Community-level bystander treatment and outcomes for witnessed out-of-hospital cardiac arrest in the state of Connecticut

Abstract

Background: Prior reports have demonstrated underutilization of bystander cardiopulmonary resuscitation (CPR) and automated external defibrillator (AED) use in patients with witnessed out-of-hospital cardiac arrest (OHCA) in Connecticut. This study aimed to identify community-level risk factors that contribute to low rates of bystander intervention to improve statewide OHCA outcomes.

Methods: We analyzed 2,789 adult patients with witnessed, non-traumatic OHCA submitted to the Connecticut Cardiac Arrest Registry to Enhance Survival (CARES) between 2013-2022. Patients were grouped by zip code, and associated municipal characteristics were acquired from 2022 United States Census Bureau data. Use of bystander CPR, attempted bystander AED defibrillation, and patient survival with favorable neurological function were determined for 19 of the 20 most populous cities and towns. Pearson correlation tests and linear regression were used to determine associations between OHCA treatment and outcomes with population size, racial/ethnic demographics, language use, income, and educational level.

Results: Bystander CPR was lower in municipalities with population size > 100,000 and in communities where > 40% of residents are non-English-speaking. AED use was also lower in these municipalities, as well as those with per capita incomes < $40,000 or > 1/3 Hispanic residents. Communities with populations > 100,000, > 40% non-English-speaking, per capita income < $40,000, and > 1/3 Hispanic residents were all associated with lower survival rates.

Conclusions: OHCA pre-hospital treatment and outcomes vary significantly by municipality in Connecticut. Community outcomes might be improved by specifically targeting urban population centers and Hispanic communities with culturally sensitive, low, or no-cost CPR and AED educational programs, using instructional languages other than English.

Keywords: Bystander intervention; Cardiopulmonary resuscitation; Community health; Health disparities; Out-of-hospital cardiac arrest; Social determinants of health.

Fig. 1

Relationships between independent municipal census statistics (population [n] and primary language other than English [%]) and dependent outcomes of bystander CPR percentage, bystander AED percentage, sustained ROSC, or survival to discharge. Red dots represent individual municipalities. Solid red lines indicate significant linear correlations by Pearson test (p < 0.05), black dashed lines indicate insignificant tests (p > 0.05). Linear fit equation, correlation coefficient (r) and p-values presented in upper left of graphs. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Relationships between independent municipal census statistics (per capita income [$] and Hispanic/Latino race/ethnicity [%]) and dependent outcomes of bystander CPR percentage, bystander AED percentage, sustained ROSC, or survival to discharge. Red dots represent individual municipalities. Solid red lines indicate significant linear correlations by Pearson test (p < 0.05), black dashed lines indicate insignificant tests (p > 0.05). Linear fit equation, correlation coefficient (r) and p-values presented in upper left of graphs. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Pooled odds ratios (ORs) between municipal risk factors and dependent outcomes. Data points represent ORs plus or minus 95% confidence intervals. Vertical line at 1 represents threshold of no association; values with an upper bound of < 1 represent statistically significant risk factors.

Fig. 4

Bystander CPR versus sustained ROSC and survival by county in Connecticut. Red dots represent individual counties. Solid red lines indicate significant linear correlations by Pearson test (p < 0.05). Vertical black line at 40% bystander CPR represents national average. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)