The Burden of Hypertension and Associated Risk for Cardiovascular Mortality in China

Importance  Hypertension is a leading cause of premature death in China, but limited evidence is available on the prevalence and management of hypertension and its effect on mortality from cardiovascular disease (CVD).

Objectives  To examine the prevalence, diagnosis, treatment, and control of hypertension and to assess the CVD mortality attributable to hypertension in China.

Design, Setting and Participants  This prospective cohort study (China Kadoorie Biobank Study) recruited 500 223 adults, aged 35 to 74 years, from the general population in China. Blood pressure (BP) measurements were recorded as part of the baseline survey from June 25, 2004, to August 5, 2009, and 7028 deaths due to CVD were recorded before January 1, 2014 (mean duration of follow-up: 7.2 years). Data were analyzed from June 9, 2014, to July 17, 2015.

Exposures  Prevalence and diagnosis of hypertension (systolic BP ≥140 mm Hg, diastolic BP ≥90 mm Hg, or receiving treatment for hypertension) and treatment and control rates overall and in various population subgroups.

Main Outcomes and Measures  Cox regression analysis yielded age- and sex-specific rate ratios for deaths due to CVD comparing participants with and without uncontrolled hypertension, which were used to estimate the number of CVD deaths attributable to hypertension.

Results  The cohort included 205 167 men (41.0%) and 295 056 women (59.0%) with a mean (SD) age of 52 (10) years for both sexes. Overall, 32.5% of participants had hypertension; the prevalence increased with age (from 12.6% at 35-39 years of age to 58.4% at 70-74 years of age) and varied substantially by region (range, 22.7%-40.7%). Of those with hypertension, 30.5% had received a diagnosis from a physician; of those with a diagnosis of hypertension, 46.4% were being treated; and of those treated, 29.6% had their hypertension controlled (ie, systolic BP <140 mm Hg; diastolic BP <90 mm Hg), resulting in an overall control rate of 4.2%. Even among patients with hypertension and prior CVD, only 13.0% had their hypertension controlled. Uncontrolled hypertension was associated with relative risks for CVD mortality of 4.1 (95% CI, 3.7-4.6), 2.6 (95% CI, 2.4-2.9) and 1.9 (95% CI, 1.8-2.0) at ages 35 to 59, 60 to 69, and 70 to 79 years, respectively, and accounted for about one-third of deaths due to CVD (approximately 750 000) at 35 to 79 years of age in 2010.

Conclusions and Relevance  About one-third of Chinese adults in this national cohort population had hypertension. The levels of diagnosis, treatment, and control were much lower than in Western populations, and were associated with significant excess mortality.

In 2010, cardiovascular disease (CVD) accounted for more than one-third of all adult deaths in China, including 1.7 million deaths due to stroke and 0.9 million deaths due to ischemic heart disease (IHD).¹ Hypertension is one of the most important causes of CVD mortality in China, because of its high prevalence and concomitant vascular risks.^2,3

Large prospective studies of mainly Western populations have shown a continuous positive association for stroke and IHD-related mortality with systolic blood pressure (SBP) down to at least 115 mm Hg, with each 20-mm Hg lower usual SBP associated with an approximate halving in risk for stroke and IHD mortality.³ Furthermore, randomized clinical trials of BP-lowering medication (antihypertensives)⁴ have demonstrated that lowering SBP by 10 mm Hg lowers the risk for stroke and IHD by about one-fourth, indicating almost complete reversibility of the excess risks for CVD associated with hypertension within a few years of treatment.

Successive population surveys of hypertension in China during the last 30 years have reported an increasing prevalence of hypertension, with the most recent surveys indicating that about one-third of all adults have hypertension.^5-7 Considerable uncertainty remains about the prevalence of hypertension by age, sex, region, and socioeconomic status; the current levels of BP management in the population; and the likely effect of this high burden of uncontrolled hypertension on CVD mortality rates in China.

We herein report findings from the recent large-scale prospective China Kadoorie Biobank Study of 500 223 adults aged 35 to 74 years (when first surveyed during 2004-2008) in 10 diverse regions of China.^8,9 The aims of the present report were to (1) examine the prevalence, diagnosis, treatment, and control of hypertension at baseline, overall and by age, sex, region, and socioeconomic and other characteristics; (2) assess prospectively the age- and sex-specific effects of hypertension on CVD mortality; and (3) estimate the number of CVD deaths attributed to hypertension in 2010 in China.

Details of the study design, survey methods, and participant characteristics have been reported previously.^8,9 Briefly, the baseline survey was conducted from June 25, 2004, to August 5, 2009, in 10 geographically defined regions (5 urban and 5 rural) of China. The regions were chosen to ensure a wide range of risk exposures and disease patterns while taking account of the quality of established disease registries for follow-up and the local capacity to conduct the study. The regions were not selected to be representative of China as a whole. In each region, temporary assessment centers were set up at local residential centers (village or street committees), and permanent residents with no major disability were invited to participate; about one-third of residents responded.⁹ Local, national, and international ethics approvals were obtained (see eMethods in the Supplement for details). All participants provided written informed consent.

At the baseline survey, trained health care workers collected detailed information on sociodemographic characteristics, lifestyle, and medical history using a laptop computer–based questionnaire. Each participant also had measurements taken, including height, weight, lung function, and BP. Blood pressure was measured twice by trained staff using a digital sphygmomanometer (Omron UA-779; Live Source) after participants had remained at rest in the seated position for at least 5 minutes. If the difference between the 2 measurements was greater than 10 mm Hg for the SBP, a third measurement was obtained and the last 2 measurements were recorded. The mean of the 2 recorded values of SBP and diastolic BP (DBP) were used in all analyses. All devices were regularly maintained and calibrated to ensure consistency of the measurements. Participants were considered to be hypertensive if they had a measured SBP of at least 140 mm Hg or a measured DBP of at least 90 mm Hg or were receiving treatment for hypertension. The latter was defined as those who reported a diagnosis of hypertension by a physician and use of antihypertensives at baseline.

Vital status of each participant was obtained through the Disease Surveillance Points system of the Chinese Centre for Disease Control and Prevention, annual checks against local residential records and health insurance records, and active confirmation through street committee or village administrators. Residential records were also used to identify those who had moved out of the area and were therefore lost to follow-up. Fatal CVD events were assigned codes from the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (I00-I25, I27-I88, or I95-I99) by trained Disease Surveillance Points staff who were blinded to baseline information. By January 1, 2014, 25 448 deaths occurred, including 7028 due to CVD, and 2411 participants (0.5%) were lost to follow-up.

Data were analyzed from June 9, 2014, to July 17, 2015. We excluded participants with a baseline age outside the age range of interest for this study (9410 persons <35 years; 3258 persons ≥75 years). We found no missing values for BP or any other key variables in the remaining 500 223 participants. Mean SBP, mean DBP, and prevalence of hypertension were calculated within each age group (35-39, 40-49, 50-59, 60-69, and 70-74 years) separately for men and women. To assess the management of hypertension (as previously defined), we calculated the proportion of those with hypertension who had received a diagnosis from a physician, the proportion of those diagnosed who were treated with antihypertensives, and the proportion of those treated who achieved BP control (ie, SBP <140 mm Hg and DBP <90 mm Hg). These proportions were stratified by age, sex, area, prior CVD (IHD or stroke or transient ischemic attack reported at baseline), the season at the time of the survey, household income, and highest level of education; all subgroups were prespecified. Analyses were adjusted, where appropriate, for age, sex, and region.

We used Cox regression analysis¹⁰ to calculate the age- and sex-specific death rate ratios and 95% CIs (adjusted for region) for CVD mortality comparing participants with uncontrolled hypertension (SBP ≥140 mm Hg or DBP ≥90 mm Hg) and those without uncontrolled hypertension. These analyses excluded individuals with a history of CVD to limit the effects of reverse causality. Participants contributed person-years until death, the date lost to follow-up, or the censoring date (December 31, 2013). We calculated the age- and sex-specific population-attributable fractions using the formula P_d(RR − 1)/RR, where P_d is the proportion of participants who died of vascular causes with uncontrolled hypertension and RR is the adjusted death rate ratio for vascular mortality in participants with vs without uncontrolled hypertension.¹¹ Population-attributable fractions were applied to the estimated age- and sex-specific number of CVD deaths in China for 2010,^12,13 to give the number of deaths attributable to hypertension for that year. The proportionality assumption of the Cox regression was checked by comparing death rate ratios in the first and second half of the follow-up period and confirmed the validity of the analyses. Analyses used SAS software (version 9.3; SAS Institute Inc), and figures were plotted using R software (version 3.0; https://www.r-project.org).

Among the 500 223 participants included in the analysis, 59.0% were women and 41.0% were men; the mean (SD) age was 52 (10) years for both sexes (Table 1). Overall, 55.9% were from rural areas, 57.0% had an annual household income of less than ¥20 000, and 18.5% had no formal schooling (men, 8.8%; women, 25.3%); 4.5% reported a history of CVD at baseline. More than half of the participants were surveyed during spring or autumn (55.4%); 20.1% were surveyed during winter; and 24.5% were surveyed during summer.

Mean SBP showed an approximately linear increase with age in both sexes, but the association was about twice as extreme in women compared with men (Figure 1). Mean levels of SBP were similar in both sexes at about 55 years of age, however, such that men had higher SBP than women at younger than 55 years and women had higher SBP at older than 55 years. The association between DBP and age was nonlinear, with a positive association in those younger than 55 years and an inverse association thereafter. The shape of this association was similar in both sexes, but mean DBP was higher at all ages in men.

Overall, 32.5% of participants (33.7% of men and 31.9% of women) had hypertension (Figure 2); 17.3% of the total population had isolated systolic hypertension, 2.0% had isolated diastolic hypertension, 11.8% had both, and 1.4% had controlled hypertension (eTable 1 in the Supplement). Overall, 20.1% of participants had stage 1 hypertension (SBP 140-159 mm Hg or DBP 90-99 mm Hg), 11.0% had stage 2 hypertension (SBP ≥160 mm Hg or DBP ≥100 mm Hg) (eTable 2 in the Supplement), and 39.5% had prehypertension (SBP 120-139 mm Hg or DBP 80-89 mm Hg). The prevalence of hypertension increased with age (from 12.6% at 35-39 years of age to 58.4% at 70-74 years of age), but the age-specific prevalence of hypertension was higher in men than women when younger (age 35-39 years, 17.9% for men and 8.8% for women) and slightly higher in women than men when older (age 70-74 years, 60.2% for women and 56.2% for men) (Figure 2 and eTables 3 and 4 in the Supplement).

Overall, only 30.5% of participants with hypertension had received a diagnosis by a physician (eTable 3 in the Supplement). This proportion increased with age and ranged from 15.2% at ages 35 to 39 years to 37.9% at ages 70 to 74 years in women and 10.8% at ages 35 to 39 years to 36.1% at ages 70 to 74 years in men. Among participants with a diagnosis of hypertension, less than half were being treated, and this proportion did not vary substantially by age or sex (48.3% for men and 45.1% for women) (eTable 3 in the Supplement; eTable 5 in the Supplement provides information on the type of antihypertensive used). Among those treated for hypertension, only 29.6% achieved control of hypertension and this proportion was similar in men and women. This proportion declined with increasing age and to a slightly greater extent in women than men. Overall, the control of hypertension as a proportion of all participants with hypertension increased with age, but never exceeded 5.0% in both sexes and at all ages (eTable 3 in the Supplement).

The prevalence of hypertension varied markedly by region, ranging from 22.7% in Haikou to 40.7% in Zhejiang (Table 2 and eTable 6 in the Supplement). The prevalence was slightly higher in rural than urban areas, but no clear difference by geography (north vs south or east vs west) was found. We detected a 2-fold variation by region in the proportions of participants with hypertension who were diagnosed (range, 21.0% in Sichuan to 43.8% in Liuzhou) and a 3-fold variation in the proportions of participants with diagnosed hypertension who received treatment (range, 18.5% in Zheijiang to 64.1% in Liuzhou) and those in whom treated hypertension was controlled (range, 13.8% in Gansuto to 39.0% in Luizhou). Overall, however, the proportion of hypertensive participants with controlled BP was low in all regions (range, 1.3% in Zhejiang to 11.0% in Liuzhou).

Hypertension was slightly more common at lower levels of education (no formal school, 34.8%; technical school or university, 30.7%), but did not vary materially by income (Table 2 and eTable 6 in the Supplement). Diagnosis, treatment, and control of hypertension, however, were substantially less common at lower levels of education and income. Those with prior CVD at baseline had a higher prevalence of hypertension than those without prior CVD (49.4% vs 31.8), but they also had higher rates of diagnosis, treatment, and control. Control of BP in participants with hypertension and prior CVD, however, was still low (13.0%). The estimated prevalence of hypertension was substantially higher in winter (39.6%) than summer (24.9%) and intermediate in spring or autumn (33.3%).

Uncontrolled hypertension was associated with relative risks for CVD mortality for both men and women combined of 4.13 (95% CI, 3.72-4.59), 2.61 (95% CI, 2.38-2.85), and 1.89 (95% CI, 1.76-2.03) at ages 35 to 59, 60 to 69, and 70 to 79 years, respectively (Figure 3). We found no evidence that further adjustment for a broad range of socioeconomic and lifestyle risk factors (listed in eTable 7 in the Supplement) materially changed the strength of these associations. Overall, uncontrolled hypertension accounted for about one-third of all CVD deaths at ages 35 to 79 years (almost half of all CVD deaths at ages 35-59 years). In 2010, approximately 750 000 CVD deaths could be attributed to uncontrolled hypertension in China.

This large study provides important new evidence about the current burden of hypertension in China, particularly the extent of its undertreatment and the consequences of uncontrolled hypertension for premature vascular death. Overall, about one-third of the adult population had hypertension. The prevalence increased steeply with age, varied substantially by region, and was higher in those with prior CVD. About one-third of those with hypertension were diagnosed; of those diagnosed, about half were treated; and, of those treated, about one-third had their hypertension controlled, resulting in less than 5% of participants with hypertension who achieved properly controlled BP. Undertreatment was observed in all regions and subgroups assessed, including those with prior CVD. Although the prevalence of hypertension was not strongly related to socioeconomic status, the control of BP was lower among participants with lower levels of education and income. Uncontrolled hypertension was estimated to cause 750 000 CVD deaths annually in China.

Although the China Kadoorie Biobank Study was not designed as a nationally representative study, the prevalence of hypertension observed in this study is consistent with previous reports of recent nationally representative hypertension surveys in China.^5,6,14 Only a few studies (all substantially smaller than the present study) have reported the age-specific proportions of persons with hypertension who have received a diagnosis and in whom the hypertension is treated and controlled. The 2007 to 2008 China National Diabetes and Metabolic Disorders Survey⁶ and the 2000 to 2001 Chinese InterAsia Survey⁵ reported higher levels of diagnosed hypertension than the present study, with about half of participants with hypertension receiving a diagnosis at 45 to 64 years of age (vs one-third in the present study). As in the present study, the participants in those studies were not randomly selected. Unlike the present study, however, the levels of diagnosed hypertension were based on diagnosis by any health care professional rather than specifically by a fully qualified physician. The InterAsia Survey⁵ also reported that at 45 to 64 years of age, two-thirds of participants with diagnosed hypertension were treated (vs half in the present study), but less than one-third of those treated achieved control of hypertension, consistent with the present study.

A systematic review in 2005 of the prevalence of hypertension in different regions of the world reported that the prevalence in China was higher than in other parts of Asia, but lower than in established market economies.² The levels of diagnosis, treatment, and control of hypertension in China, however, are substantially lower than those achieved in many high-income countries, where recent surveys indicate that more than two-thirds of the population with hypertension receive a diagnosis, more than two-thirds of those with a diagnosis are treated, and about two-thirds of those treated achieve BP control.¹⁵ Although previous reports indicated that the prevalence of hypertension was slightly lower on average in China than in the West, the present study indicates that the prevalence of hypertension in China is now comparable to that of Western populations, but the prevalence of uncontrolled hypertension is much higher in China than in Western populations. Occlusive CVD mortality has fallen significantly during the last few decades in the West, due at least in part to better management of hypertension; by contrast, occlusive CVD mortality has increased in China during the same period.^16-18

A large prospective cohort study¹⁹ reported that about 60% of CVD deaths in China at ages 55 to 64 years in 2005 were attributable to hypertension, compared with about 40% in this age group in the present study. In this study, the risk for CVD mortality in persons with hypertension was compared with that in persons with normal BP (defined as SBP <120 mm Hg and DBP <80 mm Hg). In the present study, usual BP in those who did not have hypertension was in the prehypertensive range (SBP 120-139 mm Hg and DBP 80-89 mm Hg), which is the standard target range for those with hypertension; as such, the population-attributable fractions and associated attributable deaths more closely estimate the effect of addressing hypertension than reducing BP to much lower levels. A large trial²⁰ and a meta-analysis of all available trials²¹ recently demonstrated reductions in major cardiovascular events from antihypertensive regimens targeting SBP less than 140 mm Hg even in those older than 60 years, consistent with the evidence from prospective cohort studies. A combination of antihypertensives is generally required to reach this target BP range, but only a small proportion (17.0%) of those treated for hypertension in the present study were taking more than 1 type of antihypertensive (eTable 5 in the Supplement).

The strengths of the present study include the very large sample size, the diversity (geographic and socioeconomic) of the areas sampled, the range of population subgroups assessed, and the standardization of techniques to measure BP at baseline, including the training of technicians. In addition, because the baseline survey was conducted throughout the year and for several years, we were able to examine the effect of the season on the estimates of prevalence and control, and health care professionals and policy makers should be aware of such effects when evaluating the burden and control of BP in China.²² The study was not designed to collect a representative sample of the Chinese population, which limits the generalizability of the findings to China as a whole, despite the consistency of the age-specific prevalence with recently conducted nationally representative surveys. Before the analyses in this study, the real prevalence of hypertension in China was expected, if anything, to be underestimated in this cohort of generally healthy volunteers. Another limitation is that the diagnosis of hypertension was based on serial BP measurements recorded on a single occasion (as is standard in BP surveys), whereas ideally 2 or more readings would be taken on separate occasions. The estimated prevalence of hypertension at baseline will be affected by regression to the mean of subsequent BP measurements. This regression to the mean is unlikely to affect the estimated prevalence of hypertension materially at 60 to 69 years of age (because the mean BP at this age is approximately equal to the threshold BP for hypertension), but the prevalence may be somewhat overestimated at younger ages and underestimated at older ages.

In 2013, the World Health Assembly set voluntary targets for the control of noncommunicable diseases, including a 25% reduction in the prevalence of elevated BP by 2025. We estimate that such a reduction would have prevented about 130 000 vascular deaths in China for those who were from 35 to 79 years of age in 2010 alone. Detection and adequate treatment of hypertension in primary care have a key role to play in meeting this goal, and the present study shows an enormous potential for improvement, given that no regions or subgroups had satisfactory BP control; such an approach ideally would be expanded to lower the BP of all individuals at high absolute risk for a cardiovascular event.²³ Furthermore, public health initiatives are required that focus on the major determinants of BP at a population level, including salt intake (known to be particularly high in China²⁴), harmful alcohol consumption, obesity, and poor home heating (shown to have a substantial effect on BP in China²²). Ongoing research seeks to identify the most effective population-level approaches for BP control in China.²⁵ These population-based approaches also would have the added benefit of reducing the large proportion of vascular deaths that occur in persons with normal BP. Unless concerted efforts are made to lower BP levels in China, rates of death due to CVD are likely to increase further during the next few decades.

Corresponding Author: Robert Clarke, FRCP, Clinical Trial Service Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford, Richard Doll Building, Roosevelt Drive, Oxford OX3 7LF, England (robert.clarke@ctsu.ox.ac.uk).

Accepted for Publication: January 7, 2016.

Published Online: March 14, 2016. doi:10.1001/jamainternmed.2016.0190.

Author Contributions: Drs Lewington, Lacey, Clarke, Li, and Z. Chen contributed equally to this study. Dr Lewington had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Lewington, Clarke, Li, Z. Chen.

Acquisition, analysis, or interpretation of data: Lacey, Clarke, Guo, Kong, Yang, Y. Chen, Bian, J. Chen, Meng, Xiong, He, Pang, Zhang, Collins, Peto, Li, Z. Chen.

Drafting of the manuscript: Lewington, Lacey, Clarke, Kong.

Critical revision of the manuscript for important intellectual content: Lewington, Lacey, Clarke, Guo, Yang, Y. Chen, Bian, J. Chen, Meng, Xiong, He, Pang, Zhang, Collins, Peto, Li, Z. Chen.

Statistical analysis: Lewington, Lacey, Kong, Z. Chen.

Obtained funding: Collins, Peto, Li, Z. Chen.

Administrative, technical, or material support: Lacey, Clarke, Guo, Yang, Y. Chen, Bian, J. Chen, Meng, Xiong, He, Pang, Zhang, Collins, Li, Z. Chen.

Study supervision: Lewington, Clarke, Pang, Peto, Li, Z. Chen.

Conflict of Interest Disclosures: None reported.

Funding/Support: This study was supported by the Kadoorie Charitable Foundation (China Kadoorie Biobank baseline and first follow-up surveys), by grant 088158/Z/09/Z from the Wellcome Trust (2009 to 2014 follow-up of the project), and by core funding from the UK Medical Research Council, the British Heart Foundation, and Cancer Research UK (Clinical Trial Service Unit and Epidemiological Studies Unit at Oxford University).

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Group Members: Members of the China Kadoorie Biobank Consortium include the following: International Steering Committee: Junshi Chen, MD, Zhengming Chen, DPhil (principal investigator [PI]), Rory Collins, FRS, Liming Li, PhD (PI), and Richard Peto, FRS; International Coordinating Centre, Oxford, England: Daniel Avery, MSc, Derrick Bennett, PhD, Yumei Chang, PhD, Yiping Chen, DPhil, Zhengming Chen, DPhil, Robert Clarke, FRCP, Huaidong Du, PhD, Xuejuan Fan, PhD, Simon Gilbert, MSc, Alex Hacker, BA, Michael Holmes, PhD, Andri Iona, MSc, Christiana Kartsonaki, DPhil, Rene Kerosi, MSc, Xiang Ling Kong, BSc, Om Kurmi, PhD, Garry Lancaster, BA, Sarah Lewington, DPhil, John McDonnell, GD(Comp Sc), Winnie Mei, MPhil, Iona Millwood, DPhil, Qunhua Nie, BA, Jayakrishnan Radhakrishnan, MSc, Sajjad Rafiq, PhD, Paul Ryder, BA, Sam Sansome, BSc, Dan Schmidt, MSc, Paul Sherliker, BA, Rajani Sohoni, BSc, Iain Turnbull, MBBS, Robin Walters, PhD, Jenny Wang, MSc, Lin Wang, MSc, Ling Yang, PhD, and Xiaoming Yang, PhD; National Coordinating Center, Beijing, China: Zheng Bian, MD, Ge Chen, BSc, Yu Guo, MSc, Bingyang Han, MSc, Can Hou, BSc, Jun Lv, PhD, Pei Pei, BSc, Shuzhen Qu, BSc, Yunlong Tan, MD, Canqing Yu, PhD, and Huiyan Zhou, MA; Qingdao Centre for Disease Control and Prevention (CDC): Zengchang Pang, BSc, Ruqin Gao, Shaojie Wang, Yongmei Liu, Ranran Du, Yajing Zang, Liang Cheng, Xiaocao Tian, and Hua Zhang; Licang CDC: Silu Lv, Junzheng Wang, and Wei Hou; Heilongjiang Provincial CDC: Jiyuan Yin, Ge Jiang, Shumei Liu, Zhigang Pang, BSc, and Xue Zhou; Nangang CDC: Liqiu Yang, Hui He, Bo Yu, Yanjie Li, Huaiyi Mu, Qinai Xu, Meiling Dou, and Jiaojiao Ren; Hainan Provincial CDC: Jianwei Du, Shanqing Wang, Ximin Hu, Hongmei Wang, Jinyan Chen, Yan Fu, Zhenwang Fu, Xiaohuan Wang, and Hua Dong; Meilan CDC: Min Weng, Xiangyang Zheng, Yijun Li, Huimei Li, and Chenglong Li; Jiangsu Provincial CDC: Ming Wu, Jinyi Zhou, Ran Tao, and Jie Yang; Suzhou CDC: Jie Shen, Yihe Hu, Yan Lu, Yan Gao, Liangcai Ma, Renxian Zhou, Aiyu Tang, Shuo Zhang, and Jianrong Jin; Guangxi Provincial CDC: Zhenzhu Tang, Naying Chen, and Ying Huang; Liuzhou CDC: Mingqiang Li, Jinhuai Meng, Rong Pan, Qilian Jiang, Jingxin Qing, Weiyuan Zhang, Yun Liu, Liuping Wei, Liyuan Zhou, Ningyu Chen, Jun Yang, and Hairong Guan; Sichuan Provincial CDC: Xianping Wu, Ningmei Zhang, Xiaofang Chen, and Xuefeng Tang; Pengzhou CDC: Guojin Luo, Jianguo Li, Xiaofang Chen, Jian Wang, Jiaqiu Liu, and Qiang Sun; Gansu Provincial CDC: Pengfei Ge, Xiaolan Ren, and Caixia Dong; Maiji CDC: Hui Zhang, Enke Mao, Xiaoping Wang, and Tao Wang; Henan Provincial CDC: Guohua Liu, Baoyu Zhu, Gang Zhou, Shixian Feng, Liang Chang, and Lei Fan; Huixian CDC: Yulian Gao, Tianyou He, Li Jiang, Huarong Sun, Pan He, Chen Hu, Qiannan Lv, and Xukui Zhang; Zhejiang Provincial CDC: Min Yu, Ruying Hu, Le Fang, and Hao Wang; Tongxiang CDC: Yijian Qian, Chunmei Wang, Kaixue Xie, Lingli Chen, Yaxing Pan, and Dongxia Pan; Hunan Provincial CDC: Yuelong Huang, Biyun Chen, Donghui Jin, Huilin Liu, Zhongxi Fu, and Qiaohua Xu; and Liuyang CDC: Xin Xu, Youping Xiong, Weifang Jia, Xianzhi Li, Libo Zhang, and Zhe Qiu.

Additional Contributions: Judith Mackay, FRCP (Chinese University of Hong Kong); Yu Wang, Gonghuan Yang, Zhengfu Qiang, Lin Feng, Maigeng Zhou, Wenhua Zhao, and Yan Zhang (China CDC); Lingzhi Kong, Xiucheng Yu, and Kun Li (Chinese Ministry of Health); and Yiping Chen, DPhil, Sarah Clark, DPhil, Martin Radley, BSc, Mike Hill, PhD, Hongchao Pan, PhD, and Jill Boreham, PhD (Clinical Trial Service Unit, Oxford University) assisted with the design, planning, organization, and conduct of the study (no compensation was received). We thank the participants of the study and the members of the survey teams in each of the 10 regional centers and the project development and management teams based at Beijing, Oxford, and the 10 regional centers.