Prevalence of metabolic syndrome among middle-aged and elderly adults in China: current status and temporal trends

Abstract

Background: Metabolic syndrome (MetS) is a cluster of major risk factors for cardiovascular diseases. We aimed to estimate prevalence and distribution of MetS among middle-aged and elderly adults in China.

Methods: The present analysis used data from a national study in 2014–2015. We defined MetS by different definitions, and compared results of the present study and previous nationally representative studies to illustrate possible temporal changes in MetS prevalence.

Results: The estimated prevalence of MetS was 18.4% by the ATP III criteria, 34.0% by the revised ATP III criteria, and 26.9% by IDF criteria. The prevalence was higher in women, older adults, those with lower education level, and in economically developed regions. Contrasting with previous national studies, adults in urban areas had a lower rate of MetS than those in rural areas (odds ratio 0.94; 95% CI 0.92−0.97). Rural adults had worse deterioration or less improvement in abdominal obesity, overweight, hypertension, and high fasting plasma glucose, than urban adults, which was particularly striking for women.

Conclusion: While measures to prevent and control cardiovascular diseases need to be strengthened in China, rapid increasing risk factors among rural residents and women should be prioritized in making public health policy decisions.

KEY MESSAGES

• Our study assessed prevalence and temporal changes of MetS among Chinese population with the most recently completed and the largest sample size.

• The current prevalence of MetS was higher in women, older adults, those with lower education level, and in economically developed regions and the CVD risk factors among rural residents and women should be prioritised in making public health policy decisions.

• A comparison of results of the present study and previous national studies showed that rural adults had worse deterioration or less improvement in abdominal obesity, overweight, hypertension, and high fasting plasma glucose, than urban adults, which was particularly striking for women.

Keywords:

• Metabolic syndrome
• prevalence
• trends
• epidemiology

Introduction

The morbidity and mortality of cardiovascular disease (CVD) are increasing in China, as in other low-and-middle income countries, while CVD mortality has been declining in high-income western countries since 1980s [1,2,3]. The metabolic syndrome (MetS) refers to a cluster of metabolic factors associated with the risk of CVD, including abdominal obesity, high blood pressure, dyslipidaemia and dysglycaemia [4]. The prevalence and distribution of MetS in population are important indicators for making public health decisions regarding the control of CVD epidemic [5].

There were several nationally representative studies that provided data on the prevalence of MetS in China. The International Collaborative Study of Cardiovascular Disease in Asia (InterASIA) in 2000–2001 was a cross-sectional study of a nationally representative sample of 15,540 Chinese adults aged 35–74 years [6]. In 2007–2008, a sample of 46,024 Chinese aged 20 years and over were included in the National Diabetes and Metabolic Disorders Survey (NDMDS) [7]. The China Health and Nutrition Survey (CHNS) in 2009 examined metabolic risk factors among a sample of 7488 adults [8]. The 2010 China Noncommunicable Disease Surveillance (CNCDS) included 97,098 participants from 31 provinces in China [9]. Results of these national studies showed an increasingly high prevalence of MetS in China. In addition, it was found that the prevalence of MetS was generally higher in women, among older adults, and in urban residents [6–9]. However, it is often difficult to compare the results of different studies, due to different sampling methods, participant age range, and definitions of MetS used.

The China National Stroke Prevention Project (CSPP) survey in 2014–2015 was a recently completed cross-sectional study that included 109,551 participants aged ≥40 years from 30 provinces in China. In the present study, we used data from the CSPP to estimate the current prevalence of MetS in China. We also compared results of CSPP and previous nationally representative studies to reveal changes in the prevalence of MetS over time in China.

Methods

Design and participants of the CSPP study

The CSPP was administrated by the National Project Office of Stroke Prevention and Control, and carried out in 30 provinces in China from October 2014 to November 2015. Using a two-stage stratified cluster sampling method, 200 project areas were first selected in proportion to the local population size and the numbers of counties. Then an urban community and a rural village were selected from each project area as primary sampling units according to geographical locations and suggestions from local hospitals. The cluster sampling method was used in every primary sampling unit and all residents aged ≥40 years were surveyed during the primary screening.

All participants received information on the study and provided a written informed consent to participate. Questionnaire completion and physical examination were conducted by trained staff in primary health care institutions, to collect data on demographic characteristics, medical history, behavioural risk factors, height, weight, waist circumference, and blood pressure. The assessment of stroke risk was based on the following variables: hypertension, atrial fibrillation, current smoking, dyslipidaemia, diabetes, physical inactivity, obesity or overweight (BMI ≥26 kg/m²), and family history of stroke. Physical inactivity was defined as fewer than four times regular physical exercise per week and less than 30 min per session. Participants with ≥3 risk factors or with prior TIA or stroke were considered at high risk of stroke, and those with <3 risk factor but having hypertension, diabetes, atrial fibrillation, or heart valve disorders were categorised as at intermediate risk of stroke. Low risk of stroke was defined as having <3 risk factors and without the above CVD conditions.

According to the screening plan, participants at high risk of stroke were invited for further laboratory tests, carotid ultrasound, and electrocardiogram. However, a large number of participants at intermediate or low risk of stroke were also undergone additional laboratory tests. A total of 726,451 participants were included in the CSPP survey in 2014–2015, and 109,551 of them received additional laboratory tests. In the current study, the analysis of MetS was based on data from 109,551 participants who had completed all laboratory examinations.

Data from other relevant studies

Data from previous nationally representative studies of MetS in China were obtained to compare results of different studies and to illustrate changes in the prevalence of MetS over time. We were able to access the original data from the CHNS [8,10]. For other national scale studies of MetS in China (InterASIA [6], NDMDS [7], and CNDS [9], we had to rely on results reported in published articles. We have attempted to make results from different studies as comparable as possible in terms of participant age and the definition of Mets in analysis.

Definition of metabolic syndrome

The metabolic syndrome was defined by different definitions, including National Cholesterol Education Programme (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (NCEP ATP III) criteria [11], revised NCEP ATP III for Asian-Americans [12], and the updated International Diabetes Federation (IDF) [13] definition (see Supplementary Table S1).

Data analysis

We calculated the prevalence of MetS by gender, age, location (urban or rural), and the level of socio-economic development, weighted by age and sex distribution of the 2014 population in China. Because relatively more participants at high risk of stroke were included in the analysis of MetS, the prevalence of MetS were weighted according to the level of stroke risk in all CSPP participants. The t test and Chi-square test were used for statistically testing continuous and categorical variables. Logistic regression analysis was conducted to compare the risk of MetS among different variables including age, gender, urban/rural location, education level, and economic development. The level of economic development of provinces or municipalities was defined on the basis of the gross domestic product per capita in 2014, and it was divided into three categories: developed (≥10,000 $), intermediately developed (6000–10,000$), and underdeveloped (≤6000$) (in 2014, 6.1428 yuan equalled U.S. $1.00). In addition, the discrepancy in the prevalence of MetS between urban and rural areas was measured by using prevalence rate ratio (PRR = (a/N₁)/(b/N₀)) [14]. Analyses were conducted using SPSS 19.0 (SPSS Inc., Chicago, IL).

Ethical approval

The CSPP study protocol was approved by the Ethics committee of the XuanWu Hospital Institutional Review Board, Capital Medical University (Beijing, China).

Results

Table 1 shows the basic characteristics of the 109,551 participants who were included in the analysis and 616,900 participants without laboratory tests. The included participants were relatively older, and less likely from rural areas or under developed regions. As expected, the risk of stroke was much higher for the included participants compared with those not included. The proportion of low risk participants was 40.2% for the included participants and 76.3% for those who were not included.

Table 1. The main characteristics of CSPP participants in 2014–2015.

Characteristic	Included (n = 109,551)	Not included (n = 616,900)	p value
Age (mean, SD)	59.23 ± 11.01	56.88 ± 11.44	<.001
Age groups (%)			<.001
40–49	22.8	32.3
50–59	28.6	29.5
≥60	48.6	38.2
Sex (%)			<.001
Men	45.4	47.0
Women	54.6	53.0
Education (%)			<.001
Primary school and below	42.3	41.4
Middle school	35.0	37.1
High school or the equivalent	15.5	14.8
University or other tertiary degree	7.3	6.7
Location (%)			<.001
Urban	52.3	46.6
Rural	47.7	53.4
Economic development (%)			<.001
Under developed	19.4	30.3
Intermediately developed	54.2	43.8
Developed	26.4	25.9
Stroke risk (%)			<.001
Low	40.2	76.3
Intermediate	10.0	14.5
High (≥3 risk factors)	35.9	6.6
History of TIA	5.7	1.2
History of stroke	8.3	1.4
Smoking status			<.001
Current smokers (%)	18.0	3.5
Former smokers (%)	3.4	0.5
Never smokers (%)	78.6	96.0
Current drinking (%)	13.8	1.8	<.001
Physical inactivity (%)	28.4	16.7	<.001

The estimated prevalence of MetS by the ATP III criteria without adjusting for stroke risk level is presented in Supplementary Table S2. The prevalence of MetS was clearly associated with the level of stroke risk (Supplementary Table S2). After adjusting for the stroke risk, the estimated prevalence of MetS for Chinese adults aged ≥40 in 2014–2015 was 18.4% by the ATP III criteria, 34.0% by the revised ATP III criteria, and 26.9% by the IDF criteria (Table 2). The risks of MetS were significantly higher in women, older adults, rural residents, adults with low level of education, and in economically developed regions (Table 3).

Table 2. Prevalence of MetS by different definitions in 2014–2015.

		% (95% CI)
	N	NCEP ATP III	Revised NCEP ATP III	IDF
Overall	109,551	18.4 (18.2, 18.7	34.0 (33.7, 34.3)	26.9 (26.7, 27.2)
Sex
Men	49,789	13.2 (12.9, 13.6)	27.1 (26.7, 27.6)	18.4 (18.1, 18.8)
Women	59,762	23.7 (23.3, 24.1)	40.9 (40.5, 41.3)	35.7 (35.3, 36.1)
Age group
40–49	24,937	13.3 (12.8, 13.7)	27.3 (26.7, 27.8)	21.4 (20.8, 21.9)
50–59	31,376	18.5 (18.1, 19.0)	34.2 (33.6, 34.7)	27.4 (26.8, 27.9)
> =60	53,238	22.8 (22.4, 23.2)	39.4 (38.9, 39.8)	31.3 (30.9, 31.7)
Location
Urban	57,298	17.8 (17.5, 18.2)	33.1 (32.7, 33.5)	26.7 (26.3, 27.0)
Rural	52,253	19.6 (19.3, 20.0)	35.6 (35.1, 36.0)	27.7 (27.3, 28.1)
Economic
Underdeveloped	21,272	15.1 (14.5, 15.7)	30.8 (30.2, 31.5)	23.6 (23.0, 24.2)
Intermediately	59,365	19.0 (18.6, 19.3)	34.1 (33.7, 34.5)	28.7 (28.4, 29.1)
Developed	28,914	19.7 (19.1, 20.2)	34.9 (34.3, 35.5)	26.3 (25.8, 26.9)
Smoking status
Never	86,210	17.8 (17.6, 18.1)	33.3 (32.9, 33.6)	26.3 (26.0, 26.6)
Former	3672	32.2 (30.7, 33.7)	49.7 (48.1, 51.3)	41.7 (40.1, 43.3)
Current	19,669	29.1 (28.4, 29.7)	47.2 (46.5, 47.9)	38.8 (38.1, 39.5)
Current drinking
No	94,485	17.8 (17.5, 18.1)	32.8 (32.5, 33.1)	25.8 (25.5, 26.1)
Yes	15,066	31.3 (30.6, 32.0)	51.4 (50.6, 52.2)	44.0 (43.2, 44.8)

Estimated rates of the prevalence were weighted by 2014 population in China and adjusted by level of stroke risk in all CSPP participants.

Table 3. Association between MetS (defined by ATP III) and selected basic variables – results of logistic regression analysis.

Variables	Odds ratio	95% CI	p Value
Sex
Men	1.00
Women	2.11	2.05–2.17	<.001
Age
40-49	1.00
50-59	1.65	1.58–1.72	<.001
≥60	2.05	1.97–2.13	<.001
Location
Rural	1.00
Urban	0.94	0.92–0.97	<.001
Education
Primary school and below	1.00
Middle school	1.01	0.98–1.04	.692
High school or the equivalent	0.97	0.93–1.01	.161
University or other tertiary degree	0.82	0.77–0.87	<.001
Economic development
Under developed	1.00
Intermediately developed	1.08	1.04–1.12	<.001
Developed	1.20	1.15–1.25	<.001

Comparison of results of different studies

We were able to compare sex- and age-specific prevalence of MetS (using the ATP III definition) with the results of other three nationally representative studies (InterASIA [6], CHNS [8], and CNCDS [9]) (Figure 1). The prevalence of MetS in 2014–2015 by the CSPP study was generally higher than that in 2000–2001 by InterASIA study, but often lower than that in 2009 by CHNS study and 2010 CNCDS study (Figure 1). The high prevalence of MetS estimated by the CNCDS study in 2010 is particularly noteworthy, compared with other studies in 2000–2001, 2009, and 2014–2015.

Figure 1. Rates of prevalence of MetS (ATP III criteria) estimated by different nationally representative studies. Results of CHNS [15] and CSPP were obtained by analyses of original individual participant survey data, while results of InterASIA [7] and CNCDS [11] were based on the published reports. The comparison of CSPP results and other studies for the 35–44 age group should be cautious as adults aged 35–39 were not included in the CSPP survey.

Figure 2 shows the prevalence rate ratio of MetS in urban and rural areas, according to results of the CSPP and four previous studies [6–9]. Results of previous studies indicated a higher prevalence of MetS in urban adults, with a prevalence rate ratio of 1.47 in 2000–2001, 1.27 in 2007–2008, 1.36 in 2009, and 1.18 in 2010. However, the overall prevalence rate ratio was 0.91 (95% CI: 0.89 − 0.93), indicating a lower rate of MetS in urban adults, by 2014–2015 (Figure 2(a)). While rural and urban men had similar prevalence of MetS in 2014–2015 (Figure 2(b)), rural women had a much higher prevalence than urban women in 2014–2015 (Figure 2(c)).

Figure 2. Prevalence rate ratios of MetS in urban and rural areas in China, results from five nationally representative studies between 2000/2001 and 2014/2015. The five studies are InterASIA [7], CDMDS [8], CHNS [15], CNCDS [11], and CSPP survey. The prevalence rate ratio >1 indicates a higher prevalence in urban residents, and <1 indicates a lower prevalence in urban residents, compared with that in rural areas.

Data from three nationally representative studies (InterASIA [6], CHNS [8], and CSPP) are available to compare the prevalence of MetS components in 2000–2001, 2009, and 2014–2015 in China (Table 4). The rate of abdominal obesity was increased by 134.5% between 2000–2001 and 2009, and the increase was much greater in men than in women, particularly in rural areas. The prevalence of abdominal obesity was increased by 142.4% in rural areas, much greater than that (75.6%) in urban areas, between 2000–2001 and 2009. The overall prevalence of abdominal obesity in China was somewhat reduced by −13.6% between 2009 and 2014–2015 (Table 4). However, the reduction in the prevalence of abdominal obesity between 2009 and 2014–2015 was more substantial in urban adults, particularly in men. Compared with that in urban adults, the prevalence of abdominal obesity in rural adults was lower in 2000–2001 (7.2% versus 10.1%), similarly high in 2009 (17.5% versus 17.7% or 18.7% versus 18.8%), and higher in 2014–2015 (18.7% versus 14.5%) (Table 4).

Table 4. Prevalence of MetS components and changes over time by sex and location – comparison of results of different studies.

	2000/2001 InterASIA (35–74)	2009 CHNS		2014/2015	Per cent change (p value)
		(35–74)	(40–74)	CSPP (40–74)	2000/2001–2009	2009–2014/2015
Abdominal obesity
Overall	7.7	18.1	19.1	16.5	134.5 (<.001)	−13.6 (<.001)
Men	1.7	7.9	7.8	4.8	366.5 (<.001)	−38.1 (<.001)
Women	13.9	27.4	29.5	25.9	97.3 (<.001)	−12.2 (<.001)
Rural
Overall	7.2	17.5	18.7	18.7	142.4 (<.001)	.1 (.992)
Men	1.3	6.9	6.6	5.5	431.5 (<.001)	−16.0 (.098)
Women	13.4	27.4	29.9	29.5	104.5 (<.001)	−1.2 (.756)
Urban
Overall	10.1	17.7	18.8	14.5	75.6 (<.001)	−23.0 (<.001)
Men	3.1	8.5	9.0	4.3	175.2 (<.001)	−52.1 (<.001)
Women	17.5	25.9	27.8	23.0	48.2 (<.001)	−17.2 (.001)
Hypertriglyceridemia
Overall	24.8	33.9	35.1	38.2	36.9 (<.001)	8.9 (<.001)
Men	24.9	38.7	38.5	39.9	55.4 (<.001)	3.7 (.169)
Women	24.6	29.5	32.0	36.7	20.1 (<.001)	14.8 (<.001)
Rural
Overall	23.1	33.2	34.0	38.0	43.6 (<.001)	11.9 (<.001)
Men	22.1	37.5	37.1	39.8	69.5 (<.001)	7.2 (.032)
Women	24.2	29.1	31.1	36.8	20.3 (<.001)	18.5 (<.001)
Urban
Overall	32.3	35.8	37.6	38.4	10.7 (.006)	2.2 (.499)
Men	37.0	41.4	41.7	40.3	11.8 (.022)	−3.3 (.462)
Women	27.3	30.7	33.8	36.7	12.6 (.034)	8.5 (.069)
Low HDL cholesterol
Overall	33.9	25.5	25.1	31.3	−24.8 (<.001)	24.7 (<.001)
Men	21.9	17.1	16.3	20.4	−21.7 (<.001)	25.2 (<.001)
Women	46.5	33.2	33.2	40.1	−28.5 (<.001)	20.8 (<.001)
Rural
Overall	33.5	24.1	23.7	29.3	−28.0 (<.001)	23.7 (<.001)
Men	29.7	15.8	14.8	20.9	−46.7 (<.001)	41.5 (<.001)
Women	47.3	32.0	31.9	36.1	−32.4 (<.001)	13.2 (<.001)
Urban
Overall	37.3	28.4	28.0	33.1	−23.8 (<.001)	18.3 (<.001)
Men	29.7	20.0	19.5	20.5	−32.8 (<.001)	5.0 (.504)
Women	45.3	36.0	35.7	43.4	−20.6 (<.001)	21.6 (<.001)
High blood pressure or medication use
Overall	41.2	40.9	45.6	52.8	−0.7 (.731)	15.9 (<.001)
Men	44.2	45.7	49.5	55.8	3.3 (.256)	12.8 (<.001)
Women	38.0	36.5	42.0	50.6	−3.9 (.191)	20.5 (<.001)
Rural
Overall	41.5	40.3	45.0	58.4	−2.8 (.250)	29.7 (<.001)
Men	44.2	45.1	49.0	60.8	2.0 (.568)	24.0 (<.001)
Women	38.7	35.9	41.4	56.8	−7.4 (.035)	37.2 (<.001)
Urban
Overall	40.8	42.3	46.7	48.8	3.6 (.289)	4.5 (.108)
Men	45.1	47.1	50.5	52.5	4.5 (.326)	4.1 (.300)
Women	36.2	38.0	43.3	45.8	4.9 (.325)	5.7 (.168)
High fasting glucose or medication use
Overall	12.7	14.0	15.7	13.5	10.5 (.024)	−14.0 (<.001)
Men	12.8	16.8	18.2	14.0	30.9 (<.001)	−23.1 (<.001)
Women	12.6	11.5	13.4	13.1	−8.7 (.135)	−2.2 (.686)
Rural
Overall	12.1	13.7	15.2	13.5	13.2 (.016)	−11.0 (.007)
Men	11.9	16.3	17.3	13.8	37.0 (<.001)	−20.3 (<.001)
Women	12.3	11.3	13.2	13.3	−8.5 (.231)	0.8 (.929)
Urban
Overall	15.1	14.9	16.9	13.7	−1.3 (.829)	−18.8 (<.001)
Men	16.2	17.9	20.1	14.2	10.2 (.249)	−29.4 (<.001)
Women	13.6	12.3	13.9	13.0	−9.9 (.147)	−6.7 (.409)
Overweight (BMI ≥25.0 kg/m^2)
Overall	28.9	34.1	34.9	36.7	18.1 (<.001)	5.1 (.009)
Men	26.9	34.3	33.9	36.6	27.4 (<.001)	7.9 (.007)
Women	31.1	34.0	35.8	36.7	9.4 (.008)	2.5 (.366)
Rural
Overall	26.6	33.1	34.0	38.8	24.5 (<.001)	14.2 (<.001)
Men	23.9	32.7	32.2	36.5	36.7 (<.001)	13.4 (<.001)
Women	29.5	33.6	35.6	40.7	13.7 (.002)	14.2 (<.001)
Urban
Overall	39.1	35.1	36.0	35.1	−10.4 (.001)	−2.4 (.452)
Men	39.9	36.7	36.7	37.0	−8.0 (.090)	0.9 (.873)
Women	38.1	33.6	35.3	33.5	−11.9 (.007)	−5.2 (.247)

Age range for CHNS study [7] used were different to improve the comparability between studies: the InterASIA [12] aged 35–74; CHNS aged 35–74 or 40–74, and CSPP aged 40–74. Per cent change over time was calculated by: 100 × (rate2–rate1)/rate1. ATP III definition was used: abdominal obesity: WC ≥102 cm for men and WC ≥88 cm for women; high triglyceride: ≥1·7 mmol/L; low high-density-lipoprotein cholesterol: <1·03 mmol/L for men, and <1·29 mmol/L for women; high blood pressure: SBP ≥130 or DBP ≥85 mmHg or use of medication; high fasting plasma glucose: ≥6105 mmol/L or use of medication. The estimated prevalence rates were weighted by sex and age structure of population in 2000 in China. Results of the CSPP survey were adjusted for stroke risk.

Results in Table 4 indicated the changes in the prevalence of hypertriglyceridemia, low HDL cholesterol, hypertension, high fasting glucose or diabetes, and overweight in 2000/01, 2009 and 2014/15 in China. The overall prevalence of hypertriglyceridemia was increased by 36.9% between 2000/2001 and 2009, and by 8.9% between 2009 and 2014/2015, the prevalence of low HDL cholesterol, reduced by −24.8% between 2000/2001 and 2009 and increased by 24.7% between 2009 and 2014/2015. The recent increase in the prevalence of hypertension was greater in rural adults than that in urban adults (29.7% versus 4.5%). The prevalence of high fasting glucose or diabetes was higher in urban areas in 2000/2001 and 2009, and the difference between rural and urban areas was no longer obvious by 2014/2015. And the prevalence of overweight in rural adults was increased by 24.5% between 2000–2001 and 2009 and by 14.2% between 2009 and 2014–2015, while it was reduced by −10.4% and −2.4%, respectively, in urban adults

Discussions

The estimated prevalence of MetS among adults aged ≥40 years in 2014–2015 in China was 18.4% overall (13.2% in men and 23.7% in women) by the ATP III criteria. It was considerably higher by using the revised NCEP ATP III criteria (34.0% overall, 27.1% in men, and 40.9% in women), or by using the IDF definition (26.9% overall, 18.4% in men, and 35.7% in women). The overall rate of MetS (18.4% based on the ATP III definition) in China remains lower than that in the United States (about 35% among adults aged ≥20 years in 2011–2012) [16], in Europe (about 24%) [17], and in Brail (28.9%) [15]. However, it is already a serious public health problem due to the rapid population aging, behavioural, and socioeconomic changes. According to the ATP III definition, 107 million Chinese adults aged 40–74 had metabolic syndrome. The number of adults with abdominal obesity and overweight was 96 million and 182 million, respectively, among adults aged 40–74 in 2014/2015 in China. Furthermore, among Chinese adults aged 40–74 in 2014/2015, there were 307 million with high blood pressure, 78 million with high fasting plasma glucose, 222 million with hypertriglyceridemia, and 182 million with low HDL cholesterol. The estimated number of adults with MetS will be much higher if the revised NCEP ATP III or IDF criteria were used.

Compared with results of previous cross-sectional studies, the prevalence of MetS estimated by the CSPP study in 2014–2015 was higher than that in 2000–2001, but often lower than the estimated prevalence in 2009 and 2010. A possible explanation is that public health measures taking to control cardiovascular conditions in China have some effects on reduced deterioration or increased improvement in MetS among adults. For example, the prevalence of abdominal obesity and high fasting plasma glucose among urban adults was lower in 2014–2015 than that in 2009. Because of different sampling, analysis methods, and possible differences in population and areas conducted between previous studies and CSPP (see Supplementary Table S3), the results of the across-study comparison need to be interpreted with caution and further studies are required to confirm the temporal trends in the prevalence of MetS in Chinese adults.

The CSPP study confirmed findings from previous studies that the prevalence of MetS was higher in women, older adults, those with low level of education, and in economically developed regions. Contrasting with previous nationally representative studies in China, the 2014–2015 CSPP study was the first to show that the MetS was more prevalent in rural areas than that in urban areas, which may be partly because of the improvement in healthy awareness and behaviours in urban areas. The CSPP study revealed that rural adults had worse deterioration and/or less improvement in abdominal obesity, overweight, hypertension, and high fasting plasma glucose, than urban adults. In addition, the discrepancy in MetS between rural and urban adults was most striking for women. These may be related with some complicated reasons. For instance, the higher prevalence of Mets in women may be associated with the higher prevalence of abdominal obesity in women. As to the Mets components, the higher prevalence of hypertension in rural areas may be related with a higher prevalence of overweight in rural adults, which was an important risk factor of hypertension.

Evidence from global nutrition transition suggested that urban residents generally have higher prevalence of obesity and overweight in low-income or less developed nations, but rural residents will catch up quickly along with economic development [18]. For example, although women with the highest wealth and education had the highest prevalence of overweight and obesity in 39 low- and middle-income countries, the increase of overweight was greater among women with the lowest wealth and education [19]. The present study indicated that Chinese adults in rural areas are not only catching up quickly but also possibly have overtaken those in urban areas regarding obesity and related MetS problems.

The speed of nutrition transition and socioeconomic development in China is uniquely rapid, which will lead to swift geographic and temporal changes in MetS. Cardiovascular diseases are expected to increase considerably in future due to population ageing and high prevalence of MetS. However, appropriate public health measures can be implemented to reduce the burden of cardiovascular diseases. It may be interesting to note that the CSPP study in 2014–2015 found a lower prevalence of abdominal obesity in urban adults, compared with the CHNS study in 2009. In addition, the estimated prevalence of high fasting plasma glucose for men in both rural and urban areas was lower in the CSPP study in 2014–2015 compared with that in the CHNS study in 2009. However, we should not be complacent as MetS components remain highly prevalent among Chinese adults. For example, the prevalence of high blood pressure is high in both rural and urban areas.

While efforts to prevent and control cardiovascular diseases need to be strengthened for the whole Chinese population, health decision makers need to be aware of rapid changes in the distribution of CVD risk factors by sub-population categories. Findings of the present study clearly indicate that CVD risk factors among rural residents and women should be prioritized in making public health policy decisions, such as health education and other dissemination measures, which could improve their healthy awareness and behaviours.

Strengths and limitations

The CSPP study was a nationally representative study, the most recently completed, with the largest sample size to estimate the prevalence of MetS in China. In addition to the current prevalence of MetS in Chinese adults, we were able to show the possible temporal changes in MetS prevalence by comparing results of the present study and previous nationally representative studies.

Some limitations of the study are worth mentioning. First, the present study included only participants who had additional laboratory test, and the prevalence of MetS will be over-estimated because high risk of stroke was more prevalent among them. However, we were able to cope with this problem by adjusting the estimated prevalence of MetS according to the distribution of stroke risk among all CSPP participants. Second, considered improvement, warning against complacency about inaction about risk factors may be too general due to limited data, which need to be studied further.

To show the possible temporal changes in MetS, we made cross-study comparisons. There may be many alternative explanations for the observed differences between the CSPP and previous studies. Although we have attempted to make the results of different studies as comparable as possible, it was difficult to be certain about the true causes of the difference between different cross-sectional studies. Availability of individual participant data from all relevant studies should help improve comparison of results across studies. However, so far we were able to access individual participant data for only one previous study (CHNS⁸), in addition to the CSPP study.

Conclusions

The prevalence of metabolic syndrome has been rapidly increasing and remains high, particularly in women and rural areas, among middle-aged and elderly adults in China. While efforts to prevent and control cardiovascular diseases need to be strengthened for the whole population, rapid increasing risk factors among rural residents and women should be prioritised in making public health policy decisions. It should also be a concern for health decision makers in other developing counties which are experiencing socio-economic transformation.

Acknowledgements

The authors thank the National Project Office of Stroke Prevention and Control for data support. In addition, this research uses data from China Health and Nutrition Survey (CHNS). The authors thank the National Institute for Nutrition and Health, China Center for Disease Control and Prevention, Carolina Population Center (P2C HD050924, T32 HD007168), the University of North Carolina at Chapel Hill, the NIH (R01-HD30880, DK056350, R24 HD050924, and R01-HD38700) and the NIH Fogarty International Center (D43 TW009077, D43 TW007709) for financial support for the CHNS data collection and analysis files from 1989 to 2015 and future surveys, and the China–Japan Friendship Hospital, Ministry of Health for support for CHNS 2009, Chinese National Human Genome Center at Shanghai since 2009, and Beijing Municipal Center for Disease Prevention and Control since 2011.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This study was funded by the Ministry of Finance of the People’s Republic of China (Issued by Finance and Social Security [2011] Document No. 61, Ministry of Finance), the Fundamental Research Funds for the Central Universities, Huazhong University of Science and Technology, Wuhan, China [2016YXMS215], and the China Postdoctoral Science Foundation funded project [2017M622466]. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.