Association of cardiovascular health at old age with all-cause mortality: a prospective cohort study in China

Cardiovascular disease (CVD) continues to be the leading cause of death and rising healthcare costs worldwide [1]. It increased the number of deaths from 12.1 million in 1990 to 18.6 million in 2019 [2], and this number is still increasing steadily [3]. Every year, nearly 4 million people in China died from CVD [4]. The number of prevalent cases reached 94 million in 2016, more than doubling that of 1990 [5]. With rapid aging, the number of people aged 65 and above will reach 366 million (26.1%) by 2050 [6], with a significant shift toward CVD in old age [7]. CVD events are expected to increase by 50% per year from 2010 to 2030 due to aging, and by 23% based on the current risk factor trend [8, 9]. The excess burden of CVD among the elderly in China poses a significant challenge to the medical and health care systems [10]. Despite advances in the medical system and health institutions in the treatment and care of CVD, the significant burden of cardiovascular disease risk factors in China continues to rise [11].

The American Heart Association (AHA) released 2020 Impact Goals in 2010 to emphasize the importance of public cardiovascular health prevention. It also defined the concept of Ideal Cardiovascular Health (ICH), also known as the Life's Simple 7 principles [12]. Since then, CVH has been widely adopted in numerous studies, aiming at promoting the interventions of modifiable health behaviors and CVD risk factors [13‐18]. Increasingly, studies show that having a higher level of CVH can reduce the risk of various chronic diseases (CVD events [19‐21], caners [22, 23], cognitive impairment [24‐26], and so on) and mortality [27‐29].

We conducted a thorough literature review prior to beginning this research. From inception to 2022/07/07, PubMed, Web of Science, and Embase were used to conduct searches. The following keywords were applied: (“ideal cardiovascular health” OR “cardiovascular health metrics” OR “life’s simple 7”) AND (“mortality” OR “death”). The reference lists of eligible literature were thoroughly searched. The effect of CVH on mortality risk has been studied in younger Chinese people [30‐33]: the HRs ranged from 0.54 to 0.78 (highest CVH group vs. lowest CVH group), but data for the elderly in China is limited. In view of this situation, In light of this, we intend to examine the relationship between CVH score and all-cause mortality in the elderly in China from 2009 to 2021 using data from the Beijing Elderly Comprehensive Health Cohort Study (BECHCS).

This study investigated the relationship between CVH score with all-cause mortality among Chinese elderly: the higher the CVH score in old age, the lower the risk of all-cause mortality; and a linear-dose response relationship was observed. In both male and female population, consistent relationships were found between the obtainment of per point in CVH score and the reduced risk of mortality. This study provides evidence that CVH metrics can reduce the risk of all-cause mortality in the Chinese elderly population.

The prevalence of ideal CVH in Chinese elderly remains low. Our study found that 8% of the participants scored 11–12 points in CVH, and only 1.6% got all six metrics of ideal CVH. This result was consistent with a previous study [35], it included 5026 participants aged 65 and above in rural China, and found that only 0.8% of them met the ideal levels for 6 CVH metrics. Several studies have found a low prevalence of ideal CVH: 5% [36] (participants had ≥ 5 out of 7 metrics at the ideal level), 12% [37] (participants had 9–12 out of 12 points of CVH score), 6.5% [25] (participants had 5 to 7 health metrics at optimal levels). Due to the various definitions of ideal CVH, the distribution of CVH metrics components varies from one another. In our study, the low prevalence of ideal CVH resulted in the unfavorable blood pressure level: only 16.1% of all the subjects achieved the ideal state. It could be partially due to the use of biomass fuel [38] or a diet with high sodium [39]. In China, the hypertension is common, accounting for 33.0–47.2 of atherosclerotic CVD events [40], while the control rate is low [41]. Immediate actions are required in China to improve the prevention or treatment of hypertension in the elderly. Furthermore, because the score of metabolic CVH metrics would be affected by behavior metrics, composite metrics would be expected to be lower [20].

In this study, the CVH score was generally higher among the elderly men under 75 years old who live in the rural areas. Bi et al. investigated a nationally representative sample of 96,121 Chinese adults aged 20 years and older in 2010 and discovered that being female and being younger were the protective factors for cardiovascular health [42]. While Han et al. reported men were more likely than women to achieve ideal levels on all CVH metrics, smoking was not one of them [35]. Of note, smoking metric directed primarily at the sex difference in global and behavior CVH metrics, where males outnumber females in terms of smoking prevalence.. The same patterns were observed in urban and rural disparities. The inconsistency is likely due to the fact that premenopausal women are relatively unaffected by CVD risk, whereas sex disparity diminishes with age after menopause [43]. Previous studies found that the oldest old and centenarians had a higher overall level of CVH metrics than the young elderly [11, 34, 44, 45], but the prevalence of ideal CVH still needs to be improved. Furthermore, the difference between gender, age and residence in our study could be influenced by the population characteristics in our study as well as the sample size of subgroups. More related studies are favorable.

Strong associations of the CVH metrics with all-cause mortality have been confirmed. Our study found that the risk among Chinese elderly with CVH score at 11–12 points was 41.6% lower than that of those with a CVH score of 0–4 points (HR = 0.584, 95% CI: 0.373–0.913). As the number of CVH metrics increased, the risk of all-cause mortality decreased continuously and linearly[HR = 0.925, 85%CI: (0.885,0.967); (P _nonlinear = 0.575)]. In lined with our findings, Zhou et al. found that in the Chinese population, participants with higher ideal CVH metrics had a 54% lower risk of all-cause mortality compared to those with lower ideal metrics [32]. A stronger protective effect was observed in comparison to previous studies in younger adults [30‐33]. The linear dose–response relationship was consistent with a meta-analysis [27], which included 142,137 participants to quantify the mortality risk of each unit change in ideal CVH and found that there is a linear inverse relationship. Previous studies [31, 46‐48] recognized this close relationship, which were summarized in a meta-analysis [28], and the evidence was well established among younger populations. For older age adults, Gaye et al. conducted a Three-City Study to examine the benefit of higher CVH status in terms of mortality among participants aged 65 years and over [36]. The estimated HR (95% CI) was 0.71 (0.55–0.90) of all-cause mortality among subjects with ≥ 5 vs. ≤ 2 ideal metrics was assessed. Several studies in the Korean and British elderly population [34, 37] confirmed the evidence of an association between CVH and cardiovascular events in the elderly population. Then, our findings supplemented this gap among Chinese elderly even more. With the increase of ideal CVH score, the risk of all-cause mortality reduction was consistent across all age groups, highlighting the importance of promoting CVH metrics in old age. Furthermore, because the linear relationship was shown to be statistically significant, it should be emphasized that improving every point of CVH score- that is, shifting each CVH metric from poor into intermediate- systematically yielded favorable outcomes [18]. While achieving the ideal CVH is undoubtedly a long-term goal, the immediate action or short-term goal is to encourage the attainment of per point of CVH score at a time progressively [36].

Subgroup analysis could provide supporting evidence for more efficient methods of promoting CVH metrics in the elderly. In the age-stratified analysis, the decreasing trend in the risk of all-cause mortality with increasing CVH score was statistically significant among the younger elderly (age under 75 years), but not among the very elderly (age 75 years and above). The reason accounting for this could be “survivor bias”, which means that participants had to live to reach age 75 years old or older before the assessment of CVH score [49]. This shows the importance of paying special attention to the young and elderly when providing prevention or care. In the stroke-stratified analysis, more considerations are favorable to determine the difference of pattern of CVH metrics among older adults with stroke. In our study, individuals in rural areas tended to have higher levels of unhealthy lifestyles. According to current research [11, 18, 21], the disparity in health behavior between urban and rural areas is partly explained by socioeconomic development factors such as income and education level. This suggests that public health policies and strategies should be implemented especially focusing on behavior interventions for Chinese elderly in rural areas.

There are several limitations to the study. First, the prevalence of ideal CVH metrics in this study should be generalized with caution, given the non-nationally representative study population. The primary goal of our study was to evaluate the association of CVH score with all-cause mortality in the elderly. Further research into the epidemiology of CVH among the elderly in China would benefit from larger-scale or nationwide samples of the elderly. Second, data about lifestyles and diseases were self-reported, which could lead to recall bias. To address this problem, highly stringent training procedures and quality assurance programs were implemented throughout the research process. Including recruiting trained interviewees, requiring certificate of disease diagnosis, etc. if possible. Third, individuals with prior CVD were included in the analysis throughout, which might bring over estimation of the reported associations. To address this problem, sensitivity analysis was performed to exclude subjects with CHD or stroke, ensuring robustness of results. Besides, subjects with extremely low BMI or who died two years after follow-up were also excluded to account for misclassification bias. Fourth, detailed information was absent. To avoid classification bias and remain consist with previous study, we revised Life’s Simple 7 into 6 metrics. And because all of the data was measured once, any differences between the baseline survey and the endpoint investigation may have gone unnoticed. Furthermore, the risk of CVD mortality wasn’t estimated. Fifth, the sample sizes limited further stratified analysis or potential confounders. Considering the power of statistical analysis, we did gradual adjustments with 4 models, which might explain the reason why some associations were not statically significant.

Among the Chinese elderly, there was a linear rekationship between CVH scores and the lower risk of all-cause mortality. It is critical to develop strategies to improve cardiovascular health attainment among Chinese elderly with low CVH scores.