Changes in Hypertension-Related Knowledge and Behavior and Their Associations with Socioeconomic Status among Recently Urbanized Residents in China: 2013–2016

The rapid urbanization in China has brought with it some health benefits, but it also brought about a negative influence on the lifestyle of residents. We conducted this study to assess the change in hypertension-related knowledge and behavior from 2013 to 2016 among recently urbanized residents and determine their association with socioeconomic status (SES). This research used data from two cross-sectional studies conducted in Hezuo community in Chengdu, Sichuan province of China. A total of 2268 and 2601 individuals, respectively, participated and completed standard questionnaires. According to the results, the median (IQR) scores of health knowledge was 1 (0,3) and 3 (1,5), respectively, (p < 0.001) and the median (IQR) scores of health behavior was 6 (5,6) and 5 (5,6), respectively, (p < 0.001) in 2013 and 2016. The rate of sufficient knowledge increased from 8.8% to 18.1% (p < 0.001), while the rate of correct behavior decreased from 54.5% to 45.5% (p < 0.001) in three years. Logistic regression analysis showed that higher education was associated with sufficient hypertension-related knowledge (p < 0.05), and those with higher education, unemployment, and retirement were more likely to have sufficient behavior (p < 0.05). The impact of SES on knowledge was stable between 2013 and 2016. The behavior difference between the middle school educated and the illiterate increased from 2013 to 2016 (p < 0.05), and the behavior difference between the unemployed and manual workers decreased from 2013 to 2016 (p < 0.05). Our results revealed that hypertension-related knowledge improved with no corresponding improvement in self-reported behavior among recently urbanized residents from 2013 to 2016. Organizational strategy should be implemented to improve health education on knowledge, and what is more, translate knowledge into behavior. All these measures should be given more attention to the lower educated and manual workers among recently urbanized residents to eliminate the SES disparity.


Introduction
It is estimated that 80% of deaths and 70% of disability-adjusted life-years lost are caused by non-communicable diseases (NCDs) in China [1]. Hypertension is the most common NCD and its main complication, cardiovascular diseases (CVD), has a high mortality and morbidity [1,2]. Based on and this is the first study to explore hypertension-related knowledge and behavior among recently urbanized residents.

Setting and Participants
This research used data from two cross-sectional studies conducted in Hezuo community in Chengdu, Sichuan province in 2013 and 2016. The Hezuo community is located in the Chengdu High-tech Development Zone which started to change from rural to urban in 2006, the majority of its residents are still in the rural urbanization process. The gross domestic product (GDP) of high-tech development zones was 18.22 billion yuan in 2006, and it increased to 103.97 billion yuan in 2013, and 143.65 billion yuan in 2016 [27][28][29]. Residents aged ≥18 years and lived in the Hezuo community for no less than 6 months qualified for this study. A multistage stratified random sampling survey was used to acquire the sample. First, we randomly selected six of nine community neighborhood committees in the Hezuo district. Next, five building units were randomly selected from each selected community neighborhood committee. Then, we used systematic random sampling to choose 90 and 100 representative families in the chosen building units, respectively, in 2013 and 2016. The sample of this study was based on family unit and each family randomly chose one resident as the respondent. On the basis of informed consent, each participant took a face-to-face interview guided by an appropriately-trained investigator and completed the questionnaire anonymously. There were 2268 and 2601 residents that participated in the survey in 2013 and 2016, respectively (the effective response rate was 84.0% and 86.7%). The Public Health School of Sichuan University Institutional Review Board approved the protocol (Project identification code: H160311).

Measures
The questionnaire was designed according to the community health assessment standardized questionnaires in China [30].

SES
This study selected three most representative SES measures: education, occupation, and monthly income. Education was defined as the highest level of qualification that has been completed and was divided into five levels: illiterate, elementary school, middle school, senior school, or technical secondary school, junior college, and above. Occupation was grouped into four categories, which were manual work, nonmanual work, unemployment, and retirement. Monthly income was classified as less than 1000CNY, 1000-2499CNY, 2500-3999CNY, and more than 4000CNY.

Other Sociodemographic Characteristics
There are other sociodemographic characteristics contained in study including gender, age, marital status, and self-reported hypertension.

Hypertension Related Knowledge
The instrument of hypertension related knowledge contains eight items: one question about the recommended daily intake of salt for adults, one question about whether hypertension patients should take medicine for life, and another six questions about risk factors of hypertension. Each correct answer gains 1 point and the wrong answer gains 0 points, the total score of hypertension related knowledge ranges from 0 to 8 and sufficient hypertension related knowledge was defined as the correct rate higher than 80%.

Hypertension Related Behavior
The instrument of hypertension related behavior contains seven items: smoking (regularly, quit, or never), alcohol use (regularly, occasionally, or never), salt intake (low or high), smoked or pickled food intake (low or high), exercise (regularly, occasionally, or never), weight control (regularly, occasionally, or never), and measure blood pressure (regularly, occasionally, or never). Healthy behavior included quit or never smoking, occasionally or never drinking, low salt intake, low smoked or pickled food intake, exercise regularly, weight control, and measure blood pressure regularly. Each healthy behavior gains 1 point and the unhealthy behavior gains 0 points, the total score of hypertension related behavior ranges from 0 to 7 and sufficient hypertension related behavior was defined as a healthy behavior formation rate higher than 80%.

Statistical Analysis
First, we used descriptive statistics to present the SES and other demographic characteristics of participants, including education, occupation, monthly income, gender, age, marital status, and self-reported hypertension. Second, both the overall score level of hypertension related knowledge and behavior and the correct rate of each item were presented. We used the number and percentages to describe categorical variables and the median and interquartile range to describe non-normal continuous variables. Changes in each item was analyzed by logistic regression and presented by the odds ratios (ORs) and 95% confidence intervals (CIs), changes in the overall score was tested by the Wilcoxon rank sum test. Next, we used logistic regression analysis to estimate the relationship between SES and sufficient hypertension related knowledge and behavior in two models. In model 1, we tested the main effects of education, occupation, monthly income, and other confounding influencing factors. In the model 2, the interaction effect between survey years and SES was analyzed. p < 0.05 was considered statistically significant. Data from this study was recorded with Epidata 3.1 (The Epidata Association, Odense, Denmark) and statistical analyzed by SPSS 23.0 (SPSS Inc., Chicago, IL, USA). Table 1 shows the demographic characteristics of 2268 respondents in 2013 and 2601 respondents in 2016. The average age of respondents was 46.85 ± 17.08 years in 2013 and 46.85 ± 17.13 years in 2016. About half of respondents were female (52.8% and 53.4%, respectively). Most of respondents were married (82.2% and 83.5%, respectively). Those with a degree below middle school were 39.3% and 31.9%, respectively, and those with a junior college and above degree were 10.4% and 20.5%, respectively. Most of respondents were unemployed (41.9%) in 2013 and engaged in nonmanual work (39.5%) in 2016. The monthly income level of most respondents was concentrated in 1000-2499 CNY (63.4% and 58.5%, respectively). 10.9% and 11.5% of respondents self-reported suffering from hypertension in 2013 and 2016. Differences are statistically significant in education, occupation, and monthly income between 2013 and 2016.

Results
The correct rate and changes of hypertension related knowledge and behavior among respondents are presented in Table 2. In the field of knowledge, the lowest correct question was "people should consume no more than 6 g of salt per day" which less than 10% of respondents answered correctly in 2013 (4.7%) and 2016 (7.5%). A significant increase in the correct rate of hypertension related knowledge was found for the risk factors including obesity, excessive salt intake, mental stress, smoking, and lack of exercise (p < 0.001) from 2013 to 2016. In the field of health behavior, the highest correct behavior was weight control both in 2013 (96.3%) and 2016 (93.6%). The lowest correct behavior was exercise regularly (53.4%) in 2013 and take blood pressure regularly (27.0%) in 2016. A significant increase (p < 0.001) in correct rate of hypertension related behavior was found in eating little smoked or pickled food and exercising regularly, while a significant decrease (p < 0.001) was found in drinking, weight control, and taking blood pressure regularly from 2013 to 2016.  Table 3 describes the overall score of hypertension related knowledge and behavior. The median (IQR) scores of health knowledge was 1 (0,3) and 3 (1,5) in 2013 and 2016, respectively, with statistical significance (p < 0.001). The median (IQR) scores of healthy behavior was 6 (5,6) and 5 (5,6) in 2013 and 2016, respectively, with statistical significance (p < 0.001).      Table 4 shows the association between SES and hypertension related knowledge and behavior. In the field of knowledge, respondents with higher education were more likely to have sufficient knowledge after considering gender, age, marital status, and self-reported hypertension (p < 0.001). And respondents who self-reported suffering from hypertension were more likely to have sufficient knowledge (p < 0.001). As shown in model 2, the interaction effects between SES and knowledge were not significant, suggesting that the impact of SES on knowledge was stable between 2013 and 2016. In the field of behavior, the higher educated, unemployed, and retired individuals were more likely to have sufficient behavior after considering gender, age, marital status and, self-reported hypertension (p < 0.001). Respondents who self-reported suffering from hypertension, female, older age, divorced, or widowed were more likely to have sufficient behavior (p < 0.001). At the same time, respondents with sufficient knowledge were more likely to have sufficient behavior (p < 0.001). As shown in model 2, the behavior difference between the middle school and the illiterate increased from 2013 to 2016 (p < 0.05). The behavior difference between the unemployed and manual workers decreased from 2013 to 2016 (p < 0.05).  Table 4 shows the association between SES and hypertension related knowledge and behavior. In the field of knowledge, respondents with higher education were more likely to have sufficient knowledge after considering gender, age, marital status, and self-reported hypertension (p < 0.001). And respondents who self-reported suffering from hypertension were more likely to have sufficient knowledge (p < 0.001). As shown in model 2, the interaction effects between SES and knowledge were not significant, suggesting that the impact of SES on knowledge was stable between 2013 and 2016. In the field of behavior, the higher educated, unemployed, and retired individuals were more likely to have sufficient behavior after considering gender, age, marital status and, self-reported hypertension (p < 0.001). Respondents who self-reported suffering from hypertension, female, older age, divorced, or widowed were more likely to have sufficient behavior (p < 0.001). At the same time, respondents with sufficient knowledge were more likely to have sufficient behavior (p < 0.001). As shown in model 2, the behavior difference between the middle school and the illiterate increased from 2013 to 2016 (p < 0.05). The behavior difference between the unemployed and manual workers decreased from 2013 to 2016 (p < 0.05).

Discussion
It was demonstrated that an improvement in knowledge and changes in behavior can largely prevent and control hypertension [31]. China implemented the community-based comprehensive prevention and control for hypertension involved in EPHS in urban areas which focuses on risk factor control [32]. As a major means of controlling risk factors, health education activities include effective communication between general practitioners and residents, distribution of knowledge manuals, health education lectures provided by CHCs, and so on. Along with the recent development of community health service, the recently urbanized residents can use convenient and economical hypertension services from CHCs, e.g., many CHCs provide free blood pressure measurement services to residents in China [33].
The level of hypertension related knowledge among recently urbanized residents has improved in three years (p < 0.001). With the development of urbanization and the information technology revolution mass media has transmitted and shared various health information that allow people to live in a health related knowledge-rich society [34]. Excellent medical service resources concentrated in cities has increased the availability of high quality health related knowledge [20]. However, the sufficient rate of hypertension related knowledge was only 18.8% and the correct rate of some items was still below 50% in 2016. Only 37.8% (2016) respondents were aware that mental stress was a risk factor for hypertension. As a response of the human body to the stimulation of psychological and physiological factors in the environment, negative mental stress can significantly increase hypertension risk [35]. And for recently urbanized residents, adapting to the pressures of strange societies or different lifestyles may lead to or aggravate their mental health problems [36]. Only 37.8% (2016) respondents were aware of smoking as a risk factor for hypertension. Smoking places people at serious risk of atherosclerosis disease, the recently urbanized residents probably faced a greater smoking risk after transforming into a higher income and higher consumption environment, it is also influenced by their life stress and mental problems [37,38]. There were only 35.4% (2016) respondents who recognized adequately that regular activity contributes to energy consumption and the stability of cardiovascular system [39]. Only 42.3% (2016) of respondents realized that people with hypertension should take drugs for life once they are diagnosed to influence blood pressure control [40]. Despite 62.1% (2016) of respondents recognizing the dangers of excessive salt intake, only 7.5% (2016) knew that adults should consume no more than 6 g of salt per day; this situation has not changed in three years.
The level of hypertension related behavior among recently urbanized residents has decreased in three years, the overall health behavior formation rate in 2016 (45.5%) is lower than in 2013 (54.5%) (p < 0.001). Less respondents engaged in appropriate alcohol use, weight control, and blood pressure measuring regularly in 2016 than in 2013 (p < 0.001). Under the process of urbanization, alcohol use in China is increasing rapidly, and the average annual alcohol consumption of Chinese people over the age 15 has increased from 0.4 L in 1952 to 4.9 L in 2009 [41]. One reason for the change in drinking behavior in China is that it is mainly used to promote social intercourse or business meetings which are more common in work and life in the modern city [26]. The 2013 China Chronic Disease and Risk Factor Surveillance study revealed the drinking rate of urban residents (39.1%) is higher than that of rural residents (35.4%) [42]. Weight control of residents is also constrained by more factors accompanies with urbanization for several reasons. Living in urban areas can increase the odds of doing nonmanual work and using vehicles for transportation which reduce energy consumption [43,44]. Mobile phones, television, and the Internet fill the leisure time of urban residents and increase their sedentary behavior [45]. In addition, the measurements of blood measure decreased strikingly in three years, the possible reasons for this situation should be further explored in the later research.
Overall, hypertension related knowledge has improved for the recently urbanized residents, while the related behavior gets worse from 2013 to 2016. This result verifies that the effect of urbanization on residents is double-edged. In addition, this result is consistent with the theory that it is the ability to respond to knowledge, rather than the knowledge itself, that promotes changing behavior [13]. Health knowledge forms the basis for behavior, but behavior can also be influenced by many factors, such as health belief, self-efficacy, and the individual's social environment [46][47][48].
There were 39.3% and 31.8% of respondents respectively in 2013 and 2016 with a lower education than middle school, which showed lower degree of education among recently urbanized residents. After adjusting for potential confounding factors, the higher education respondents were found more likely to have the sufficient hypertension related knowledge and behavior which is consistent with previous studies [49,50]. Education develops the ability of individuals to access, evaluate, and use information, enabling them to seek valuable health knowledge [51]. Well-educated people are easier to adopt health behavior based on their conscious of prevention [52,53]. Education also brings resources to improve health behavior including a better social support and a feeling of being in control of one's life [54]. Poor education not only has an effect on smoking, alcohol use, salt intake, and other risk behaviors, but also has difficulty in overcoming problems existed in the process of changing behavior, for example, Naik et al. found that poor education contributed to continued smoking and continued physical inactivity [55].
The result of this study also examined the relationship between occupation and hypertension related behavior. Unemployment and retirement respondents had higher odds of hypertension related behavior than manual workers. Existing studies of the influence of unemployment and retirement on health showed conflicting results [56,57]. Although employment is always thought to have a positive impact on health, some studies reflected that unemployment is not always harmful for health [58]; retirement and unemployment release individuals from occupational stress, at the same time, increased leisure time to give them more opportunities to form health promoting behavior [59].
We also find the impact of SES on hypertension-related knowledge was stable while it was unstable on behavior. The behavior difference between middle school educated and illiterate individuals increased from 2013 to 2016 (p < 0.05), and the behavior difference between unemployed individuals and manual workers decreased from 2013 to 2016 (p < 0.05), this suggests that we should pay more attention to the role of education in our actions to promote hypertension-related behavior. In addition to the SES factors, other social demographic influencing factors were also assessed in this study. Our findings reported that self-reported hypertension was associated with hypertension related knowledge which is consistent with the results of Xia Li et al. [14]. Respondents who self-reported suffering from hypertension, who were female, older age, divorced, or widowed were more likely to have sufficient behavior. These findings suggest that measures to improve hypertension-related knowledge and behavior should consider the difference in age, gender, marital status, and hypertensive status.
In general, the performance of community-based comprehensive prevention of hypertension was relatively effective in improving hypertension related knowledge while ineffective in correcting behavior from 2013 to 2016. On one hand, we attributed this result to the double influence of urbanization on recently urbanized residents, and on the other we have to reconsider the measures taken by EPHS. Continued health education is essential to enrich knowledge as education is the key-influencing factor, however, EPHS should take more effective measures to promote residents' ability to turn knowledge into action. In addition, the present study focused on differences in hypertension related knowledge and behavior among those in various SES and other demographic factors. Since the results reflected social inequality, the most important thing is that EPHS should maintain its welfare character as a fundamental policy of the state by paying more attention to recently urbanized residents, especially individuals with lower education and manual workers.
The major limitations of this study should be addressed. Firstly, it is difficult to prove a causal relationship among the factors based on a cross-sectional study. Secondly, the self-reported hypertension data may be lower than the actual ones.

Conclusions
From 2013 to 2016, hypertension related knowledge improved while behavior had no significant improvement among recently urbanized residents. Despite the improvement, the hypertension related knowledge of residents is still poor, especially in daily salt intake, medication regimens, and some risk factors of hypertension (mental stress, smoking, and lack of exercise). In the context of urbanization, socioeconomic inequality acted on hypertension related knowledge and behavior. The measures devoting to strengthening health education and transforming knowledge into behavior should be taken into account in the recently urbanized residents. At the same time, emphasis should be placed on the individuals with a low level of education and engaged in manual work.