Results from our study indicated that both social isolation and MCDs were independent risk factors for functional disability in elderly stroke patients. Moreover, the hypothesis that a combination of these factors was associated with a greater extent of risk for functional disability than single factors alone, was successfully assessed. The coexistence of social isolation and MCDs ≥ 3 was associated with a 38 times higher risk for severe disability, a nearly 10 times higher risk for moderate disability than the counterparts.
This study, including 4281 elderly stroke patients from 103 clinical settings of 23 cities, was a nationally representative study with a large sample size. To the best of our knowledge, our study is the first to address the combined association of social isolation and MCDs with functional disability in this special population in China. We believe that these findings will provide insights that future prevention and intervention strategies for functional disability may integrate the social content with combined treatments of MCDs to achieve more effectiveness. However, the interpretation of these results also needs to consider the following limitations. First, the cross-sectional design of this study was restricted to drawing associations between the causal relationships among the variables. Second, the sampling method was non-random, and included hospitals that collaborated with our departments. Nonetheless, the inherent bias may have been avoided since our study covered over 100 hospitals in 23 sites to ensure its generalisability. Moreover, our study only selected stroke patients aged over 60 years, therefore our results may not be applicable to other aged populations. Third, the majority of variables were self-reported; we invited experienced investigators to assess functional disability and combine their assessments with the medical records to reduce recall bias as much as possible. Moreover, considering that this study covered a large sample size and many organisations, the investigation of lifestyles selected only smoking and drinking as confounders [32] ; presented as a simple question to facilitate easy answering by the participants which resulted in the loss of information, including diet, physical activity, and ex-smokers. Fourth, our study included both first-ever and recurrent stroke, and social isolation could not be differentiated into pre-stroke or post-stroke social isolation. Thus, long-term prospective studies are needed to identify the differences between these aspects to provide more precise intervention strategies.
BI is the most popular method to measure functional disability in stroke patients [33]. Although the BI cut-off point for categorising stroke patients with an unfavourable outcome was different from study to study, each study stressed the fact that the prevalence of functional disability was severe in stroke patients [33, 34]. There is a paucity of studies that consider geriatric stroke patients in China. To the best of our knowledge, only Li et al. used the same classification criterion of BI (severe disability ≤ 55) in disabled elderly stroke patients with a small sample size (n = 158) and showed that the prevalence of severe disability was more than half [26]. Our study’s results, similar to Li’s research, was 71.5% scaled up to 76.2% after excluding patients without disability. Additionally, the prevalence of severe disability in our study was more than double that of a Brazilian study of 260 middle-aged and elderly stroke survivors (31.5%) [35], more than six times that of the Northern Manhattan Stroke Study of 340 elderly stroke patients [36]. Although the difference in the prevalence of severe disability was potentially attributed to ethnicity [33], the prevention of functional disability is important in Chinese elderly stroke patients.
The identification of risk factors for functional disability is key to the formulation of intervention strategies in elderly stroke patients. In our study, MCD status was found to be a risk factor for functional disability. Studies focusing on elderly stroke patients were limited, and we found some similar findings in a population with a single condition (geriatric or stroke) [37]. A recent systematic review of 3339 stroke patients from seven hospital-based cohort studies, indicated that MCDs are longitudinally associated with an 11% increase in the risk of functional disability after stroke, irrespective of the different methods used for assessing MCDs and functional disability [37]. Meanwhile, a large cohort study including a general geriatric population also showed that MCDs significantly increased the risk of functional disability by 117% [38]. In contrast, another cohort study with a small sample size but with a mean age over 85 years found no significant association between them [39]. We speculated that the inconsistency was attributed to age. On the one hand, based on trajectory analysis [4], functional disability tended to be stable beyond the age of 80 years; on the other hand, the combination of MCDs and the age over 80 years had a high probability of leading to death [38]. In fact, MCD status was inevitable to some extent in the ageing process, [14] and nearly half of the participants in our study suffered from two and more chronic diseases, which is consistent with a recent systematic review [14]. Many possible explanations have addressed this issue. Sharing similar risk factors, including genetic, environmental, social, physiological, and psychosocial factors, was the common one [14, 40]. For example, chronic inflammation and oxidative stress played an important role in the initiation and progression of MCDs related to ageing [40, 41]. Besides, the shared risk factors were more likely to be the mechanisms by which MCDs synergistically increased the functional disability [41]. Furthermore, another systematic review proposed that certain chronic disease clusters were internally and highly correlated [28]. As for stroke, we found that hypertension (77%), diabetes (32%), and heart diseases (33%) were more common in stoke patients than other two chronic diseases. A review similarly indicated that hypertension was the most important risk factor for stroke and sustained the highest population-attributable risk (34.6%) among obesity, several classic chronic diseases (diabetes, hyperlipidaemia, coronary artery disease, and arterial fibrillation) and lifestyles (smoking and physical inactivity).[42] Thus, the prevention of post-stroke functional disability requires the control and management of hypertension, diabetes, and heart diseases.
Another strong risk factor for functional disability in our study was social isolation, which has been well documented in many cohort studies. [43–45] Although these studies were conducted in older adults without referring to stroke, they consistently indicated that social activity can reduce or delay the decline of functional disability caused by ageing [43–45]. Social isolation may have a deleterious effect on neural networks and decreased musculoskeletal function, accelerating the functional disability related to ageing [21]. In addition, many other behavioural and psychosocial pathways underlined the improvement in depression and cognitive decline associations with social activity; they all are important psychological risk factors for functional disability and stroke [22, 44]. Social networks change to smaller and close-knit ones during ageing [19] and after stroke. [46] A clinical study has reported that baseline social isolation robustly predicted the stroke and prognosis outcome. [46] Thus, both primary prevention and secondary therapies for stroke and functional disability should consider the promotion of social engagement among older adults.
Importantly, the novel finding of our study is that the combination of social isolation and MCDs is associated with a much higher risk for functional disability than a single condition. Although the exact mechanisms regarding the relationship of their combination with functional disability have yet to be explained sufficiently, we can understand its plausibility from several dimensions: 1) both ageing and stroke are the common risk factors for them [4, 7, 14, 20], although they represent social and biological levels, respectively; 2) social isolation may interact with MCDs through biological, behavioural, and neurological links. A previous study reported that a series of downstream biological changes, including abnormalities in cortisol secretion and a deterioration in immune and anti-oxidative stress functions, was easy to detect in people with social isolation [47–49]. From a behavioural point of review, social isolation constantly leads to the development of many unhealthy lifestyles, which accordingly increases the risk for MCDs [48]. As for the neurology aspect, a recent review synthesised the evidence and found that social isolation combined with other mental disorders (depression and anxiety) to accelerate the activation of the hypothalamic pituitary adrenocortical axis and subsequently moderates the biological process [49]. Additionally, we observed that the OR of social isolation reduced slightly after excluding patients with heavy hurt fall. Obviously, since many factors can directly influence social activity [50], future studies related to social activity should consider more potential mediators. It should be noted that the OR of social isolation in the comparison between the no disability and severe group was still much higher than that of the other two comparisons. Given that the intervention on social isolation through encouraging the participation of social activities was much more available than MCD management, future interventions aiming to increase functional ability and quality of life of elderly stroke patients should pay more attention to social activity.