Cheima Amrouch ^1,2* Davide Liborio Vetrano ^3,4 Cecilia Damiano ⁵ Lu Dai ³ Amaia Calderón-Larrañaga ^3,4 Maxim Grymonprez ^6,7 Marco Proietti ^8,9 Gregory Y. Lip ^10,11 Søren P. Johnsen ¹¹ Jonas W. Wastesson ^12,3 Kristina Johnell ¹² Delphine De Smedt ² Mirko Petrovic ⁶

• ¹ Department of Internal Medicine and Paediatrics, Ghent University, Ghent, Belgium
• ² Department of Public Health and Primary Care, Faculty of Medicine and Health Sciences, Ghent University, Ghent, East Flanders, Belgium
• ³ Karolinska institutet, Aging Research Center (ARC), Stockholm, Stockholm, Sweden
• ⁴ Stockholm Gerontology Research Center Foundation, Stockholm, Stockholm, Sweden
• ⁵ Department of Cardiovascular, Endocrine-Metabolic and Ageing-Associated Diseases, National Institute of Health (ISS), Rome, Lazio, Italy
• ⁶ Department of Internal Medicine and Pediatrics, Faculty of Medicine and Health Sciences, Ghent University, Ghent, East Flanders, Belgium
• ⁷ Ghent university, Department of Bioanalysis, Pharmaceutical Care Unit, Ghent, Belgium
• ⁸ Department of Clinical Sciences and Community Health, University of Milan, milan, Italy
• ⁹ Division of Subacute Care, IRCCS Istituti Clinici Scientifici Maugeri, Milan, Italy
• ¹⁰ Liverpool Centre for Cardiovascular Science at University of Liverpool, Liverpool John Moores University and Liverpool Heart & Chest Hospital, Liverpool, United Kingdom
• ¹¹ Danish center for health services research, Department of clinical medicine, Aalborg University, Aalborg, Denmark
• ¹² Department of Medical Epidemiology and Biostatistics, Karolinska Institutet (KI), Stockholm, Stockholm, Sweden

Current research on potentially inappropriate prescribing (PIP) in polymedicated older adults with atrial fibrillation (AF) and multimorbidity is predominantly focused on PIP of oral anticoagulants (OAC). Our study aimed to assess (i) the prevalence of PIP in older multimorbid adults with AF, (ii) potential associated factors of PIP, and (iii) the association of PIP with adverse health outcomes in Swedish older adults nationwide. Swedish national registries were linked to establish a cohort with a two-year follow-up of older adults (≥ 65y) who, on 1 st January 2017, had a diagnosis of AF and had at least one comorbidity (n=203,042). PIP was assessed using the reduced STOPP/START version 2 screening tool. The STOPP criteria identify potentially inappropriate prescribed medications (PIM), while the START criteria identify potential prescribing omissions (PPO). PIP is identified as having at least one PIM and/or PPO. Cox regression analyses were conducted to examine the association between PIP and adverse health outcomes: mortality, hospitalisation, stroke, bleeding, and falls. PIP was highly prevalent in older adults with AF, with both polypharmacy (69.6%) and excessive polypharmacy (85.9%). In the study population, benzodiazepines (22.9%), hypnotic Z-medications (17.8%) and analgesics (8.7%) were the most frequent PIM. Anticoagulants (34.3%), statins (11.1%), vitamin D and calcium (13.4%) were the most frequent PPO. Demographic factors and polypharmacy were associated with different PIM and PPO categories, with the nature of these associations differing based on the specific type of PIM and PPO. The co-occurrence of PIM and PPO, compared to appropriate prescribing, was associated with an increased risk of adverse health outcomes compared to all appropriately prescribed medications: cardiovascular (CV) (Hazard ratio (HR) [95% confidence interval] = 1.97 [1.88-2.07]) and overall mortality (HR= 2.09 [2.03-2.16]), CV (HR= 1.34 [1.30-1.37]) and overall hospitalisation (HR= 1.48 [1.46-1.51]), stroke (HR= 1.93 [1.78-2.10]), bleeding (HR= 1.10 [1.01-1.21]), and falls (HR= 1.63 [1.56-1.71]). The present study reports a high prevalence of PIP in multimorbid polymedicated older adults with AF. Additionally, a nuanced relationship between prescribing patterns, patient characteristics, and adverse health outcomes was observed. These findings emphasise the importance of implementing tailored interventions to optimise medication management in this patient population.