Association of Socioeconomic Status With Ischemic Stroke Survival

Rosa Maria Vivanco-Hidalgo, MD, PhD, MPH; Aida Ribera, PhD; Sònia Abilleira, MD, PhD


Stroke. 2019;50(12):3400-3407. 

In This Article


Principal Findings

We analyzed the association between the socioeconomic status (at individual and PCSA levels) and all-cause deaths in patients with ischemic stroke in a nationwide population-based cohort. We found evidence of an individual socioeconomic status gradient in long-term survival that persisted after adjustment for comorbidities and cardiovascular risk factors. PCSA index seemed to influence short-term survival only but not long-term survival.

Comparison With Other Studies

The association between socioeconomic status and mortality after stroke has been widely described in the literature. Studies show that low socioeconomic status is associated with increased stroke mortality,[12] although evidence is inconsistent in some countries with universal healthcare systems.[13–18]

Area-based and individual socioeconomic indicators independently contribute to several important health outcomes, such as mortality and cardiovascular risk burden.[19,20] Many studies assessing stroke mortality have used either area-based or individual measures of socioeconomic status to assess the association with survival after stroke, which leads to inadequate adjustment for confounding and different interpretation and implication of the findings depending on the type of measure. The individual socioeconomic status indicators used in previous studies varied from educational, occupational, income, and medical insurance status measurements, with its corresponding disadvantages and different association with outcomes.[2] Individual income seems to be the individual socioeconomic characteristic that better correlates with mortality in patients with stroke.[21,22] We found that patients in the lowest individual income category had a 52% higher risk of death in the long-term follow-up compared with those in the highest category.

Most studies that have applied area-based measurements used index composites of several variables corresponding to small geographic areas. Those indexes usually reflect an overall marker of neighborhood conditions and may even determine access to primary healthcare.[23] We decided to use an area-based deprivation index that reflects the socioeconomic differences in health necessities among PCSA. This is currently being used to assign resources to improve territorial equity in Catalonia.[10] We found that this index contributed to short-term survival in our cohort of patients but not to long term.

The majority of previous studies have focused on in-hospital and short-term mortality, and few studies have assessed long-term survival, showing heterogeneous results.[4,24,25] The original contribution of the present study is that we analyzed the influence of both socioeconomic indexes (individual and PCSA) in short- and long-term survival in the same cohort of patients. We found that, in our population, socioeconomic factors might have a different influence on ischemic stroke survival depending on the length of the follow-up. That would mean that short-term case fatality might be influenced by healthcare system factors related to PCSA, and those health factors that depend more directly on individual socioeconomic factors would determine the long-term prognosis. Additional research is needed to determine which individual (behavioral) factors mediate in this relation. Our results of short-term survival are in line with those from a previous study set in France, where there is also universal access and where the influence of the neighborhood socioeconomic status was evident only in the latter period of the acute care.[26] However, the role of the individual socioeconomic status was not assessed.

Stroke subtype is another important factor to take into account. It is well described that prognosis, and risk factors associated, vary between ischemic and hemorrhagic stroke,[27] with higher short-term mortality rate and higher prevalence of hypertension in patients with intracranial hemorrhage than in ischemic strokes. Therefore, because stroke subtype is important to study the influence of socioeconomic status in survival, we focused on patients with ischemic stroke only.

We observed that individual cardiovascular risk factors, and especially comorbidity, mildly attenuated the magnitude of the relation between socioeconomic status and survival. Cardiovascular risk factors play an important role in stroke incidence.[7] Some studies have shown in other contexts that the burden of cardiovascular risk factors is higher among patients with lower socioeconomic status[28,29] and there are multiple potential explanations for this finding, including its association with lifestyle factors. However, we found that these factors explained only partially the mechanism of worse outcomes in our population.

Strengths and Limitations

This study has several strengths. Its nationwide population-based approach reduces the risk of ascertainment bias. Analyzing individual socioeconomic status clustering by PCSA avoids the ecological fallacy that is present in several studies.[2] A priori models allow us to analyze the influence of comorbidity and cardiovascular risk factors in this association.

Limitations must also be acknowledged. We did not have data regarding initial stroke severity—an important predictor of acute survival—although its impact is of less importance in predicting long-term survival.[27] Besides, the negative influence of socioeconomic deprivation on initial stroke severity has been described previously.[30] To overcome this limitation, we excluded those patients who died during acute hospitalization in a sensitivity analysis, and results did not change meaningfully. We did not have information about patients attending private health services. As in Catalonia, there is a universal healthcare coverage, and >90% of the population uses the public system, we would estimate a low number of lost cases. We have to mention that individuals are assigned to a particular copayment group on the basis of the income stated in individual income tax declarations. Thus, within a family unit, there could be different incomes. Besides, intragroup differences might exist in the ≥€18 000 income per year category that could not be detected with this classification. Nevertheless, the distribution of the stroke patients in the copayment categories is similar to the general population of Catalonia.[9]

Conclusions and Implications

This nationwide study provides further evidence of the relation between socioeconomic status and short- and long-term survival in patients with ischemic stroke and reinforces the idea that factors other than clinical or healthcare related do play a role in the survival after the disease. Both primary healthcare area and individual socioeconomic aspects should be addressed to achieve equal outcomes in populations with universal healthcare coverage.