123 Utilization Patterns by Socioeconomic Status in a Large Urban Emergency Medical Service

Abstract

Relationships between aggregate volumes of Emergency Medical Services (EMS) utilization and socioeconomic status (SES) have been previously described, most often demonstrating an inverse relationship between the 2. Epidemiologic research from emergency department or hospital records have demonstrated increased prevalence of certain medical conditions in low-income populations. In this study, we sought to describe the relationship between SES and utilization of EMS in New York City (NYC) across a spectrum of specific medical concerns, and determine if there are differences in the patterns of utilization across different SES groups. This retrospective cohort study was performed using data from the electronic Patient Care Report database from the Fire Department of the City of New York (FDNY), the municipal provider of EMS for NYC, accounting for approximately 64% of 911 EMS responses. The FDNY patient care report database was queried for all patient contacts between November, 2005 and December, 2009. The patient's medical concerns, as well as home zip code, were extracted. Each unique patient encounter was assigned an income value as a proxy for SES by correlating the patient's home zip code to published United States Census Bureau median income data for 2008. Patients were then divided into deciles according to the assigned income. Using this method, the relative frequency of each medical concern in each decile was tabulated. Statistical correlations between the relative frequency of each medical concern and income decile were measured using Spearman's rank coefficients. Individual medical concerns were categorized as either medical/surgical or trauma related and similarly correlated to income decile. All data analysis was performed using the Statistical Package for the Social Sciences (SPSS) v18. A total of 2,402,675 unique subject encounters were analyzed. Statistically significant negative correlations (lower SES associated with higher EMS utilization) were identified for the principal concerns of fever (-0.988), sore throat (-0.988), abdominal pain (-0.976), asthma (-0.927) and stabbings (-0.782) [all p<0.01]. Statistically significant positive correlations (higher SES associated with higher EMS utilization) were found for the principal concerns of syncope (0.988), motor vehicle accidents (0.988), falls (0.976), and respiratory arrest (0.939) [all p<0.001]. No significant trends were seen for assaults, gastrointestinal bleeds, chest pain, or allergic reactions. Overall, there was a positive correlation for all combined trauma related concerns (0.939, p<0.001) and a negative correlation for all combined medical/surgical concerns (-0.927, p<0.001). Using income level as a surrogate for SES, our study demonstrates very strong relationships between SES and EMS utilization among specific medical concerns. While aggregate volumes of EMS calls are known to be higher among low SES populations, not all types of medical concerns rise uniformly. The distribution of medical concerns varies considerably at different income levels, with lower income levels tending to utilize EMS in greater proportions for chronic conditions such as asthma. This out-of-hospital data supports findings from hospital studies demonstrating the need for increased primary and preventative care in low-income communities.