than Kazaks (P < 0.0001).

Table 2 shows the odds ratio (OR) of simple and multiple logistic regressions of important factors. Factors associated with hypertension are age, overweight and obesity, plasma TC level and current high alcohol intake (≥30 g/d). Multiple logistic regression showed that, after adjustments for important factors from the simple logistic regression, the ethnic attribution plays an additional role. In particular, the prevalence of hypertension in Mongolians (as well as in Han) was significantly smaller than that found in Kazaks. Within Kazaks, similar factors were detected associating with hypertension as dependent variables: age (OR [95%CI]:1.08 [1.06-1.09], P < 0.0001), total cholesterol (1.38 [1.17-1.64], P = 0.0002), BMI (1.06 [1.02-1.10], P = 0.0048), and more than 30 g alcohol intake per day (4.04 [1.03-15.89], P = 0.0455).

According to the results showed in Table 1, statistically significant differences were detected in nearly all variables between the four different ethnic groups, except for sex and age. Further comparison between Kazaks and the other three groups was performed (Table 3), where the average SBP, DBP, PP, MAP, PPI and PP/MAP was significantly higher in Kazaks. Though the mean BMI in Kazaks and Uygurs was not statistically different (P = 0.9816), the average plasma lipid profile level was significantly higher in Kazaks than in Uygurs. The mean TC level between Kazaks and the other ethnic groups was not different, but the LDL-c level in Kazaks was significantly the highest. There was some interaction between ethnics and overweight/obesity as determinants of hypertension. The effect of overweight/obesity on both SBP and DBP in Mongolians was significantly smaller than that in Kazaks (P = 0.001 and P = 0.026 respectively). Furthermore, the effect of overweight/obesity on SBP was significantly smaller in Hans than in Kazaks (P < 0.0001).

This epidemiological survey mainly reported considerable prevalence of hypertension (about 52%), overweight/obesity (near 48%) and dyslipidemia (over 50%), combined with low awareness of hypertension (15% only) in the representative sample of the population living in the pasture area of Xinjiang.

The Kazaks, Mongolians and Uygurs recruited in this study are herdsmen or peasants. Their ancestral allocation to the pasture area of Xinjiang goes back for hundreds of years, and they form a relatively homogeneous group with regard to socioeconomic status, occupational and dietary exposures. Few large-scale epidemiological studies have focused on the prevalence of hypertension, overweight/obesity and dyslipidemia among the minorities from the pasture area of Xinjiang in the last decade. Although the large-scale population-based survey in 1995 on the prevalence of hypertension in different ethnic groups in China [11] reported much lower prevalence of the above-mentioned conditions, to our knowledge, most Kazaks and Uygurs involved in the study came from big cities or towns in Xinjiang, while the Mongolians were from the Inner Mongolia Autonomous Region. Therefore, different prevalences were reported in diverse populations and regions, which is a common result of epidemiological investigations. In some small-scale surveys on hypertension in Kazaks, Uygurs and Mongolians of Xinjiang, the results varied greatly. For example, a study including 119 cases [12] showed that the prevalence of hypertension in residential Kazaks of Urumqi was 42.0%. Liu et al [9] data showed the prevalence of Uygurs as 24.3% (264 cases) and Kazaks as 37.9% (99 cases). Data from the Hebukesel pastoral area of Xinjiang demonstrated that the detection rate of hypertension in Kazaks and Mongolians was 55.09% in 2003[13]. In 2005, Wang et al [8] reported in a study of 3732 Mongolians, Kazaks, Uygurs and Hans over 30 years of age in Bortala Prefecture of Xinjiang that the prevalence of overweight and obesity were 36.02% and 27.39% respectively, which was lower than that in our data. Additionally, the low awareness of hypertension displayed in the Guideline for the Prevention and Treatment of Hypertension in China (2005) was confirmed in the present study. Our data calls for the urgent need to develop effective strategies for prevention and treatment of hypertension in Xinjiang.

Our study confirmed the conventional risk factors for hypertension - age, increased BMI, hypercholesterolemia and high alcohol intake (≥30 g/d) - in all participants and specifically in the Kazak population. It is well understood that the aging process affects hypertension [14]. Also positive association between BMI and blood pressure has been well documented [15]. Our study demonstrates that obese individuals, classified by BMI, have significantly larger odds ratio for hypertension than overweight ones. However, we observed significant interaction effects between ethnics and overweight/obesity both for SBP as well as DBP. In particular Mongolians have the highest prevalence in overweight/obesity but the lowest prevalence of hypertension. However, the number of Mongolians included in this study is relatively small, and further research will be necessary to figure out the reasons for this observation.

The results of a 7-year follow-up study on Finnish men [16] suggests that dyslipidemia characteristic of the metabolic syndrome predicts the development of hypertension. Although TG, HDL-C and LDL-c levels were significantly different in our four ethnic groups (Table 3) multiple logistic regression suggested that hypercholesterolemia was positively associated with hypertension in the whole population, not only in Kazaks, which indicates that hypercholesterolemia is a potential determinants of hypertension.

Numerous studies have suggested that excessive alcohol intake causes an increase in blood pressure [17‐21]. But in recent years, some researchers claimed that light or moderate alcohol intake might actually benefit hypertension patients [22, 23]. Interestingly, in this study, ≥30 g/d alcohol intake was associated with hypertension adjusted for age, sex, smoking and plasma lipid level while the positive association could not be observed for an intake of alcohol <30 g/d. However, only large-scale, prospective and randomized trials might elucidate the actual role of alcohol in hypertension.

Apart from the direct influence of alcohol on blood pressure levels, one might consider that, traditionally among Kazaks, Uygurs and Mongolians in Xinjiang, alcohol consumption is associated with an increased consumption of animal fat or salted food, which could lead to an increase in fibrinogen levels. The male population in particular customarily drinks spirit to deal with the cold weather, as well as having an increased intake of animal fat and salt. Additionally, salted milk tea is enjoyed in large quantities coupled with a low consumption of vegetables, which are scarcely available in the area. Here it should be mentioned that 307 missing values of alcohol intake involved a special alcoholic beverage brewed from horse milk. Over-consumption of this may cause drunkenness, but it is difficult to assess the amount of alcohol within this specific beverage. These missing values may affect the logistic model in the statistical analysis.

Smoking is a well-known risk factor for hypertension and CVD. Association between smoking habits and BP in Japanese men [24], negative dose-effect relationships between the amount of smoking and SBP [25], DBP[26] or both [27] have been reported. On the other hand, some studies have failed to observe a significant dose-effect relationship [28, 29]. In this study, no association was found between cigarette smoking and hypertension, for which two potential reasons should be mentioned:

Some general limitations should be also mentioned. First, the number of minority participants except for Kazaks was too small to allow further analysis. This is a problem when considering the differences between ethnic groups. For example, Han population of Xinjiang mostly consists of immigrants from other provinces, different in cultural background, language, reasons for migration, duration of stay in Xinjiang and age when migrating. Therefore, our conclusions cannot automatically be attributed to all Han immigrants. Second, the study was not primarily designed to compare multiple ethnics and Kazak participants, but the high percentage of residents with different ethnicities in the Xinjiang pasture area made this contingently possible. The data from Mongolians might be biased due to the rather small sample size in this study. Third, blood pressure was measured three times, but only on one occasion, which may overestimate the prevalence of hypertension. Such an over-estimation should, however, be the same in all studied groups, and single-occasion blood pressure measurements are common practice in epidemiological studies [5, 30, 31]. Finally, we unfortunately did not have data on intake of salt and macronutrient factors that may have an impact on blood pressure [32]. With respect to the high prevalence of hypertension, obesity and dyslipidemia found in Kazaks and Uygurs, there may be a risk of residual confounding by factors we were not able to adjust for, like e.g. the physical activity level of participants. However, keeping these limitations in mind, our study may provide a baseline database and preliminary results for further studies on the association of hypertension and its risk factors with lifestyle in different ethnic groups from the pasture area of Xinjiang.