Article Text
Abstract
Introduction Cardiovascular disease (CVD) continues to pose a significant burden among the elderly population in China. There is a knowledge gap in the temporal trends, regional variations and socioeconomic inequalities among this vulnerable population.
Methods This study conducted cross-sectional and cohort analyses based on four survey waves of the China Health and Retirement Longitudinal Study among adults aged ≥60 years spanning 2011–2018 across 28 provinces. Cross-sectional analyses examined temporal trends, regional variations and socioeconomic inequalities in CVD prevalence. Cohort analyses identified individuals without CVD in 2011 and followed them up until 2018 to calculate CVD incidence. Generalised estimating equations (GEE) were employed to identify associated factors.
Results A total of 5451, 7258, 8820 and 11 393 participants were eligible for cross-sectional analyses, and 4392 and 5396 participants were included in cohort analyses of CVD and comorbid CVD. In 2018, the age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD was 31.21% (95% CI 27.25% to 35.17%) and 3.83% (95% CI 2.85% to 4.81%), respectively. Trend analyses revealed a significant increase in the adjusted prevalence from 2011 to 2018 (p for trend <0.001). There were substantial provincial variations in the adjusted prevalence of CVD and comorbid CVD. Higher socioeconomic status (SES) participants exhibited higher prevalence, and the concentration curves and concentration indices suggested persistent but narrowing inequalities in CVD and comorbid CVD across survey waves. Cohort analyses from 2011 to 2018 yielded overall CVD and comorbid CVD incidence densities of 17.96 and 2.65 per 1000 person-years, respectively. GEE results indicated increased CVD risks among older individuals, women, higher SES participants and northern residents.
Conclusion More efforts should be taken to optimise strategies for high-quality CVD prevention and management in China’s elderly population. Future interventions and policies should address age-specific and gender-specific, geographical, and socioeconomic disparities to ensure equitable access and outcomes for all.
- Cardiovascular disease
- Public Health
- Indices of health and disease and standardisation of rates
- Epidemiology
- Descriptive study
Data availability statement
Data are available in a public, open access repository. Publicly available datasets were analysed in this study. The data can be accessed on the CHARLS database (https://charls.charlsdata.com).
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- Cardiovascular disease
- Public Health
- Indices of health and disease and standardisation of rates
- Epidemiology
- Descriptive study
WHAT IS ALREADY KNOWN ON THIS TOPIC
There are limited studies examining the temporal trends and regional variations in CVD among the elderly population in China.
Socioeconomic status (SES) plays a significant role in the occurrence of CVD. However, with the rapid changing of China’s socioeconomic structure, it remains unclear on socioeconomic inequality and its temporal trend concerning CVD among the elderly population in China.
Cardiovascular disease (CVD) has remained the leading cause of death and disability-adjusted life-years in China, and the burden of CVD is even greater among the Chinese elderly population.
WHAT THIS STUDY ADDS
This is the first study to examine the temporal trends and regional variations in CVD, especially comorbid CVD, among the elderly population in China. We found that there was a significant increase in the age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD among the elderly population in China from 2011 to 2018, and the prevalence and changing patterns of CVD and comorbid CVD varied widely among provinces.
This study provides valuable insights into the interplay between SES and CVD prevalence among the Chinese elderly population. CVD and comorbid CVD tended to be more common among the elderly with higher SES, and socioeconomic inequalities were persistent but narrowed from 2011 to 2018.
This study offers a unique opportunity to evaluate the associaiton between key demographic factors and CVD/comorbid CVD. This assessment is particularly valuable given the significant demographic shifts that have transpired in China over recent decades.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
More efforts should be taken to optimise the strategy and provide high-quality healthcare to prevent and manage CVD for the elderly population in China.
Geographical variations and socioeconomic inequalities should be considered and addressed in future interventions and policies.
Introduction
Cardiovascular disease (CVD) has remained the leading cause of death and disability-adjusted life-years globally over the last 20 years and is killing more people than ever before, especially in the Western Pacific Region.1 In China, two out of five deaths are attributed to CVD,2 and the rapid ageing of the population further exacerbates the burden of CVD.3 Evidence indicates that the deaths from CVD among elderly individuals in China increased by 251% from 1990 to 2019,4 and the burden is compounded when the elderly individuals have one or more cardiac diseases.5 Additionally, with the rapid changing of China’s socioeconomic structure, socioeconomic disparities have contributed to new patterns in CVD prevalence.6 Therefore, it is crucial to comprehensively understand the epidemiology and socioeconomic characteristics of CVD among the elderly population in China to identify the key issues to be addressed and to implement evidence-based health strategies to effectively alleviate the substantial burden of CVD in China.
Although some studies have estimated the prevalence of CVD in China, most of them have only provided age-standardised data, without specifically addressing the CVD situation among the elderly population.3 6 7 As for those analysing CVD among elderly individuals in China, the results have limitations in terms of analysing comorbid cardiac diseases, exploring temporal trends and capturing regional variations within the country.8 9 Moreover, while studies suggest that socioeconomic status (SES) plays an important role in the prevalence of CVD,10 the existing studies conducted among Chinese elderly individuals focused on the relationship between SES and overall health status.11 12 It remains unclear about the socioeconomic inequality and its secular trend concerning CVD among the Chinese elderly population. To address the gap, we used data from the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative resource to explore the temporal trends in CVD prevalence, to investigate the regional variations and socioeconomic inequalities and to evaluate the association between pivotal demographic factors and incident CVD in China from 2011 to 2018.
Methods
Data sources and participants
The CHARLS is a nationally representative longitudinal study beginning in 2011 that was designed to provide reliable data for understanding the health status and socioeconomic determinants of individuals over age 45 in China.13 Employing a consistent, multistage, stratified probability-proportionate-to-size sampling strategy, samples with national representativeness were selected from 28 provinces in China and then were followed every 2–3 years.14 Three follow-up surveys were conducted in 2013, 2015 and 2018. In each follow-up survey, the original participants were revisited, and new participants were selected using the same methods mentioned above to make up for people who dropped out and ensure representativeness. Data were collected via one-to-one interviews with a structured questionnaire, and the overall response rate of the CHARLS was 80.5%. A detailed description of the CHARLS has been reported elsewhere.15
The sample selection process is outlined in figure 1. Participants were initially extracted from each CHARLS survey wave conducted in 2011, 2013, 2015 and 2018. Missing age and sex data were subsequently imputed using information from other survey waves. Then, in each survey wave, individuals were excluded if they had missing age and sex data or if they were aged <60 years. Furthermore, participants with missing CVD data across all survey waves were excluded, with any gaps in CVD data filled using information from previous survey waves when available. Finally, a total of 5451, 7258, 8820 and 11 393 participants were eligible for inclusion in the cross-sectional analyses examining CVD prevalence, temporal trends, regional variations and socioeconomic inequalities. To examine potential biases due to exclusions related to missing data on CVD, a χ2 test was employed to compare the socioeconomic distribution between excluded and included participants. A significant difference existed between the two groups, except for the socioeconomic distribution in 2011 (online supplemental table 1). Considering the potential non-response biases stemming from the different distribution between the excluded and included participants, we applied individual weights with the household and individual non-response adjustments provided by the CHARLS to ensure the generalisability of the estimates to the targeted population.
Supplemental material
Measures
CVD, encompassing a class of diseases, predominantly contributes to the mortality burden, with heart disease and stroke being the primary causes.16 Information regarding heart disease and stroke in the CHARLS was collected through the following questions: ‘Have you been diagnosed with heart attack, coronary heart disease, angina, congestive heart failure, or other heart problems by a doctor?’ and ‘Have you been diagnosed with stroke by a doctor?’ In this study, we defined CVD as the presence of at least one diagnosed condition of heart disease or stroke and defined comorbid CVD as having been diagnosed with both heart disease and stroke.
Health regions refer to legislated administrative provinces defined by the State Council of China to administer and deliver healthcare services to all residents. The CHARLS was conducted in 28 out of 34 provinces of China.
Previous studies mostly used education, consumption, income and occupation to construct SES.17–19 Considering the study’s focus on participants aged ≥60 years, the majority of whom were retirees, we employed education and consumption as indicators to measure SES. Education was used as a proxy for conveying a degree of social status, and consumption was used as an indicator of economic status.19 In the CHARLS, education was harmonised according to the 1997 International Standard Classification of Education (ISCED-97) codes,20 and coded as three categories (1, 2 and 3) corresponding to less than lower secondary education, upper secondary and vocational training, and tertiary education. As for consumption, we used the monthly per capita household consumption as the metric. Data on household consumption were collected from various sources, including both food consumption in the past month and non-food consumption in the past year, as investigated in the CHARLS. Monthly household consumption was calculated by aggregating these data. Thereafter, monthly per capita household consumption was derived by dividing monthly household consumption by the number of individuals residing in each household. For analytical purposes, monthly per capita household consumption was stratified into quartiles for each survey wave, with 1–4 representing the lowest (quartile 1) to the highest (quartile 4). Following the methodology employed in previous studies,21 22 we constructed SES using a composite score that summed education levels (1, 2 or 3) and monthly per capita household consumption quartiles (1, 2, 3 or 4). This approach assumed exchangeability between education and monthly per capita household consumption. The summed score, ranging from 2 to 7, was used to represent SES, with smaller values indicating lower SES. To further enhance the efficiency of analyses, we categorised the summed SES score into four groups: low,2 lower middle,3 4 upper middle5 6 and high SES.7
Statistical analyses
The statistical analyses were conducted by using Stata (V.17.0; StataCorp). A two-tailed significance level of less than 0.05 was considered statistically. To enhance comparability across diverse subgroups and survey waves, age-adjusted and sex-adjusted CVD prevalence was calculated through direct standardisation to the 2010 China Census population using the joint age (10-year interval) and sex groups,23 aligning prevalence estimates with the demographic structure of the entire population. Temporal trends in CVD prevalence were assessed using multivariable Poisson regression analysis,24–26 involving adjustments for age and sex in different configurations. We adjusted for age and sex simultaneously for overall participants, as well as independently adjusted for age within each sex group and for sex within each age group. P values for trend were then calculated using the contrast postestimation command in Stata. Temporal variations in regional CVD prevalence from 2011 to 2018 were quantified by computing the difference between the prevalence in 2018 and that in 2011 for each province.
Socioeconomic inequality in CVD was assessed using the concentration index, a recognised metric with the dual capacity to reflect health indicators within distinct socioeconomic strata and to adeptly capture shifting inequality trends over time.27–31 The concentration index ranges from −1 to 1, with negative values indicating a higher prevalence of CVD among lower SES groups and positive values denoting concentration among higher SES groups, while 0 suggests no SES-related inequality. To visually represent the concentration index, we employed the concentration curve (CC), which plots cumulative CVD prevalence against the cumulative proportion of the population ranked by SES.28 When the CC lies below the equality line, it signifies elevated CVD prevalence among higher SES groups.28 31 The concentration index and CC were computed using the conindex command in Stata. A joinpoint regression analysis was employed to assess significant changes in socioeconomic inequality trends for CVD prevalence (2011–2018), by estimating the average annual percentage change (AAPC) of concentration indices.32 33 The changes in concentration indices were determined through permutation test, and p values and the 95% CI by Monte Carlo method.32–34
We conducted a cohort analysis to assess the CVD incidence and explore the effects of pivotal demographic factors on CVD from 2011 to 2018. The set of factors investigated in this analysis included age, gender, region and SES. Participants without baseline CVD diagnosis (n=4 392) and comorbid CVD (n=5 396) were longitudinally followed up throughout 2013, 2015 and 2018 to track the incidence of CVD and comorbid CVD (figure 2). The incidence density of CVD and comorbid CVD was calculated, accounting for participants who remained in the study for all survey waves. We employed a generalised estimating equation (GEE) to examine the temporal association between baseline exposure factors and new cases of CVD and comorbid CVD identified in subsequent survey waves.35 Considering provinces with 0% prevalence of comorbid CVD, we used Firth’s penalised likelihood logistic regression to reevaluate the association between exposure factors and incident CVD,36 and performed a sensitivity analysis to compare the outcomes of Firth’s approach with those of GEE. The absence of statistically significant differences between the two methods reaffirmed the stability and reliability of the associations (p=0.142).
Patient and public involvement
The data analysed in this study were exclusively sourced from the CHARLS, conducted by the China National Development Research Institute of Peking University. Patients or the public were not involved in the design, conduct, reporting or dissemination plans of our research. Stringent measures were applied to safeguard data confidentiality and anonymity. Ethical protocols were strictly followed to protect participant privacy, with analyses conducted on aggregated and anonymised data. The findings of this study have important implications for the public, as they highlight the need for enhanced strategies and high-quality healthcare to prevent and manage CVD among the elderly population in China.
Results
In total of 5451, 7258, 8820 and 11 393 participants aged ≥60 years were included in cross-sectional analyses in 2011, 2013, 2015 and 2018, respectively. The largest age group was those aged 60–69 years (eg, 2018 wave: 62.39% (95% CI 62.38% to 62.39%)). The sample size and characteristics of the study sample by age group and sex for each survey wave were presented in online supplemental table 2. A total of 4392 and 5396 participants at risk (ie, were not diagnosed with CVD and comorbid CVD, respectively) in 2011 were included in the cohort analyses.
The CVD prevalence and trend analysis
At the most recent wave in 2018, the overall crude prevalence of CVD and comorbid CVD was 31.21% (95% CI 27.25% to 35.17%) and 3.83% (95% CI 2.85% to 4.81%), respectively. The age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD overall and by age group and sex was shown in figure 3. The overall age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD was 31.01% (95% CI 28.54% to 33.48%) and 3.79% (95% CI 3.22% to 4.36%) in 2018, respectively; and both of them tended to be higher among women. Additionally, in 2018, the adjusted prevalence of CVD and comorbid CVD was the highest among participants aged 70–79 years, followed by the adjusted prevalence among participants aged ≥80 years.
From the baseline 2011 to the most recent 2018 survey wave, there were statistically significant increases in the age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD. Temporal trends in the age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD by age group and sex were also shown in figure 3. In terms of CVD, the overall age-adjusted and sex-adjusted prevalence increased from 20.77% (95% CI 18.99% to 22.55%) in 2011 to 31.01% (95% CI 28.54% to 33.48%) in 2018 (P for trend<0.001). In terms of comorbid CVD, the overall age-adjusted and sex-adjusted prevalence increased from 1.22% (95% CI 0.83% to 1.61%) in 2011 to 3.79% (95% CI 3.22% to 4.36%) in 2018 (p for trend<0.001). From 2011 to 2018, the estimated total number of older adults diagnosed with CVD and comorbid CVD increased by 18.19 million (49.30%) and 4.56 million (210.66%), respectively.
Regional variations
In 2018, the province-level age-adjusted and sex-adjusted prevalence of CVD varied widely from 13.08% (95% CI 5.65% to 20.51%) in Guangdong to 64.66% (95% CI 56.55% to 72.77%) in Heilongjiang (figure 4). Between 2011 and 2018, 27 out of 28 provinces experienced an increase in the prevalence rate over this period. The top three provinces with the most rapid increase in the age-adjusted and sex-adjusted prevalence of CVD were Xinjiang (33.55%), Anhui (21.09%) and Shanxi (18.36%). Guangdong was the only province experiencing a decline in the age-adjusted and sex-adjusted prevalence of CVD (−7.07%).
The province-level age-adjusted and sex-adjusted prevalence of comorbid CVD in 2018 ranged from 0.00% (95% CI 0.00% to 4.72% and 95% CI 0.00% to 1.16%, respectively) in Beijing and Shanghai to 15.80% (95% CI 8.25% to 23.35%) in Heilongjiang (figure 5). Between 2011 and 2018, 25 out of 28 provinces experienced an increase in the prevalence rate over this period. The top three provinces with themost rapid increase were Heilongjiang (10.89%), Shaanxi (7.78%) and Qinghai (7.20%). In contrast, Tianjin and Beijing experienced a decline in the age-adjusted and sex-adjusted prevalence (−1.95% and −1.38%, respectively). The age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD in each province in 2011 was presented in online supplemental table 3. More details about age-adjusted and sex-adjusted prevalence change of CVD and comorbid CVD in each province between 2011 and 2018 were presented in online supplemental table 4.
Socioeconomic inequalities
The distribution of age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD by SES groups and sex was shown in figure 6. In all survey waves, both CVD and comorbid CVD tended to be more common among those in higher SES groups. Trends analysis showed that there were significant increases in the adjusted prevalence of CVD in the low, lower-middle and upper-middle SES groups from 2011 to 2018 (eg, the prevalence of CVD in both sexes combined, upper-middle SES group, 2011:28.42% (95% CI 24.70% to 32.14%), 2018: 37.04% (95% CI 33.61% to 40.47%), p for trend<0.001). As for the adjusted prevalence of comorbid CVD, a significant increase was observed across all SES groups for both sexes combined and men, while the significant increase for the women was found specifically within the low and low-middle SES groups.
The CCs presented in figures 7 and 8 and online supplemental figures 1 and 2 demonstrate consistent relative socioeconomic inequalities in CVD and comorbid CVD across all survey waves. The CCs consistently remained below the equality line from 2011 to 2018, with the 2011 curve furthest from the equality line, suggesting that the age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD was higher for higher SES groups. In addition, the gap between the CC and the equality line narrowed over time, with decreasing concentration indices (eg, the concentration index of CVD in both sexes combined decreasing from 0.14 (95% CI 0.10 to 0.18) in 2011 to 0.08 (95%CI 0.06 to 0.10) in 2018 and the concentration index of comorbid CVD in both sexes combined decreasing from 0.37 (95% CI 0.19 to 0.55) in 2011 to 0.13 (95% CI 0.07 to 0.19) in 2018). The concentration indices presented in figure 9 showed the magnitude of socioeconomic inequalities in CVD and comorbid CVD across all survey waves. According to the joinpoint regression analysis, the concentration indices of age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD showed a significantly decreasing trend in both sexes combined from 2011 to 2018, with an AAPC of −5.44% (95% CI −5.71% to −5.16%) and −21.50% (95% CI −37.12% to −5.88%), respectively. The details of AAPCs of concentration indices are shown in online supplemental table 5.
Cohort analysis of CVD incidence
From 2011 to 2018, the overall incidence density of CVD and comorbid CVD among the elderly population was 17.96 and 2.65 per 1000 person-years, respectively. Furthermore, table 1 and online supplemental table 6 summarise the results of the GEE, with table 1 showing provinces with statistically significant relative risks of CVD and online supplemental table 6 encompassing the entire array of provinces. As shown in table 1, compared with adults aged 60–69 years, those aged 70–79 and ≥80 years displayed a significantly heightened risk of both CVD and comorbid CVD (all Ps<0.001). Women exhibited an elevated susceptibility to CVD and comorbid CVD when contrasted with men (p<0.001 and p=0.001, respectively). Using Henan Province of Central China as the reference, Sichuan and Hunan exhibited a decreased risk of both CVD and comorbid CVD (all p<0.05). Remarkably, Heilongjiang showcased an elevated risk (p<0.001). When contrasting elderly adults with low SES, those positioned in the upper-middle and high SES groups manifested an increased risk of CVD and comorbid CVD, while the lower-middle SES group exclusively displayed a statistically significant association with comorbid CVD.
Discussion
This study conducted cross-sectional and cohort analyses, focusing on examining the temporal trends, regional variations, SES inequalities and the influence of key demographic factors on CVD among Chinese individuals aged ≥60 years from 2011 to 2018. The main findings are the age-adjusted and sex-adjusted prevalence of CVD increased significantly from 20.77% to 31.01%, and the adjusted prevalence of comorbid CVD increased significantly from 1.22% to 3.79% over the study period. Geographically, we observed striking variations among provinces in the age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD, although the temporal patterns in almost all provinces followed the national trends. Furthermore, we found persistent but decreasing relative socioeconomic inequalities in the prevalence of CVD and comorbid CVD among the elderly across all surveys, and a higher prevalence was observed among individuals in higher SES groups. Moreover, the cohort analysis indicated increased risks of incident CVD and comorbid CVD among older individuals, women, those with higher SES, and residents of northern China.
Similar to the global trend, the adjusted prevalence of CVD among the elderly tended to increase in China.16 Additionally, the total number of elderly individuals with comorbid CVD, defined in this study as individuals diagnosed with both heart disease and stroke, was also on an increasing trend. Previous studies have shown that heart disease complicated by stroke is associated with great suffering for patients and imposes a substantial burden on society.37 Therefore, more and higher-quality healthcare services for the prevention and treatment of CVD should be required as the number of elderly individuals with CVD and comorbid symptoms is increasing in China. Concurrently, the significance of proactive health promotion initiatives should be underscored. Such initiatives can empower individuals to adopt healthier lifestyles, cultivate cardiovascular wellness and mitigate the risk factors associated with CVD and its comorbidities.38 39 Different from other countries such as Canada, our study reveals a notable gender difference in CVD prevalence, with a higher incidence observed among women than men among individuals aged ≥60 years.10 This difference highlights the significance of considering unique population characteristics and healthcare contexts in CVD research. Clinical evidence indicates that as age increases, the prevalence of CVD in women significantly rises, with older women outnumbering men in CVD patients.40–43 This phenomenon can be elucidated by the elevated expression of CVD when risk factors are present among elderly women.43 44 For example, previous studies have found that in China elderly women who smoked or consumed alcohol faced a 1.5-fold higher risk of CVD (p<0.001), but no statistical differences were found in elderly men.45 Additionally, in China, women are at particular risk of suboptimal secondary prevention, and this disparity does not diminish among urban residents and those with higher education levels.46 Therefore, gender differences should be considered to improve the management and treatment of CVD risk factors in women.
We found that most provinces experienced an increase in CVD prevalence among elderly individuals over a fairly short period in China, which indicates the importance of CVD prevention and health promotion strategies because the temporal changes might be driven mostly by modifiable risk factors rather than genetic factors.3 Furthermore, we confirmed substantial variations in the prevalence among provinces. Consistent with previous studies, CVD prevalence was higher in northern regions compared with southern regions of China.3 47 These findings are particularly useful for local health authorities to assess existing CVD prevention and control programmes and develop more localised prevention strategies targeting key CVD risk factors in different regions. The China Cardiovascular Health Index provides a comprehensive dimension of CVD risk factors, including the smoking rate, the physical inactivity rate, the salt intake level, the intake rate of vegetables and fruit, obesity, hypertension, diabetes, hyperlipidaemia and concentration of PM2.5.48 These factors may provide at least partial explanations for the regional disparities in CVD prevalence and contribute to guiding localised prevention goals and strategies.49 It is notable that within our study, certain provinces displayed a 0% prevalence of comorbid CVD among individuals aged ≥60 years between 2011 and 2018. This phenomenon may be attributed to a multifaceted interplay between China’s distinctive CVD occurrence and regional disparities in health factors and behaviours.49 50 China’s CVD occurrence exhibits a significant pattern characterised by a high stroke prevalence with low heart disease prevalence in almost every population.49 51 This national trend sheds light on the observed low comorbidity rates of heart disease and stroke within the country. Furthermore, it is noteworthy that some provinces, including those reporting 0% comorbid prevalence in this study, demonstrate better health factors and behaviours, which may account for their 0% comorbid CVD prevalence. For instance, provinces in western China, such as Chongqing, Qinghai and Guizhou, exhibited more favourable health factors profiles, and those in the southern and eastern regions, such as Guangdong, Shanghai and Zhejiang, demonstrated better health behaviours.50However, it is important to acknowledge that while the CHARLS data we used is a reliable resource, the quality of specific CVD information might have limitations due to its secondary nature. Therefore, we recognise the need for further investigations using nationally representative first-hand data to comprehensively explore comorbid CVD prevalence in China.
According to the Fundamental Cause Theory, higher SES is associated with better health outcomes, which has been supported by abundant evidence from Western countries.10 52 However, because the SES-health associations are context-dependent and vary by country-level characteristics, we suspect the association between SES and CVD in China. Therefore, we analysed the CVD prevalence among different SES groups and found that elderly individuals with higher SES had a higher prevalence of CVD, which is consistent with findings from another study.53 This is partly because deleterious health behaviours, embodying strong cultural meanings, are socially accepted and even encouraged among older individuals with higher SES in China. As the backbone of China’s rapid economic growth beginning in the 1990s, adults aged ≥60 years were among the first to benefit from the sudden increase in personal wealth and the availability of consumer goods in China.54 While their Western counterparts became increasingly aware of the negative health impacts of tobacco, alcohol and high-calorie foods, this group of Chinese individuals was enjoying the newly widespread availability of these commodities.55–58 The critical roles of smoking and alcohol consumption in people’s social lives, coupled with the enduring image of ‘affluence’ attached to excess weight, may outweigh the protective effect of higher SES.59–61 Some studies have found that higher wealth and education levels predict higher tobacco and alcohol use, and are associated with higher obesity rates.58 61 Therefore, the higher prevalence of CVD in the high SES population may be attributed to the higher risk factors present in this group.53 Additionally, the higher prevalence of CVD among elderly individuals with higher SES may partially be the result of higher healthcare utilisation. Participants with higher SES may have better access to healthcare services, greater capacity and higher health literacy to seek healthcare and undergo regular screening for CVD.62 These contribute to higher diagnosis rates of CVD among elderly individuals with higher SES, reflecting higher prevalence.
The cohort analyses were designed to investigate the dynamic impact of key demographic factors—age, sex, province and SES—on CVD incidence over time. Age, as a pivotal risk factor contributing to the escalating burden of CVD, has been largely ignored in China.3 Sex differences in CVD incidence among the elderly in China remain insufficiently studied, while international research indicates that women ‘catch up’ with men after menopause.63 As for provinces, previous studies were limited in their scope, primarily focusing on broad geographical analyses across major regions of China,50 which constrained their ability to investigate disparities across provinces. Although SES is acknowledged as a significant determinant of overall health,11 12 its evolving impact on CVD incidence in elderly Chinese populations remained unexplored. Given the critical importance of these demographic factors, and in light of significant demographic shifts occurring in China over recent decades,64 we examined the evolving association between these pivotal demographic factors and CVD incidence. The results of cohort analyses align with those identified in our cross-sectional analyses. Specifically, the results showed an increased risk of CVD with age increasing. This finding is particularly significant in light of the ongoing ageing of China’s population. Projections indicate an alarming 50% annual surge in CVD incidence between 2010 and 2030,65 emphasising the critical need for a comprehensive understanding of the age-related dynamics of CVD incidence in China. Additionally, the cohort analyses reaffirmed the significant regional variations in cardiovascular health within China. Specifically, we selected Henan Province, located in Central China, as our reference point, revealing a distinct north-south gradient in CVD risks. The identification of such disparities underscores the critical importance of prioritising increased investments in health promotion, as well as proactive strategies for CVD prevention and treatment, particularly in provinces where these challenges are more pronounced.
Strengths and limitations
This study has several important strengths, most importantly its novelty, to our knowledge, this study addresses a research gap by providing a comprehensive estimation of the temporal trends and regional variations and recognising the importance of SES in CVD, especially comorbid CVD, among the elderly population in China. Moreover, this study used data from the CHARLS, a nationally representative database of high quality. Therefore, the findings of this study exhibit strong national representativeness, underpinned by the reliability of the CHARLS survey data. Additionally, the comprehensive cross-sectional and cohort analyses provide robust evidence concerning the association between the key demographic factors and CVD.
Several limitations of this study should be properly acknowledged. First, limited by the CHARLS data, CVD only included heart disease and stroke, which are the primary causes of mortality burden, and did not include other cardiac conditions such as aortic aneurysm and peripheral arterial disease. Second, like other studies based on the CHARLS data, the data on CVD in this study were obtained through self-report questionnaires.9 66 Although medical records were not available in the CHARLS, its sister survey with similar survey protocols, the English Longitudinal Study of Ageing, found good consistency between self-reported coronary heart disease events and medical records (with an accuracy of 77.5%).67 Third, to gain insights into broader trends of CVD incidence across provinces, we categorised each province as a distinct category. However, it is noteworthy that Beijing and Shanghai had no cases of comorbid CVD, resulting in empty standard errors in the GEE estimates. This could potentially affect the model’s robustness. To address this concern and ensure the reliability of our findings, we performed a sensitivity analysis, which confirmed the associations even in the presence of zero counts. Fourth, while we used individual weights with household and individual non-response adjustments and poststratification techniques to enhance generalisability, it is important to acknowledge that perfect generalisation is challenging to achieve in large-scale surveys due to inherent variability and complexities.
Conclusion
From 2011 to 2018, there were statistically significant increases in the age-adjusted and sex-adjusted prevalence of CVD and comorbid CVD. Geographically, there were substantial variations in the prevalence and temporal changes of CVD and comorbid CVD across provinces. Furthermore, the prevalence of CVD and comorbid CVD was higher among older individuals of higher SES and socioeconomic inequalities were persistent but narrowing during the study period. Additionally, the cohort analysis indicated increased risks of CVD and comorbid CVD among older individuals, women, those with higher SES and residents of northern China. More efforts, including robust health promotion initiatives, focused on the prevention and management of CVD, should be taken to optimise the strategy and provide high-quality healthcare for the elderly population in China. Furthermore, future interventions and policies should consider and resolve age-specific and gender-specific, geographical, and socioeconomic disparities to ensure equitable access and outcomes for all.
Data availability statement
Data are available in a public, open access repository. Publicly available datasets were analysed in this study. The data can be accessed on the CHARLS database (https://charls.charlsdata.com).
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and the data used in this study were from the CHARLS, which was reviewed and approved by Peking University Biomedical Ethics Review Committee (IRB00001052-11015). The patients/participants provided their written informed consent to participate in the CHARLS. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
We commend the research team of the CHARLS for their commendable efforts in assembling a high-quality dataset, encompassing nationally representative participants, and making the data accessible to the public.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Handling editor Seema Biswas
Contributors QW contributed to the study concept and design, drafting the manuscript, data acquisition, and statistical analysis. YZ revised the manuscript. LL supervised the study. LL is the guarantor, who had full access to the data, controlled the decision to publish, and accepts full responsibility for the work. All authors have made contributions to the manuscript, and have read and approved the final version.
Funding This study was funded by the National Key Research and Development Program of China (2020YFC2003402) and the Fundamental Research Funds for the Central Universities (3332021052).
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.