The opioid epidemic in the U.S has gotten payers, prescribers, and policymakers alike interested in trends in opioid use. Despite no recognized opioid crisis in Europe, several countries have reported an increase in opioid-related deaths, which has further prompted discussion on the need of monitoring of opioid prescriptions. This study was conducted to offer information on opioid use during the escalation of the U.S. opioid epidemic in Finland, a Nordic country with universal tax-based health care. This is a nationwide retrospective register-based cohort study on all individuals in Finland who were dispensed opioids in 2009-2017 (n of unique patients = 1,761,584). By using the unique personal identification code assigned to every Finnish resident, we linked data from nationwide registers on dispensed drugs, medical history, and socio-demographic parameters. We report a wide set of patient demographics, dispensing trends for all opioid Anatomical Therapeutic Chemical (ATC) classes, and reasons for opioid initiation based on diagnostic coding for the most recent health care visit. For a cohort of incident opioid users with a four-year wash-out period (n = 1 370 057), we also present opioid use patterns in a three-year follow-up: the likelihood of becoming a persistent user or escalating from weak to strong opioids. A steady 7% of the Finnish population were dispensed opioids annually in 2009-2017. The mean annual quantity of dispensed opioids per opioid patient increased between 2009 and 2017 by 33%, reaching 2 583 oral morphine equivalent mg (OMEQ)/patient/year in 2017. The median quantity of dispensed opioids was lower: 315 OMEQ/year/patient. Depending on the opioid ATC class, there were either increasing or decreasing numbers of patients who had been dispensed said opioid class, and also in the mean quantity. The most common reason for opioid initiation was post-surgical pain (20%), followed by musculoskeletal pain (15%), injury (8.3%), and non-postsurgical dental pain (6.2%). 94% of new opioid initiators started with a weak opioid, i.e. codeine or tramadol. 85% of the patients who had been dispensed a weak opioid were not dispensed an opioid subsequently 3-6 months after the first one, and 95% of them had not escalated to a strong opioid in a 3-year follow-up. The number of patients dispensed opioids in Finland did not change during the escalation of the opioid epidemic in the U.S., but there were changes in the quantity of opioids dispensed per patient. Opioid therapy was typically initiated with weak opioid, the initial dispensed prescription was relatively small, and escalation to strong opioids was rare. A considerable share of patients had been prescribed opioids for chronic non-cancer pain – a type of pain where the risk-benefit ratio of opioids is controversial.