Mass gathering medicine: event factors predicting patient presentation rates

This study was conducted to identify the event characteristics of mass gatherings that predict patient presentation rates held in a southeastern US university community. We conducted a retrospective review of all event-based emergency medical services (EMS) records from mass gathering patient presentations over an approximate 23 month period, from October 24, 2009 to August 27, 2011. All patrons seen by EMS were included. Event characteristics included: crowd size, venue percentage filled seating, venue location (inside/outside), venue boundaries (bounded/unbounded), presence of free water (i.e., without cost), presence of alcohol, average heat index, presence of climate control (i.e., air conditioning), and event category (football, concerts, public exhibitions, non-football athletic events). We identified 79 mass gathering events, for a total of 670 patient presentations. The cumulative patron attendance was 917,307 persons. The patient presentation rate (PPR) for each event was calculated as the number of patient presentations per 10,000 patrons in attendance. Overdispersed Poisson regression was used to relate this rate to the event characteristics while controlling for crowd size. In univariate analyses, increased rates of patient presentations were strongly associated with outside venues [rate ratio (RR) = 3.002, p < 0.001], unbounded venues (RR = 2.839, p = 0.001), absence of free water (RR = 1.708, p = 0.036), absence of climate control (RR = 3.028, p < 0.001), and a higher heat index (RR = 1.211 per 10-unit heat index increase, p = 0.003). The presence of alcohol was not significantly associated with the PPR. Football events had the highest PPR, followed sequentially by public exhibitions, concerts, and non-football athletic events. In multivariate models, the strong predictors from the univariate analyses retained their predictive significance for the PPR, together with heat index and percent seating. In the setting of mass event medical care, we note that several factors are strongly associated with an increased patient census, including outside (external) or unbounded venues, the absence of fee water (i.e., without cost), no climate control, percent (occupied) seating, and increasing heat index. Although the presence of alcohol is noted to increase patient needs, it does not do so significantly. Regarding event type, collegiate football games have the highest patient census among the range of other events studied. These findings should be considered during the process of EMS resource planning for mass gatherings.