BACKGROUND: In varying doses, opioids, including heroin, produce analgesia, impaired consciousness, respiratory depresssion, and death. We have seen a worrying nationwide increase in the incidence of opioid overdose, and a better understanding of the problem is fundamental to remedying it. Naloxone is a medicine that has one and only one use in clinical medicine—the treatment of opioid overdose. Thus naloxone administration by emergency medical services (EMS) personnel can serve as a useful surveillance indicator for opioid overdose. Using areal data aggregated to administrative spatial units (e.g. census tracts), two previous studies have demonstrated a concentration of EMS naloxone administrations in certain parts of two cities (Dublin, Ireland and Providence, Rhode Island.) The research question in the present study was whether EMS calls for opioid overdose, represented by the administration of naloxone, demonstrate spatial clustering over and above that expected from EMS calls in general, using their individual locations and the techniques of point pattern analysis
METHODS: The Susquehanna EMS Region has a population of about 300,000 and a surface area of about 6000 sq km. The Regional Emergency Medical Services Council (REMSCO) maintains an electronic database of patient care reports from all 71 EMS agencies in the Region. Thirteen agencies participated in the study; together they account for the vast majority of all Regional EMS calls. The study period was 9 September 2012 to 9 February 2014. A retrospective chart review was conduced to explore the spatial distribution of EMS calls in which naloxone was administered. The F, G, K, and L functions were used as measures of clustering or aggregation. Random samples from the distributiion of all EMS calls were used in Monte Carlo simulation to represent the background inhomogeneity of the population.
RESULTS: 10,826 usable patient care reports were identified; 183 involved administration of naloxone. Cases of naloxone administration demonstrated spatial clustering, especially around 2000 meters, over and above what would be attributable to spatial heterogeneity of EMS calls in general.
CONCLUSIONS: The point pattern of opioid overdose EMS calls, as represented by naloxone administrations, demonstrates spatial clustering. Further research could yield insights into opioid overdose as a patial point process, perhaps leading to innovative public health interventions to control the epidemic.