Aim of the research: The aim of this study was to determine the frequency and characterize spontaneously reported adverse drug reactions (ADRs) to Croatian Agency for Medicinal Products and Medical Devices (HALMED) in 2012 and 2016 in persons over 65 years of age, related to the use of the medicinal products outside the terms of the marketing authorisation. Additionally, the aim was to determine the cause of these ADRs. Materials and methods: A retrospective observational study of spontaneously reported suspected ADRs to HALMED in 2012 and 2016 in persons over 65 years of age. The Summary of Product Characteristics was used as a reference document to determine the association of the ADRs with the use of the medicinal product outside the terms of the marketing authorization. ADRs were analyzed according to patient age, reporter's qualification, seriousness of ADR, MedDRA system organ class (SOC), Anatomical Therapeutic Chemical (ATC) classification, number of used drugs, and number of comorbidities. Statistical evaluation data was performed using IBM SPSS v. 20.0 statistical software. Results: A total of 196 reports of ADRs to HALMED in persons over 65 years of age were included in the analysis. Out of the 48 (24,5 %) ADR reports associated with the use of the medicinal product outside the terms of the marketing authorization, majority were due to medication errors (N = 39). There were 69 serious ADR reports in persons over 65 years of age, and 19 of them were caused by the use of the medicinal product outside the terms of the marketing authorization. Therapeutic drug classes the most frequently involved in ADRs caused by the use of the medicinal product outside the terms of the marketing authorization included cardiovascular system agents (ATC code C). Skin and subcutaneous tissue disorders (15,9 %) and gastrointestinal disorders (13,6 %) were the most commonly reported ADRs aroused from the medicinal product use outside the terms of the marketing authorization. In 66,7 % of ADR reports arising from the use of drugs outside the terms of the marketing authorization were ADRs type A, and in 33,3 % of ADR reports were ADRs type B. Conclusion: The results of this study will complement existing evidence on the frequency and characteristics of ADRs in elderly patients. In addition, it will allow the development of a strategy to prevent ADRs that arise from the use of medicinal products outside the terms of the marketing authorization.