Geospatial Analysis for Targeting Out-of-Hospital Cardiac Arrest Intervention
This paper examines the rates of cardiac arrest and bystander CPR for the City of Houston. The paper grew out of a larger body of work examining the link between air quality and health endpoints. As we were studying this link, we observed the unequal distribution of cardiac arrest rates in the City leading to the question of the corresponding bystander CPR rates. It is important to remember that this study is based on the 911 data for the City of Houston, and thus misses some of the smaller cities surrounding the City of Houston. These municipalities operate their own emergency medical services and this information is not included in our study.
The distribution of the cardiac arrest rates by census tract was developed using a normal prior distribution coupled with the observational data to obtain the final estimates of the out-of-hospital-cardiac arrest rates within each census tract. The Bayesian smoothing minimizes spurious observations due to low population levels or simple random variation within a census tract. A corridor of high-risk census tracts was then identified. We were able to describe and compare the corridor and non-corridor demographics.
Our findings relating cardiac arrest and bystander CPR rates are summarized very nicely in one of the paper’s figures (Figure 1 here). We first generated this figure to try to better understand our data, however, it proved highly beneficial in communicating our findings as well. Through this graph we were able to identify census tracts that had unusual circumstances, such as a large senior citizen complex, a prison, as well as community venues such as stadiums and concert halls. These particular census tracts were excluded from our recommendations as these communities already have mechanisms in place to support its members.
The graph provides three key pieces of information by census tract, namely, the out-of-hospital-cardiac-arrest rate, the rate at which bystanders administered CPR, and finally the population size within the census tract. Using these three factors we identified census tracts where community education resources would reach the most people and have the biggest impact.
The targeted census tracts and demographics of those communities were communicated to the City officials. The City in conjunction with community leaders and the American Heart Association quickly launched a neighborhood outreach program to educate citizens of these communities on the importance of bystander CPR and how to perform CPR. At the time of publication of our findings, over 3,000 individuals in these communities have been trained in CPR.
The opportunity to bring quality science expeditiously to the benefit of the community is a noted outcome of this work as well as our ongoing collaboration.
We would like to thank Houston Endowment, the City of Houston and Rice University for supporting this collaboration. We would also like to take this opportunity to acknowledge and thank the Houston Firefighters for their daily dedication to our community.
— Katherine Ensor, PhD, Rice University
Full AJPM Article:
Raun LH, Jefferson LS, Persse D, Ensor KB. Geospatial Analysis for Targeting Out-of-Hospital Cardiac Arrest Intervention. Am J Prev Med 2013;45(2):137-42.