Lifestyle, clinical, and occupational risk factors of recurrent stroke among the working-age group: A systematic review and meta-analysis

Background Stroke recurrence is increasing in the working-age population. This study aimed to summarize and evaluate the risk factors for recurrent stroke among the working-age population. Methods Relevant studies were extracted from several databases following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Fixed- or random-effects estimates of the pooled odds ratio (OR) and 95% confidence interval (CI) of risk factors for recurrent stroke were generated based on heterogeneity. Subgroup and publication bias analyses were also performed. Results Seventeen studies were included in the meta-analysis. Pooled effects results revealed that the risk of recurrent stroke in the working-age group was as follows: Diabetes (OR = 1.85, 95% CI: 1.47, 2.32), hypertension (OR = 1.27, 95% CI: 1.12, 1.44), smoking (OR = 1.52, 95% CI: 1.27, 1.81), history of cardiac disease (OR = 2.86, 95% CI: 2.22, 3.67), history of stroke (OR = 2.45, 95% CI: 1.81, 3.31), and National Institutes of Health stroke severity score (OR = 1.09, 95% CI: 1.03, 1.15). Conclusion These findings suggest that several factors, such as diabetes, hypertension, smoking, history of cardiac disease and stroke, and severity of a stroke, are potential risk factors for recurrent stroke in the working-age group. Therefore, strategies to reduce those risk factors should be adopted and attention paid to prevent recurrent stroke among working-age populations.


Introduction
Stroke is a major global public health concern for the 21st century, as the number of stroke patients is expected to rise in the future [1]. Stroke burden is associated with high mortality and morbidity; it is the second leading cause of death worldwide, with a mortality rate of 5.5 million per year [2].
Approximately half of the patients surviving an initial stroke have a significantly increased risk of recurrent stroke within 30 days to 5 years, which leads to increased mortality [3][4][5]. Recognizing warning signs, such as sudden unilateral weakness, difficulty speaking, difficulty walking, loss of balance, or dizziness [6], may reduce the time delay between the occurrence of an event and hospital

Eligibility criteria and study selection
An initial eligibility screening of all titles and abstracts was performed, and studies reporting recurrent stroke were selected for further review. The following criteria were used for study selection: (1) studies reporting recurrent stroke after any subtype of stroke; (2) studies in adults aged 15-64 years, as this is considered the working-age population [22]; (3) studies using a prospective design; and (4) studies that provided an odds ratio (OR)/risk ratio (RR)/hazard ratio (HR) estimates. Recurrent stroke was defined as a focal neurological deficit lasting more than 24 h and occurring after the initial stroke. Patients with subsequent stroke within 24 h, experimental studies, systematic reviews, and meta-analyses were excluded. Two authors (CS and YW) independently examined eligible studies, and any disagreement was mediated by a third author (AC).

Data extraction and quality assessment
Data extraction was performed by two authors (CS and YW) and included first authors' name, year of publication, study location, age range, recurrence rate, stroke type, and factors of recurrent stroke. The Newcastle-Ottawa Scale (NOS) [23] was used to assess the methodological quality of all included studies, and any disagreement was resolved by discussion. Quality assessment scores were defined as poor (0-3), fair (4-6), or good (7-9).

Statistical analysis
The association of lifestyle and occupational factors with recurrent stroke was assessed using either OR/RR/HR with 95% confidence interval (CI). As the incidence of recurrent stroke was relatively low, OR/HR was close to RR and was assumed to be the same measure. In the meta-analysis, the fixed-or random-effects models were used to generate OR with 95% CI. Cochran's Q test and I 2 statistics were used to assess heterogeneity across studies [24]. We chose the random-effects model if the heterogeneity I 2 statistic was significant; otherwise, the fixed-effects model was chosen. To identify potential sources of heterogeneity, subgroup analysis was performed based on the study location, recurrence rates, stroke type, and study quality. Begg's rank correlation test was used to assess publication bias. All statistical analyses were performed using STATA 17 software (StataCorp, College Station, TX, USA). Note: IS = ischemic stroke, Any = any stroke type, NOS = Newcastle-Ottawa Scale, NIHSS: National Institutes of Health Stroke Severity.

Included studies and study characteristics
In this systematic review, we identified 5212 articles through database searches. We assessed the eligibility of 136 full-text articles. After evaluating full-text articles, 17 studies were included in the systematic review and meta-analysis (Fig. 1). The characteristics of the included studies are summarized in Table 1. These studies were published between 2008 and 2021. Of the 17 included studies, a total of 164,491 stroke patients were involved, and 3788 (2.3%) patients had recurrent stroke. Sixteen studies were cohort studies and one study used a case-control design. Ten studies were conducted in Western countries and 7 in Eastern countries. The studies were conducted across working ages ranging from 15 to 69 years. Stroke recurrence was reported to range from 0.3% to 36.2%. According to the NOS, 7 studies scored 7 or more (good), 9 studies scored 4-6 (fair), and only one study had poor quality.

Occupational risk factors
As shown in Table 2

Subgroup analysis
Diabetes and hypertension showed a high degree of heterogeneity (I 2 > 50%). We then conducted a meta-regression analysis to check the sources of heterogeneity. Study location significantly explained the variance in the effects of diabetes on recurrent stroke (p < .05) and stroke type significantly explained the variance in the effects of hypertension on recurrent stroke (p < .01).

Publication bias
Based on the results of the Begg's rank correlation test (diabetes, p = .06, hypertension, p = 1.00; smoking, p = .35, history of cardiac disease, p = .06, history of stroke, p = .30, NIHSS stroke severity, p = 1.00), there was no evidence of publication bias.

Discussion
This systematic review and meta-analysis evaluated various risk factors for recurrent stroke among working-age groups. Based on the 17 included studies, three lifestyle risk factors (diabetes, hypertension, and smoking) and three clinical risk factors (history of cardiac disease, history of stroke, and NIHSS stroke severity) were found to be associated with recurrent stroke in the working-age group. However, no association was found between occupational risk factors and recurrent stroke. According to the subgroup analysis, there was an association between recurrent stroke and diabetes or hypertension, regardless of study location or stroke type. Furthermore, there was no evidence of publication bias based on Begg's rank correlation test.     Lifestyle risk factors such as diabetes, hypertension, and smoking were associated with recurrent stroke in the working-age group. These results were in line with previous studies that reported the risk of recurrent stroke associated with diabetes, hypertension, and smoking in young adults [26,37]. Healthy lifestyle was associated with reduced risk of recurrent stroke [18,41]. Moreover, pharmacological treatments (such as antihypertensive and antidiabetic medications) were associated with a lower risk of recurrent stroke [42,43]. Therefore, to reduce and prevent stroke recurrence in the working-age group, young adults need to be considered for treatment and lifestyle modification (such as antidiabetic medication, antihypertensive medication, smoking cessation, eating a healthy diet, and performing exercise activities) to reduce the risk of recurrent events. Clinical risk factors such as history of cardiac disease, history of stroke, and NIHSS stroke severity were associated with recurrent stroke in the working-age group. Consistent with previous studies, these clinical risk factors were associated with a higher risk of recurrent stroke in young adults [15]. Other researchers have also found that these clinical risk factors are associated with stroke recurrence among the general population, including older individuals, and are not specific to the working-age group. Fu et al. [15] also found that a previous history of stroke and coronary heart disease in the elderly was related to recurrent stroke. Wu et al. [44] showed an increased risk of recurrent stroke in patients with moderate and severe stroke relative to those with mild stroke. These results confirm that clinical risk factors (history of cardiac disease, previous stroke history, and stroke severity) could be used as predictors of recurrent stroke. There is no association between occupational status and recurrent stroke indicating there is no evidence to support an association between occupational risk factors and recurrent stroke in the working age group. A possible reason for this phenomenon is that young adults who have an occupation receive better healthcare benefits. Returning to work is the primary goal in the rehabilitation process of the working-age group [45]. However, a number of studies have investigated occupational risk factors and stroke risk. Long working hours [46] and stress at work [47] are associated with an increased risk of stroke. Perceived psychosocial work environment is also associated with an increased stroke risk [48]. However, it is reasonable to assume that occupational risk factors, especially long working hours, stress at work and psychosocial environment, may have affected the development of stroke recurrence in the working-age group.
The results of this study indicate that diabetes is a risk factor for recurrent stroke among working-age groups in both Western and Eastern populations. This finding is consistent with previous meta-analysis results; the effect of diabetes on stroke recurrence showed no regional differences [49]. From the results of our investigation, recurrent stroke associated with diabetes was greater among Eastern populations than among Western populations. A possible explanation for this phenomenon is that Eastern populations have small sample sizes; therefore, there is a possibility that the findings are affected by small-study effects [50]. Previous studies have shown that the prevalence of diabetes in stroke patients was relatively high in Eastern populations [51][52][53]. Therefore, further research is needed to explore the similarities and differences concerning findings observed in Western and Eastern populations.
Furthermore, this study found that hypertension is a risk factor for recurrent stroke among working-age groups for both stroke type and ischemic stroke. In fact, the risk of recurrent stroke associated with hypertension is greater among working-age groups with any type of stroke than among those with ischemic stroke. A possible reason for this phenomenon is that studies included any stroke type, including hemorrhagic stroke. Hemorrhagic strokes cause a severe increase in blood pressure [54] and lead to worse functional and clinical status compared to ischemic stroke [55,56]. A previous study found that the effect of hypertension on stroke was greater in hemorrhagic stroke than in ischemic stroke [57]. However, management of blood pressure in hemorrhagic stroke should be considered to prevent recurrent events [58,59]. Therefore, blood pressure-lowering treatment is an effective measure to reduce recurrent stroke in all stroke types.
However, this study had several limitations. First, most of the studies were conducted in high-and middle-income countries. Thus, more research evidence from low-income countries is required. Second, the recurrence rate in Eastern populations was significantly higher than that in Western populations (t = 2.252, p = .04). A possible reason is that the Eastern population used smaller cohorts than the Western population. Results bias may be relatively high for small cohorts, but lower for larger samples. Bias may also be due to loss to follow-up, which represents a threat to the internal validity of effect estimates from cohort studies [60]. Another possible reason is that larger samples increase the power of the study compared to smaller ones. The effect sizes of large samples have sufficient statistical power to become more significant than small ones [61]. Therefore, a larger cohort study should be conducted to obtain a more precise estimate. Third, the included studies reported early and late recurrence and type of recurrence, which might affect the results. It is suggested that the differences in early and late recurrence and type of recurrence should be explored in further studies to reduce heterogeneity. Finally, some studies were adjusted with control variables; hence, confounding factors in some studies might bias the true association.

Conclusion
This meta-analysis summarizes and evaluates lifestyle, clinical, and occupational risk factors for recurrent stroke among workingage individuals. Diabetes, hypertension, smoking, history of cardiac disease and stroke and the severity of a stroke are potential risk factors for recurrent stroke in working-age groups. Strategies for those risk factors should be adopted and focus on preventing recurrent stroke among the working-age population.

Author contribution statement
All authors listed have significantly contributed to the development and the writing of this article.

Funding statement
This work was supported by Navamindradhiraj University Research Fund (RESEARCH.NMU 62/2564).

Data availability statement
Data will be made available on request.

Declaration of interest's statement
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information
No additional information is available for this paper.