Geographical Association Between Basic Life Support Courses, Bystander Cardiopulmonary Resuscitation and Survival

Background Denmark is known world-wide for our ability to examine out-of-hospital cardiac arrest (OHCA) on a population scale and for the high number of laypersons educated in cardiopulmonary resuscitation (CPR). A nationwide digital dispatch- and patient registry for the respective EMS providers was implemented in 2016. Since then all OHCAs were registered electronically via the EMS digital system and stored nationally. Several characteristics is coupled with each registered OHCA, including bystander-initiated CPR and survival.

Denmark has approximately 5400 OHCAs each year nationally. There has been a significant rise in survival after OHCA in Denmark the last two decades. From 2001 30-day survival rate, from hospital discharge, has increased fourfold from around 4% in 2001 to 16% in 2018. This is likely in part due to a significant rise in bystander cardiopulmonary resuscitation (CPR). Since 2001 bystander-initiated CPR rate has increased from 20% to 77% in 2018. In this same period Denmark has had a threefold increase in the number of Basic Life Support (BLS) courses taken by the public. This is largely due to several central initiatives. To increase bystander CPR rates in Denmark, CPR courses have been made mandatory in all primary schools since January 2005 and when taking driver's license since October 2006. Further, several large-scale national campaigns have been initiated to increase bystander CPR frequency and AED use. In an upcoming study of a representative part of the entire population 4 out of 5 respondents answered that they had participated in some kind of BLS course in their lifetime.

Currently in Denmark, approximately 250.000-300.000 citizens attend certified CPR and automatic external defibrillator (AED) courses annually. Most courses (>97%) are provided by one of the organizations of the Danish First Aid Council. Only a minor proportion of approximately 2.500-4.000 courses annually is provided by the Danish/European resuscitation council. Both recognized structures intend to follow the standards of the ERC guidelines. Since 2016 the Danish first aid council, who conduct most courses, have registered all certificates electronically with personal identification number, type of course and date of passed course for each of the approximately 250.000-300.000 participants. The courses include 4 hours training in BLS including CPR and AED use.

In 2017 a french observational study evaluating the association between BLS education, Public Access Defibrillation programs and survival after OHCA in conjunction with sports in 51 of 96 French districts from 2005-2010 was published. The study concluded that there was a positive association between BLS education and survival even when adjusted for relevant confounders. The authors gained insight to BLS educational level through two sources data. Firstly, from national records of individuals educated by national initiatives from five governmental agencies and a through information from one organization. Secondly, from local initiatives by a media search through web-based media and newspapers with keywords. The authors gained insight into survival through a register based on OHCA related to sports in each in French districts. That is within a district all sports-related OHCA, defined as cardiac arrests occurring during or within 1 hour of cessation of sports activity, whether in a sports facility or outside.

The french observational study provides a concrete basis for further investigations but presents several obvious shortcomings when investigating national BLS educational initiatives.

Firstly, data is not national, and the regions are not geographically aggregated, which presents obvious challenges to the magnitude of the conclusions on a national level. Secondly, there is not a geographically or quantitatively accurate count of BLS educated citizens. The number of citizens with BLS education is recorded over every 5 years with no account of whether the BLS course is attended post or prior to relevant OHCA sport related arrest. Further it seems logical to assume, that laypersons do not respect district borders when assisting for OHCA. Since only about half of the French districts are included, the perspective of persons trained in the other half of the districts is not accounted for. This can be overcome with a comprehensive national geographical setting. Thirdly, the study only compares BLS education with overall survival throughout the district on sports related OHCA. The obvious association to bystander initiated BLS rates is not examined even though bystander-initiated CPR proved the strongest predictor for survival in the study. Hence a causal chain from intervention to survival is not well described. Lastly, as is covered in the article, survival data were taken only from the subgroup of the national sports OHCA registry that represents potentially the best situation for AED use and bystander CPR, due to the high frequency of bystander presence.

These fundamental problems can all be overcome in the Danish setting as there is precise and complete national data with geographical location and other attributes to both BLS education and all OHCA (not only sports related).

Geodata analysis has recently been utilized to compare AED location and OHCA via the Danish Cardiac Arrest registry. The technology is available and applicable to both OHCA locations and BLS education certificates. This enables a precise analysis of both temporal and spatial association between OHCA and BLS education on a national level.

Aim:

1. To examine to association between BLS course certificates, bystander initiated BLS and survival via geographical analysis of distribution
2. To identify geographical areas with low frequency of BLS education and low level of bystander initiated BLS to direct further efforts

Relevance The present study will greatly add to our knowledge about which factors influence outcome after OHCA. It will also provide knowledge that can help optimization of public efforts to increase survival.

Methods The study is a registry-based follow-up study.

OHCA data All cases of OHCA in which a resuscitative attempt is initiated prior to EMS arrival, have specific data recorded to the Danish Cardiac arrest registry. Most variables have been collected since 2001, but GPS coordinates were only recently added nation-wide. Data will be used from all verified OHCA from the national Danish Cardiac Arrest Register from the period mid-2016 to mid-2019.

BLS education data All data concerning BLS education level is derived from the Danish First Aid Council Digital Data base on course certificates from early 2016 to early 2019. The approximately half a year different in start point from OHCA data is to ensure that certificates are issues prior to OHCA. That is to ensure some measure of cause and effect of intervention.

The database is national owned by the Danish First Aid council. In the database, all certificates have attributed time of issue, social security number and certificate type. With personal identification number (PIN) and Central Person Registry the public Danish Statistical Agency can locate official current address of each certificate holder and convert addresses into anonymized geographical coordinates (European Terrestrial Reference System 1989 (ETRS89) or GPS).

Geodata-analysis A collaborating group from Aalborg University Hospital have divided the entire Danish geography into squares of 100*100 meters based on the UTM-system, and the group have obtained data on population density, mean income, educational level, age and several additional variables within each square. This data can be coupled with national administrative parish of Denmark comprising some 2201 geographical units of approximately equal size allowing for a wide range of data coupling through national registries.

In this setting each OHCA is marked with a coordinate and have been tied to the appropriate square along with data on the individual OHCA, that being both pre-hospital factors, burden of disease prior to the event, and long-term outcome. In addition, data on large public facilities, airport, schools etc. where the daytime population will be artificially high does no skew the results. Further adjusted measures will be created on factors possible presenting a skewed OHCA survival profile (i.e. nursing homes, rehabilitation centers and areas with high proportion of elderly). This will allow characterisation of each square by means of general-population-data and by overall OHCA characteristics within the square.

With the data from all certifications of BLS education in the same period all dara can be coupled in the danish administrative parish units. This allows for a statistical association calculation and a direct pinpoint of geographical areas in need of further educational interventions.

Statistical analysis data storage and anonymization Descriptive analysis will be conducted for identifying areas with low BLS educational level and low level of bystander initiated BLS. Geographical illustrations will be completed using ArcGIS. Pearson's correlations or other relevant measure between survival, BLS education level, and bystander initiated BLS will be calculated as a measure of association.

It is further sought to investigate a link between survival and BLS educational level with a Pearson's correlation or a Spearmans Rho test and regression analysis to portray strength. Outcome on survival will, as far as data allows, be adjusted for characteristics of the individual geographical area. Examples of such are population age, gender distribution, mean income and educational level. Furthermore, it is considered relevant to examine potentially relevant sub-group analysis, e.g. after work hours OHCAs. Data management and statistical analyses will be performed using R-static™ (The R Foundation for Statistical Computing, Austria).

All data will be stored in an anonymous form in secure folders in the regional data storage system. Data access and analysis on OHCA was approved by the Danish Data Protection Agency (reference 2007-58-0015, GEH-2014-019, I-suite 02737). A formal application to Danish data Protection Agency regarding usage, transformation and storage of certificate data will be formulated.

Ethics Because this is a Registry-based study no ethical approval is required.

List of abbreviations:

AED: automatic external defibrillator CPR: Cardiopulmonary Resuscitation PIN: Personal Identification Number BLS: Basic Life Support OHCA: out-of-hospital cardiac arrest ETRS89: European Terrestrial Reference System 1989 UTM: Universal Transverse Mercator