Purpose: The purpose of this study was to report the locality of out-of-hospital cardiac arrest (OHCA) in the city of Göteborg and to identify implications for public access defibrillation (PAD). Methods: Ambulance run reports for the years 1994–2002 were studied retrospectively and manually to establish the location of the cardiac arrest. Results: The location could be identified in 2194 of 2197 patients (99.9%). One thousand four hundred and twenty-nine (65%) of the arrests took place in the victims’ homes. Two hundred eighty-five (13%) were outdoors and 57 (3%) in cars. Fifty-one (2%) took place en route in ambulances. These arrests were regarded not to be generally suitable for PAD. One hundred thirty-five (6%) of the arrests happened in a public building. Eighteen of these 135 were in 15 different general practitioners’ offices. A ferry terminal had 11 cardiac arrests. One hundred fifty (7%) of the arrests took place in different care facilities. Twenty-one (1%) patients had their cardiac arrest in public transport locations. Twenty-two (1%) patients arrested at work in 20 different sites. In total, 17% of the cardiac arrests were regarded as generally suitable for PAD. Several sites with more than one cardiac arrest in five years could be identified and 54 patients (2.5%) had their cardiac arrest in these high-incidence sites. Conclusion: Among patients suffering from out-of-hospital cardiac arrest in Göteborg in whom resuscitation efforts were attempted 17% of all cardiac arrests were regarded as generally suitable for PAD. According to previous suggestions, the indication for public access defibrillation is in a place with a reasonable probability of use of one AED in 5 years. Several high-incidence sites that probably would benefit from defibrillator availability could be identified, and 54 patients (2.5%) arrested in these sites.