High risk neighbourhoods: The effect of neighbourhood level factors on cardiac arrest incidence. (April 2020)

• Record Type: Journal Article
• Authors: Nascimiento, Barto; Scales, Damon; Ko, Dennis; Hutchison, Jamie; Dainty, Katie; Morrison, Laurie; Dorian, Paul; Swartz, Richard; Verbeek, Richard; Rizoli, Sandro; Cheskes, Sheldon; Brooks, Steven; Lin, Steve; Chan, Tim; Allan, Katherine S.; Ray, Joel G.; Gozdyra, Peter; Morrison, Laurie J.; Kiss, Alexander; Buick, Jason E.; Zhan, Cathy C.; Dorian, Paul
• Is Part Of: Resuscitation. Volume 149(2020)
• Page Start: 100
• Page End: 108
• Publication Date: 2020-04
• Subjects: Out-of-hospital cardiac arrest -- Geocoding -- Incidence -- Residence characteristics -- Census tracts -- Socioeconomic status; Resuscitation -- Periodicals; Resuscitation -- Periodicals; Réanimation -- Périodiques; Electronic journals; 616.025
• DOI: 10.1016/j.resuscitation.2020.02.002
• Languages: English
• ISSNs: 0300-9572

Abstract

Abstract: Background: Numerous studies have shown significant neighbourhood level variation in out-of-hospital cardiac arrest (OHCA) incidence rates, however, few have provided an explanation for these disparities beyond traditional socioeconomic measures. Methods: This was a retrospective study using data from a large population-based OHCA database (Rescu Epistry). We included adults ≥20 years who experienced a non-traumatic OHCA and were treated by emergency medical services within Toronto, Canada between 2006–2012. The residential address of each OHCA patient was spatially mapped to 1 of 517 Toronto census tracts (CTs). Patient and CT level characteristics were included in multivariate regression models to assess their association with OHCA incidence per 100, 000 persons. Results: Of the 7775 OHCAs occurring in the study area, 7692 (98.9%) were eligible for inclusion. OHCA incidence rates varied widely across CT quintiles, with rates differing almost 4-fold (109.1 per 100, 000 yearly Q5 most deprived vs. 30.0 per 100, 000 yearly Q1 least deprived p < 0.0001). Numerous areas of high incidence adjacent to areas of low incidence were observed. After adjustment, all variables except the Activity Friendly Index showed highly significant linear trends, with increasing age, sex ratio, diabetes prevalence, material deprivation and ethnic concentration being independently associated with increasing OHCA incidence. In contrast, we did not observe a linear relationship between high OHCA incidence and median household income. Conclusions: This study showed almost 4-fold OHCA incidence variability across a large metropolitan area. This variability was partially correlated with population and health data, but not typical socioeconomic predictors, such as median household income.