Prevalence, awareness, treatment, and control of dyslipidemia in Chinese adults: a systematic review and meta-analysis

Background Researchers have conducted a considerable number of epidemiological studies on dyslipidemia in China over recent years. Nevertheless, a representative study to comprehensively appraise for the epidemiological status of dyslipidemia is still lacked. This meta-analysis is intended to explore the pooled prevalence, rates of awareness, treatment, and control of dyslipidemia among adults in Chinese Mainland. Materials and methods A systematic review was performed on relevant cross-sectional studies published since January 2012 by searching six authoritative literature databases. Meta-analyses were conducted in included studies based on a random-effect model to summarize the epidemiological status of dyslipidemia in China. A potential source of heterogeneity was detected by subgroup analysis and meta-regression. Publication bias was assessed by Egger's test and funnel plots. A sensitivity analysis was conducted to examine the study quality's influence on the pooled estimate of prevalence and rates of awareness, treatment, and control. Results Forty-one original researches with a total of 1,310,402 Chinese participants were finally included in the meta-analysis. The prevalence, rates of awareness, treatment, and control of dyslipidemia were 42.1%, 18.2%, 11.6%, and 5.4%, respectively. With a pooled prevalence estimate at 24.5%, low HDL-C was the most prevalent among various dyslipidemia types, followed by hypertriglyceridemia (TG) (15.4%), hypercholesterolemia (TC) (8.3%), and high LDL-C (7.1%). The pooled prevalence of elevated serum lipoprotein(a) [Lp(a)] was 19.4%. By gender, the prevalence of dyslipidemia was 47.3% in males and 38.8% in females. Subgroup analyses revealed that the prevalence in southern and urban areas were higher than their counterparts. Females and population in urban areas tended to possess higher rates of awareness, treatment, and control. Meta-regression analyses suggested that the year of screening influenced prevalence estimates for dyslipidemia. The impact of the study's quality on the pooled estimates is insignificant. Conclusion Our study suggested a severe epidemic situation of dyslipidemia among adults in Chinese Mainland. More importantly, the awareness, treatment, and control rates were extremely low, revealing that dyslipidemia is a grave health issue. Consequently, we should attach more importance to the management of dyslipidemia, especially in economically underdeveloped areas. Systematic review registration PROSPERO [CRD42022366456].


Introduction
Cardiovascular disease (CVD) is one of the prior sources of global disease burden, as well as the main cause of premature death for Chinese residents (1)(2)(3). In China, cardiovascular disease is the leading cause of total deaths among residents while its prevalence and mortality are still increasing (4, 5). In 2019, cardiovascular diseases were responsible for 46.76% of the total deaths in rural areas and 44.26% in urban areas of China (6), which means that approximately two in five deaths were caused by cardiovascular diseases on average. Against the background of population aging trend and progressively prevalent metabolic risk factors such as hypertension, hyperglycemia, central adiposity, and dyslipidemia (7,8), the disease burden caused by cardiovascular diseases kept increasing, which has developed into a critical concerning public health issue (9)(10)(11)(12). Atherosclerotic cardiovascular disease (ASCVD) includes ischemic heart disease and ischemic stroke (13)(14)(15). Because of the same arterial pathological characteristics and risk factors, ASCVD is increasingly regarded as a special type of cardiovascular disease in Chinese and international cardiovascular disease prevention guidelines (16)(17)(18)(19)(20). In addition, ASCVD is the pattern that causes the most deaths among all kinds of cardiovascular diseases. In 2016, it caused about 2.4 million deaths in China, which amounts to more than 60% of all cardiovascular disease deaths as well as 25% of all causes of death (21,22). Therefore, prevention and treatment of ASCVD are the top priorities of management for cardiovascular disease.
Characterized by hypercholesterolemia (high TC), hypertriglyceridemia (high TG), low HDL-C, or high LDL-C, dyslipidemia is a crucial risk factor for ASCVD and one of the three major risk factors that the Healthy China 2030 plan focuses on (23)(24)(25)(26)(27), which emphasizes the strategic role of health in China's development and outlines the major principles to achieve this. The disease burden caused by dyslipidemia in China has shown a significant growth trend in recent years (28)(29)(30). In 2017, a study showed that 862,759 deaths in China could be attributed to high LDL-C, accounting for 8.25% of all causes of death and 19.71% of cardiovascular disease deaths (31). Economist Intelligence Unit (EIU) report (2018) indicated that CVD had the economic burden of USD 21.7 billion in direct and indirect costs annually in China, of which more than 12% is due to dyslipidemia (32).
Although "co-management of hypertension, diabetes, and hyperlipidemia" was clearly proposed in the Healthy China 2030 plan, the management of dyslipidemia is far from ideal for hypertension and diabetes. The 2017 China Cardiovascular Health Index study showed the prevalence of dyslipidemia (33.7%) exceeded hypertension (26.0%), and diabetes (9.7%), while its awareness rate (14.5%), treatment rate (7.9%), and control rate (5.4%) were all below the corresponding levels of hypertension and diabetes (33). Compared with the standardized management of hypertension and hyperglycemia, that of dyslipidemia is still in a neglected position, and the public's awareness and attention to it need to be strengthened (34)(35)(36)(37)(38).
In recent years, researchers have carried out surveys on the epidemiological status of dyslipidemia in China, and many related data have been disclosed. Due to divergent research backgrounds and other reasons, the results varied widely between different studies (39). A comprehensive evaluation on the dyslipidemia epidemiology nationwide in China which may promote our understanding of the epidemiological status of dyslipidemia as well as benefit future research and policy formulation is needed. Consequently, we performed the meta-analyses to comprehensively synthesize the prevalence and management status of dyslipidemia among adults in Chinese Mainland.

Search strategy
Based on and Meta-analysis of Observational Studies in Epidemiology (MOOSE) guidelines (40) and the PRISMA statement (41), we conducted the systematic review. Studies on dyslipidemia from six databases were searched, including Web of Science, Embase, PubMed, WanFang, CNKI, and Chinese BioMedical Literature Database. The search strategy was based on a conjunction of "dyslipidemias," "hyperlipidemias," "epidemiology," "prevalence," "awareness rate," "treatment rate," "control rate," "cross-sectional study," and so on. The detailed literature retrieval strategy for each database can be found in Supplementary Material. Only studies published in English and Chinese between 1 January 2012 and 31 January 2023 were included.

Inclusion and exclusion criteria
The following information concerning dyslipidemia in Chinese population must be included in eligible studies: prevalence, awareness rate, treatment rate, and control rate. In addition, studies reported that the prevalence of elevated lipoprotein(a) [Lp(a)] was also included.
Eligibility criteria were set as followed: (1) research types: original cross-sectional studies; (2) study participants: Chinese adults; (3) the criteria of prevalence, awareness, treatment, and control rates of dyslipidemia were set based on the latest guideline of Prevention and Treatment for Dyslipidemia in Chinese adults (42).
The exclusion criteria we applied were the following: (1) animal research, (2) non-cross-sectional study, (3) study with a sample size lower than 500, and (4) study on specific occupational groups.

Study identification and data extraction
After combining all searched articles in EndNote X9 and removing duplicates, two reviewers (QX and YC) independently scanned the titles and abstracts of the retrieved studies for possible eligible studies. Then, the two reviewers independently evaluated the full texts of all potential eligible studies. After discussing and determining the final list of the included studies, QX and YC extract the relevant information by a standardized Xia et al. 10.3389/fcvm.2023.1186330 Frontiers in Cardiovascular Medicine data collection form, respectively. The extracted information was cross-checked, and a third reviewer (ZY) was responsible for determining any unsettled discrepancies. The two reviewers (QX and YC) independently extracted the data from studies including but not limited to regular information (such as title, author, and the publication year), characteristics of study (such as study area and number of participants), and characteristics of participants (such as age range, gender, and residential area). Then, the core information was extracted, i.e., the prevalence and rates of awareness, treatment, and control reported in each study. We also extracted these data stratified by age group, sex, region area, and year of screening.
According to the observational study criteria recommended by AHRQ, QX and YC independently evaluated the quality of the included studies (43). With a full score at 11 points, each study was grouped according to its own score. The grouping rules are as follows: good (above 7 points), medium (4-7 points), and poor (below 4 points). In addition, the risk of bias (ROB) of the included studies was evaluated based on the results' quality.

Statistical analysis
We calculated the pooled rates of prevalence, awareness, treatment, and control and corresponding 95% confidence intervals (95% CI) with a systematic analysis approach. Heterogeneity among studies was examined by Cochran's Q test and I 2 statistic. The I 2 statistic value at 25%, 50%, and 75% indicated a low, moderate, and high degree of heterogeneity, respectively (44,45). The random-effect model will be used if the result suggested a high degree of heterogeneity (I 2 > 50%). Otherwise, the fixed-effect model will be used (46).
To address heterogeneity between studies, subgroup analyses by age group, sex, geographic region, and the year of screening were performed. Afterward, a meta-regression was conducted, and the added variables include sex ratio (males vs. females), the year of screening, geographic area (southern vs. northern China), studies' quality score, and sample size. Finally, we performed a sensitivity analysis by evaluating the influence of the study's quality on the pooled estimates. Funnel plots and Egger's test were used to assess the risk of publication bias. We set the significance level at a P < 0.05. Stata and SPSS software were used to perform statistical analyses.

Search results
A total of 10,379 studies from all databases were searched and gathered in EndNote X9 software. Initially, 3,674 duplicates were removed, and then 6,161 studies were eliminated after reading the titles and abstracts, leaving 544 possibly qualified articles. Studies using other diagnostic criteria, reporting incomplete data or with a sample size of <500, were further excluded after referring to the full text. Finally, 41 studies  were included for analysis which involved a total of 1,310,402 Chinese adults. The search selection process is displayed in Figure 1.  The results of quality evaluation of the included studies were displayed in Table 2. A total of 29 studies scored above 7 and were consequently rated as high-quality, while 12 studies were rated as medium-quality, and no studies of low-quality were observed. The studies' average quality score was 8.07, while the standard deviation was 1.09. Since no study was evaluated as low-quality, all studies were included for further analyses. Figure 2 showed the summary plots of assessment for risk bias.

Prevalence of dyslipidemia
Twenty-five articles detailed the dyslipidemia prevalence of their population (see Table 3). The meta-analysis suggested that the pooled prevalence of dyslipidemia among adults in Chinese Mainland was 42.1% (95% CI: 39.2%-44.9%), while the heterogeneity between studies was extremely high (I 2 = 99.8%, P < 0.001). The forest plots of the pooled prevalence and rates of awareness, treatment, and control were displayed in Figure 3, and the corresponding funnel plots were shown in Figure 4. The Egger's test and funnel plots ( Figure 3A) demonstrated that no significant publication bias on the prevalence of dyslipidemia was found (P = 0.996).  Summary plot of risk bias. The pooled prevalence of different types of dyslipidemia varied widely. HDL-C was the highest at 24.5% (95% CI: 21.0%-28.1%), followed by hypertriglyceridemia at 15.4% (95% CI: 12.5%-18.3%), and the pooled prevalence of hypercholesterolemia and LDL-C were lower at 8.3% (95% CI: 6.9%-9.6%) and 7.1% (95% CI: 5.9%-8.2%), respectively. The forest plots for the pooled prevalence of different types of dyslipidemia were displayed in Figure 5. Corresponding funnel plots were shown in Figure 6. We also calculated the pooled prevalence of elevated Lp(a), which was defined as a serum Lp(a) value of >30 mg/dl according to the latest guideline of Prevention and Treatment for Dyslipidemia in Chinese adults (42

Dyslipidemia awareness, treatment, and control
According to the related information released in 12 surveys, we arrived at pooled rates of awareness, treatment, and control of dyslipidemia at 18.2%, 11.6%, and 5.4%, respectively (see Table 4). The funnel plots (see Figure 4) and the Egger's tests demonstrated that no significant publication bias on the rates was observed (P = 0.072, 0.110 and 0.958, respectively).
The pooled rates of awareness, treatment, and control in females (18.5%, 12.3%, and 6.4%, respectively) were higher than those in males (15.7%, 9.7%, and 4.2%, respectively). Urban residents had higher pooled rates of awareness, treatment, and control than the rural residents (P < 0.001). Southern residents had higher pooled rates of treatment and control (13.3% and 6.8%, respectively) than the northern populations (9.4% and 4.6%, respectively), while the difference of awareness rates was not statistically significant among them (P = 0.055). The pooled rates of awareness and control

Sensitivity analysis and meta-regression
In the sensitivity analysis, we excluded four citations with a quality score of 6 which was the lowest among all studies. After omitting these studies, we discovered a slight decrease in the pooled prevalence (from 42.1% to 41.0%, P = 0.653). Funnel plot combined with Egger's test (P = 0.639) indicated that no significant publication bias was noted. The pooled prevalence of hypercholesterolemia, hypertriglyceridemia, HDL-C, and LDL-C changed to 8.7%, 15.9%, 23.5%, and 7.6% from 8.3%, 15.4%, 24.5%, and 7.1%, respectively. The pooled rates of awareness, treatment, and control also changed slightly (from 18.2%, 11.6%, and 5.4% to 18.8%, 11.8%, and 5.2%, respectively) after omitting these studies. The results indicated that the pooled prevalence and rates of awareness, treatment, and control had good stability.
Finally, a meta-regression was performed to address the high level of heterogeneity between studies (I 2 = 98.7%-100.0%). Five variables (sex ratio, the year of screening, geographic area, studies' quality score, and sample size) were included in the analyses. The results of meta-regression demonstrated that only the year of screening variable had a significant impact on the heterogeneity (P = 0.011) (see Table 5).

Discussion
For the first time, this study comprehensively summarized and analyzed the epidemiological studies on dyslipidemia in recent years and explored its potential influencing factors. With a total number of 1,310,402 participants from different cross-sectional studies, our systematic review incorporated 41 studies conducted in multiple provinces of Chinese Mainland over the past decade. The results of the meta-analyses suggested a high-degree of prevalence of dyslipidemia in Chinese adults along with unacceptably low rates of awareness, treatment, and control. Results from this study would be a timely alarm since a comprehensive evaluation on the dyslipidemia epidemiology nationwide in China which may promote our understanding of its epidemiological status and benefit future research and policy formulation is still lacking. In 2014, the pooled prevalence of dyslipidemia among adults in Chinese Mainland is 42.1%, which is very close to the research results of Huang et al. (41.9%) (88). The deepening of aging population degree and the great changes   (65,98,99), the more prevalent types of dyslipidemia in China were still HDL-C and hypertriglyceridemia. As the risk of ASCVD will be increased by all types of dyslipidemia, it is essential to treat them with applicable interventions, both clinical and non-clinical (25, 88).
The pooled estimates of dyslipidemia prevalence between different genders, age groups, and regions were calculated and analyzed. We found that males had higher pooled prevalence than females (47.3% vs. 38.8%; P < 0.001), which is consistent with the previous studies' results (100)(101)(102)(103). Many factors can attribute to such difference, for instance, males are more likely to possess unhealthy behaviors than females, including but not limited to lack of vegetables and fruits in diet and fewer physical activity. Such factors could contribute to higher prevalence of many metabolic diseases in males than females, including dyslipidemia (104)(105)(106)(107)(108). Researches showed that estrogen changes the vascular permeability by increasing nitrous oxide production which retains a healthful lipoprotein profile (109,110). Nonetheless, such protective mechanisms will disappear after menopause, resulting in an ascending risk for suffering Frontiers in Cardiovascular Medicine cardiovascular diseases. For example, the levels of TC, TG, LDLC, and VLDL-C aming postmenopausal women appear an upward trend, while that of HDL-C significantly decreases. Consequently, the prevalence of dyslipidemia in women often showed a dramatic increase after menopause, even surpassing that of men of the same age (60, 62, 65, 99). The systematic review revealed that southern and urban residents were more likely to suffer from dyslipidemia than their counterparts, which was a result in line with previous findings (111,112). Another study in China suggested that economically developed areas (e.g., southeast area of China) tend to possess higher burden of dietary chronic conditions, such as dyslipidemia and obesity (113). In addition, healthcare facilities are more accessible for residents in highly urbanized and economically vibrant cities, resulting in more diagnosis and hence higher reporting rate of dyslipidemia in urban and southern areas. It is worth noting that in recent years, dyslipidemia has become more and more prevailing in Chinese youngsters. A cross-sectional study conducted in Wenzhou, Zhejiang Province, showed a prevalence at 34.11% among 7,859 young adults (114), which was much higher than the figure in Mainland China 10 years before that, as well as other Asian developing countries (115,116). In Beijing, a study of 3,249 children aged 6-18 years showed that the prevalence of dyslipidemia was 28.9%, higher than 18.8% in 2004 (117). The younger trend of dyslipidemia may be partly attributed to the westernization of young people's lifestyle, such as dietary patterns (114). It is indisputable that corresponding risk factor intervention project should be developed based on the characteristics of young people and further screening and management programs should be strengthened. One of the key findings of this study is a summary estimation of the pooled prevalence of elevated Lp(a) in Chinese adults, although it is not included in the definition of dyslipidemia. According to the latest guideline of Prevention and Treatment for Dyslipidemia in Chinese adults (42), we set the cutoff value at >30 mg/dl and got a pooled prevalence at 19.4%, which was much lower than that in the United States (35%) (118). As a novel lipid biomarker which could promote the formation of atherosclerosis and thrombosis, Lp(a) was considered the core pathogenesis of ASCVD and deemed as a reversible risk factor (119). Based on seven randomized controlled trials and 29,069 patients taking statin medication, one meta-analysis found that elevated Lp(a) can still increase the risk of CVD, despite a controlled LDL-C level (120). Due to the fact that the concentration level of Lp (a) in plasma is mainly caused by genetic factors, it is relatively stable throughout life (121). Funnel plots for the pooled prevalence of different types of dyslipidemia.

Frontiers in Cardiovascular Medicine
Chinese researchers recommended that people consider at least once measurement of Lp(a) to identify individuals who have inherited extremely increased levels of Lp(a) (≥180 mg/dl), which may bring an extremely high lifetime risk of ASCVD (119).
Publicizing the harm of dyslipidemia to the general public along with encouraging a healthier lifestyle is the basic strategy to prevent dyslipidemia and ASCVD. For patients with dyslipidemia, the focus is to control the blood lipid level to the normal range through therapeutic approaches, such as taking statins. Therefore, improving residents' awareness, treatment, and control of dyslipidemia is the key to effective management of dyslipidemia. The pooled rates of awareness, treatment, and control of dyslipidemia among Chinese adults are 18.2%, 11.6%, and 5.4%, respectively, which are slightly higher than the research results of Zhao et al. (10.9%, 6.8%, and 3.5%, respectively) in Chinese Mainland in 2010 but much lower than that reported in the United States (73.3%, 54.1%, and 35.7%) and also lower than the results from Argentina (37.3%, 36.6%, and 20.0%) and South Korea (29.4%, 17.0%, and 13.9%, respectively) (79,122,123). In accordance with previous studies (80,88), our study suggested that women had higher awareness rates of dyslipidemia than men and tend to receive corresponding treatment, which further contributed to better control of their condition. Various factors can account for such phenomenon, for instance, studies demonstrated that women more frequently seek medical services than men (124). In addition, women are more likely to have a health insurance as well as an ongoing source for primary care than men (125). Therefore, we suggested strengthening the promotion and management of dyslipidemia among male residents in China. In line with the previous studies (62, 80), our results showed that urban residents had significantly higher rates of awareness, treatment, and control than rural residents. Similarly, such urban-rural difference was found in a couple of findings from other countries (119,126,127). The differences in rates between urban and rural areas may be partly attributed to the gap between residents' economy and education   (129,130). Therefore, it is suggested to attach more importance to the dyslipidemia management in rural areas, such as increasing the availability of drugs. The significance of this study has been highlighted by the extremely low rates of awareness, treatment, and control discovered, which may imply some disparities and deficiencies in dyslipidemia management in China.  (129). The PURE study evaluated the usage of secondary prevention drugs for patients with CVD in communities of countries with different income levels (133). Results signified a positive relation between the rates of use for statins and the economic level of country. In addition, the use rate of statins in patients with coronary heart disease or stroke in China was the lowest (1.7%) among all secondary prevention drugs, which is lower than the level in North America, Europe, the Middle East, South America, Malaysia, and South Asia and only slightly higher than the level in Africa. Consequently, at the first step, we suggested to determine a patient journey of dyslipidemia with clear process and easy understanding and operation. Second, by improving the allocation of manpower and resources in primary medical institutions, it may help in making up for their vacancy in the management of dyslipidemia and gradually play their role in patient screening, initial diagnosis, referral, and follow-up management. Finally, a comprehensive management system should be established to prevent and control various cardiovascular diseases and their risk factors.
The quality of the studies included in this systematic review is generally high; consequently, the results of sensitivity analysis revealed that when the lowest score studies were excluded, only a slight impact on the results was achieved. Data from multiple provinces in Chinese Mainland with a large sample size was incorporated in our meta-analysis. However, some limitations in this study are shown as well. Firstly, due to limitations in data availability, the relationship between the prevalence of dyslipidemia and some factors cannot be explored by subgroup analyses and meta-regression. Therefore, this study is limited in exploring the influencing factors of dyslipidemia. Secondly, only epidemiological studies were included in our systematic review and meta-analysis, and the high degree of heterogeneity between studies is inevitable (134). However, subgroup analysis and meta-regression alleviate this issue to some extent. Finally, we cannot infer causality between dyslipidemia and other factors due to the cross-sectional design in all included studies.

Conclusion
Nearly half of Chinese adults suffered from dyslipidemia, while the most prevalent type of dyslipidemia was low levels of highdensity lipoprotein cholesterol. Males and urban residents had a higher prevalence of dyslipidemia than their counterparts. This study further suggested extremely low rates of awareness, treatment, and control for dyslipidemia in Chinese adults. The government should increase the financial and manpower support for primary medical institutions and implement effective programs to prevent and control dyslipidemia.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, and further inquiries can be directed to the corresponding author.

Funding
This research was funded by the Capital health development scientific research project (2021-1g-4251). The funding bodies did not play any role in the design of the study and collection, analysis, and interpretation of data and in writing this manuscript.

Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of Xia et al. 10.3389/fcvm.2023.1186330 Frontiers in Cardiovascular Medicine their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.