Regional variation in functional outcome after out-of-hospital cardiac arrest across 47 prefectures in Japan

Abstract

Background: Although prior work reported regional variation in survival after out-of-hospital cardiac arrest (OHCA), mechanisms of the variation have not been fully investigated. We sought to evaluate regional variation in favourable functional outcome after OHCA across 47 prefectures in Japan as our primary aim. We also evaluated the associations between favourable functional outcome and the numbers of basic life support (BLS) providers and public access automated external defibrillators (AEDs) within each prefecture as our secondary aim.

Methods: Using the All-Japan Utstein Registry, a nationwide prospective, population-based OHCA database, we identified 97,408 patients with OHCA of medical origin across 47 prefectures in 2014. Primary outcome was 1-month survival with favourable functional outcome, defined as Cerebral Performance Category (CPC) scale 1 or 2. We fitted multivariable hierarchical logistic regression models (patients nested within prefectures) to adjust for potential confounding factors at patient- and prefecture-level and clustering of patients within prefectures. We calculated median odds ratios (ORs) from the hierarchical models to quantify the outcome variation at prefecture-level. We also evaluated the associations between OHCA outcome and the numbers of BLS providers and public access AEDs within each prefecture, using the hierarchical models.

Results: A total of 2246 patients (2.3%) had 1-month survival with favourable functional outcome. The unadjusted rates of 1-month survival with favourable functional outcome in each prefecture ranged from 1.1% to 4.1% (median OR = 1.29; 95% credible interval, 1.20-1.40) and the adjusted rates varied from 0.9% to 3.5% (median OR = 1.34; 95% credible interval, 1.24-1.48). We observed no associations between 1-month survival with favourable functional outcome and the numbers of BLS providers (correlation coefficient = -0.25; 95% confidence interval [CI], -0.50 to 0.04; p = 0.09) and public access AEDs (correlation coefficient = -0.27; 95% CI, -0.51 to 0.02; p = 0.07) within prefectures.

Conclusions: We found substantial regional variation in favourable functional outcome after OHCA of medical origin that was not explained by the numbers of BLS providers and public access AEDs within each prefecture.

Keywords: Automated external defibrillator; Bystander cardiopulmonary resuscitation; Cardiopulmonary resuscitation; Out-of-hospital cardiac arrest; Regional variation.