The Lancet Regional Health - Western Pacific 15 (2021) 100227

Contents lists available at ScienceDirect

The Lancet Regional Health - Western Pacific

journal homepage: www.elsevier.com/locate/lanwpc

Research paper

Changes in the prevalence of obesity and hypertension and

demographic risk factor profiles in China over 10 years: two national
cross-sectional surveys
Yongze Li, Ph.D., Di Teng, Ph.D., Xiaoguang Shi, Ph.D., Xiaochun Teng, Ph.D.,
Weiping Teng, M.D.∗, Zhongyan Shan, Ph.D.∗, Yaxin Lai, Ph.D.∗ , for the China National
Diabetes and Metabolic Disorders Study Group and the Thyroid Disorders, Iodine Status
and Diabetes Epidemiological Survey Group
Department of Endocrinology and Metabolism and the Institute of Endocrinology, First Hospital of China Medical University, Shenyang, 110001, Liaoning,
China

a r t i c l e i n f o a b s t r a c t

Article history: Background: Previous studies have shown increases in the prevalence of obesity and hypertension, but
Received 17 May 2021 nationally representative data on recent changes in prevalence adjusted for population structure changes
Revised 25 June 2021
are lacking. Two nationwide surveys were conducted in 2007 and 2017 to assess the prevalence changes
Accepted 5 July 2021
of these conditions in China.
Methods: A multistage stratified random sampling method was used to obtain a nationally representative
Keywords:
sample of adults aged 20 years and older in mainland China in 2007 and 2017. Temporal changes in the
Obesity
Hypertension prevalence of hypertension and obesity were investigated. Changes in blood pressure, body mass index
Prevalence (BMI) and waist circumference were also assessed. Logistic regression models were constructed to assess
China the changes in prevalence over time.
Findings: The weighted prevalence of hypertension (25.7% vs. 31.5%, P=0.04), high-normal blood pres-
sure (11.7% vs. 14.3%, P<0.0 0 01), general obesity (31.9% vs. 37.2%, P=0.008), and central obesity (25.9%
vs. 35.4%, P=0.0 0 02) was significantly higher in 2017 (n=72824) than in 2007 (n=45956) in the overall
population. No significant changes in the prevalence of overweight and grade 1 or grade 2 hypertension
were observed in the overall population, but a significantly higher prevalence was observed among par-
ticipants aged 20-29 years for grade 1 hypertension (P=0.002) and among participants aged 70 years and
older for grade 2 hypertension (P=0.046) in 2017.
Interpretation: Compared with 2007, the prevalence of hypertension and obesity was significantly higher
among adults in mainland China after adjusting for demographic confounding factors in 2017. More tar-
geted interventions and prevention strategies are needed to offset the increasing risk of cardiovascular
disease due to increases in the prevalence of hypertension and obesity.
Funding: The Clinical Research Fund of the Chinese Medical Association (Grant No. 15010010589), the
National Natural Science Foundation of China (Grant No. 820 0 0753), and the Chinese Medical Association
Foundation and Chinese Diabetes Society (Grant No. 07020470055)
© 2021 The Author(s). Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

∗
Address for correspondence: Yaxin Lai, Ph.D. Department of Endocrinology and Metabolism, The Institute of Endocrinology, The First Hospital of China Medical University,
Shenyang, P.R. China. Weiping Teng, M.D. Department of Endocrinology and Metabolism, The Institute of Endocrinology, The First Hospital of China Medical University,
Shenyang, P.R. China. Zhongyan Shan, Ph.D. Department of Endocrinology and Metabolism, The Institute of Endocrinology, The First Hospital of China Medical University,
Shenyang, P.R. China.
E-mail addresses: [email protected] (W. Teng), [email protected] (Z. Shan), [email protected] (Y. Lai).

https://doi.org/10.1016/j.lanwpc.2021.100227
2666-6065/© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

Research in Context tension prevalence of 44.7%.[10] In 2019, NCDs accounted for ap-
proximately 95% of all deaths and 90% of all disability-adjusted
Evidence Before this Study life years lost in China, which increased from approximately 80%
A growing body of literature has documented increas- and 60% in 1990, respectively.[11] The increase in obesity and hy-
ing changes in the prevalence of obesity and hypertension pertension is expected to continue to affect the future burden of
among adults in China. However, the majority of these stud- NCDs.[12] A growing body of literature has documented increas-
ies were limited to certain age groups or regions or in- ing trends in the prevalence of obesity and hypertension among
volved nonrepresentative sampling. Dynamic changes in pop- adults in China.[12-14] However, the majority of these studies were
ulation structure, economic development, education levels, limited to certain age groups or regions or involved nonrepresen-
and lifestyles should be taken into consideration when as- tative sampling. In addition, dynamic changes in population struc-
sessing the changes in prevalence over time. In addition, the
ture, economic development, education levels, and lifestyles should
change in the prevalence of high-normal blood pressure in
mainland China is unknown. be taken into consideration when assessing the changes in preva-
Added Value of this Study lence over time.
Two large-sample nationally representative surveys indi- In June 2020, the International Society of Hypertension (ISH)
cated that the prevalence of general obesity, central obesity, published new guidelines for the management of patients with
hypertension, and high-normal blood pressure was higher in arterial hypertension.[15] Compared with the previous guidelines,
2017 among Chinese people aged 20 years or older than in the 2020 ISH guidelines have a simplified definition of the blood
2007. pressure categories, making classification and risk stratification of
The higher prevalence of hypertension and obesity shifted people with hypertension more feasible for clinicians.[16] Aware-
from urban to rural populations over the course of a decade.
ness of the risks associated with high-normal blood pressure needs
Body mass index, waist circumference, and systolic blood
to be promoted so individuals with this condition can delay or pre-
pressure have increased slightly, with relatively larger in-
creases in systolic blood pressure in men, rural residents, and vent incident hypertension through the early adoption of healthy
young adults. lifestyle interventions that lower blood pressure levels and reduce
Implications of All the Available Evidence the risk of cardiovascular disease.[15] However, the changes in the
This study revealed increasing changes in the prevalence prevalence of high-normal blood pressure in mainland China are
of obesity and hypertension in Chinese adults, as well as the unknown.
prevalence of high-normal blood pressure, indicating a sub- To obtain a more accurate and comprehensive understanding of
stantial future burden of cardiovascular disease in China. the changes in obesity and hypertension in mainland China over
The changes in the populations most commonly affected the decade between 2007 and 2017, this analysis presents nation-
by high-normal blood pressure suggest that increased atten-
ally representative data from two population-based cross-sectional
tion should be given to men, young adults, and rural resi-
dents. surveys. In addition, we determined the temporal changes in the
More targeted interventions and prevention strategies are prevalence of different categories of hypertension and obesity both
needed to offset the increasing risk of cardiovascular disease in the overall population and within subgroups defined by sociode-
due to increases in the prevalence of hypertension and obe- mographic and behavioural characteristics after adjusting for de-
sity. mographic confounding factors.

2. Methods

1. Introduction 2.1. Study Population and Survey Design

Obesity and hypertension, which are two major risk factors for The first national cross-sectional study (China National Diabetes
noncommunicable diseases (NCDs), contribute to global health and and Metabolic Disorders Study) was carried out in 20 07-20 08 to
economic burdens.[1,2] The prevalence of obesity has increased evaluate the status of major metabolic risk factors, including blood
worldwide in the past 50 years, and it is often referred to as an glucose, blood pressure, and blood lipids, within the adult pop-
epidemic. Obesity represents a major health challenge because it ulation of mainland China. Details of the study design are pre-
substantially increases the risk of diseases such as type 2 diabetes sented elsewhere.[17] In brief, a multistage stratified random sam-
mellitus, fatty liver disease, and several cancers, thereby contribut- pling method was used to select a nationally representative sam-
ing to a decline in both quality of life and life expectancy.[1] More- ple of the general population aged 20 years or older in China (Sup-
over, overweight and obesity are associated with hypertension, and plementary Figure 1). An additional cross-sectional survey (Thyroid
hypertension is considered to be the leading cause of cardiovascu- Disorders, Iodine Status and Diabetes Epidemiological Survey) was
lar disease and premature death worldwide.[3] Over the past few carried out in 2015-2017. We previously described the study design
decades, obesity and hypertension have increased rapidly in Asian in detail, and a detailed flowchart of the study design can also be
countries due to the westernization of lifestyles.[4-7] In India, found in Supplementary Figure 1.[18] Briefly, the same multistage
more than 135 million individuals are affected by obesity.[4] As stratified random sampling method was applied in urban and rural
a developing country, Vietnam is also facing several environmen- locations to obtain nationally representative samples (Supplemen-
tal and health problems, including hypertension and obesity.[5,6] tary Figure 1). The inclusion criteria for this study were as follows:
Even in Japan, a country with a highly developed economy, hyper- age 20 years or older, having lived in the selected community for
tension is highly prevalent, affecting up to 60% of men and 45% of at least five years, and not pregnant. Ultimately, 45956 participants
women.[7] in 2007 and 72824 participants in 2017 were eligible for inclusion
With the acceleration of China’s economic development and ur- in the analysis after the exclusion of those with missing informa-
banization, the ageing of the population and the ongoing epidemic tion on sex, age, body mass index (BMI), waist circumference, sys-
of obesity, hypertension has become a major public health prob- tolic blood pressure (SBP), and diastolic blood pressure (DBP) (Sup-
lem affecting Chinese residents.[8] Data from the China National plementary Figure 1). The numbers of participants with missing in-
Nutrition Surveys in 2015 indicate that the prevalence of obesity formation were 283 (0.6%) in 2007 and 524 (0.6%) in 2017. These
among adults in China was 16.4%.[9] The previous national sur- missing data were not associated with either the specific value that
vey of hypertension in China, conducted in 2017, found a hyper- was supposed to be obtained or the set of observed responses. The

2
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

analysis thus remains unbiased. Power may be slightly lost in the the geographic regions of the adults living in China. Weighting co-
design, but the estimated parameters are not biased due to miss- efficients were derived from the Chinese population census data,
ing data.[19] The research protocols were approved by the medi- and the sampling scheme of the two surveys was used to obtain
cal ethics committees of China Medical University and China–Japan a national estimate. Briefly, the weighting coefficient was the in-
Friendship Hospital. All the participants provided written informed verse of the adjusted probability of obtaining the data for the re-
consent after receiving a thorough explanation of the research pro- spondent; each individual case in the analysis was assigned a spe-
cedures. cific coefficient (individual weight), by which the value was mul-
tiplied to represent the actual population with the same charac-
2.2. Measurements teristics of sex, age, province, and location. Standard errors were
estimated by Taylor series linearization. To counteract the effect of
For each participant, a trained interviewer used a detailed ques- the changes in population structure from 2007 to 2017, age- and
tionnaire to collect information about demographic variables, be- sex-specific adjustments were performed using direct standardiza-
havioural factors, and personal medical history. Current smoking tion, with the standard being all adults across the entire period;
was defined as having smoked at least 100 cigarettes in one’s life the age- and sex-specific standardized coefficients were based on
and currently smoking cigarettes. An identical protocol was used to the 2010 Chinese population census data. Categorical data are pre-
measure body weight, height, and waist circumference in 2007 and sented as percentages and 95% confidence intervals (CIs) and were
2017. Body weight and height were measured according to the 3rd analysed by a χ 2 test or Fisher’s exact test, as appropriate. Contin-
edition of Cardiovascular Survey Methods from the World Health uous data are described as the means and 95% CIs and were anal-
Organization (WHO).[20] BMI was calculated by dividing the body ysed with t tests. Logistic regression models were used to examine
weight in kg by the square of the height in metres. Waist cir- the changes in the prevalence of obesity and hypertension between
cumference was measured in upright participants midway between 2007 and 2017. Linear regression models were used to estimate the
the lower edge of the costal arch and the upper edge of the iliac changes in mean SBP, DBP, BMI, and waist circumference between
crest.[20] 2007 and 2017. To further test the stability of the results, two sets
In the first study, blood pressure was measured using a stan- of sensitivity analyses for odds ratios (ORs) were undertaken. First,
dardized calibrated mercury sphygmomanometer (regular adult, three models with progressively increased adjustment of risk fac-
large, or thigh) in the seated position after five minutes of rest.[20] tors among all participants were applied. Second, considering that
Two consecutive readings were taken on the nondominant arm. the prevalence of obesity and hypertension differs according to de-
In the second study, blood pressure was measured by a validated mographic background, we stratified participants according to sub-
electronic blood pressure monitor (Omron HEM-7430, Omron Cor- groups for analysis. Statistical significance was defined by a 2-sided
poration) on the nondominant arm with the participant in a seated P value <0.05. All the statistical analyses were conducted using
position after five minutes of rest.[20] Two consecutive measure- SAS, version 9.3 (SAS Institute Inc, Cary, NC) and SUDAAN, version
ments were taken with a 10-minute interval between measure- 10.0 (Research Triangle Institute).
ments. The mean of the two consecutive measures was used in
both studies for analysis. 2.5. Role of the Funding Source

2.3. Definitions of Obesity and Hypertension The funders had no role in the execution of this study or the
interpretation of the results.
According to the International Diabetes Federation diagnostic
criteria, we defined central obesity as a waist circumference of 90
cm or greater for men and 80 cm or greater for women.[21] Ac- 3. Results
cording to the Asian-specific cut-off points, overweight was de-
fined as a BMI from 23 kg/m2 to less than 25 kg/m2 , and gen- 3.1. Characteristics of the Study Participants
eral obesity was defined as a BMI of 25 kg/m2 or greater for both
men and women.[22] According to the 2020 ISH guidelines, hy- Table 1 presents the characteristics of the respondents in each
pertension was defined as an SBP of 140 mmHg or greater, a DBP survey. Significant differences were observed in the mean age, sex,
of 90 mmHg or greater, or the self-reported use of antihyperten- income level, education level, BMI, waist circumference, and SBP
sive medication within the previous two weeks.[15] Normal BP levels between 2007 and 2017. Compared with 2007, the mean age
was defined as an SBP less than 130 mmHg, a DBP less than 85 was younger (44.8 years vs. 43.8 years, P=0.02) in 2017; the pro-
mmHg and no use of antihypertensive medicines.[15] High-normal portion of men was higher (49.4% vs. 50.2%, P=0.0 0 01) in 2017.
BP was defined as an SBP from 130-139 mmHg, a DBP from 85- Higher income levels, education levels, BMI values, waist circum-
89 mmHg, and no use of antihypertensive medicines.[15] Grade 1 ferences, and SBP levels were seen in 2017 (P<0.05 for all).
hypertension was defined as an SBP from 140-159 mmHg and/or
a DBP from 90-99 mmHg.[15] Grade 2 hypertension was defined 3.2. Changes in the Prevalence of Obesity
as an SBP of 160 mmHg or greater and/or a DBP of 100 mmHg or
greater.[15] Figure 1 presents the changes in the age- and sex-standardized
prevalence of overweight, general obesity, and central obesity in
2.4. Statistical Analysis mainland China. Compared with 2007, the prevalence of general
obesity (31.9% vs. 37.2%, P=0.008) and central obesity (25.9% vs.
An identical statistical plan was used to account for the com- 35.4%, P=0.0 0 02) was significantly higher in 2017 among the over-
plex sampling design of the two studies; we used SUDAAN all population. Figure 2 shows the changes in the mean BMI
software (Research Triangle Institute) to obtain prevalence esti- value and waist circumference between the first and second stud-
mates and standard errors according to the Taylor linearization ies, with adjustment for age, sex, urbanization, ethnicity, income
method.[23] The Taylor series (linearization) method is the most level, education level, and smoking status. Substantial increases
commonly used method to estimate the covariance matrix of the in waist circumference were found consistently across all sex and
regression coefficients for complex survey data.[23] Estimates were age groups (except age≥70 years) and among rural residents (ad-
weighted to reflect the age, sex, and urban-rural distributions of justed change, 3.4 cm; 95% CI, 0.6 to 6.2 cm; P=0.02). The mean

3
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

Table 1
Sample characteristics (weighted) by survey wave. Values are percentages (95% CI) unless stated otherwise.

Characteristics 2007 2017 P for difference

No. of participants 45956 72824

Mean age at survey (standard error), years 44.8 (0.14) 43.8 (0.42) 0.02
Sex
Men 49.4 (49.3 to 49.5) 50.2 (49.9 to 50.5)
Women 50.6 (50.5 to 50.7) 49.8 (49.5 to 50.2) 0.0001
Urbanization
Urban 45.8 (18.6 to 75.8) 52.1 (31.7 to 71.8)
Rural 54.2 (24.3 to 81.4) 47.9 (28.2 to 68.3) 0.75
Ethnicity
Han 86.4 (69.2 to 94.7) 95.6 (93.9 to 96.8)
Non-Han 13.6 (5.3 to 30.8) 4.4 (32 to 6.1) 0.14
Income per year
≤30000 yuan 82.8 (77.7 to 87.0) 44.7 (38.0 to 51.5)
>30000 yuan 17.2 (13.0 to 22.4) 55.3 (48.5 to 62.0) <0.0001
Education
Less than college 79.6 (72.5 to 85.2) 66.2 (59.1 to 72.7)
College and above 20.4 (14.8 to 27.5) 33.8 (27.3 to 40.9) 0.006
Current cigarette smoker 26.1 (23.5 to 28.8) 26.7 (25.4 to 27.9) 0.69
Physical examination, mean (standard error)
Mean body mass index, kg/m2 23.7 (0.17) 24.1 (0.06) 0.04
Mean waist circumference, cm 80.7 (0.58) 83.5 (0.37) 0.0002
Mean systolic blood pressure, mm Hg 121.7 (0.79) 126.6 (0.67) <0.0001
Mean diastolic blood pressure, mm Hg 77.5 (0.61) 78.6 (0.57) 0.23

BMI value (adjusted change, 0.2 kg/m2 ; 95% CI, 0.01 to 0.4 kg/m2 ; levels are provided in the Supplement. The changes in the preva-
P=0.04) increased significantly in the overall population. lence of overweight, general obesity, central obesity, hypertension,
normal blood pressure, high-normal blood pressure, and grade 2
3.3. Changes in the Prevalence of Hypertension hypertension remained stable in three logistic regression models
with adjustment for different numbers of demographic risk factors
Figure 3 presents the changes in the age- and sex-standardized in the overall population (Supplementary Tables 1-8). In addition,
prevalence of hypertension, normal blood pressure, and high- the increases in mean BMI, waist circumference, and blood pres-
normal blood pressure in mainland China. Compared with 2007, sure levels remained stable in the two linear regression models af-
the prevalence of hypertension (25.7% vs. 31.5%, P=0.04) and high- ter adjustment for different numbers of demographic risk factors
normal blood pressure (11.7% vs. 14.3%, P<0.0 0 01) was higher, in the overall population (Supplementary Tables 9-10).
while the prevalence of normal blood pressure (62.6% vs. 54.2%,
P=0.001) was lower among the overall population in 2017. A signif- 4. Discussion
icantly higher prevalence of hypertension was seen in those aged
20-29 years (OR, 1.77; 95% CI: 1.19-2.64; P=0.006), men (OR, 1.29; In the current large-sample, population-based cross-sectional
95% CI: 1.05-1.60; P=0.02), and rural residents (OR, 1.37; 95% CI: study, we found that the age- and sex-standardized weighted
1.01-1.85; P=0.04). No significantly increased prevalence of grade prevalence of hypertension and obesity was higher among adults
1 or grade 2 hypertension was observed in the overall population in 2017 in mainland China after adjusting for demographic con-
(Supplementary Tables 1-2). For grade 1 hypertension, a signifi- founding factors compared to that in 2007. In addition, significant
cantly higher prevalence was seen only among participants aged increases in the mean BMI, waist circumference, and SBP occurred
20 to 29 years (6.0% vs. 8.7%, P=0.002) in 2017. For grade 2 hy- in adults over the decade after 2007. Moreover, a higher prevalence
pertension, a significantly higher prevalence was observed among of high-normal blood pressure was observed in 2017 among men,
participants aged 70 years and older (16.2% vs. 23.8%, P=0.046), urban residents, and young individuals. We used a nationally rep-
among those who were overweight (6.1% vs. 8.5%, P=0.02), and resentative sample for large-scale recruitment, which could be gen-
among those without central obesity (5.1% vs. 7.3%, P=0.04) in eralized to adults aged 20 years and older in China.
2017. Our study expands the existing literature on changes in obe-
Figure 4 shows the change in the mean SBP and DBP strati- sity and hypertension in several ways. First, to our knowledge,
fied by sex, age group, and location between the two surveys, with our study is one of the largest to describe the changes in preva-
adjustment for age, sex, urbanization, ethnicity, income level, ed- lence and blood pressure levels among adults in mainland China,
ucation level, smoking status, BMI, and waist circumference. Sig- which allowed us to explore associations across a variety of di-
nificant increases were found consistently across all sexes, age verse subgroups. We found increasing changes in the prevalence
groups, and regions for SBP (adjusted change, 4.5 mmHg; 95% CI, of obesity and hypertension, which is consistent with previously
3.4 to 5.7 mmHg, in the overall population), with greater increases reported changes in the Chinese population.[13,14] This is similar
among men (adjusted change, 5.9 mmHg; 95% CI, 4.6 to 7.2 mmHg; in India and Vietnam, whose prevalence of obesity and hyperten-
P<0.0 0 01), participants aged 20-29 years (adjusted change, 4.8 sion has shown upward trends.[24-26] However, the time-related
mmHg; 95% CI, 4.1 to 5.5 mmHg; P<0.0 0 01), and rural residents trends in developing countries are different from those in Asian
(adjusted change, 5.6 mmHg; 95% CI, 3.1 to 8.2 mmHg; P=0.0 0 01). developed countries. The prevalence of hypertension in Japanese
men dropped from 54.2% in 1980 to 50.1% in 2010, while that
3.4. Sensitivity Analysis of women dropped from 47.4% to 37.8%.[27] In South Korea, the
prevalence of hypertension in men dropped from 33.3% in 1998 to
The results of the sensitivity analysis of the changes in the 30.3% in 2014 and in women dropped from 28.7% to 22.7%.[28] Fur-
prevalence and mean BMI, waist circumference, and blood pressure thermore, the United States, which also has a substantial burden of

4
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

Figure 1. Changes in age- and sex-standardized prevalence of overweight, general obesity, and central obesity between 2007 and 2017 in adults in China.
Values are weighted percentages (95% confidence intervals) unless stated otherwise. Logistic models were adjusted for age, sex, urbanization, ethnicity, income level, educa-
tion level, and smoking status from 2007 to 2017.

NCDs, has seen plateaus or even decreasing trends in the preva- young adults had a greater increase in SBP levels than somewhat
lence of obesity and hypertension in recent years.[29,30] In the older individuals. Trends in blood pressure levels in young adults
past decade, rapid economic growth, which brought with it cer- are a marker of the future population burden of cardiovascular dis-
tain unhealthy lifestyles, especially a higher level of dietary sodium ease and may be particularly relevant in areas with high disease
intake, is another new and crucial factor related to the increased rates.[32] This phenomenon might be partially explained by the
prevalence of hypertension in China.[31] In addition, we found that fact that in the Chinese population, the later the year in which

5
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

Figure 2. Adjusted increases in mean body mass index and waist circumference over the course of 10 years in adults in mainland China.
Values are means (95% confidence intervals). Adjusted for age, sex, urbanization, ethnicity, income level, education level, and smoking status.

an individual was born is, the higher their risk of developing hy- and less motivated to make lifestyle changes.[33-35] Furthermore,
pertension is; while the development of the economy has grad- the consumption of processed and packaged foods and beverages is
ually improved the standard of living of the Chinese people, the on the rise among the young generation of China, and these types
burdens of work and stress have increased.[13] Younger adults are of foods usually contain higher levels of saturated fat, salt, and
somewhat more difficult to reach through traditional clinic-based sugar.[36] Evidence from longitudinal studies has shown adverse
preventive programs because they may be less aware of the long- effects of reductions in physical activity on weight change due to
term benefits of the early control of cardiovascular risk factors and the use of occupational and household technology in China.[37-39]
are therefore less likely to be in contact with the health system Young adults have increasingly entered the middle class in China,

6
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

Figure 3. Changes in age- and sex-standardized prevalence of hypertension, normal blood pressure, and high-normal blood pressure between 2007 and 2017 in adults in
China.
Values are weighted percentages (95% confidence intervals) unless stated otherwise. Logistic models were adjusted for age, sex, urbanization, ethnicity, income level, educa-
tion level, smoking status, body mass index, and waist circumference from 2007 to 2017.

7
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

Figure 4. Adjusted increases in mean blood pressure over the course of 10 years in adults in mainland China.
Values are means (95% confidence intervals). Adjusted for age, sex, urbanization, ethnicity, income level, education level, smoking status, body mass index, and waist circum-
ference.

which might further expand the demand for convenient products Second, our study is the first, to our knowledge, to describe the
that could reinforce unhealthy lifestyles.[40,41] In addition, given changes in the prevalence of high-normal blood pressure based
that the clinical importance of the treatment of hypertension in on data from a national survey of the Chinese population. High-
younger adults has been questioned in the past and that most pre- normal blood pressure is associated with increased risks of hy-
vious studies of hypertension have focused on older individuals, pertension and cardiovascular disease and can be reduced through
there are limited recommendations for the management of hyper- lifestyle modifications and the use of antihypertensive medica-
tension in younger adults.[42-44] Thus, our findings may be a re- tion.[15] We found that the prevalence of high-normal blood pres-
flection of the lack of clinical data on this population and highlight sure was higher in 2017 in the overall population. A previous
the need for clinical trials in this population. study indicated that adults with prehypertension had risk factors

8
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

for incident hypertension and had not made lifestyle modifica- although the survey staff were highly trained, their efficacy or skill
tions.[45] Importantly, low-cost interventions for preventing hyper- level may have resulted in some misclassification errors. The lim-
tension have been shown to be effective in all age groups, ethnic- itations of the current analysis also warrant discussion. The first
ities, and sexes.[46] This indicates that there is a substantial op- and second surveys used different types of blood pressure mon-
portunity to reduce the incidences of hypertension and cardiovas- itoring, which would produce systematic error, although previous
cular disease through lifestyle changes. However, novel approaches studies have proven good agreement for blood pressure measure-
for maintaining lifestyle modifications may be needed because in- ments between mercury sphygmomanometers and electronic de-
creasing changes in central obesity and general obesity were also vices.[54,55]
found in the current study. In addition, we found that the some- In conclusion, the prevalence of hypertension, high-normal
what higher prevalence of hypertension and obesity shifted from blood pressure, and obesity was significantly higher among adults
urban to rural populations over the decade from 2007 to 2017. in mainland China after adjustment for demographic confound-
This result seems to be consistent with previous studies that found ing factors in 2017. The BMI, waist circumference, and SBP levels
that the prevalence of central obesity of residents in rural areas in- increased slightly, with a greater increase in SBP in men, young
creased more rapidly than that of residents in urban areas.[47,48] adults, and rural residents. More targeted interventions and pre-
These findings may be related to the changes in socioeconomic vention strategies are needed to offset the increasing risk of car-
structure led by urbanization in China.[48] A previous observation diovascular disease due to increases in the prevalence of hyperten-
confirmed that chronic health conditions are related to moderniza- sion and obesity.
tion and affluence and that the emergence of these problems is no
longer limited to urban populations.[49] The per capita food con-
Author Contributions
sumption of Chinese rural households increased by 2.6 times from
1997 to 2012. The Engel coefficient of urban and rural households
Yaxin Lai, Zhongyan Shan, Weiping Teng and Yongze Li had full
dropped by 10.4 and 15.8 percentage points from 1997 to 2012,
access to all the data in the study and take responsibility for the
respectively.[50] With the advancement of urbanization, the food
integrity of the data and the accuracy of the data analysis.
consumption ability of rural adults has developed rapidly. There-
Concept and design: Yaxin Lai, Zhongyan Shan, Weiping Teng.
fore, high-fat diets and reduced physical activity may exacerbate
Acquisition, analysis, and interpretation of the data: Yaxin Lai,
health deterioration, such as the higher prevalence of obesity in
Zhongyan Shan, Weiping Teng and Yongze Li.
more urbanized regions.[51] In addition, high sodium intake was
Drafting of the manuscript: Yongze Li.
associated with a higher risk of central obesity than general obe-
Statistical analysis: Yongze Li.
sity.[52] Significant differences in the prevalence of central obe-
Obtaining funding: Zhongyan Shan and Weiping Teng.
sity among rural residents may be due to increasingly high dietary
Administrative, technical and material support: All authors.
sodium in rural areas in China.[53] China has a large rural popu-
Study supervision: Zhongyan Shan, Weiping Teng, Yongze Li,
lation, and sanitation is lacking in rural areas; thus, an increased
Yaxin Lai, Di Teng, Xiaochun Teng, Xiaoguang Shi.
prevalence of obesity and hypertension in rural areas will lead to
increased incidences of NCDs. Given the greater increases in SBP
and waist circumference in rural populations, a large number of Declaration of Competing Interest
people are at risk of developing hypertension in the absence of the
implementation of effective preventive measures. The authors declare no conflict of interests.
Several recommendations for national policies and efforts may
potentially combat the further development of obesity and hyper-
Data Sharing Statement
tension in China.[12] First, to establish fiscal policies to prevent
and control obesity and levy taxes on unhealthy foods and bev-
The data used during the current study are available from the
erages, subsidies should be provided to promote healthy diets and
corresponding author upon reasonable request.
healthy lifestyles. Second, activity centres, indoor and outdoor fit-
ness venues, and self-service health management inspection points
equipped with height, weight, and blood pressure measurements Funding
should be established. Third, obesity treatment should be included
in the coverage of health insurance, and medical expense reim- This work is supported by the Clinical Research Fund of the
bursement should be correlated with the results of weight man- Chinese Medical Association (Grant No. 15010010589), the National
agement of obese patients. Fourth, obesity prevention policies and Natural Science Foundation of China (Grant No. 820 0 0753), and the
strategies should take the inequalities found in this study into full Chinese Medical Association Foundation and Chinese Diabetes So-
consideration and be tailored to high-risk groups to prevent a fur- ciety (Grant No. 07020470055). The authors have no relationships
ther gap in obesity prevalence among subgroups and ensure health relevant to the contents of this paper to disclose.
equity.
The 2007 and 2017 studies have potential limitations, some of
Acknowledgements
which have been mentioned in previous studies.[17,18] First, they
did not assess dietary intake, alcohol consumption, and physical
We thank the participants of this study. For continuous sup-
activity. Therefore, we were not able to determine the associations
port, assistance, and cooperation, we thank the investigators for
between these factors and the changes in the prevalence of obe-
the China National Diabetes and Metabolic Disorders Study Group
sity and hypertension. Second, the 2020 ISH guidelines recommend
and the Thyroid Disorders, Iodine Status and Diabetes Epidemio-
longitudinal and three measurements of blood pressure levels for
logical Survey Group.
the diagnosis of hypertension.[15] Because the two studies were
large-scale population-based cross-sectional surveys, blood pres-
sure was only measured in the participants two times in a sin- Supplementary materials
gle day. Considering the effect of regression to the mean, this may
have overestimated the prevalence of hypertension. However, the Supplementary material associated with this article can be
effect of regression on the mean should not be substantial. Third, found, in the online version, at doi:10.1016/j.lanwpc.2021.100227.

9
Y. Li, D. Teng, X. Shi et al. The Lancet Regional Health - Western Pacific 15 (2021) 100227

References [31] Huang C, Yu H, Koplan JP. Can China diminish its burden of noncommunicable
diseases and injuries by promoting health in its policies, practices, and incen-
[1] Blüher M. Obesity: global epidemiology and pathogenesis. Nat Rev Endocrinol tives? Lancet 2014;384:783–92.
2019;15(5):288–98. [32] McCarron P, Smith GD, Okasha M, McEwen J. Blood pressure in young adult-
[2] Mills KT, Stefanescu A, He J. The global epidemiology of hypertension. Nat Rev hood and mortality from cardiovascular disease. Lancet 20 0 0;355:1430–1.
Nephrol 2020;16(4):223–37. [33] Allen NB, Siddique J, Wilkins JT, et al. Blood pressure trajectories in early adult-
[3] Rahmouni K. Obesity-associated hypertension: recent progress in deciphering hood and subclinical atherosclerosis in middle age. JAMA 2014;311:490–7.
the pathogenesis. Hypertension 2014;64:215–21. [34] Bucholz EM, Gooding HC, de Ferranti SD. Awareness of cardiovascular risk fac-
[4] Ahirwar R, Mondal PR. Prevalence of obesity in India: A systematic review. tors in US young adults aged 18-39 years. Am J Prev Med 2018;54:e67–77.
Diabetes Metab Syndr 2019;13(1):318–21. [35] Mahajan S, Zhang D, He S, et al. Prevalence, Awareness, and Treatment of Iso-
[5] Bui Van N, Vo Hoang L, Bui Van T, et al. Prevalence and Risk Factors of lated Diastolic Hypertension: Insights From the China PEACE Million Persons
Hypertension in the Vietnamese Elderly. High Blood Press Cardiovasc Prev Project. J Am Heart Assoc 2019;8:e012954.
2019;26(3):239–46. [36] Zhen S, Ma Y, Zhao Z, Yang X, Wen D. Dietary pattern is associated with obe-
[6] Hanh NTH, Tuyet LT, Dao DTA, Tao Y, Chu DT. Childhood Obesity Is a High- sity in Chinese children and adolescents: data from China Health and Nutrition
-risk Factor for Hypertriglyceridemia: A Case-control Study in Vietnam. Osong Survey (CHNS). Nutr J 2018;17(1):68.
Public Health Res Perspect 2017;8(2):138–46. [37] Monda KL, Adair LS, Zhai F, Popkin BM. Longitudinal relationships between oc-
[7] Shimamoto K, Ando K, Fujita T, et al. The Japanese Society of Hypertension cupational and domestic physical activity patterns and body weight in China.
Guidelines for the Management of Hypertension (JSH 2014). Hypertens Res Eur J Clin Nutr 2008;62:1318–25.
2014;37(4):253–390. [38] Chen C, Chou SY, Thornton RJ. The effect of household technology on weight
[8] Bundy JD, He J. Hypertension and Related Cardiovascular Disease Burden in and health outcomes among Chinese adults: evidence from China’s “Home Ap-
China. Ann Glob Health 2016;82:227–33. pliances Going to the Countryside” policy. J Hum Cap 2015;9:364–401.
[9] The State Council Information Office of the People’s Republic of China. Press [39] Huang C-C, Yabiku ST, Kronenfeld JJ. The effects of household technology on
briefing for the Report on Chinese Residents’ Chronic Diseases and Nutrition body mass index among Chinese adults. Popul Res Policy Rev 2015;34:877–99.
2020. (in Chinese). http://www.gov.cn/xinwen/2020-12/24/content_5572983. [40] Zhang X, Dagevos H, He Y, van der Lans I, Zhai F. Consumption and corpulence
htm (Accessed 21 June 2021). in China: A consumer segmentation study based on the food perspective. Food
[10] Lu J, Lu Y, Wang X, et al. Prevalence, awareness, treatment, and con- Policy 2008;33:37–47.
trol of hypertension in China: data from 1•7 million adults in a popula- [41] Bonnefond C, Clement M. Social class and body weight among Chinese urban
tion-based screening study (China PEACE Million Persons Project). Lancet adults: the role of the middle classes in the nutrition transition. Soc Sci Med
2017;390(10112):2549–58. 2014;112:22–9.
[11] Institute for Health Metrics and Evaluation. Global Health Data Exchange. [42] Thomopoulos C, Parati G, Zanchetti A. Effects of blood pressure-lowering
GBD results tool. http://ghdx.healthdata.org/gbd-results-tool (Accessed 21 June treatment on cardiovascular outcomes and mortality: 13 - benefits and ad-
2021). verse events in older and younger patients with hypertension: overview,
[12] Pan XF, Wang L, Pan A. Epidemiology and determinants of obesity in China. meta-analyses and meta-regression analyses of randomized trials. J Hypertens
Lancet Diabetes Endocrinol 2021;9(6):373–92. 2018;36:1622–36.
[13] Fang L, Song J, Ma Z, Zhang L, Jing C, Chen D. Prevalence and characteristics of [43] Liu LS. Writing Group of 2010 Chinese Guidelines for the Management of Hy-
hypertension in mainland Chinese adults over decades: a systematic review. J pertension. [2010 Chinese guidelines for the management of hypertension].
Hum Hypertens 2014;28:649–56. Zhonghua Xin Xue Guan Bing Za Zhi 2011;39:579–615 Chinese.
[14] Shen C, Zhou Z, Lai S, et al. Urban-rural-specific trend in prevalence of general [44] Whelton PK, Carey RM, Aronow WS, et al. PCNA Guideline for the Pre-
and central obesity, and association with hypertension in Chinese adults, aged vention, Detection, Evaluation, and Management of High Blood Pressure in
18-65 years. BMC Public Health 2019;19:661. Adults: Executive Summary: A Report of the American College of Cardiol-
[15] Unger T, Borghi C, Charchar F, et al. 2020 International Society of Hypertension ogy/American Heart Association Task Force on Clinical Practice Guidelines.
Global Hypertension Practice Guidelines. Hypertension 2020;75:1334–57. Hypertension 2018;71:1269–324 http://www.ACC/AHA/AAPA/ABC/ACPM/AGS/
[16] Verdecchia P, Reboldi G, Angeli F. The 2020 International Society of Hyperten- APhA/ASH/ASPC/NMA/.
sion global hypertension practice guidelines - key messages and clinical con- [45] 3rd Booth JN, J Li, Zhang L, Chen L, Muntner P, Egan B. Trends in Prehyper-
siderations. Eur J Intern Med 2020;82:1–6. tension and Hypertension Risk Factors in US Adults: 1999-2012. Hypertension
[17] Yang W, Lu J, Weng J, et al. Prevalence of diabetes among men and women in 2017;70:275–84.
China. N Engl J Med 2010;362:1090–101. [46] Whelton PK, He J, Appel LJ, et al. Primary prevention of hypertension: clinical
[18] Li Y, Teng D, Shi X, et al. Prevalence of diabetes recorded in mainland China and public health advisory from The National High Blood Pressure Education
using 2018 diagnostic criteria from the American Diabetes Association: na- Program. JAMA 2002;288:1882–8.
tional cross sectional study. BMJ 2020;369:m997. [47] Du T, Sun X, Yin P, Huo R, Ni C, Yu X. Increasing trends in central obesity
[19] Kang H. The prevention and handling of the missing data. Korean J Anesthesiol among Chinese adults with normal body mass index, 1993-2009. BMC Public
2013;64(5):402–6. Health 2013;13:327.
[20] Luepker Russell V, Evans Alun, McKeigue Paul, Reddy KSrinath. Car- [48] Xi B, Liang Y, He T, Reilly KH, Hu Y, Wang Q, Yan Y, Mi J. Secular trends
diovascular survey methods. Third Edition. Geneva, Switzerland: World in the prevalence of general and abdominal obesity among Chinese adults,
Health Organization; 2004 https://apps.who.int/iris/bitstream/handle/10665/ 1993-2009. Obes Rev 2012;13(3):287–96.
42569/9241545763_eng.pdf;sequence=1 . [49] Van de Poel E, O’Donnell O, Van Doorslaer E. Urbanization and the spread of
[21] Alberti KG, Zimmet P, Shaw J. International Diabetes Federation: a consensus diseases of affluence in China. Econ Hum Biol 20 09;7(2):20 0–16.
on Type 2 diabetes prevention. Diabet Med 2007;24:451–63. [50] National Bureau of Statistics. Average food consumption per person in rural
[22] Pan WH, Yeh WT. How to define obesity? Evidence-based multiple action households. http://data.stats.gov.cn/easyquery.htm?cn=C01&zb=A0A01&sj=2016
points for public awareness, screening, and treatment: an extension of Asian– (Accessed 31 May 2018).
Pacific recommendations. Asia Pac J Clin Nutr 2008;17:370–4. [51] Ouyang Y, Wang H, Su C, Du W, Wang Z, Zhang B. Why is there gender dispar-
[23] SAS/STAT(R) 9.22 User’s Guide. https://support.sas.com/documentation/cdl/en/ ity in the body mass index trends among adults in the 1997-2011 China health
statug/63347/HTML/default/viewer.htm#statug_surveylogistic_a0 0 0 0 0 0 0386. and nutrition surveys? Asia Pac J Clin Nutr 2015;24(4):692–700.
htm (Accessed 21 June 2021). [52] Zhang X, Wang J, Li J, Yu Y, Song Y. A positive association between dietary
[24] Gupta R, Gaur K, Ram CV S. Emerging trends in hypertension epidemiology in sodium intake and obesity and central obesity: results from the National
India. J Hum Hypertens 2019;33(8):575–87. Health and Nutrition Examination Survey 1999-2006. Nutr Res 2018;55:33–44.
[25] Gulati S, Misra A. Sugar intake, obesity, and diabetes in India. Nutrients [53] Du S, Wang H, Zhang B, Popkin BM. Dietary Potassium Intake Remains
2014;6(12):5955–74. Low and Sodium Intake Remains High, and Most Sodium is Derived
[26] Tuan NT, Tuong PD, Popkin BM. Body mass index (BMI) dynamics in Vietnam. from Home Food Preparation for Chinese Adults, 1991-2015 Trends. J Nutr
Eur J Clin Nutr 2008;62(1):78–86. 2020;150(5):1230–9.
[27] Nagai M, Ohkubo T, Murakami Y, et al. Secular trends of the impact of [54] Rotch AL, Dean JO, Kendrach MG, Wright SG, Woolley TW. Blood pressure
overweight and obesity on hypertension in Japan, 1980-2010. Hypertens Res monitoring with home monitors versus mercury sphygmomanometer. Ann
2015;38:790–5. Pharmacother 2001;35(7-8):817–22.
[28] Kim TJ, Lee JW, Kang HT, et al. Trends in Blood Pressure and Prevalence of [55] Topouchian JA, El Assaad MA, Orobinskaia LV, El Feghali RN, Asmar RG. Valida-
Hypertension in Korean Adults Based on the 1998-2014 KNHANES. Yonsei Med tion of two automatic devices for self-measurement of blood pressure accord-
J 2018;59:356–65. ing to the International Protocol of the European Society of Hypertension: the
[29] Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity Omron M6 (HEM-7001-E) and the Omron R7 (HEM 637-IT). Blood Press Monit
among US adults, 1999-2008. JAMA 2010;303:235–41. 2006;11(3):165–71.
[30] Guo F, He D, Zhang W, Walton RG. Trends in prevalence, awareness, manage-
ment, and control of hypertension among United States adults, 1999 to 2010. J
Am Coll Cardiol 2012;60:599–606.

10