The aim of this study was to identify if there are ‘place’ or area-level differences that impact on out-of-hospital cardiac arrest (OHCA) incidence and outcomes:
• Explore area-level differences in Irish OHCA
• Compare OHCA incidence and outcomes between Sweden and Ireland
• Explore the issue of best achievable outcomes in a ‘place’ that is common to most countries i.e. international airports.
Using OHCA registry data Bayesian Conditional Autoregressive (BCAR) modelling was applied to area level Irish data to produce ‘smoothed’ local incidence rates. A comparison of incidence and outcomes between Sweden and Ireland was then carried out using logistic regression techniques was then performed. Finally multilevel logistic regression was used to investigate differences best achievable outcomes in international airports.
BCAR modelling revealed a significant association between Irish OHCA rates and deprivation, and specific areas where the OHCA rates were highest. As well as confirming that the common Utstein predictors had a similar impact on survival in Sweden and Ireland, results of logistic regression analysis showed the degree of variation that was not explained. The final part of our analysis reported some of the best OHCA survival rates published to date, but indicated that there was country-level variation between international airports.
1. Local area analysis runs the risk of a small number of cases having a disproportionate effect on estimates of incidence and outcome. BCAR modelling can be used to adjust for small numbers and provide more accurate estimation
2. There are unmeasured differences between countries that are not accounted for in routine inter-country comparisons of OHCA incidence and outcome
3. Despite sharing universal characteristics, there is country-level variation in OHCA outcomes international airports
4. The impact of ‘place’ should be considered when investigating OHCA variation, whether at local, national or international levels.