Association of blood pressure trajectories with coronary heart disease among the disabled population in Shanghai, China: a cohort study of 7 years following up

According to the latest report on disability by the World Health Organization [1], more than one billion individuals, around 15% of the global population, lived with some form of disability. Meanwhile, a recent report indicated that the number of people with disabilities has reached nearly 7% of China's total population [2]. Studies have shown a strong link between the increased incidence of disability and chronic diseases, such as hypertension, diabetes, and stroke [3, 4]. In addition, the huge health disparity between people with and without disabilities has been widely demonstrated [5, 6]. People with disabilities may be more susceptible to various chronic diseases due to greater social isolation and less access to preventive care [7, 8]. However, few studies have analyzed the epidemic of the coronary heart disease (CHD) that accompany individuals who are already disabled.

Blood pressure (BP) is a leading treatable risk factor for CHD, a common chronic cardiovascular disease (CVD) that exists widely in the world, and has been deeply studied in the general population [9, 10]. Most studies focusing on the relationship between BP and the onset of CHD were mainly based on BP data from single measurements at baseline or the variability of multiple measurements [11‐14]. Results from such approaches have suggested that the risk of incident CHD increases with increasing baseline BP and variability in BP. However, the above BP indicators may not adequately reflect longitudinal BP changes in individuals. Several recent studies have explored the role of BP patterns over time (i.e., BP trajectories) in predicting CVD outcomes, dynamically reflecting the impact of arterial BP on the development of diseases [15‐19]. In particular, high-level and rapidly increasing BP patterns are associated with a high risk of stroke over an age range from 55 to 106 years. Nevertheless, much less is known about the importance of BP trajectories over time concerning the incidence of CHD in people with disabilities.

Thus, this study aimed to identify distinct longitudinal trajectories of systolic BP (SBP) and diastolic BP (DBP) over 7 years and to investigate the association of these trajectories with the risk of the onset of CHD among the disabled population. Moreover, we compared other BP-related indicators concerning the incident CHD.

Although previous studies have discussed the relationship between BP trajectories and CVDs risk in the general population, the association among the disabled population has rarely been elucidated. In the current study, we explored the association between BP trajectories and CHD risk in the disabled population using a substantial population-based cohort. First, the trajectories of SBP and DBP were categorized into three classes by LCGMM, respectively. Second, the disabled individuals in stage 1 hypertension-decreasing SBP group and prehypertension-stable SBP group had higher CHD risk relative to the normotensive-stable SBP group. Third, the SBP trajectory may be a stronger predictor of CHD than single baseline SBP, average SBP, and CV of SBP.

Our findings identified three SBP and DBP trajectories separately, these overall trend waves were consistent with previous cohort studies on BP trajectories associated with CVDs during follow-up in the general population but the number of trajectories ranged from three to five classes [15, 24, 25]. The main difference is that we did not find a trajectory of BP that rose significantly from normal levels to high values, which may be due to population differences and limitations of the LCGMM fitting subgroup sample. In addition, subgroup analysis revealed almost identical trajectories to the primary results in the no-hypertension group, but only two BP trajectories were found in the hypertensive group without the normotensive-stable BP class (< 130/80 mmHg).

The relationship between BP trajectories and CVD is not identical due to the heterogeneity in the characteristics of trajectories and populations in relevant studies. Our study suggests that the disabled individuals in the stage 1 hypertension-decreasing SBP group (160–120 mmHg) and the prehypertension-stable group (130–140 mmHg) may both increase the risk of future CHD, although a statistically significant association was not observed between DBP trajectories and CHD after multivariate adjustment. Among the general population, the Framingham Heart Study explored 16-year trajectories of BP and its association with the risk of atrial fibrillation in 4351 participants and found that the risk was doubled in participants who had initially elevated levels of SBP and then either decreased or increased SBP, yet the trajectory of DBP was not an independent predictor [18]. A 7-year cohort study determined that class 2 (SBP persistently around 133 mmHg) and class 3 (initially elevated SBP decreasing to normal) were associated with increased CVD risk relative to class 1 (initially normal levels increasing to above 140 mmHg) among 4067 individuals [15]. In China, five distinct BP trajectories over 4 years were identified in the Kailuan Study, the prehypertension-stable SBP group had higher stroke risk and individuals in the stage 2 hypertension-stable group (175–179 mmHg) had the highest risk of intracerebral hemorrhage and cerebral infarction relative to the normotensive-stable group (< 130 mmHg), and trajectories of DBP was also significantly associated with the risks of above outcomes [26]. Moreover, a meta-analysis among 123 studies with 613,815 participants revealed that lowering SBP to less than 130 mmHg significantly reduced the risk of CHD [11]. Our results indicated that people with disabilities may have an impact on health outcomes once they are chronically exposed to pre-hypertension or poorly controlled BP over their life course, and a more stringent BP control strategy may be more effective for primary CHD prevention. However, the optimal BP goal for primary CHD prevention remains unclear in the disabled population.

The association between single baseline BP as well as long-term variability and CHD risk have been previously demonstrated [9, 11‐13, 27]. These findings were also revealed in our study, in addition to the fact that SBP trajectory was considered as an independent better predictor among disabled subjects. More information was added to the existing knowledge regarding risk factors and prevention of CHD as they suggest that effects of elevated BP levels may take several years to appear, and that longitudinal analysis of risk factors can improve the prediction of outcomes compared to single measurements or averages [26, 28]. Group-based trajectory modeling takes into account the averages and variations of multiple measurements to distinguish dynamic exposure in BP over time [29], which could reflect the potential impact of long-term BP on target organs. Available evidence showed that consecutive years of hypertensive trajectories are strongly correlated with subclinical atherosclerosis, intima-media thickness, and left ventricular mass index [30, 31]. Assessing BP trajectories provided a more nuanced understanding of the association between BP and CHD and improved prediction of CVD, our findings could help provide new insights into the prevention and management of CVDs in the disabled population.

Our study has several limitations. First, due to difficulties in collecting self-reported data from disabled people in the early study, we were unable to assess information on their diet, physical activity, the accessibility of medical resources, socioeconomic status, and care for them. However, all SDHS participants have access to Shanghai Disability Living Guarantee, which can reduce this gap to some extent. Second, low/high density lipoprotein cholesterol tests were not included in the examination due to limited funding for SDHS. Therefore, we attempted to mitigate this effect by entering trajectories of other objective indicators affecting BP in the multivariate model. Third, the lack of detailed information on antihypertensive treatment may lead to non-differential misclassification, but a sensitivity analysis stratified by hypertension was performed. Fourth, the subjects in this study may have overrepresented those who lived longer, because participants with poor cardiovascular profiles may have died before their next physical examination, which may have led to an underestimation of the association between BP trajectory and CHD risk.

Our study observed three distinct BP trajectories in people with disabilities and these trajectories were associated with future CHD risk. Monitoring BP trajectories over time could provide a more important approach to identifying these individuals with higher risk and preventing CHD. Future research needs to explore the optimal BP goal for primary CVDs prevention in the disabled population.