Cluster Analysis as a Statistical Method for Planning the Optimal Placement of Automated External Defibrillators

<b>Background/Objectives:</b> Out-of-hospital cardiac arrest (OHCA) remains a major public health challenge, with survival rates significantly dependent on early defibrillation. In Bialystok, Poland, the bystander usage rate of automated external defibrillators (AEDs) is extremely low, and the current distribution of public-access AEDs may not support optimal response times. The aim of this study was to identify an effective AED placement strategy using spatial analysis. <b>Methods:</b> We retrospectively analyzed 49,649 emergency dispatch records from 2018 to 2019, identifying 787 patients with OHCA within Bialystok's city limits. After excluding ineligible records, 766 cases were geolocated and subjected to cluster analysis using the K-means algorithm. The goal was to determine optimal AED locations based on the geographic distribution of OHCA cases in both public and residential settings. <b>Results:</b> AEDs were used in only 0.51% of all cases of OHCA. Most cardiac arrests occurred in private homes (80.05% of cases). Cluster analysis identified 18 to 36 optimal AED locations, revealing significant mismatches with the current AED network. Notably, grocery store chain "PSS Spolem" emerged as an ideal AED deployment partner due to alignment with identified high-incidence clusters. <b>Conclusions:</b> The current AED distribution in Bialystok is inadequate for an effective response to OHCA. Geographic cluster analysis can significantly improve placement strategies. Priority should be given to residential areas and commonly accessed sites. Enhanced public education, a national AED registry, and improved accessibility are essential for increasing AED use and survival rates.