China by News Hour

Articles

Prevalence, awareness, treatment, and control of hypertension in China: data from 1·7 million adults in a population-based screening study (China PEACE Million Persons Project) Jiapeng Lu*, Yuan Lu*, Xiaochen Wang, Xinyue Li, George C Linderman, Chaoqun Wu, Xiuyuan Cheng, Lin Mu, Haibo Zhang, Jiamin Liu, Meng Su, Hongyu Zhao, Erica S Spatz, John A Spertus, Frederick A Masoudi, Harlan M Krumholz†, Lixin Jiang†

Summary

Background Hypertension is common in China and its prevalence is rising, yet it remains inadequately controlled. Few studies have the capacity to characterise the epidemiology and management of hypertension across many heterogeneous subgroups. We did a study of the prevalence, awareness, treatment, and control of hypertension in China and assessed their variations across many subpopulations. Methods We made use of data generated in the China Patient-Centered Evaluative Assessment of Cardiac Events (PEACE) Million Persons Project from Sept 15, 2014, to June 20, 2017, a population-based screening project that enrolled around 1·7 million community-dwelling adults aged 35–75 years from all 31 provinces in mainland China. In this population, we defined hypertension as systolic blood pressure of at least 140 mm Hg, or diastolic blood pressure of at least 90 mm Hg, or self-reported antihypertensive medication use in the previous 2 weeks. Hypertension awareness, treatment, and control were defined, respectively, among hypertensive adults as a self-reported diagnosis of hypertension, current use of antihypertensive medication, and blood pressure of less than 140/90 mm Hg. We assessed awareness, treatment, and control in 264 475 population subgroups—defined a priori by all possible combinations of 11 demographic and clinical factors (age [35–44, 45–54, 55–64, and 65–75 years], sex [men and women], geographical region [western, central, and eastern China], urbanity [urban vs rural], ethnic origin [Han and non-Han], occupation [ farmer and non-farmer], annual household income [< ¥10 000, ¥10 000–50 000, and ≥¥50 000], education [primary school and below, middle school, high school, and college and above], previous cardiovascular events [yes or no], current smoker [yes or no], and diabetes [yes or no]), and their associations with individual and primary health-care site characteristics, using mixed models. Findings The sample contained 1 738 886 participants with a mean age of 55·6 years (SD 9·7), 59·5% of whom were women. 44·7% (95% CI 44·6–44·8) of the sample had hypertension, of whom 44·7% (44·6–44·8) were aware of their diagnosis, 30·1% (30·0–30·2) were taking prescribed antihypertensive medications, and 7·2% (7·1–7·2) had achieved control. The age-standardised and sex-standardised rates of hypertension prevalence, awareness, treatment, and control were 37·2% (37·1–37·3), 36·0% (35·8–36·2), 22·9% (22·7–23·0), and 5·7% (5·6–5·7), respectively. The most commonly used medication class was calcium-channel blockers (55·2%, 55·0–55·4). Among individuals whose hypertension was treated but not controlled, 81·5% (81·3–81·6) were using only one medication. The proportion of participants who were aware of their hypertension and were receiving treatment varied significantly across subpopulations; lower likelihoods of awareness and treatment were associated with male sex, younger age, lower income, and an absence of previous cardiovascular events, diabetes, obesity, or alcohol use (all p<0·01). By contrast, control rate was universally low across all subgroups (<30·0%). Interpretation Among Chinese adults aged 35–75 years, nearly half have hypertension, fewer than a third are being treated, and fewer than one in twelve are in control of their blood pressure. The low number of people in control is ubiquitous in all subgroups of the Chinese population and warrants broad-based, global strategy, such as greater efforts in prevention, as well as better screening and more effective and affordable treatment. Funding Ministry of Finance and National Health and Family Planning Commission, China.

Introduction Blood pressure control is a national public health priority in China.1 Surveys in China show that high blood pressure is common, but hypertension treatment and control rates are less than 50% and 20%, respectively, across different studies.2–9 Findings from previous studies have estimated China’s average burden of hypertension,2–9 but national

data on hypertension treatment are scarce, and how hypertension awareness, treatment, and control rates vary geographically and across population subgroups is uncertain. Because of their small sample sizes, previous studies examined hypertension measures in only a few subgroups, and to our knowledge none had the capacity to create a wide variety of discrete subgroups to investigate

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Published Online October 25, 2017 http://dx.doi.org/10.1016/ S0140-6736(17)32478-9 See Online/Comment http://dx.doi.org/10.1016/ S0140-6736(17)32743-5 See Online/Articles http://dx.doi.org/10.1016/ S0140-6736(17)32476-5 *Joint first authors listed alphabetically †Joint senior authors National Clinical Research Center of Cardiovascular Diseases, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China (J Lu PhD, C Wu MS, H Zhang MD, J Liu MD, M Su PhD, Prof L Jiang MD); Center for Outcomes Research and Evaluation, Yale University/Yale-New Haven Hospital, New Haven, CT, USA (Y Lu ScD, G C Linderman BS, L Mu BS, E S Spatz MD, Prof H M Krumholz MD); Department of Mathematics (X Cheng PhD) and Department of Biostatistics (X Wang BS, X Li MS, Prof H Zhao PhD), New Haven, CT, USA; Health Outcomes Research, Saint Luke’s Mid America Heart Institute/University of Missouri-Kansas City, Kansas City, MS, USA (Prof J A Spertus MD); and Division of Cardiology, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (Prof F A Masoudi MD).

Articles

Correspondence to: Prof Lixin Jiang, National Clinical Research Center of Cardiovascular Diseases, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100037, China. jiangl@fwoxford.org

Research in context Evidence before this study We searched PubMed and the China National Knowledge Infrastructure databases for articles about the burden of hypertension in China published before July 1, 2017, without language restrictions. Our search terms were “burden”, “awareness”, “treatment”, “control”, “hypertension”, and “China”. We excluded studies that were hospital-based or done in one province. We identified 30 articles (published from 1989 to 2014) that showed a high prevalence of hypertension in the Chinese population, but with low awareness, treatment, and control. Of these 30 studies, two reported geographical variations in prevalence, awareness, treatment, and control, two compared these indicators between China and other countries, and only one reported the number of medications used by treated hypertensive patients but did not include detail about treatment patterns. No previous study enrolled patients from every province, investigated variations in treatment and control according to both geographical and individual characteristics, or assessed use of antihypertensive medications. Added value of this study Our study was the most recent nationwide report of the burden of hypertension across all 31 provinces in mainland China and in 264 475 population subgroups that were based on all possible combinations of 11 demographic and clinical variables

See Online for appendix

variations in hypertension measures. Additionally, most previous studies did not link blood pressure control with diverse individual and primary health-care sites,10–13 and did not present data on the intensity and classes of medications used by patients, who often require several drugs to control blood pressure. Accordingly, we report here the results from 1·7 million participants in our China Patient-Centered Evaluative Assessment of Cardiac Events (PEACE) Million Persons Project, a government-funded, large-scale populationbased screening project in China. We assessed the prevalence, awareness, treatment, and control of hypertension by geography and subpopulations; the number and classes of medications used by treated patients; and the association of hypertension awareness, treatment, and control with both individual and primary health-care site characteristics.

Methods

Study design and participants The China PEACE Million Persons Project has been described previously.14 Briefly, the pilot phase was successfully done between Oct 1, 2014, and June 30, 2015, to screen 0·4 million community-dwelling residents in four Chinese provinces (Jilin, Liaoning, Zhejiang, and Guangxi).14 Data were missing for less than 5% of participants, and audit results showed that the overall 2

and included more than 500 people. The study assessed geographical and individual variations in hypertension treatment and control, and antihypertensive medication use among treated patients. We showed that, despite the high prevalence of hypertension, awareness, treatment, and control were low across all population subgroups. People who were male, younger, with lower incomes, without previous cardiovascular events, and without coexisting conditions (diabetes, obesity, or alcohol use) were less likely to be aware of, or treated and controlled for, hypertension. Most people with hypertension were either not treated or treated inadequately. Implications of all the available evidence Our study identified the need for broad-based reform and interventions to mitigate the burden of hypertension in China, and a strategy to prevent and control hypertension that includes all provinces and subpopulations. Future research should focus on assessing the causes of low awareness, treatment, and control of hypertension from the perspectives of patients, doctors, and the health system, and on assessments of interventions and treatment strategies to control blood pressure at the population level. Nationally integrated strategies such as health education, free blood-pressure screening, and improved access to affordable medications are urgently needed to improve the prevention and control of hypertension in China.

quality of the data was high. From Sept 15, 2014, to June 20, 2017, we used a convenience sampling strategy to select 141 sites (88 rural counties and 53 urban districts) from all 31 provinces in mainland China (see appendix for more detail). At each site, we selected about five towns or subdistricts according to their size and population stability, in which potential partici pants were invited to the trial by local staff via extensive publicity campaigns on television and in newspapers.14 Participants were enrolled if they were aged 35–75 years and registered in the selected region’s Hukou (a record officially identifying area residents). During 2015–17, residents aged 35–75 years who had lived in the selected regions for at least 6 of the previous 12 months were enrolled. Of 1 765 425 enrolled participants, 26 539 (1·5%) were excluded because of missing data for blood pressure and covariates (appendix). The central ethics com mittee at the China National Center for Cardio vascular Disease approved this project. All enrolled participants provided written informed consent.

Data collection and variables For each participant, we measured blood pressure two times on their right upper arm after 5 min of rest in a seated position using an electronic blood pressure monitor (Omron HEM-7430; Omron Corporation, Kyoto, Japan) and a standard protocol (appendix). During a

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

standardised in-person interview done by trained personnel (appendix), participants were asked if they had used or taken a prescription drug for antiplatelet, blood pressure, lipid, or glucose control in the past 2 weeks. Those who answered “yes” and knew the drug names were asked to report the name, dose, and frequency of each drug. Those who did not remember the exact dose stated the number of tablets or pills taken (appendix). Consistent with the US Joint National Committee and Chinese definitions,15–17 hypertension was defined as an average systolic blood pressure (SBP) of at least 140 mm Hg or an average diastolic blood pressure (DBP) of at least 90 mm Hg, or self-reported use of an antihypertensive drug in the past 2 weeks. Stage 2 and above hypertension was defined as an average SBP of at least 160 mm Hg or an average DBP of at least 100 mm Hg.17 Participants were deemed to be aware of hypertension if they responded “yes” when asked “have you ever been diagnosed with hypertension?” Treatment of hypertension was defined as current use of antihypertensive medication. Hypertension control was defined as an average SBP of less than 140 mm Hg and an average DBP of less than 90 mm Hg over two readings in people with hypertension; patients not meeting these criteria were considered uncontrolled. We also collected information on participants’ socio demographic characteristics, health behaviours, and medical history during the in-person interviews (appendix). We took physical measurements, including height and weight, following standard protocols. Bodymass index (BMI) was defined as weight (kg) divided by height² (m²); obesity was defined as a BMI of at least 28 kg/m², based on the recommendations of the Working Group on Obesity in China.18

Statistical analysis We estimated the prevalence of hypertension among all participants, as well as awareness, treatment, and control of hypertension among hypertensive participants. To syste matically examine how hypertension awareness, treatment, and control varied by subpopulations, we did a compre hensive analysis of population subgroups. These were defined a priori by all possible combinations of 11 selected characteristics: age (35–44, 45–54, 55–64, and 65–75 years), sex (men and women), geographical region (western, central, and eastern China), urbanity (urban vs rural), ethnic origin (Han and non-Han), occupation (farmer and non-farmer), annual household income (<¥10 000 [equiv alent to US$1452], ¥10 000–50 000 [$1452–7259], and ≥¥50 000 [$7259]), education (primary school and below, middle school, high school, and college and above), previous cardiovascular events (yes or no), current smoker (yes or no), and diabetes (yes or no). We retained 264 475 subgroups of more than 500 people, calculated the prevalence of hypertension, and the proportions of those who were aware of their condition, receiving treatment, and had achieved control. We used histograms to show the distributions (appendix), and repeated the analysis by

restricting the subgroups to 130 085 that included at least 3000 participants each. To characterise the pattern of antihypertensive medication use, we assessed the number and classes of medications used by treated patients. Medication classes were the six defined by the 2010 Chinese guidelines for the management of hypertension,16 including angiotensin-convertingenzyme (ACE) inhibitors or angiotensin-receptor blockers (ARBs), β-blockers, calcium-channel blockers, diuretics, α-blockers, and fixed-dose combination drugs. We also collected information on traditional Chinese medicines used for hypertension (appendix). Finally, we developed multivariable mixed models with a logit link function and township-specific random intercepts, taking into account spatial autocorrelation, to identify both individual and primary health-care site characteristics independently associated with hyper tension awareness, treatment, and control. We calculated two sets of models: the first, using 1·7 million participants in the Million Persons Project, assessed individual characteristics associated with hypertension awareness, treatment, and control. The explanatory variables included participants’ age, sex, geographic regions, education, annual household income, health insurance status, smoking, drinking, obesity, physiciandiagnosed diabetes mellitus, and previous cardiovascular events (myocardial infarction or stroke). In a sensitivity analysis, we tested all 146 two-way interaction terms between covariates and used p=0·0003 as the Bonferroni corrected threshold for significance. In the second set of models, we linked the Million Persons Project with another study, the Primary Health Care Survey (done in a subset of Million Persons Project sites) of 1·03 million participants at 113 sites (appendix). In the sensitivity analyses, we calculated agestandardised and sex-standardised rates of hypertension prevalence, awareness, treatment, and control at the national and regional levels, using data from all 31 provinces in the 2010 Chinese census.19 We assigned individuals different weights so that the age and sex distributions matched the census data. All analyses were done with SAS 9.3 and R 3.02.

Role of the funding source The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results Our sample contained 1 738 886 people with a mean age of 55·6 years (SD 9·7), and 1 035 026 (59·5%, 95% CI 59·4–59·6) were women. Compared with the population aged 35–75 years in the 2010 Chinese census (46·6%), our sample contained more people aged 50 years and older (70·1%; table 1).

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

Overall, 44·7% (777 637, 95% CI 44·6–44·8) of the study population had hypertension, including 15·2% (264 822, 15·2–15·3) with stage 2 and above hypertension (table 1). Hypertension prevalence increased with age in both men and women (figure 1). People with higher blood pressure were more likely to have lower annual household income, lower educational status, currently smoke, currently drink, have diabetes, obesity, and previous cardiovascular events than those with lower blood pressure (table 1). Fewer than half (347 755, 44·7% [95% CI 44·6–44·8]) of the people with hypertension and just over half Overall (n=1 738 886) Prevalence

100·0% (100·0–100·0)

(147 887, 55·8% [55·7–56·0]) of those with stage 2 and above hypertension were aware of their diagnosis. Among the people with hypertension, 233 933 (30·1% [30·0–30·2]) were taking prescribed medications to lower blood pressure, and 55 876 (7·2% [7·1–7·2]) had achieved control (figure 1). Among the 15·2% of people with stage 2 and above hypertension, more than 60% were not treated and only a third took allopathic drugs for hypertension. The 264 475 population subgroups varied with respect to hypertension prevalence (range 11·4–88·3%),

No hypertension (n=961 249) 55·3% (55·2–55·4)

All hypertension (n=777 637) 44·7% (44·6–44·8)

Stage 2 and above hypertension (n=264 822) 15·2% (15·2–15·3)

Age (years) 35–39

77 668 (4·5%)

64 096 (6·7%)

13 572 (1·7%)

4391 (1·7%)

40–44

184 594 (10·6%)

140 398 (14·6%)

44 196 (5·7%)

13 625 (5·1%) 26 397 (10·0%)

45–49

256 922 (14·8%)

174 123 (18·1%)

82 799 (10·6%)

50–54

304 060 (17·5%)

177 001 (18·4%)

127 059 (16·3%)

41 005 (15·5%)

55–59

248 157 (14·3%)

127 944 (13·3%)

120 213 (15·5%)

39 336 (14·9%)

60–64

299 540 (17·2%)

136 111 (14·2%)

163 429 (21·0%)

56 108 (21·2%)

65–69

224 432 (12·9%)

89 541 (9·3%)

134 891 (17·3%)

48 699 (18·4%)

70–75

143 513 (8·3%)

52 035 (5·4%)

91 478 (11·8%)

35 261 (13·3%)

Sex Men Women

703 860 (40·5%)

373 961 (38·9%)

329 899 (42·4%)

111 004 (41·9%)

1 035 026 (59·5%)

587 288 (61·1%)

447 738 (57·6%)

153 818 (58·1%)

675 339 (38·8%)

388 598 (40·4%)

286 741 (36·9%)

92 866 (35·1%)

1 063 547 (61·2%)

572 651 (59·6%)

490 896 (63·1%)

171 956 (64·9%)

Urbanity Urban Rural Geographical region of China Eastern

576 110 (33·1%)

293 056 (30·5%)

283 054 (36·4%)

93 725 (35·4%)

Western

675 880 (38·9%)

401 480 (41·8%)

274 400 (35·3%)

98 380 (37·1%)

Central

486 896 (28·0%)

266 713 (27·7%)

220 183 (28·3%)

72 717 (27·5%)

1 529 611 (88·0%)

833 104 (86·7%)

696 507 (89·6%)

233 173 (88·0%)

207 376 (11·9%)

127 268 (13·2%)

80 108 (10·3%)

31 255 (11·8%)

1022 (0·1%)

394 (0·1%)

Ethnic group Han Non-Han Unknown*

1899 (0·1%)

877 (0·1%)

Education Primary school or lower

769 511 (44·3%)

387 976 (40·4%)

381 535 (49·1%)

Middle school

558 880 (32·1%)

322 264 (33·5%)

236 616 (30·4%)

77 708 (29·3%)

High school

258 905 (14·9%)

154 575 (16·1%)

104 330 (13·4%)

33 203 (12·5%)

College or above

125 113 (7·2%)

81 215 (8·4%)

43 898 (5·6%)

14 430 (5·4%)

26 477 (1·5%)

15 219 (1·6%)

11 258 (1·4%)

3946 (1·5%)

<10 000

390 948 (22·5%)

208 204 (21·7%)

182 744 (23·5%)

67 295 (25·4%)

10 000–50 000

958 190 (55·1%)

530 784 (55·2%)

427 406 (55·0%)

143 234 (54·1%)

>50 000

229 483 (13·2%)

131 188 (13·6%)

98 295 (12·6%)

30 146 (11·4%)

Unknown*

160 265 (9·2%)

91 073 (9·5%)

69 192 (8·9%)

24 147 (9·1%)

1 615 561 (92·9%)

902 844 (93·9%)

712 717 (91·7%)

241 472 (91·2%)

100 412 (5·8%)

45 531 (4·7%)

54 881 (7·1%)

19 595 (7·4%)

22 913 (1·3%)

12 874 (1·3%)

10 039 (1·3%)

Unknown*

135 535 (51·2%)

Household income (¥/year)

Marital status Married Widowed, separated, divorced, or single Unknown*

3755 (1·4%) (Table 1 continues on next page)

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

Overall (n=1 738 886)

No hypertension (n=961 249)

All hypertension (n=777 637)

1 701 087 (97·8%)

939 638 (97·8%)

761 449 (97·9%)

Stage 2 and above hypertension (n=264 822)

(Continued from previous page) Health insurance status Insured

259 234 (97·9%)

Uninsured

10 083 (0·6%)

6099 (0·6%)

3984 (0·5%)

1306 (0·5%)

Unknown*

27 716 (1·6%)

15 512 (1·6%)

12 204 (1·6%)

4282 (1·6%)

Medical history Myocardial infarction

12 649 (0·7%)

4560 (0·5%)

8089 (1·0%)

2611 (1·0%)

Stroke

40 555 (2·3%)

10 879 (1·1%)

29 676 (3·8%)

11 568 (4·4%) 24 989 (9·4%)

Cardiovascular disease risk factors Diabetes mellitus

105 379 (6·1%)

35 377 (3·7%)

70 002 (9·0%)

Current smoker

340 219 (19·6%)

185 504 (19·3%)

154 715 (19·9%)

52 340 (19·8%)

Current drinker

418 818 (24·1%)

217 158 (22·6%)

201 660 (25·9%)

69 463 (26·2%)

Obesity (body-mass index ≥28 kg/m²)

272 796 (15·7%)

98 203 (10·2%)

174 593 (22·5%)

66 675 (25·2%)

Data are % (95% CI) or n (%). No hypertension: systolic blood pressure <140 mm Hg, diastolic blood pressure <90 mm Hg, and not taking antihypertensive medication. All hypertension: systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg, or taking antihypertensive medication. Stage 2 and above hypertension: systolic blood pressure ≥160 mm Hg or diastolic blood pressure ≥100 mm Hg. *Participants either refused to answer the question or did not know the answer.

Table 1: Characteristics of the study population by blood pressure levels

Men 100

Women Prevalence

Participants (%)

80 70 60 50 40 30 20 10 0 Men 100

Women Awareness Treatment Control

90 80 Participants (%)

70 60 50 40 30 20 10

–3 9 40 –4 4 45 –4 9 50 –5 4 55 –5 9 60 –6 4 65 –6 9 70 –7 4

awareness (3·8–90·9%), and treatment (2·2–76·2%), but hypertension control was poor (<30%) in all subgroups (appendix). Subgroups with at least 3000 participants had similar results (appendix). Compared with people of younger age, with no previous cardiovascular events, and non-Han ethnicity, older people, with previous cardiovascular events, and Han ethnicity had higher awareness, treatment, and control rates for their hypertension (figure 2). Additionally, regional variations occurred for hypertension prevalence, awareness, treatment, and control; prevalence was highest in the eastern region, but awareness, treatment, and control were highest in the central region, followed by the eastern and western regions (appendix). Compared with urban areas, rural areas had slightly higher hypertension prevalence (rural 46·1% [95% CI 46·1–46·3] vs urban 42·5% [42·3–42·6]) but lower awareness, treatment, and control (urban 46·3% [46·1–46·5], 33·4% [33·2–33·5], and 9·1% [9·0–9·2] vs rural 43·8% [43·7–43·9], 28·2% [28·0–28·3], and 6·1% [6·0–6·2], respectively). In all different age groups, rural areas had higher prevalence, but lower awareness, treatment, and control rates of hypertension compared with urban areas (appendix). In sensitivity analyses, we standardised our national census-based estimates and noted lower prevalence, awareness, treatment, and control: 37·2% (95% CI 37·1–37·3), 36·0% (35·8–36·2), 22·9% (22·7–23·0), and 5·7% (5·6–5·7), respectively. Among patients treated for hypertension, 189 740 (81·1% [95% CI 81·0–81·3]) reported using one medication, 39 832 (17·0% [16·9–17·2) reported two, and 4361 (1·9% [1·8–1·9]) reported three or more (appendix). Among patients treated with one, two,

Age (years)

Figure 1: Prevalence, awareness, treatment, and control of hypertension among study participants Data are shown stratified by age and sex.

and three or more medications, respectively, 44 694 (23·6% [23·4–23·7]), 9979 (25·1% [24·6–25·5]), and 1203 (27·6% [26·3–28·9]) achieved control. Of the treated patients, 178 057 (76·1% [75·9–76·3]) did not have their blood pressure under control. Among

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

7·5 5·0 2·5

7·5 5·0 2·5 0

100

Awareness (%)

10·0 7·5 5·0 2·5

8 Estimated probability density

Estimated probability density

Age (years) 35–44 45–54 55–64 65–75

Control (%)

100

5·0 2·5

50 Control (%)

100

Ethnic group Non-Han Han

3 2 1

100

Treatment (%)

7·5

100

Previous cardiovascular events No Yes

10·0

12·0 Estimated probability density

Estimated probability density

Awareness (%)

100

Age (years) 35–44 45–54 55–64 65–75

Treatment (%)

100

Previous cardiovascular events No Yes

Treatment (%)

Ethnic group Non-Han Han

Awareness (%)

12·0

Estimated probability density

Previous cardiovascular events No Yes

100

25 Estimated probability density

10·0

Estimated probability density

Age (years) 35–44 45–54 55–64 65–75

Estimated probability density

10·0

Ethnic group Non-Han Han

20 15 10 5 0

100

Control (%)

Figure 2: Density plots of awareness, treatment, and control of hypertension in 264 475 subgroups Data are shown stratified by age, previous cardiovascular disease, and ethnic origin. The density means estimated probability of the corresponding awareness, treatment, and control rate. The area under the curve is 1.

patients who were treated but not controlled, 145 046 (81·5% [81·3–81·6]) used one medication, 29 853 (16·8% (16·6–16·9]) used two, and 3158 (1·8% [1·6–1·9]) used three or more. Overall, calcium-channel blockers were the most commonly used medication class (116 806, 55·2% [95% CI 55·0–55·4]), followed by ACE inhibitors or ARBs (60 361, 28·5% [28·3–28·7]) and diuretics (19 851, 9·4% [9·3–9·5]; appendix), and 19 180 (8·2% [8·1–8·3]) used traditional Chinese medicines. Among patients using only one medication, nifedipine was the most common agent, followed by amlodipine and indapamide (table 2). Calcium-channel blockers plus ACE inhibitors or ARBs was the most common combination therapy (13 812, 6·5% [6·4–6·6]), followed by ACE inhibitors or ARBs plus a diuretic (5745, 2·7% [2·6–2·8]) and calciumchannel blockers plus a β-blocker (4582, 2·2% [2·1–2·3]). Among patients using two separate medications, 6

nifedipine plus metoprolol was the most frequent combination with 897 (6·7% [6·3–7·1]) using this in urban areas and 787 (4·8% [4·5–5·1]) in rural areas. Multivariable mixed models identified several characteristics associated with hypertension prevalence, awareness, treatment, and control (table 3). People who were older, men, those with lower education, nonfarmers, current smokers, with previous cardiovascular events, and coexisting conditions (diabetes or obesity) were more likely to have a higher risk of hypertension. Among people with hypertension, those who were older, women, those with higher education, higher annual household income, with prior cardiovascular events, and coexisting conditions (diabetes or obesity) were more likely to be aware of, treated for, and controlled for hypertension. Those without health insurance, farmers, and current drinkers were less aware of, treated for, or controlled for hypertension. Sex, age, previous

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

Overall

35–44 years

45–54 years

55–64 years

65–75 years

Among adults using one medication Nifedipine

49 034 (26·2%)

1398 (22·1%)

9926 (24·2%)

19 516 (26·5%)

18 194 (27·5%)

Amlodipine

14 390 (7·7%)

476 (7·5%)

3240 (7·9%)

5594 (7·6%)

5080 (7·7%)

Indapamide

11 598 (6·2%)

349 (5·5%)

2712 (6·6%)

4664 (6·3%)

3873 (5·9%)

9944 (5·3%)

227 (3·6%)

2036 (5·0%)

3922 (5·3%)

3759 (5·7%)

Compound reserpine* Nitrendipine Total

9107 (4·9%)

193 (3·0%)

1565 (3·8%)

3490 (4·7%)

3859 (5·8%)

187 320 (100%)

6335 (100%)

40 995 (100%)

73 749 (100%)

66 241 (100%)

Among adults using two medications Nifedipine and metoprolol

1684 (5·7%)

48 (4·2%)

369 (5·3%)

621 (5·5%)

646 (6·2%)

Captopril and nifedipine

1441 (4·8%)

52 (4·5%)

279 (4·0%)

582 (5·2%)

528 (5·0%)

Nifedipine and irbesartan

911 (3·1%)

42 (3·7%)

217 (3·1%)

342 (3·0%)

310 (3·0%)

Amlodipine besylate and irbesartan

893 (3·0%)

32 (2·8%)

235 (3·4%)

330 (2·9%)

296 (2·8%)

Nifedipine and telmisartan Total

711 (2·4%) 29 792 (100%)

38 (3·3%)

183 (2·6%)

265 (2·4%)

1143 (100%)

6913 (100%)

11 263 (100%)

225 (2·1%) 10 473 (100%)

Among adults using three medications Nifedipine, metoprolol, and irbesartan

61 (2·5%)

3 (3·0%)

6 (1·1%)

23 (2·6%)

29 (3·1%)

Amlodipine, metoprolol, and irbesartan

54 (2·2%)

4 (4·0%)

9 (1·6%)

18 (2·1%)

23 (2·5%)

Nifedipine, telmisartan, and metoprolol

41 (1·7%)

13 (2·4%)

12 (1·4%)

16 (1·7%)

549 (100%)

876 (100%)

929 (100%)

Total

2453 (100%)

99 (100%)

Data are n (%) stratified by age and number of medications. *A fixed-dose combination drug consisting of reserpine (0·032 mg), hydrochlorothiazide (3·1 mg), potassium chloride (30 mg), dihydralazine sulfate (2·1 mg), and promethazine (2·1 mg).

Table 2: The most commonly used medications among treated adults with hypertension

cardiovascular events, diabetes, and obesity were stronger predictors for hypertension awareness than for treatment and control, whereas education, annual household income, farming occupation, smoking, and drinking were stronger predictors for control than for treatment and awareness. In a sensitivity analysis, 21 interaction terms were significant for hypertension awareness, 24 significant for hypertension treatment, and 20 significant for hypertension control (appendix). Among these, for more than 90% of the interaction terms (20, 23, and 18 for awareness, treatment, and control, respectively), their odds ratios (ORs) fell between 0·75 and 1·25. Except for geographical regions, the only primary health-care site characteristic associated with higher hypertension awareness, treatment, and control rate was having more primary health-care physicians with at least a bachelor’s degree, but the absolute association was not strong (OR 1·00, 95% CI 1·00–1·01 for awareness; OR 1·01, 95% CI 1·00–1·01 for treatment; OR 1·01, 95% CI 1·00–1·02 for control). Compared with the western region, the central and eastern regions were associated with a higher proportion of people who had hypertension awareness, treatment, and control rates (appendix).

Discussion In this study, hypertension was highly prevalent in China but remained undertreated and uncontrolled. Although 45% of middle-aged adults had hypertension (15% with

stage 2 and above hypertension), most were untreated; even among people with stage 2 and above hypertension, only a third were treated. People who were male, younger, lower income, without prior cardiovascular events, and without coexisting conditions were less likely to be aware of, treated for, and controlled for hypertension. However, variation in awareness and treatment did not reflect a similar pattern in control. We showed remarkably low control rates (<30%) that were consistent across more than 200 000 subgroups, highlighting China’s need for a national hypertension strategy. We have added to the scientific literature in several ways with this study, although our findings of poor management of hypertension are consistent with previous studies in China.3,4,6–9 First, the size of our study, the largest on hypertension in China, allowed us to study a wide variety of subgroups with a substantial number of people. We analysed 264 475 subgroups defined by all possible combinations of 11 selected characteristics to systematically examine how hypertension awareness, treatment, and control rates vary within the Chinese population. Our results reveal that hypertension control in community-dwelling residents in China is poor overall and across diverse population subgroups. The most direct implication is that China needs a universal, rather than a targeted, approach to hypertension and that the impediments to control must be illuminated. Second, we provide new information about individual and primary health-care site characteristics associated

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

Prevalence

Awareness

Treatment

Control

1·35 (1·35–1·35)

1·20 (1·20–1·20)

1·19 (1·19–1·20)

1·07 (1·07–1·08)

Men

1 (ref)

Women

0·97 (0·96–0·98)

1·18 (1·16–1·19)

1·16 (1·14–1·18)

1·11 (1·09–1·14)

Non–Han

1 (ref)

Han

1·00 (0·98–1·01)

1·00 (0·97–1·02)

1·03 (1·01–1·06)

1·06 (1·01–1·12)

Age (per 5 years) Sex

Ethnic origin

Marital status Not married

1 (ref)

Married

0·93 (0·92–0·94)

0·97 (0·95–0·99)

1·01 (0·99–1·03)

1·08 (1·05–1·12)

Annual household income (¥) <10 000

1 (ref)

10 000–50 000

1·00 (0·98–1·01)

1·02 (1·01–1·04)

1·04 (1·02–1·06)

1·09 (1·06–1·12)

>50 000

1·00 (0·99–1·02)

1·09 (1·07–1·12)

1·10 (1·08–1·13)

1·22 (1·18–1·27)

Lower than college

1 (ref)

College or above

0·93 (0·91–0·96)

1·16 (1·12–1·21)

1·14 (1·10–1·19)

1·22 (1·15–1·30)

Education level

Occupation Not a farmer

1 (ref)

Farmer

0·96 (0·95–0·97)

0·90 (0·89–0·92)

0·83 (0·82–0·85)

0·80 (0·77–0·82)

Health insurance status Insured

1 (ref)

Uninsured

0·99 (0·93–1·05)

0·78 (0·70–0·85)

0·76 (0·69–0·85)

0·75 (0·62–0·91)

Cardiovascular disease risk factors Current smoker

1·03 (1·02–1·04)

1·03 (1·02–1·05)

0·99 (0·98–1·01)

1·13 (1·10–1·16)

Current drinker

1·50 (1·48–1·52)

0·94 (0·93–0·96)

0·87 (0·85–0·88)

0·74 (0·72–0·77)

Diabetes mellitus

2·59 (2·57–2·62)

2·25 (2·21–2·29)

1·79 (1·76–1·83)

1·37 (1·34–1·41)

Obesity (body–mass index ≥28 kg/m²)

1·80 (1·77–1·82)

1·66 (1·64–1·68)

1·55 (1·53–1·57)

1·02 (0·99–1·04)

Previous cardiovascular diseases

2·20 (2·15–2·24)

3·20 (3·12–3·29)

2·60 (2·54–2·67)

2·07 (2·00–2·13)

Geographical region of China Western

1 (ref)

Central

1·52 (1·44–1·60)

1·02 (0·93–1·11)

0·87 (0·77–0·98)

0·93 (0·80–1·08)

Eastern

1·34 (1·26–1·42)

1·10 (0·91–1·19)

1·08 (0·94–1·23)

1·06 (0·90–1·25)

Data are odds ratios (95% CI).

Table 3: Individual characteristics associated with prevalence, awareness, treatment, and control of hypertension

with hypert ension awareness, treatment, and control, and formally test interactions between covariates. Lower awareness, treatment, and control rates are associated with younger men and people with less education, lower income, and no cardiovascular risk factors—people who might have little contact with the health-care system. However, our analysis revealed that no group had substantially better control, even though awareness and treatment did vary. We have also assessed primary healthcare site characteristics by linking the current study with the Primary Health Care Survey, but we could not identify strong explanations for the low control. Ultimately, differences in awareness and treatment are not 8

translating into control, and thus improving both are necessary, but not sufficient, to achieve better control. There is no evidence of any population subgroup excelling in hypertension control and no evidence that a particular biological or societal factor is associated with achieving high control. This suggests a fundamental structural change is needed to address hypertension in China, and education and screening will not be sufficient without parallel efforts to improve treatments. The issues might include the need for screening and education, supplemented by protocols and policies that ensure provision of adequate treatment and access to affordable medications. Third, our assessment of antihypertensive medications highlights substantial opportunities to intensify treat ment among the few patients whose hypertension was treated but not controlled. Taking advantage of the medication data, our study showed that only a quarter of treated patients had their hypertension under control and most used only one antihypertensive drug. Because hypertension control often requires multiple medi cations, our finding that use of multiple medications is rare in China might help to partly explain the few people with optimal control. Compared with national data from high-income countries such as the USA,20–22 our data showed a fourtimes higher prevalence of stage 2 and above hypertension but substantially fewer people with hypertension awareness, treatment, or control in China. Using data from the 2013–14 US National Health and Nutrition Examination Survey, we estimated that 87% of adults aged 40–75 years with hypertension were diagnosed, 73% were treated, and 61% achieved blood pressure goals. By contrast, the corresponding percentages are only 44·7%, 30·1%, and 7·2% in China. Moreover, the classes of medications used differed greatly between the two countries. ACE inhibitors or ARBs, β-blockers, and diuretics are commonly used in the USA but less so in China despite guideline recommendations.15,16 China’s lower use of diuretics might reflect that new guidelines recommend other medication classes (calcium-channel blockers, ACE inhibitors, and ARBs) as initial therapy and that insurance coverage for these drugs has been expanded.15,16,23 Moreover, China is commonly using a diuretic (indapamide) that is different from those used in the USA and Europe. Additionally, combination therapy is used much more frequently in the USA than in China.21 Many factors contribute to these differences, including primary care physicians’ prescription habits, physicians’ knowledge of and willingness to adhere to new hypertension treatment guidelines, and affordability of newer drugs. Finally, traditional Chinese medicines are still widely used, but these might not be effective for treating hypertension. Our study has several policy implications. Unlike many other countries, stroke is the leading cause of death in China, causing 20% of deaths annually.24

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

Because high blood pressure is a major attributable risk factor of stroke,5 uncontrolled hypertension could largely contribute to the high stroke mortality in China. This fact is even more relevant because blood pressure levels in China are increasing, a trend presumably related to increases in population ageing, urbanisation, dietary changes, and obesity.25 Therefore, an important step to mitigate the burden of cardiovascular diseases in China is to continue previous efforts to reduce blood pressure at the population level. A few key actions might be effective, as highlighted in the 2016 Lancet Commission report on hypertension.26 These include promoting free blood pressure screening and public awareness programmes, strengthening health systems, integrating hypertension management into routine primary care practice, promoting structured physicianeducation programmes to reduce clinical inertia and improve guideline adherence,27–29 removing financial barriers to health care, and improving access to affordable drugs.30 Additionally, given the high prevalence of hypertension, China must respond with high-quality individual clinical care as the basis for public health interventions, including diet and physical activity, to prevent the development of hypertension in unaffected individuals. Our study had several potential limitations. First, our study did not use representative sampling because it was not possible with such rapid, large-scale recruitment. We corrected for the characteristics of the underlying population at the national and regional levels using the 2010 Chinese census data and found that the corrected estimates were lower for awareness, treatment, and control. Moreover, we would expect that our sampling biases would result in overestimates of awareness, treatment, and control as study participants by virtue of responding to this screening project would be more likely connected with the health system. Second, we did not obtain information about patients’ adherence to medication, access to regular physicians, or physicians’ knowledge of antihypertensive medication use, which reduced our ability to investigate some potential reasons for suboptimal treatment. Third, we did not capture or classify non-pharmacological treatment strategies, such as dietary modification, which might yield an under estimation of hypertension treatment but would not affect our assessments of awareness and control. Fourth, people who used an antihypertensive medication might not have reported that use, possibly yielding an under estimation of the hypertension treatment rate. Finally, the blood pressure levels used to define hypertension control in this paper are largely consistent with European, Canadian, and British guidelines1–3 but might underestimate control if higher goal blood-pressure levels are used as in the eighth Joint National Committee.4 We conclude that hypertension is a major public health challenge in China. Despite its high prevalence, hypertension control in community-dwelling residents

is poor, both overall and in diverse population subgroups. Our findings support broad-based opportunities to mitigate the burden of hypertension and suggest the need for a national strategy on hypertension prevention and control. Contributors LJ and HMK conceived the China PEACE Million Persons Project and take responsibility for its all aspects. JLu led the data collection. HMK, LJ, and JLu designed the study. HMK, LJ, YL, and JLu conceived this article. HMK, LJ, YL, and JLu wrote the manuscript, with further contributions from HZhang, JLiu, MS, ESS, JAS, and FAM. XW and XL completed all the statistical analysis supported by CW, XC, LM, GCL, and HZhao. All authors interpreted data, contributed to critical revisions, and approved the final version of the Article. Declaration of interests HMK is a recipient of research agreements from Medtronic, and Johnson & Johnson (Janssen) through Yale, to develop methods of clinical trial data sharing; received a grant from the US Food and Drug Administration and Medtronic, through Yale, to develop methods for post-market surveillance of medical devices; works under contract with the US Centers for Medicare & Medicaid Services to develop and maintain performance measures; chairs a cardiac scientific advisory board for United Health; is a participant or participant representative of the IBM Watson Health Life Sciences Board; is a member of the Advisory Board for Element Science and the Physician Advisory Board for Aetna; and is the founder of Hugo, a personal health information platform. All other authors declare no competing interests. Acknowledgments This project was supported by the CAMS Innovation Fund for Medical Science (2017-I2M-2-002, 2016-I2M-1-006, 2016-I2M-2-004); the Ministry of Finance of China and National Health and Family Planning Commission of China; the China-WHO Biennial Collaborative Projects 2016-2017 (2016/664424-0); the National Key Technology R&D Program (2015BAI12B01, 2015BAI12B02); Research Special Fund for Public Welfare Industry of Health (201502009); the 111 Project from the Ministry of Education of China (B16005); and the PUMC Youth Fund and the Fundamental Research Funds for the Central Universities (2017330003). We thank all study individuals for their participation, and appreciate the contributions made by project teams at the National Center for Cardiovascular Diseases and the YaleNew Haven Hospital Center for Outcomes Research and Evaluation in the realms of project design and operations, particularly data collection by Yaping Cao, Li Li, Yang Yang, Jianlan Cui, Xin Sun, Wei Xu, Bo Gu, Xi Li, Hao Dai, Hui Zhong, and Minghui Zhang; and the Ministry of Finance of China and National Health and Family Planning Commission of China for funding support. We thank all provincial and regional officers and research staff in all 31 provinces in China for their collection of data and biosamples, and appreciate the advice from Yun Wang (Harvard TH Chan School of Public Health, MA, USA) and Steven DeMaio (Yale-New Haven Hospital Center for Outcomes Research and Evaluation, MA, USA). Members of the provincial coordinating office of the China PEACE Million Persons Project are in the appendix. References 1 Chen Z. Launch of the health-care reform plan in China. Lancet 2009; 373: 1322–24. 2 Yang ZJ, Liu J, Ge JP, et al. Prevalence of cardiovascular disease risk factor in the Chinese population: the 2007–2008 China National Diabetes and Metabolic Disorders Study. Eur Heart J 2012; 33: 213–20. 3 Li D, Lv J, Liu F, et al. Hypertension burden and control in mainland China: analysis of nationwide data 2003–2012. Int J Cardiol 2015; 184: 637–44. 4 Feng XL, Pang M, Beard J. Health system strengthening and hypertension awareness, treatment and control: data from the China Health and Retirement Longitudinal Study. Bull World Health Organ 2014; 92: 29–41. 5 He J, Gu D, Chen J, et al. Premature deaths attributable to blood pressure in China: a prospective cohort study. Lancet 2009; 374: 1765–72.

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

6 7

8 9 10 11

12 13

14 15

16 17

Yang G, Wang Y, Zeng Y, et al. Rapid health transition in China, 1990–2010: findings from the Global Burden of Disease Study 2010. Lancet 2013; 381: 1987–2015. Wang J, Zhang L, Wang F, Liu L, Wang H, China National Survey of Chronic Kidney Disease Working Group. Prevalence, awareness, treatment, and control of hypertension in China: results from a national survey. Am J Hypertens 2014; 27: 1355–61. Li W, Gu H, Teo KK, et al. Hypertension prevalence, awareness, treatment, and control in 115 rural and urban communities involving 47 000 people from China. J Hypertens 2016; 34: 39–46. Lewington S, Lacey B, Clarke R, et al. The burden of hypertension and associated risk for cardiovascular mortality in China. JAMA Intern Med 2016; 176: 524–32. Wang Z, Wang X, Chen Z, et al. Hypertension control in community health centers across China: analysis of antihypertensive drug treatment patterns. Am J Hypertens 2014; 27: 252–59. Yang L, Yan J, Tang X, Xu X, Yu W, Wu H. Prevalence, awareness, treatment, control and risk factors associated with hypertension among adults in southern China, 2013. PLoS One 2016; 11: e0146181. Gao Y, Chen G, Tian H, et al. Prevalence of hypertension in china: a cross-sectional study. PLoS One 2013; 8: e65938. Meng XJ, Dong GH, Wang D, et al. Prevalence, awareness, treatment, control, and risk factors associated with hypertension in urban adults from 33 communities of China: the CHPSNE study. Journal Hypertens 2011; 29: 1303–10. Lu J, Xuan S, Downing NS, et al. Protocol for the China PEACE (Patient-centered Evaluative Assessment of Cardiac Events) Million Persons Project pilot. BMJ Open 2016; 6: e010200. James PA, Oparil S, Carter BL, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014; 311: 507–20. Liu LS, Writing Group of Chinese Guidelines for the Management of H. 2010 Chinese guidelines for the management of hypertension. Chin J Cardiol 2011; 39: 579–615. Chobanian AV, Bakris GL, Black HR, et al. The seventh report of the joint national committee on prevention, detection, evaluation, and treatment of high blood pressure: the JNC 7 report. JAMA 2003; 289: 2560–72. Zhou BF. Predictive values of body mass index and waist circumference for risk factors of certain related diseases in Chinese adults—study on optimal cut-off points of body mass index and waist circumference in Chinese adults. BES 2002; 15: 83–96.

19 National Bureau of Statistics of China. 2010 Population Census of People’s Republic of China. http://www.stats.gov.cn/english/ Statisticaldata/CensusData/rkpc2010/indexch.htm (accessed April 17, 2017). 20 Egan BM, Zhao Y, Axon RN. US trends in prevalence, awareness, treatment, and control of hypertension, 1988–2008. JAMA 2010; 303: 2043–50. 21 Gu Q, Burt VL, Dillon CF, Yoon S. Trends in antihypertensive medication use and blood pressure control among United States adults with hypertension: the National Health And Nutrition Examination Survey, 2001 to 2010. Circulation 2012; 126: 2105–14. 22 Yoon SS, Gu Q, Nwankwo T, Wright JD, Hong Y, Burt V. Trends in blood pressure among adults with hypertension: United States, 2003 to 2012. Hypertension 2015; 65: 54–61. 23 Taylor J. 2013 ESH/ESC guidelines for the management of arterial hypertension. Eur Heart J 2013; 34: 2108–09. 24 GBD Mortality Causes of Death Collaborators. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2015; 385: 117–71. 25 Danaei G, Finucane MM, Lu Y, et al. National, regional, and global trends in fasting plasma glucose and diabetes prevalence since 1980: systematic analysis of health examination surveys and epidemiological studies with 370 country-years and 2·7 million participants. Lancet 2011; 378: 31–40. 26 Olsen MH, Angell SY, Asma S, et al. A call to action and a lifecourse strategy to address the global burden of raised blood pressure on current and future generations: the Lancet Commission on hypertension. Lancet 2016; 388: 2665–712. 27 Luders S, Schrader J, Schmieder RE, Smolka W, Wegscheider K, Bestehorn K. Improvement of hypertension management by structured physician education and feedback system: cluster randomized trial. Eur J Cardiovasc Prev Rehabil 2010; 17: 271–79. 28 Onysko J, Maxwell C, Eliasziw M, et al. Large increases in hypertension diagnosis and treatment in Canada after a healthcare professional education program. Hypertension 2006; 48: 853–60. 29 Johnson W, Shaya FT, Khanna N, et al. The Baltimore Partnership to Educate and Achieve Control of Hypertension (The BPTEACH Trial): a randomized trial of the effect of education on improving blood pressure control in a largely African American population. J Clin Hypertens 2011; 13: 563–70. 30 Su M, Zhang Q, Bai X, et al. Availability, cost, and prescription patterns of antihypertensive medications in primary health care in China: a nationwide cross-sectional survey. Lancet 2017; published online Oct 25. http://dx.doi.org/10.1016/S0140-6736(17)32476-5.

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32478-9

Articles

Availability, cost, and prescription patterns of antihypertensive medications in primary health care in China: a nationwide cross-sectional survey Meng Su*, Qiuli Zhang*, Xueke Bai, Chaoqun Wu, Yetong Li, Elias Mossialos, George A Mensah, Frederick A Masoudi, Jiapeng Lu, Xi Li, Sebastian Salas-Vega, Anwen Zhang, Yuan Lu, Khurram Nasir, Harlan M Krumholz†, Lixin Jiang†

Summary

Background Around 200 million adults in China have hypertension, but few are treated or achieve adequate control of their blood pressure. Available and affordable medications are important for successfully controlling hypertension, but little is known about current patterns of access to, and use of, antihypertensive medications in Chinese primary health care. Methods We used data from a nationwide cross-sectional survey (the China Patient-Centered Evaluative Assessment of Cardiac Events Million Persons Project primary health care survey), which was undertaken between November, 2016 and May, 2017, to assess the availability, cost, and prescription patterns of 62 antihypertensive medications at primary health-care sites across 31 Chinese provinces. We surveyed 203 community health centres, 401 community health stations, 284 township health centres, and 2474 village clinics to assess variation in availability, cost, and prescription by economic region and type of site. We also assessed the use of high-value medications, defined as guideline-recommended and low-cost. We also examined the association of medication cost with availability and prescription patterns. Findings Our study sample included 3362 primary health-care sites and around 1 million people (613 638 people at 2758 rural sites and 478 393 people at 604 urban sites). Of the 3362 sites, 8·1% (95% CI 7·2–9·1) stocked no antihypertensive medications and 33·8% (32·2–35·4) stocked all four classes that were routinely used. Village clinics and sites in the western region of China had the lowest availability. Only 32·7% (32·2–33·3) of all sites stocked high-value medications, and few high-value medications were prescribed (11·2% [10·9–11·6] of all prescription records). High-cost medications were more likely to be prescribed than low-cost alternatives. Interpretation China has marked deficiencies in the availability, cost, and prescription of antihypertensive medications. High-value medications are not preferentially used. Future efforts to reduce the burden of hypertension, particularly through the work of primary health-care providers, will need to improve access to, and use of, antihypertensive medications, paying particular attention to those with high value. Funding CAMS Innovation Fund for Medical Science, the Entrusted Project from the China National Development and Reform Commission, and the Major Public Health Service Project from the Ministry of Finance of China and National Health and Family Planning Commission of China.

Introduction An estimated 200 million adults in China have hypertension but fewer than 15% are treated.1–4 Moreover, among those who are treated, about two-thirds do not achieve adequate control of their blood pressure.5–8 Inadequate management of patients with hypertension might have substantial health and economical consequences; the sequelae of hypertension, including stroke and heart disease, are the leading causes of morbidity and mortality in China,9,10 and are associated with substantial expense to patients and the health system. The successful management of hypertension in China requires, in addition to lifestyle and behavioural modifications, affordable and widely available anti hypertensive medications prescribed appropriately in primary health-care settings, a main point of contact with the health system in China. Findings from studies have suggested low availability and high costs as major barriers

to optimum use and adherence to essential anti hypertensive medications, especially in low-income rural areas in China.3,11 The Chinese health reform in 2009 strengthened the role of primary health care that serves as gatekeeper to the health-care system.12 The reform also introduced the National Essential Medicine Program that was designed to provide affordable and equitable basic health care for all by 2020. Pharmaceutical policy has also evolved quickly, such as allowances for primary health-care sites to procure non-essential medicines in 2014 and abolishment of the government price ceiling in 2015 (appendix). Although availability of medications increased after the National Essential Medicine Policy was launched,13–15 little is known about current patterns of access to antihypertensive medications across Chinese primary health-care settings, where higher financial burdens and few medication choices might

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Published Online October 25, 2017 http://dx.doi.org/10.1016/ S0140-6736(17)32476-5 See Online/Comment http://dx.doi.org/10.1016/ S0140-6736(17)32743-5 See Online/Articles http://dx.doi.org/10.1016/ S0140-6736(17)32478-9 *Joint first authors, listed alphabetically †Joint senior authors National Clinical Research Center of Cardiovascular Diseases, State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China (M Su PhD, Q Zhang PhD, X Bai MS, C Wu MS, Y Li BS, J Lu PhD, X Li PhD, Prof L Jiang MD); Usher Institute of Population Health Sciences & Informatics, University of Edinburgh, Edinburgh, UK (Q Zhang); Department of Health Policy, London School of Economics and Political Science, London, UK (Prof E Mossialos PhD, S Salas-Vega MSc, A Zhang PhD); Center for Translation Research and Implementation Science, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD, USA (Prof G A Mensah MD); Division of Cardiology, Department of Medicine, University of Colorado Anschutz Medical Campus, Aurora, CO, USA (F A Masoudi MD); Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, Yale University, New Haven, CT, USA (Y Lu ScD, Prof H M Krumholz MD); and Center for Healthcare Advancement & Outcomes/ Miami Cardiac & Vascular Institute, Baptist Health South Florida, Miami, FL, USA (K Nasir MD)

Articles

Research in context Evidence before this study To assess treatment and control of hypertension in China, we searched PubMed for English-language published articles and the China National Knowledge Infrastructure for Chinese-language articles, published before May, 2016, using the terms “primary health care”, “pharmaceutical policy”, “essential medication”, “essential medicine”, “hypertensive medication”, OR “hypertension”. We also reviewed references from the identified articles and other relevant articles and reports. Awareness, treatment, and control of hypertension were found to be low. The Chinese National Essential Medicine Program, a zero mark-up policy for essential medications, was implemented by all primary health-care sites in 2009 to meet residents’ medication needs and reduce their out-of-pocket costs. However, we noted conflicting results for availability and cost, and few studies reported prescription patterns of antihypertensive medications. Previous studies were limited to specific regions, populations, and data sources. Added value of this study We used data from a national primary health-care study and a screening study of high risks for cardiovascular disease in China

result in lower treatment and control rates. This information is essential for developing targets for interventions that are designed to improve national hypertension treatment and control. Accordingly, to address the need for information about the availability, cost, and prescription of antihypertensive medications in primary health-care settings across China, we analysed data from a national, government-funded study of the primary health-care system and a large national cardiovascular screening project.16 Our study focused on all four types of primary health-care sites in China—community health centres and community health stations in urban areas, and township health centres and village clinics in rural areas—and also examined how the costs of antihypertensive medications were associated with the availability and prescription of antihypertensive medications, with a particular focus on lower-cost, guideline-recommended treatments.

Methods

Data source and study sample We derived data on the availability, cost, and prescription of antihypertensive medications from the nationwide China Patient-Centered Evaluative Assessment of Cardiac Events (PEACE) Million Persons Project (MPP) primary health-care survey, undertaken from November, 2016 through May, 2017. The design of this survey has been described previously.17 Briefly, we established a nationwide epidemiological collaborative network of the China PEACE MPP, which consisted of 2

to depict the current status of availability, cost, and prescription patterns of antihypertensive medications in primary health-care settings in China. We collected the data directly from primary health-care sites, rather than analysing secondary data from other reports. We noted marked deficiencies in availability, cost, and prescription of hypertensive medications, and a few health-care pharmacies did not stock any antihypertensive medications. Village clinics and sites in the western region of China had the lowest availability. High-cost medications were more likely to be prescribed than low-cost medications, and high-value medications, those that are guideline-recommended and low-cost, were not preferentially prescribed. Implications of all the available evidence Our findings suggest that interventions to improve hypertension treatment and control in China will need to ensure that low-cost antihypertensive medications are adequately available in primary health-care settings. Implementation of the essential medicines policy at the local level is currently inadequate. Use of high-value medications might help to reduce patients’ cost burden of hypertension treatment.

141 county or district-level regions from all 31 provinces in mainland China.16 The MPP enrolled the eligible study sites according to the number of residents of the catchment area, population stability, local economical conditions, and geographical location. The PHC services are provided by community health centres and community health stations (one level below) in urban areas, and township health centres and village clinics (one level below) in rural areas (appendix). We surveyed 203 community health centres, 401 community health stations, 284 township health centres, and 2474 village clinics to quantify the care-delivery capacity and the quality of primary health care. The distribution of primary health care study sites sampled across rural and urban areas reflects the national ratio.18 Data on hypertension treatment and control were derived from the cardiovascular risk screening programme of the China PEACE MPP, which up to June 20, 2017, had enrolled 1·7 million residents, aged 35 to 75 years, who had lived in 141 selected county-level or district-level regions for at least 6 months in the past year. The interview data from China PEACE MPP included information about the history of hypertension diagnosis and treatment, collected by face-to-face administration of a questionnaire. In total, 1·09 million people with information in the MPP lived in the 435 townships or communities that are the focus of this study. We linked the characteristics of primary healthcare sites with local population epidemiological data at the township or community level (ie, one level below county or district), China’s 2015 National Census data,

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

and geographical location information from a Chinese web mapping service (AMAP).19

Data collection and definitions We obtained, copied, and reviewed lists of medications in stock at the time of the survey from Nov 1, to Dec 31, 2016, from each participating primary health-care site. For each medication on the list, we obtained information about the generic name, brand name, dosage form, specification, manufacturer, and retail price per sale unit. We cleaned and checked the reported generic names against China Pharmacopoeia 2015,20 achieving 95% accuracy. For outpatient prescriptions (from April 1, 2015, to March 31, 2016), we collected information about each patient’s age, sex, and diagnosis, which is routinely listed on the prescriptions, as well as the medication name, dosage, and administration. In primary healthcare sites with an electronic prescription system, relevant information was directly imported into a digital database. For sites without an electronic prescription system, we included the first 100 outpatient prescriptions in each 10-day period for the 12 months in our analysis (3600 total prescriptions during the study period). We scanned copies and trained staff to manually abstract relevant information, with double-entry methods included as a quality check to ensure an accuracy standard of at least 98%. In MPP, blood pressure was measured twice during the interview, using a unified electronic blood pressure monitor (Omron HEM-7430; Omron Corporation, Kyoto, Japan) with standardised procedures.21 Treatment was defined as the proportion of patients in the China PEACE MPP population with hypertension who selfreported use of antihypertensive medications at the baseline interview. Control was defined as the proportion of these patients with a blood pressure of less than 140/90 mm Hg at the baseline interview. Treatment and control of hypertension were assessed by aggregating participant data at the township or community level. We obtained information on the characteristics of each primary health-care site, including workforce and use of information technology systems, through a survey of site leaders and health-care professionals. We focused on 62 oral medications by generic names listed in the database of guidelines for hypertension management, regardless of whether they were recommended,2,22,23 and the essential medicine lists.24 We categorised each of these medicines as angiotensin-converting-enzyme inhibitors (ACE inhibitors), angiotensin-receptor blockers, betablockers, calcium-channel blockers, diuretics, fixed-dose combination medications, traditional compound drugs, or centrally active drugs; each category was mutually exclusive (ie, a medicine in one category could not be categorised in another). We did not include alpha-blockers or alpha-beta blockers because of their scarcity in the primary care sector.

We ascertained the availability of each antihypertensive medication at any dose in the site pharmacy, calculated as the proportion of all participating sites with a specific antihypertensive medication or medication Rural

Urban

Total

Primary health-care site Number Township health centre Village clinic

··

3362

284 (8·4%)

··

2474 (73·6%)

··

Community health centre

··

203 (6·0%)

··

Community health station

··

401 (11·9%)

··

Eastern region

··

31·9% (30·3–33·5)

Township health centre

28·5% (23·3–34·2)

··

Village clinic

27·8% (26·0–29·6)

··

Community health centre

··

38·9% (32·2–46·0)

··

Community health station

··

56·4% (51·3–61·3)

··

Central region

··

35·6% (34·0–37·2)

Township health centre

31·0% (25·7–36·7)

··

Village clinic

39·9% (37·9–41·8)

··

Community health centre

··

29·1% (22·9–35·8)

··

Community health station

··

15·7% (12·3–19·6)

··

Western region

··

32·5% (30·9–34·1)

Township health centre

40·5% (34·7–46·5)

··

Village clinic

32·4% (30·5–34·3)

··

Community health centre

··

32·0% (25·7–38·9)

··

Community health station

··

27·9% (23·6–32·6)

··

Linked MPP population* Number

613 638

478 393

1 092 031

Age (years)

56·2 (9·8)

55·8 (9·8)

56·0 (9·8)

Women

59·6% (59·4–59·7)

60·9% (60·8–61·0)

60·1% (60·1–60·2)

Men

40·4% (40·3–40·6)

39·1% (39·0–39·2)

39·9% (39·8–39·9)

Ethnic Han

88·8% (88·7–88·9)

93·3% (93·2–93·3)

90·8% (90·7–90·8)

9·4% (9·3–9·5)

32·3% (32·2–32·5)

19·5% (19·4–19·5)

21·7% (21·6–21·8)

18·4% (18·3–18·5)

20·2% (20·2–20·3)

12 years of education or more Current smoker No insurance

0·2% (0·2–0·2)

Last-year income >¥50 000

8·6% (8·5–8·6)

18·3% (18·2–18·4)

12·8% (12·7–12·9)

Patients with hypertension

46·4% (46·2–46·5)

44·1% (43·9–44·2)

45·4% (45·3–45·5)

Awareness†

43·8% (43·6–44·0)

48·7% (48·5–48·9)

45·9% (45·8–46·0)

Treatment†

27·1% (27·0–27·3)

34·5% (34·3–34·7)

30·3% (30·1–30·4)

5·5% (5·4–5·6)

9·8% (9·7–9·9)

7·3% (7·3–7·4)

Control†

1·1% (1·1–1·1)

0·6% (0·6–0·6)

Data are n (%), % (95% CI), or mean (SD). MPP=Million Persons Project. ··=not applicable. *The population is linked at the township or community level. †Among all patients with hypertension. Awareness was defined as self-reported diagnosis of hypertension at the baseline interview.

Table 1: Characteristics of primary health-care sites and Million Persons Project population included by rural and urban area

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

class in stock. We calculated the annual median cost for each medication, using its median price across different primary health-care sites and the Essential medicines list*

Guidelines

WHO

NEM

PSEM

Guideline in primary health care 2014

JNC 8 2014

Chinese guideline 2010

ACE inhibitors

··

Captopril

··

Enalapril

Benazepril

··

Ramipril

guideline-recommended dose. We defined high-value medications as those that were both recommended by the Chinese Guideline for Hypertension Management

Availability, % (95% CI), (n=3362)

Prescription frequency, % (95% CI), (n=29 171)

··

44·0% (42·3–45·7)

··

38·9% (37·3–40·6)

··

225 (160–435)

··

8·4% (7·5–9·4)

··

1066 (748–1144)

1·2% (1·1–1·3)

··

0·4% (0·2–0·6)

··

1819 (1819–2021)

Perindopril

··

1·0% (0·7–1·4)

··

1255 (1172–1261)

0·2% (0·2–0·3)

Fosinopril

··

3·6% (3·0–4·3)

··

1022 (723–1054)

0·7% (0·6–0·8)

Lisinopril

··

3·3% (2·8–4·0)

··

659 (469–678)

0·1% (0·1–0·2)

Quinapril

··

<0·1% (<0·1–0·2)

··

1166 (1166–1166)

<0·1% (<0·1–0·1)

Imidapril

··

0·3% (0·1–0·5)

··

1123 (1123–1123)

Cilazapril

··

Valsartan

··

Losartan

··

Irbesartan

··

Candesartan

··

Eprosartan

··

Telmisartan

··

Olmesartan

··

ACE inhibitors or angiotensin receptor blockers

··

Beta-blockers

··

Bisoprolol

Metoprolol

··

62·6% (61·0–64·3)

Annual cost per patient (RMB), median (IQR)

Angiotensin receptor blockers

··

Class availability, % (95% CI), (n=3362)

34·4% (32·7–36·0)

··

16 (11–43)

··

2·5% (2·3–2·7) 4·3% (4·1–4·5)

0 0

··

21·4% (20·0–22·8)

··

663 (340–1028)

7·8% (6·9–8·7)

··

1306 (942–1883)

2·1% (1·9–2·3)

21·1% (19·7–22·5)

··

850 (610–1101)

6·7% (6·4–7·0)

4·2% (3·5–4·9)

··

589 (468–751)

··

10·3% (9·3–11·4)

··

516 (298–801)

4·3% (4·0–4·5)

··

0·9% (0·6–1·2)

··

1833 (1761–2660)

0·2% (0·2–0·3)

··

8·8% (7·9–9·8)

··

69·0% (67·4–70·6)

··

47·2% (45·5–48·9)

··

7·4% (7·1–7·7)

1·0% (0·9–1·1) 0

··

791 (549–1080)

0·9% (0·8–1·0)

··

41·0% (39·4–42·7)

··

251 (171–281)

8·0% (7·7–8·4)

Metoprolol extended release ··

··

3·6% (3·0–4·3)

··

438 (401–440)

0·2% (0·2–0·3)

Atenolol

··

5·1% (4·4–5·9)

··

7 (5–8)

0·2% (0·1–0·2)

Propranolol

··

7·6% (6·7–8·5)

··

26 (16–86)

<0·1% (<0·1–0·1)

Betaxolol

··

Amlodipine

33·8% (32·2–35·4)

··

Nifedipine

··

34·4% (32·8–36·0)

··

Nifedipine extended release

··

41·2% (39·5–42·9)

··

413 (266–468)

7·4% (7·1–7·7)

Nifedipine controlled release ··

··

15·0% (13·8–16·2)

··

1012 (890–1526)

6·0% (5·7–6·3)

Felodipine extended release

··

19·2% (17·9–20·6)

··

425 (313–1049)

3·7% (3·5–3·9)

Levamlodipine

··

11·8% (10·7–12·9)

··

688 (437–904)

7·7% (7·4–8·0)

Nitrendipine

··

30·3% (28·7–31·9)

··

Verapamil

··

1·4% (1·1–1·9)

··

302 (288–360)

Verapamil extended release

··

<0·1% (<0·1–0·2)

··

460 (460–460)

Diltiazem

··

1·9% (1·5–2·4)

··

507 (269–1117)

Diltiazem extended release

··

0·7% (0·5–1·1)

··

583 (527–869)

0·1% (<0·1–0·1)

Lacidipine

··

4·0% (3·4–4·8)

··

228 (225–273)

0·3% (0·3–0·4)

Nicardipine

··

Benidipine

··

0·1% (<0·1–0·3)

··

999 (908–1610)

Lercanidipine

··

0·3% (0·1–0·5)

··

1530 (1509–1530)

Calcium channel blockers

·· ··

75·5% (74·0–76·9)

··

369 (206–565)

16·8% (16·3–17·2)

9 (5–19)

2·7% (2·5–2·8)

12 (7–44)

··

1·0% (0·9–1·1)

(Table 2 continues on next page)

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

Essential medicines list*

Guidelines

Availability, % (95% CI), (n=3362)

Class availability, % (95% CI), (n=3362)

Annual cost per patient (RMB), median (IQR)

Prescription frequency, % (95% CI), (n=29 171)

WHO

NEM

PSEM

Guideline in primary health care 2014

JNC 8 2014

Chinese guideline 2010

··

Indapamide extended release ··

··

Hydrochlorothiazide

··

Bendroflumethiazide

··

Chlortalidone

··

Spironolactone

··

22·3% (20·9–23·7)

··

41 (28–56)

0·5% (0·5–0·6)

Furosemide

··

14·1% (13·0–15·4)

··

17 (16–25)

0·5% (0·5–0·6)

Triamterene

··

3·3% (2·7–4·0)

··

115 (58–158)

Amiloride

··

1·2% (0·9–1·7)

··

232 (232–232)

··

Compound reserpine triamterene

··

36·6% (35·0–38·2)

··

Compound reserpine

··

34·9% (33·3–36·6)

··

80 (54–158)

0·6% (0·5–0·7)

Compound dihydralazine sulfate

··

1·2% (0·9–1·7)

··

137 (137–158)

<0·1% (<0·1–0·1)

Compound trizin and rutinum ··

··

0·1% (<0·1–0·2)

··

Compound kendir

··

2·5% (2·0–3·1)

··

65 (63–194)

Zhenju jiangya tablet

··

8·2% (7·3–9·2)

··

178 (130–269)

Fixed-dose combination drugs ··

··

(Continued from previous page) Diuretics Indapamide

Traditional compound drugs†

··

Amiloride compounds

··

Losartan potassium and hydrochlorothiazide

··

Valsartan and hydrochlorothiazide

··

Irbesartan and hydrochlorothiazide

··

Captopril and hydrochlorothiazide

··

Nitrendipine and atenolol

··

59·9% (58·2–61·5)

··

32·5% (31·0–34·2)

··

42 (18–99)

1·2% (1·1–1·3)

12·6% (11·5–13·8)

··

279 (197–389)

1·3% (1·2–1·4)

34·7% (33·1–36·3)

··

3 (1–5)

1·0% (0·9–1·1)

··

55·6% (53·9–57·3)

10·4% (9·4–11·5)

·· 414 (299–435)

··

·· 2·7% (2·5–2·9)

0 0 0·8% (0·7–0·9) ··

0·1% (<0·1–0·3)

··

116 (116–116)

<0·1% (<0·1–0·1)

2·7% (2·2–3·3)

··

1983 (895–2427)

1·6% (1·4–1·7)

··

1·2% (0·9–1·7)

··

3750 (1726–4720)

<0·1% (<0·1–0·1)

··

6·1% (5·3–6·9)

··

1136 (805–1609)

3·1% (2·9–3·3)

··

5·5% (4·8–6·4)

··

147 (66–197)

0·2% (0·1–0·2)

··

Centrally active drugs

··

Rauwofia

··

Clonidine

··

Reserpine

··

·· ··

2·3% (1·9–2·9)

·· ··

0 ··

0·1% (<0·1–0·2)

··

588 (588–588)

0·1% (0·1–0·1)

1·5% (1·1–2·0)

··

104 (104–104)

0·4% (0·2–0·7)

··

45 (45–45)

0·2% (0·2–0·3)

·· =not applicable. NEM=National Essential Medicine List. PSEM=provincial supplementary essential medicine lists (value refers to the number of provinces that had this medication in its provincial list). *Essential lists: WHO=WHO Model Lists of Essential Medicines. †Ingredients per tablet forTraditional compound drugs: compound reserpine triamterene: reserpine 0·1 mg, triamterene 12·5 mg, hydrochlorothiazide 12·5 mg, dihydralazine 12·5 mg. Compound reserpine: reserpine 0·032 mg, hydrochlorothiazide 3·1 mg, dihydralazine 4·2 mg, promethazine 2·1 mg. Compound dihydralazine sulfate: dihydralazine sulfate 10·0 mg, hydrochlorothiazide 12·5 mg, reserpine 0·1 mg. Compound trizin and rutinum: hydrochlorothiazide 2·0 mg, dihydralazine sulfate 1·5 mg, rutinum 5·0 mg, reserpine 0·03 mg. Compound kendir: kendir 220 mg, dihydralazine sulfate 1·6 mg, hydrochlorothiazide 1·6 mg, promethazine 1·05 mg. Zhenju jiangya tablet: clonidine hydrochloride 0·03 mg, hydrochlorothiazide 5·0 mg, rutinum 20·0 mg, chrysanthemum indicum powder, nacre powder.

Table 2: Availability of individual and classes of medications among all primary health-care sites

in Primary Health Care 2014 and cost no more than 200 Chinese RMB, a threshold corresponding to 1% of the average annual disposable income per person in China in 2015.25,26 The guideline recommends medicines based on clinical effectiveness. We used the 2014 guideline as a framework to guide our investigation of a wide range of antihypertensive medications expected for routine use.

Statistical analysis First, to examine availability, cost, and prescription of antihypertensive medications at different types of primary health-care sites, we calculated percentages for categorical variables, and mean (SD) or median (IQR) for continuous variables, as appropriate. To determine site-specific characteristics associated with the availability of antihypertensive medications, we fitted a mixed model

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

with township or community as random effects and a logit-link function. The model included a spherical covariate structure to account for spatial autocorrelation and differences among townships or communities. The A Type of site Type of site

Community health centre Township health centre Community health station Village clinic Aggregate

Type of site

Community health centre Township health centre Community health station Village clinic Aggregate

Type of site

Stocked at least one class of medications Community health centre Township health centre Community health station Village clinic Aggregate

Community health centre Township health centre Community health station Village clinic Aggregate

89·7

93·7 93·8

87·2

88·7

Stocked at least two classes of medications

69·6

87·2 85·6 85·3 73·9

Stocked at least three classes of medications 78·3 76·1 72·3

48·0 55·1 Stocked all four classes of medications 63·5 26·7 0

55·6

46·9 33·8

40 60 80 Availability of antihypertensive medications (%)

100

B Economic region Economic region

Eastern Central Western Aggregate

Economic region

Eastern Central Western Aggregate

Economic region

Stocked at least one class of medications Eastern Central Western Aggregate

Eastern Central Western Aggregate

91·2 91·4 83·3

88·7

Stocked at least two classes of medications 83·6 79·9 57·8

72·9 56·5 55·1

Stocked all four classes of medications 29·8 18·6 0

53·8 33·8

40 60 80 Availability of antihypertensive medications (%)

Figure 1: Availability of antihypertensive medicines in China by type of site and economic region Data are percentage (95% CI). The four classes of antihypertensive medicines assessed were angiotensinconverting-enzyme inhibitors or angiotensin-receptor blockers, beta-blockers, calcium-channel blockers, and diuretics. Definitions of the three economic regions are in the appendix.

Role of the funding source The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding and lead authors had full access to all the data in the study, and all authors had final responsibility for the decision to submit for publication.

Results

73·9

Stocked at least three classes of medications

36·1

final model included six characteristics other than the type of sites and regions—ie, the number of health-care professionals, licensed physicians, physicians with a medical bachelor degree (5-year medical education), physicians who took a continuing-education course in the past year, social insurances for contracted health-care professionals, and healthcare professionals who routinely used IT systems (appendix). Second, we modelled the prescribed medication as a function of its cost, both overall and in rural or urban subgroups. To address potential sampling variation and imbalances in numbers between electronic prescriptions and prescriptions taken from scanned copies, we adapted a resampling approach to undertake a simulation analysis with a non-parametric bootstrap, with replacement, method.27,28 Specifically, for prescription data taken from scanned copies, we randomly selected records with a sample size equal to the total number of records; for electronic prescription data, we randomly selected a sample size equal to the sample size of the prescription data from the scanned copy. We then appended the two resampled datasets together and fitted the mixed model to estimate the association between the medication’s prescription and its cost. We repeated this process 10 000 times to obtain the distributions of the estimated associations and their 95% CIs. All analyses were done using SAS 9.4, and all statistical testing was two-sided, at a significance level of 0·05. The Fuwai Hospital Institutional Review Board approved the study; the site survey was deemed exempt; and informed consent was obtained from all MPP study participants.

100

The study sample included 3362 primary health-care sites (18·0% urban, 82·0% rural) across China (appendix). Township health centres constituted 8·4% of sites, village clinics 73·6%, community health centres 6·0%, and stations 11·9% (table 1). The 435 townships or communities, which collectively enrolled 1·09 million people in the China PEACE MPP, served as the study sample for determining hypertension treatment and control rates. The median sample size of participants at the township or community level was 2128 (IQR 1165–3103). Among the 3362 PHC sites, the most widely available antihypertensives were calcium-channel blockers, of which nifedipine extended release was the most commonly stocked (table 2). The patterns of medication availability varied by site. Overall, 8·1% (273/3362, 95% CI 7·2–9·1) of primary health-care pharmacies did not have any antihypertensive

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

≤200 RMB

Availability

>200 RMB

medications; 88·7% (2982/3362, 87·6–89·7) stocked either ACE inhibitors or angiotensin-receptor blockers, beta-blockers, calcium-channel blockers, or diuretics; and 33·8% (1136/3362, 32·2–35·4) had all four classes of antihypertensive medications (figure 1). Availability, defined as having any class of medication, was associated with the type of site and economic region (appendix)— township health centres were more likely, and sites in the western region were less likely, to stock any kind of antihypertensive medication, and village clinics and sites in the western region were also less likely to have all four classes (figure 1; appendix). No characteristic specific for primary health-care sites other than the type of sites and region was strongly associated with antihypertensive availability (appendix). Median annual costs per patient for individual medications varied substantially (table 2). Figure 2 shows the medications in value quadrants, according to their guideline-recommended status and cost. Only 32·7% (95% CI 32·2–33·3) of sites stocked medications in the high-value category. Across 396 townships or communities, we sampled 26 159 of 518 915 hypertension prescriptions. The most commonly prescribed individual medication was amlodipine (16·8% [95% CI 16·3–17·2]), and the most frequently prescribed antihypertensive classes were calcium-channel blockers (45·6% [45·1–46·2]), angiotensin-receptor blockers (21·7% [21·3–22·2]) and beta-blockers (9·4% [9·1–9·7]; appendix). Overall, 86·2% (85·7–86·6) of prescriptions were for one medication, whereas less than 1% were for three or more medications. When two medications were used, fixed-dose combinations (38·9% [95% CI 37·2–40·5]), ACE inhibitors plus calcium-channel blockers (18·3% [17·0–19·7]), and angiotensin-receptor blockers plus calcium-channel blockers (17·0% [15·8–18·3]) were most commonly prescribed together (appendix). 2234 prescription records (7·7% [95% CI 7·4–8·0]) were for non-guidelinerecommended medications, 3276 (11·2% [10·9–11·6]) for high-value medications, and 23 603 (81·1% [80·6–81·5]) for higher-cost, guideline-recommended medications (figure 2). The sites varied by treatment and control (appendix). The median risk-standardised treatment and control rates were 35·7% (IQR 19·6–52·1) and 8·3% (3·6–16·1), respectively. The worst 10% of sites had rates of 4·1% (1·6–5·8) and 1·1% (0·8–1·5), respectively, whereas the best 10% had rates of 72·5% (69·3–77·3) and 30·6% (26·5–35·4). The cost of a medication was associated with being prescribed (figure 3, appendix), but not with its availability (figure 3). On average, high-cost medications were more likely than low-cost medications to be prescribed in primary health-care sites. Low-cost medications accounted for 40·5% (95% CI 39·9–41·1) of the medications in the pharmacies. Of all the prescriptions, 12·6% (12·3–13·0) were for low-cost

Prescription

32·7% (32·2–33·3)

7·8% (7·4–8·1)

11·2% (10·9–11·6)

1·4% (1·3–1·5)

53·8% (53·2–54·4)

5·7% (5·4–6·0)

81·1% (80·6–81·5)

6·3% (6·0–6·6)

Guideline recommended

Not guideline recommended

Guideline recommended

Not guideline recommended

Figure 2: Availability and prescription of medications in China by value quadrants Data are percentage (95% CI). x-axis: Chinese Guideline for Hypertension Management in Primary Health Care, 2014. y-axis: Annual cost of medication per patient (RMB).

medications and 4·7% (4·4–4·9) were for diuretics, the lowest-cost medication.

Discussion This national study of antihypertensive medications in China revealed marked deficiencies in their availability, cost, and prescription. First, hypertension medications are inconsistently available in primary health-care pharmacies across China, and one in 12 did not stock any antihypertensive medications. Second, despite the availability of low-cost antihypertensive medications, higher-cost medications were more often prescribed. In fact, the higher the cost of the medication, the more likely that it was prescribed; however, this did not necessarily mean that the most expensive drug was the one that was most frequently prescribed. The higher-cost medications did not represent medications with higher effectiveness. This study adds to the scientific literature in several ways: it is the first national study of the availability, cost, and prescription of antihypertensive medications in China, involving all provinces. National policies with regard to essential medications and reimbursement might aim to improve access;29 this study provides a contemporary assessment of the availability and use of antihypertensive therapies in primary health-care sites around the country and shows that deficiencies exist at the point of care. This study was based on actual investigations of the pharmacies and inspection of the prescriptions, and required government support, partnership with providers and administrators of primary health care, and site access to inspect pharmacies and examine prescriptions. The study did not depend on reports from the sites but, rather, involved direct data collection. Previous studies were limited to specific regions, populations, and data sources.14,30–32 The reasons for gaps in the availability and prescription of antihypertensive medications are not clear. Chinese national policies state that essential medications should be available and affordable. However, we uncovered problems

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

0·35 (0·22–0·51)

Areas

Aggregate

0·43 (0·16–0·74)

Rural

0·34 (0·19–0·52)

Urban –1·00

–0·75

–0·50

–0·25

0·25

0·50

0·75

1·00

Percentage point increase in prescription for every 10 RMB increase in cost of medication 5000

Cost per year (RMB, log scale)

1000

33 14

200 54

100

13 29

28 11 24 1626

40 53

7 52

12 57 18 3

58 61 49

39 44

Not guideline-recommended Guideline-recommended and cost per year ≤200 RMB Guideline-recommended and cost per year >200 RMB 0·1

100

Availability of medications at primary health-care site level in urban areas (%, log scale) 5000 17

Cost per year (RMB, log scale)

1000

56 38

9 55 6

12 7

200

13 11

35 50

100

27 29 18

19 2

22 45 30

0·1

100

Availability of medications at primary health-care site level in rural areas (%, log scale) 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61

Captopril Ramipril Lisinopril Cilazapril Irbesartan Telmisartan Metoprolol Propranolol Nifedipine Felodipine extended release Verapamil Diltiazem extended release Benidipine Indapamide extended release Chlortalidone Triamterene Compound reserpine Compound kendir Losartan potassium and hydrochlorothiazide Captopril and hydrochlorothiazide Clonidine

2 5 8 11 14 17 20 23 26 29 32 35 38 41 44 47 50 53 56 59 62

Enalapril Perindopril Quinapril Valsartan Candesartan Olmesartan Metoprolol extended release Betaxolol Nifedipine extended release Levamlodipine Verapamil extended release Lacidipine Lercanidipine Hydrochlorothiazide Spironolactone Amiloride Compound dihydralazine sulfate Zhenju jiangya tablet Valsartan and hydrochlorothiazide Nitrendipine and atenolol Reserpine

3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60

Benazepril Fosinopril Imidapril Losartan Eprosartan Bisoprolol Atenolol Amlodipine Nifedipine controlled release Nitrendipine Diltiazem Nicardipine Indapamide Bendroflumethiazide Furosemide Compound reserpine triamterene Compound trizin and rutinum Amiloride compound Irbesartan and hydrochlorothiazide Rauwolfia

Figure 3: Annual costs per patient compared with availability and prescription patterns of antihypertensive medications Data are (A) percentage point increase (95% CI) and (B) and (C) percentages.

in the inventory of antihypertensive medication that can be supplied to patients at primary health-care sites. Despite the national focus on blood pressure control—and the responsibility of many of the sites for hypertension management33—a substantial proportion of these pharmacies had no or only a few antihypertensive medications, especially the low-cost drugs.14 The zero mark-up policy might be a possible contributor to our finding that high-value medications are not frequently prescribed. Initiated in 2009, this policy prohibits healthcare providers from selling essential medicines at prices higher than their wholesale cost. This policy reportedly exerted a large net effect on the revenue of village clinics, despite government subsidies being increased to compensate for revenue loss.34,35 These reforms have been suggested to have led to a less-reliable drug supply system in China; for instance, village clinics might no longer provide essential medicines at zero mark-up because of the absence of a profit.36 With respect to drug-prescribing patterns, there has been an increase in the use of expensive medications since implementation of the Chinese National Essential Medicine Program.11 Refinements to this policy might provide stronger incentives for the use of low-cost medications. Mandating the availability of medicines might not be sufficient to improve the prevalence of hypertension, but their availability is arguably a necessary component of disease management programmes. Additional studies are needed to carefully examine the effect of zero mark-up policies on access to antihypertensive medicines throughout Chinese primary health-care centres.12,37 Additionally, some patients and doctors might prefer antihypertensive traditional Chinese medicines, though their use was generally low in our study. The implications of this study for hypertension management in China are substantial. The reality of care delivery in the clinics is not consistent with the health needs of the nation, and the deficiencies in primary care pharmacies have implications for patient health, as evidenced by suboptimal treatment and control rates. As such, interventions to improve hypertension treatment and control will need to focus not only on bolstering education, screening, and protocols, but also on ensuring that antihypertensive medications are adequately stocked by primary health-care pharmacies.38 The adequacy of the medication inventory is not sufficient for progress in hypertension treatment and control, but it is certainly a fundamental component. Policy makers will need to ascertain why the aspiration of national policies is being stymied at a local level and probably thwarting efforts by practitioners to address hypertension in their patients. The study has other important implications. The use of high-value medications—those that are guidelinerecommended and reasonably priced—should be a priority for all countries, especially those with few resources.39 Our study showed that high-value medications are not preferentially used in Chinese primary health-care settings, even though there is no evidence that

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

higher-priced medications are more effective.40 A greater emphasis on high-value antihypertensive medications has the potential to mitigate the cost burden of increasing the rates of treatment and providing more value to the country. In this respect, diuretics might be particularly cost-effective. Findings from previous studies have suggested that drugs such as chlorthalidone might even be better than drugs from other classes.23 However, few comparative effectiveness studies of antihypertensive drugs have been done,41 and it might be beneficial for China to prioritise such studies. Identifying the higherpriced medications with known marginal benefits over lower-cost alternatives would provide the basis for highquality and cost-efficient care. The availability of antihypertensive medications varied among types of sites and economic regions, but inadequacies were not confined to certain types of centres. Site characteristics were not strongly associated with the availability of antihypertensive medications. This finding indicates the need for a broad-based strategy that would address problems that almost all types of primary health-care centres throughout China face. Our study had several limitations. First, the study sites were not a representative sample despite spanning the entire country geographically and being so large in number. The treatment and control rates, however, were very similar to national estimates.5 Second, we used a convenience sample and excluded people who were not residents. Those who were excluded would probably have had less access to care and would likely have had even lower control rates. Nevertheless, any inaccuracies in this study would be a bias toward the null, suggesting that our findings might have even underestimated the relationship. Of note, the prescription information reflected all prescriptions, including those provided to migrants. Third, this study focused on pharmacies in primary health-care sites, and people might go elsewhere for their prescriptions. However, half of private pharmacies impose fees that individuals must pay out of pocket.42 Therefore, we expected that most patients would have a strong preference to obtain their medications from the clinic pharmacy. Future studies should build on these and other emerging primary datasets in China to examine the association between access to antihypertensive medicines and clinical outcomes, including control rates. Fourth, our choice of the 200 RMB threshold for cost might be arbitrary (ie, it was a general value that was not region-specific), and the threshold of 1% of annual disposal income might be higher for rural populations. Applying a lower-cost threshold, however, would further restrict the sample of medicines that could be defined as high-value, and lower the percentage of high-value drugs prescribed, further strengthening our findings. Finally, the inventory and prescription data that were collected covered slightly different periods. If any large-scale change in prescription drug inventories and prescribing behaviours are assumed

to be marginal over the span of several months, the effect on this study from similar, albeit non-overlapping periods for data collection might be negligible. In conclusion, this study revealed key obstacles to progress in mitigating hypertension in China. Despite advances in health-care coverage and policy to limit financial risks and improve health outcomes,43 this study highlighted deficiencies in the availability, cost, and prescription of antihypertensive medications. Future policies aimed at alleviating the burden of hypertension in China, particularly through the work of primary health-care providers, will need to improve access to high-value antihypertensive medications. Contributors LJ and HMK conceived the study and take responsibility for all its aspects. LJ, HMK, MS, and QZ initially designed the survey, with support from EM, GAM, XL, and JL. MS, JL and XL led the data collection. HMK, LJ, MS, and QZ conceived this article. HMK, LJ, MS, and QZ wrote the manuscript, with further contributions from LJ, HMK, EM, GAM, FAM, SS-V, AZ, KN, and YLu. XB, CW, YLi, SS-V, and AZ provided data management and statistical analysis. All authors interpreted data, contributed to critical revisions, and approved the final version of the Article. Declaration of interests HMK is a recipient of research agreements from Medtronic and from Johnson & Johnson (Janssen), through Yale, to develop methods of clinical trial data sharing; is the recipient of a grant from the US Food and Drug Administration and Medtronic, through Yale, to develop methods for post-market surveillance of medical devices; works under contract with the Centers for Medicare & Medicaid Services to develop and maintain performance measures; chairs a cardiac scientific advisory board for UnitedHealth; is a participant or participant representative of the IBM Watson Health Life Sciences Board; is a member of the Advisory Board for Element Science and the Physician Advisory Board for Aetna; and is the founder of Hugo, a personal health information platform. All other authors declare no competing interests. Acknowledgments This project was partly supported by the CAMS Innovation Fund for Medical Science (2017-I2M-2-002, 2016-I2M-2-004, 2016-I2M-1-006); the Entrusted Project from the China National Development and Reform Commission; the Major Public Health Service Project from the Ministry of Finance of China and National Health and Family Planning Commission of China; the China-WHO Biennial Collaborative Projects 2016â&#x20AC;&#x201C;2017 (2016/664424-0); the Entrusted Project from the China National Health and Family Planning Commission; the National Key Technology R&D Program (2015BAI12B01, 2015BAI12B02) from the Ministry of Science and Technology of China; Research Special Fund for Public Welfare Industry of Health (201502009) from the National Health and Family Planning Commission of China; the 111 Project from the Ministry of Education of China (B16005); and the PUMC Youth Fund and the Fundamental Research Funds for the Central Universities (2017330003). We appreciate the contributions made by the study teams at the National Center for Cardiovascular Diseases, and all the local collaborative sites during the study design and operations (appendix), particularly data collection by Na Tian, Yaping Cao, Chongxin Chen, Jianlan Cui, Xin Sun, Wei Xu, and Bo Gu; Winnie Yip from Harvard TH Chan School of Public Health, Kar Keung Cheng from University of Birmingham, Jan De Maeseneer from Ghent University, Qingyue Meng from Peking University, Dong Roman Xu from Sun Yat-sen University, and Richard Peto from University of Oxford, for their support in study design; Yun Wang from Harvard TH Chan School of Public Health and Hongzhao Zhang from National Center for Cardiovascular Diseases for their support in data cleaning and analysis; Beibei Yuan, Dan Wang, and Huiwen Li from Peking University Health Science Center, who searched the published work, extracted data, and summarized findings; and the advice of Steven DeMaio from The Yale-New Haven Hospital Center for Outcomes Research and Evaluation.

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Articles

References 1 Lewington S, Lacey B, Clarke R, et al. The burden of hypertension and associated risk for cardiovascular mortality in China. JAMA Intern Med 2016; 176: 524–32. 2 Wang JG. Chinese hypertension guidelines. Pulse 2015; 3: 14–20. 3 Guan X, Liang H, Xue Y, et al. An analysis of China’s national essential medicines policy. J Public Health Policy 2011; 32: 305–19. 4 Feng XL, Pang M, Beard J. Health system strengthening and hypertension awareness, treatment and control: data from the China Health and Retirement Longitudinal Study. Bull World Health Organ 2014; 92: 29–41. 5 Li Y, Yang L, Wang L, et al. Burden of hypertension in China: a nationally representative survey of 174 621 adults. Int J Cardiol 2017; 227: 516–23. 6 Yang L, Yan J, Tang X, et al. Prevalence, awareness, treatment, control and risk factors associated with hypertension among adults in southern China, 2013. PLoS One 2016; 11: e0146181. 7 Li D, Lv J, Liu F, et al. Hypertension burden and control in mainland China: analysis of nationwide data 2003–2012. Int J Cardiol 2015; 184: 637–44. 8 Lu J, Lu Y, Wang X, et al. Prevalence, awareness, treatment, and control of hypertension in China: data from 1·7 million adults in a population-based screening study (China PEACE Million Persons Project. Lancet 2017; published online Oct 25. http://dx.doi. org/10.1016/S0140-6736(17)32478-9. 9 Zhou M, Wang H, Zhu J, et al. Cause-specific mortality for 240 causes in China during 1990–2013: a systematic subnational analysis for the Global Burden of Disease Study 2013. Lancet 2016; 387: 251–72.  10 Yang G, Kong L, Zhao W, et al. Emergence of chronic non-communicable diseases in China. Lancet 2008; 372: 1697–705. 11 Song Y, Bian Y, Petzold M, et al. The impact of China’s national essential medicine system on improving rational drug use in primary health care facilities: an empirical study in four provinces. BMC Health Serv Res 2014; 14: 507. 12 Chen Z. Launch of the health-care reform plan in China. Lancet 2009; 373: 1322–24. 13 Liu Q, Tian X, Tian J, et al. Evaluation of the effects of comprehensive reform on primary healthcare institutions in Anhui Province. BMC Health Serv Res 2014; 14: 268. 14 Fang Y, Wagner AK, Yang S, et al. Access to affordable medicines after health reform: evidence from two cross-sectional surveys in Shaanxi Province, western China. Lancet Glob Health 2013; 1: e227–37. 15 Xi X, Li W, Li J, et al. A survey of the availability, prices and affordability of essential medicines in Jiangsu Province, China. BMC Health Serv Res 2015; 15: 345. 16 Lu J, Xuan S, Downing NS, et al. Protocol for the China PEACE (Patient-centered Evaluative Assessment of Cardiac Events) million persons project pilot. BMJ Open 2016; 6: e010200. 17 Su M, Zhang Q, Lu J, et al. Protocol for a nationwide survey of primary health care in China: the China PEACE (Patient-centered Evaluative Assessment of Cardiac Events) MPP (million persons project) Primary Health Care Survey. BMJ Open 2017; 7: e016195. 18 National Health and Family Planning Commission of the People’s Republic of China. Number of health institutions by the end of April 2017. 2017. http://www.nhfpc.gov.cn/mohwsbwstjxxzx/s7967/2 01706/41573016be1b41719c8ca68dfab05e9d.shtml (accessed Aug 20, 2017). 19 AMAP. https://restapi.amap.com/v3/geocode/geo?key=678c8e39ee 3f8fc598c659b4ea84cc5c&address (accessed Sept 25, 2017). 20 Chinese Pharmacopoeia. Chinese Pharmacopoeia Commission; 2015 edn. http://eng.sfda.gov.cn/WS03/CL0757/122060.html (accessed Sept 22, 2017). 21 Instructions for users of Omron HEM-7430. http://www.healthcare. omron.co.jp/support/download/manual/pdf/HEM-7430_m.pdf (accessed Sept 22, 2017) 22 Liu LS, Writing Group of Chinese Guidelines for the Management of Hypertension. 2010 Chinese guidelines for the management of hypertension. Zhonghua Xin Xue Guan Bing Za Zhi 2011; 39: 579–615. 23 James PA, Oparil S, Carter BL, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014; 311: 507–20.

24 National Health and Family Planning Commission of China. National Essential Medicine List (2012 version). 2013. http://www.moh.gov.cn/wsb/pwsyw/201303/ f01fcc9623284509953620abc2ab189e.shtml (accessed May 30, 2017). 25 Revision Committee of the Chinese Guideline for Hypertension Management in Primary Care 2014. Chinese guideline for hypertension management in primary care 2014. Chin J Hypertens 2015; 23: 24–43. 26 National Bureau of Statistics of China. China Statistical Yearbook 2016. http://www.stats.gov.cn/tjsj/ndsj/2016/indexch.htm (accessed May 30, 2017). 27 Zheng Z, Zhang H, Yuan X, et al. Comparing outcomes of coronary artery bypass grafting among large teaching and urban hospitals in China and the United States. Circ Cardiovasc Qual Outcomes 2017; 10: e003327. 28 Wang Y, Eldridge N, Metersky ML, et al. National trends in patient safety for four common conditions, 2005–2011. N Engl J Med 2014; 370: 341–51. 29 General Office of the Chinese State Council. Opinions on key tasks for deepening the health system reform 2016. 2016. http://www.nhfpc.gov.cn/tigs/s7846/201604/ ede9ab7526aa4222a56c7b906ae334af.shtml (accessed July 7, 2017). 30 Yang L, Liu C, Ferrier JA, et al. Organizational barriers associated with the implementation of national essential medicines policy: a cross-sectional study of township hospitals in China. Soc Sci Med 2015; 145: 201–08. 31 Cheng H. Prescribing pattern of antihypertensive drugs in a general hospital in central China. Int J Clin Pharm 2011; 33: 215–20. 32 Yang H, Dib HH, Zhu M, et al. Prices, availability and affordability of essential medicines in rural areas of Hubei Province, China. Health Policy Plan 2010; 25: 219–29. 33 Ministry of Health of the People’s Republic of China; Treasury Department; State Administration of Traditional Chinese Medicine. Announcement on the National Basic Public Health Service Project in 2016. 2016. http://www.nhfpc.gov.cn/jws/s3577/201606/ f29a4659c7f4455ca6f62f8d14eb4b02.shtml (accessed July 7, 2017). 34 Hu S. Essential medicine policy in China: pros and cons. J Med Econ 2013; 16: 289–94. 35 Mossialos E, Ge Y, Hu J, et al. Pharmaceutical policy in China: challenges and opportunities for reform 2016. http://www.euro. who.int/__data/assets/pdf_file/0020/320465/Pharmaceutical-policyChina-challenges-opportunities-reform.pdf?ua=1 (accessed Sept 22, 2017). 36 Mao W, Chen W. The zero mark-up policy for essential medicines at primary level facilities. Geneva: World Health Organization, 2013. 37 Yip WC, Hsiao WC, Chen W, et al. Early appraisal of China’s huge and complex health-care reforms. Lancet 2012; 379: 833–42. 38 Hogerzeil HV, Liberman J, Wirtz VJ, et al. Promotion of access to essential medicines for non-communicable diseases: practical implications of the UN political declaration. Lancet 2013; 381: 680–89. 39 Porter ME. A strategy for health care reform—toward a value-based system. N Engl J Med 2009; 361: 109–12. 40 Prasad V, De Jesús K, Mailankody S. The high price of anticancer drugs: origins, implications, barriers, solutions. Nat Rev Clin Oncol 2017; 14: 381–90. 41 Jiang L, Krumholz HM, Li X, et al. Achieving best outcomes for patients with cardiovascular disease in China by enhancing the quality of medical care and establishing a learning health-care system. Lancet 2015; 386: 1493–505. 42 Ministry of Human Resources and Social Security of the People’s Republic of China. Achievements of China’s social insurance reform have attracted worldwide attention. 2017. http://www. mohrss.gov.cn/SYrlzyhshbzb/dongtaixinwen/buneiyaowen/201705/ t20170525_271399.html (accessed June 28, 2017). 43 Meng Q, Xu L, Zhang Y, et al. Trends in access to health services and financial protection in China between 2003 and 2011: a cross-sectional study. Lancet 2012; 379: 805–14.

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32476-5

Comment

Hypertension in China: the gap between policy and practice annual—health screenings also take place.8 So why the low figures for treatment and control? There appear to be three fundamental problems—first, in the screening process itself; second, related to payments within the health system; and third, the underlying failure of primary prevention measures. In much of China, screening is simply a process and does not lead to follow-up, treatment, or referral. Individuals are often expected to act upon the results of a screening report themselves, without knowledge of normal values of blood pressure. In our experience this expectation results either in failure to access treatment altogether, or seeking treatment from more than one provider to try to get the best treatment, with no follow-up. As elsewhere in the world, China is struggling with making the shift from a health system designed for acute care, to one focusing on chronic conditions, which requires continuity, monitoring, and adjustment of treatments as necessary. The second reason for poor control of hypertension lies with the perverse incentives offered within the health system, as referred to by Su and colleagues.3 Many Chinese people have health insurance, (albeit with copayments), therefore appropriate drugs are affordable for most patients with hypertension. There are some good practices in China: some counties in Zhejiang province, for example, offer lower copayments for chronic conditions including hypertension to increase the treatment rates.9 But the 2009 health reforms have also had unintended consequences for primary care. The

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32743-5

Published Online October 25, 2017 http://dx.doi.org/10.1016/ S0140-6736(17)32743-5 See Online/Articles http://dx.doi.org/10.1016/ S0140-6736(17)32478-9 and http://dx.doi.org/10.1016/ S0140-6736(17)32476-5

Lane Oatey/Blue Jean Images/Getty Images

The high prevalence of hypertension in China is well known, with stroke being the most common cause of death and disability.1 Two large nationwide studies reported in The Lancet2,3 highlight that although the prevalence of hypertension in China is similar to that suggested in previous studies, it is simple deficiencies in the country’s health system that make a large contribution to the disease burden. Both studies used data from the PEACE (Patient-Centred Evaluative Assessment of Cardiac Events) Million Persons Project (MPP), which enrolled 1·7 million adults aged 35–75 years from across China. This is a population-centred, national screening initiative to detect populations at high risk of cardiovascular disease. In the first Article,2 Jiapeng Lu and colleagues estimated the age-standardised and sex-standardised prevalence of hypertension (defined as individuals with either blood pressure >140/90 mm Hg or self-reported use of antihypertensive medication in the previous 2 weeks) to be 37·2% (95% CI 37·1–37·3), which is similar to estimates from previous studies.4,5 Crucially, only 22·9% (22·7–23·0) of these individuals were receiving treatment, and only 5·7% (5·6–5·7) were achieving control. In the second Article,3 Meng Su and colleagues expand on the issue of antihypertensive treatments, highlighting patterns in availability, cost, and prescriptions. Drawing on data from the PEACE MPP study and from a nationwide sample of 3362 primary care facilities (613 638 people at 2758 rural sites and 478 393 people at 604 urban sites), they show that 8·1% of facilities (95% CI 7·2–9·1) did not stock any hypertensive drugs, and only 33·8% (32·2–35·4) stocked all four standard classes of antihypertensive drugs. Officially recommended lower-cost drugs (so-called high-value medications) were prescribed at only 11% of facilities, and expensive drugs were more likely to be prescribed, irrespective of availability of the cheaper drugs. This situation is worrying, not least because prevention and control of hypertension have been a high priority in China for more than two decades,6 and the two studies illustrate well the gap between policy and practice in this area. From 2009, efforts were intensified with the Chinese Basic Public Health Service Programme, which provides a package of care at community health facilities that includes hypertension and diabetes screening and management.7 In many workplaces, regular—usually

Comment

reforms introduced a zero-profit drugs policy, designed to reduce financial incentives for doctors to prescribe specific drugs, which had in the past led to massive over-prescribing of those drugs. But the zero-profit policy applies only to drugs on the Essential Drug List, with other drugs still liable to mark-up, therefore, doctors are incentivised to prescribe more expensive drugs. However, as made clear from both Articles, guidelines exist for hypertension management, but they seem to be ignored with impunity—this should not be the case. With computerisation of prescribing in many primary care facilities now, an audit of adherence to the guidelines could be carried out, with educational and even punitive measures taken if necessary. This has been done with success for antibiotic stewardship in some primary care settings,10 and could be easily duplicated. Finally, primary prevention of hypertension must be promoted. This should include education about the dangers of hypertension, the need for lifelong treatment if diagnosed, and the importance of lifestyle for prevention and treatment. Especially relevant for China is the need for salt reduction in the diet. The average daily salt intake in northern China is 12–18 g/day, compared with the globally recommended limit of no more than 5 g/day.11 Awareness about the dangers of dietary salt is low and doctors do not emphasise salt reduction when advising patients.11 Unlike many other countries, in China dietary salt comes mainly from salt products added to food during cooking, and not from food processing, therefore the public health message to reduce salt in cooking should be emphasised. Findings from a study in northern China12 showed that a schools-based programme reducing salt in school meals, alongside education of parents, was effective and cost-effective, with the potential for substantial benefits if the programme is introduced nationwide. It is not clear from either Article how many of the participants had been screened for hypertension

previously, or about their awareness of the importance of screening. However, the failure to act on the findings from the screenings, and not providing appropriate treatment, clearly lead to poor blood pressure control in a huge population. The findings of these papers should inform measures that could help reduce the huge burden of hypertension in China. *Therese Hesketh, Xudong Zhou UCL Institute for Global Health, University of London, London, WC1N 1EH, UK (TH) and Zhejiang University School of Public Health, Hangzhou, China (TH, XZ). t.hesketh@ucl.ac.uk We declare no competing interests. 1

GBD 2013 Mortality and Causes of Death Collaborators. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2015; 385: 117–71. 2 Lu J, Lu Y, Wang X,et al. Prevalence, awareness, treatment, and control of hypertension in China: data from 1·7 million adults in a population-based screening study (China PEACE Million Persons Project). Lancet 2017; published online Oct 25. http://dx.doi.org/10.1016/S01406736(17)32478-9. 3 Su M, Zhang Q, Bai X, et al. Availability, cost, and prescription patterns of antihypertensive medications in primary health care in China: a nationwide cross-sectional survey. Lancet 2017; published online Oct 25. http://dx.doi.org/10.1016/S0140-6736(17)32476-5. 4 Li D, Lv J, Liu F, et al. Hypertension burden and control in mainland China: analysis of nationwide data 2003–2012. Int J Cardiol 2015; 184: 637–44. 5 He J, Gu D, Chen J, et al. Premature deaths attributable to blood pressure in China: a prospective cohort study. Lancet 2009; 374: 1765–72. 6 Wang L, Kong L, Wu F, Bai Y, Burton R. Preventing chronic diseases in China. Lancet 2005; 366: 1821–24. 7 National Health and Family Planning Commission of China. 2016. http://www.nhfpc.gov.cn/jws/s3577/201606/ f29a4659c7f4455ca6f62f8d14eb4b02.shtml (accessed Sept 26, 2017). 8 Sun X, Chen Y, Tong X, et al. The use of annual physical examinations among the elderly in rural China: a cross-sectional study. BMC Health Serv Res 2014; 14: 16. 9 Taizhou Bureau of Human Resource and Social Security. 2016. http://www.tzhrss.gov.cn/zcfg/html/?5983.html (accessed Sept 26, 2017). 10 Yip W, Powell-Jackson T, Chen W, et al. Capitation combined with pay-for-performance improves antibiotic prescribing practices in rural China. Health Aff (Millwood) 2014; 33: 502–10. 11 Liu Y, Li H, Hong S, Yin X. Salt reduction and hypertension in China: a concise state-of-the-art review. Cardiovasc Diagn Ther 2015; 5: 191. 12 Li X, Jan S, Yan LL, et al. Cost and cost-effectiveness of a school-based education program to reduce salt intake in children and their families in China. PLoS One 2017; 12: e0183033.

www.thelancet.com Published online October 25, 2017 http://dx.doi.org/10.1016/S0140-6736(17)32743-5