We used data from the China Chronic Disease and Risk Factors Surveillance (CCDRFS) programme conducted from August 2018 to June 2019 according to a standard protocol. The CCDRFS was used to evaluate the prevalence of major chronic diseases and the associated behavioral and metabolic risk factors. The multistage stratified sampling method was used to sample, and the protocol for collecting the CCDRFS data has been described elsewhere [22,23,24]. The representative survey collected data from 298 counties (districts) in 31 provinces (autonomous regions, municipalities) in China [25, 26]. A total of 184,876 participants (living at their current residence for at least six months within the year before the survey) aged 18 and above were investigated in the CCDRFS. All participants signed informed consent.

The surveillance included questionnaire investigation, medical examination, and laboratory tests. All these contents followed a unified and standard protocol for each site. Demographic characteristics, prevalence, treatment and control of major chronic diseases, health management services, socioeconomic and behavioral factors information of participants were collected by trained staff through questionnaire investigation. Blood pressure, height, and weight were measured by trained staff with standard tools. A digital blood pressure monitor (OMRON HBP1300) was used to measure systolic blood pressure (SBP) and diastolic blood pressure (DBP) three times, respectively. The averages of SBP and DBP were calculated for analysis. The laboratory tests included blood and urine sample tests. Blood sample tests contained fasting blood glucose(FBG), blood glucose 2 h after taking 75 g oral glucose sugar(participants without self-reported diabetes history), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), serum creatinine, and albumin. Blood samples were collected after at least 8 h fast from every participant and participants taking 75 g oral glucose sugar. Morning urination samples were collected to detect urine creatinine and microalbumin. Plasma glucose was measured by hexokinase method, TC was measured by cholesterol oxidase p-aminophenazone (CHOD-PAP) method, HDL-C was measured by homogeneous enzyme colorimetry, and serum creatinine and urine creatinine were measured by enzyme coupled creatine oxidase.

Demographic Characteristics comparisons were made between PCs and NPCs. Demographic included age, sex (male or female), education (< primary school, primary school, middle school, ≥ high school), marital status (married, single/widow/divorce/separated), medical insurance status (yes or no), location (urban or rural), and geographic region [eastern (Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong and Hainan), central (Shanxi, Inner Mongolia, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei, Hunan, Guangxi), western (Sichuan, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Qinghai, Ningxia, Xinjiang)]. Household income per capita per year was converted into quartiles (≤ $727.8, $727.8 ~ 1455.5, $1455.5 ~ 2765.5, > $2765.5, unknown).The State Council had determined 832 national PCs and 334 deep PCs in China [27, 28]. The national PCs were determined according to the per capita county GDP (gross domestic product), per capita general budget income of county finance, per capita net income of county farmers, and other factors. The deep PCs were determined according to the poverty population scale, the economic development level, the difficulty of poverty alleviation, and other factors.

Diagnosed hypertension patients were divided into three groups (non-standardized, standardized or no) according to whether the patients participated in the hypertension health management services involved in the HPAP. The “standardized management” group included diagnosed hypertension patients participating in health management who have received at least four blood pressure measurements per year provided by doctors in primary medical and health institutions, as well as guidance and suggestions on medication, diet, physical activity, tobacco and alcohol control. The “non-standardized management” group included diagnosed hypertension patients without enough blood pressure measurements or guidance and suggestions on medication, diet, physical activity, tobacco and alcohol control. The “no” group included diagnosed patients without hypertension health management services.

Behavioral influencing factors of hypertension control contained smoking status (current smoker, former smoker, or nonsmoker), current alcohol use (within 30 days, out of 30 days, or nondrinker), low physical activity (< 150 min/week), low fruit and vegetable intake (< 400 g/d), high red meat intake (≥ 100 g/d), high salt intake (≥ 5 g/d), and high oil intake (≥ 25 g/d). Body-mass index (BMI) was calculated as weight (kg) divided by height (m) squared and divided into four groups (< 18.5 kg/m2, 18.5–25 kg/m2, 25–30 kg/m2 or > 30 kg/m2) according to WHO criteria. Diabetes were defined as participants with self-reported diabetes diagnosed by a health professional medical institution or with FBG ≥ 7 mmol/L or 2-h plasma glucose ≥ 11.1 mmol/L after an oral 75 g anhydrous glucose. The urine microalbumin to creatinine ratio (ACR) was calculated using urinary microalbumin divided by urinary creatinine. The glomerular filtration rate (eGFR) was calculated using the CKD-EPI equation [29]. Chronic kidney disease (CKD) was defined as eGFR < 60 mL/(min/1.73 m2) or ACR ≥ 30 mg/g according to the Clinical Practice Guideline for Evaluation and Management of CKD by Kidney Disease Improving Global Outcomes (KDIGO). Non-HDL-C was calculated by TC subtracting HDL-C and Non-HDL-C ≥ 4.90 mmol/L, defined as high Non-HDL-C [30].

Primary health outcomes were hypertension prevalence, hypertension control, and health management prevalence. The hypertension population was divided into two categories: diagnosed hypertension patients (self-reported hypertension diagnosed by a health professional medical institution) or new found population (SBP ≥ 140 mmHg or DBP ≥ 90 mmHg in this survey but not self-reported hypertension). Hypertension control was defined as SBP < 140 mmHg and DBP < 90 mmHg among diagnosed patients under 65 years old and SBP < 150 mmHg and DBP < 90 mmHg among diagnosed patients over 65 years old. Hypertension control prevalence was only calculated in diagnosed patients.

Secondary outcomes were physical examination proportion and hypertension treatment prevalence. Physical examination was collected through the question “When was your last physical examination?” in the questionnaire and divided into three categories: one year and below, more than one year, or no. The hypertension treatment was defined as the diagnosed hypertension patients using antihypertensive drugs. Hypertension treatment prevalence was only calculated among diagnosed patients.

We used count and proportions to describe qualitative data, and χ2 or Wilcoxon rank sum tests to examine differences in categorical variables. Hypertension control prevalence was calculated separately by sex (men vs. women) and locality of residence (urban vs. rural). Before multivariate analysis, we examined differences in influencing factors between controlled and uncontrolled hypertension patients by PCs and NPCs, respectively. We used a multivariate logistic regression model to explore the association between hypertension control and health management services. Models were built in NPCs (n = 14,497) and PCs (n = 4932), respectively. Only diagnosed hypertension patients were involved in regression model analysis. P values were two sided and P < 0.05 was considered statistically significant. The analyses were calculated using SAS ver. 9.4 (SAS institute).