Examining Community Involvement among Stroke Survivors in Northern Manhattan: A Model to Predict Participation

Background: The relationship between social isolation in older adults and increased morbidity and mortality is well documented. Utilization of community organizations (CO) is one way to decrease social isolation. Variation in participation by ethnic group in multi-ethnic communities is not well documented.

Methods: Data was gathered through the SWIFT (Stroke Warning Information and Faster Treatment) study of stroke and TIA survivors in Northern Manhattan. The type and frequency of participation in CO prior to stroke/TIA event was collected. Risk factors, social resources, demographics, health literacy and language are gathered. Multivariable regression models were created to examine factors predicting participation.

Findings: The SWIFT cohort enrolled 1068 participants: 49% female; 28% white; 21% Black; 51% Hispanic; 63 years mean age. Overall, 53% reported participation in religious organizations (RO), 30% reporting ≥ once a week. Only 10% reported participation in any CO. Predictors of participation in CO excluding RO included: private/Medicare insurance (OR: 1.8, CI: 1.2, 2.7); education (OR: 1.8, CI: 1.1, 2.9); social support (OR: 2.4, CI: 1.6, 3.4). In this model, Whites (OR: 4.6, CI: 2.9, 3.7) and Blacks (OR: 2.9, CI: 1.8, 4.5) were significantly more likely to participate than Hispanics. RO participation was associated with social support (OR: 1.6, CI: 1.2, 2.1) and gender (OR: 1.5, CI: 1.1, 2.0).

Conclusion: Findings suggest CO participation is low and differs by race/ethnicity. Social support may be a factor in participation overall. Examining variables associated with community participation provides insight into social structure in ethnically diverse communities and can inform targeted interventions.