Economic Evaluation Study Medicine ® OPEN Cost-effectiveness of direct oral anticoagulants versus vitamin K antagonist in atrial fibrillation A study protocol using Real-World Data from Catalonia (FantasTIC Study) ∗ Zojaina Hernández Rojas, MDa,b,c,d, Maria Rosa Dalmau Llorca, MD PhDa,b,c,d, , Carina Aguilar Martín, MD, PhDb,d,e, Alessandra Queiroga Gonçalves, PhDb,d,f, Marc Casajuana, MScg,h, José Fernández-Sáez, PhDb,d,i,j, Dolores Rodríguez Cumplido, PhDd,k, Emma Forcadell Drago, MD PhDd,l, Noèlia Carrasco-Querol, PhDb, Josep Maria Pepió Vilaubí, MD PhDd,l, Josep M. Alegret, MD PhDm,n Abstract Background: Anticoagulant therapy is used for stroke prevention and proved to be effective and safe in the long term. The study aims to analyse the cost-effectiveness relationship of using of direct-acting oral anticoagulants vs vitamin K antagonists to prevent ischaemic stroke in patients with nonvalvular atrial fibrillation, including all the active ingredients marketed in Spain, prescribed for 2 years in the Primary Care service of the Institut Català de la Salut. Methods: Population-based cohort study, in which the cost of the 2 treatment groups will be evaluated. Direct costs (pharmacy, primary care, emergency and hospitalization) and indirect costs (lost productivity) will be included from a social perspective. Editor: Leonardo Roever. Maria Rosa Dalmau Llorca, Carina Aguilar Martín and Josep M. Alegret Co-senior authors. The database was obtained through a SIDIAP grant (2017 Competitive Call), reference SIDIAP 02/17 (resolution date 04/27/18). The project will be developed within the framework of the 2018 Call for Industrial Doctorates through the Agència de Gestió d‘Ajuts Universitaris i de Recerca of the Generalitat de Catalunya, reference 2018 DI 044 and co-financed with a grant from the Fundació Dr. Ferran (2018 call) awarded on 17 May 2018. In addition, the study obtained a competitive, public funding of the Department of Health of the Generalitat de Catalunya, in the call corresponding to the year 2019 of concession of grants of the Strategic Plan of Research and Innovation in Health (PERIS) 2016-2020, modality Intensification of Specialist Physicians, with the file code SLT008/18/00021. The institutions will finance the project but will not participate in the research or have authority over any of its activities. Z Hernández Rojas declares having received help to attend conferences and to give presentations from Almirall and Boehringer, and partial sponsorship of a study funded by Bayer. M Rosa Dalmau Llorca declares having received help to attend Bayer conferences, and partial sponsorship of a study funded by Bayer. C Aguilar Martín declares having received partial sponsorship of a study funded by Bayer. A Queiroga Gonçalves declares having received partial sponsorship of a study funded by Bayer. J Fernández Sáez declares having received partial sponsorship of a study funded by Bayer. D Rodriguez Cumplido declares having received partial sponsorship of a study funded by Bayer. E Forcadell Drago declares having received help to attend congresses and meetings sponsored by MSD and GlaxoSmithKline and having received partial sponsorship of a study funded by Bayer. JM Pepió Vilaubí declares having received partial sponsorship of a study funded by Bayer. JM Alegret declares having received lecture fees from Daiichi Sankyo. M Casajuana and N Carrasco-Querol declare no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analyzed during the current study. a Equip d’Atenció Primària Tortosa Est, Institut Català de la Salut, Tortosa, Tarragona, Spain, b Unitat de Suport a la Recerca Terres de l’Ebre, Fundació Institut Universitari per a la Recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Tortosa, Tarragona, Spain, c Universitat Rovira i Virgili, Tarragona, Spain, d GAVINA Research Grup, e Unitat d’Avaluació, Direcció d’Atenció Primària Terres de l’Ebre, Institut Català de la Salut, Tortosa, Tarragona, Spain, f Unitat Docent de Medicina de Familia i Comunitària, Tortosa-Terres de l’Ebre, Institut Català de la Salut, Tortosa, Tarragona, Spain, g Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Barcelona, Spain, h Universitat Autònoma de Barcelona, Bellaterra (Cerdanyola del Vallès), Spain, i Unitat de Recerca, Gerència Territorial Terres de l’Ebre, Institut Català de la Salut, Tortosa, Tarragona, Spain, j Facultat d’Enfermeria, Campus Terres de l’Ebre, Universitat Rovira i Virgili, Tortosa, Tarragona, Spain, k Hospital Universitari de Bellvitge, Institut Català de la Salut, Barcelona, Spain, l Equip d’Atenció Primària Tortosa Oest, Institut Català de la Salut, Tortosa, Tarragona, Spain, m Grup de Recerca Cardiovascular, Departament de Cardiologia, Hospital Universitari de Sant Joan, Institut de Investigació Sanitaria Pere Virgili (IISPV), Reus, Spain, n Departament de Medicina i Cirurgia, Universitat Rovira i Virgili, Reus, Spain. ∗ Correspondence: Maria Rosa Dalmau Llorca, Av. Cristòfol Colom 14-20, 43500 Tortosa, Tarragona (e-mail: rdalmau.ebre.ics@gencat.cat). Copyright © 2020 the Author(s). Published by Wolters Kluwer Health, Inc. This is an open access article distributed under the Creative Commons Attribution License 4.0 (CCBY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. How to cite this article: Hernández Rojas Z, Dalmau Llorca MR, Aguilar Martín C, Gonçalves AQ, Casajuana M, Fernández-Sáez J, Rodríguez Cumplido D, Forcadell Drago E, Carrasco-Querol N, Pepió Vilaubí JM, Alegret JM. Cost-effectiveness of direct oral anticoagulants versus vitamin K antagonist in atrial fibrillation: A study protocol using Real-World Data from Catalonia (FantasTIC Study). Medicine 2020;99:36(e22054). Received: 28 April 2020 / Received in final form: 22 July 2020 / Accepted: 2 August 2020 http://dx.doi.org/10.1097/MD.0000000000022054 1 Hernández Rojas et al. Medicine (2020) 99:36 Medicine Effectiveness (assessed as the occurrence of a health event, the 1 of primary interest being stroke) will be determined, with a 2-year time horizon and a 3% discount rate. The average cost of the 2 groups of drugs will be compared using a regression model to determine the factors with the greatest influence on determining costs. We will carry out a univariate (‘one-way’) deterministic sensitivity analysis. Discussion: We hope to provide relevant information about direct and indirect costs of oral anticoagulants, which, together with aspects of effectiveness and safety, could help shape the consensual decision-making of evaluating bodies. Abbreviations: DOACs = direct-acting oral anticoagulants, ICS = Institut Català de la Salut, NVAF = nonvalvular atrial fibrillation, PC = primary care, VKAs = vitamin K antagonists. Keywords: atrial fibrillation, cost effectiveness, primary health care, anticoagulants, oral drug administration 1. Introduction However, the most recent study carried out in Spain does not show significant differences between dabigatran and AVKs.[26] A Atrial fibrillation (AF) is the most common arrhythmia in western recent study of edoxaban in Spain concluded that it is a cost- countries and is associated with high morbidity and mortality, effective alternative for this indication.[27] However, we did not thereby constituting a major public health problem.[1,2] It is 1 of find any cost-effectiveness studies that included all the anti- the most frequent causes of ischaemic stroke.[3,4] Anticoagulant coagulants currently available on the market, with a population therapy is used for stroke prevention, for which there are 2 main base (without economic models) and real world data (RWD). groups of drugs —vitamin K antagonists (VKAs) and direct- The main objective of the study will be to analyse the cost- acting oral anticoagulants (DOACs)— which have both proved effectiveness of using DOACs in comparison with VKAs to effective and safe in the long term.[5] prevent ischaemic stroke in patients with NVAF, including all the VKAs have been widely used for a long time and their efficacy active ingredients marketed in Spain prescribed for 2 years in the in preventing strokes and systemic embolism in patients with Primary Care (PC) service of the Institut Català de la Salut (ICS). nonvalvular atrial fibrillation (NVAF) is well established.[6] As secondary objectives, we intend to: However, most of the studies have involved warfarin, and the results have been generalized to all VKAs.[7] The price of these (1) determine the appearance of a health event, according to the drugs is low (compared with that of DOACs),[8] although they type of anticoagulation used (VKA or DOAC), and the generate direct and indirect costs. Furthermore, they have factors related to this; limitations such as interindividual and intraindividual variability (2) evaluate the cost in patients who are poorly controlled by of response, a slow onset of action, a narrow therapeutic margin, VKAs and under-treated with DOACs; and a need for dose-adjustment through periodic controls of the (3) analyse the factors associated with increased costs. international normalized ratio, and interactions with some foods The present study protocol, named FantasTIC (Non-valvular and medications.[9–11] atrial fibrillation and treatment, health assessment, Information In recent years, DOACs have emerged that have more and Communication Technologies) study has been designed to predictable anticoagulant effects and fewer interactions with address these aspects. other drugs, and that allow a fixed dosage regimen, without the need for monitoring.[12] Overall, DOACs present a favorable risk prevention profile, with significantly lower levels of strokes, 2. Methodology intracranial haemorrhages and mortality compared with VKAs, but they have a higher risk of gastrointestinal bleeding[7] and are 2.1. Design and study period expensive.[8] It is also important to take into account the This is a population-based cohort study designed to evaluate adjustment of the dosage according to age, weight, and renal and compare the cost-effectiveness of the 2 types of function, in addition to monitoring compliance, because highly anticoagulant treatment (VKA and DOAC) in patients with variable therapeutic adherence rates have been noted.[7,13] NVAF, with a time horizon of 2 years (from 1 January 2017 until The European and American guidelines recommend the use of 31 December 2018). The flow diagram of the study is presented any DOACs rather than VKAs because of their net clinical in Figure 1. benefit, except in selected cases.[14–16] However, the use of DOACs in clinical practice is determined by the recommenda- 2.2. Study population tions of the health authorities of each country. In Spain, VKAs are recommended as the first choice, with DOACs being used when Our study population will be that treated by the PC service of the specifically recommended,[17] although currently it has been ICS. The Autonomous Community of Catalonia has a health observed that more than 80% of physicians are starting service whose main provider is the ICS, in which PC is organized treatments with DOACs .[18] into teams made up of family doctors, nurses, paediatricians, Previous studies found that DOACs tend to be more cost- social workers, dentists and support staff. The ICS manages, effective than VKAs for the treatment of AF.[19–23] One of these among others, 287 PC centres (PCCs) to which 5,564,292 studies suggest that rivaroxaban is cost-effective compared to citizens are assigned (approximately 80% of the Catalan VKA, although the type of AVK and the dose thereof are not population).[28] specified.[23] A comparative study of the cost-utility of the first 3 As established by the Sistema de Información para el DOACs to be marketed in Spain (apixaban, dabigatran and Desarrollo de la Investigación en Atención Primaria (SIDIAP; rivaroxaban),[24] found dabigatran to be the most cost-useful. [25] Information System for the Development of Research in PC), 2 Hernández Rojas et al. Medicine (2020) 99:36 www.md-journal.com SIDIAP database NVAF patients receiving anticoagulant treatment Exclusion criteria: • Patients with indications of anticoagulation other than NVAF • Pregnant women Study population Distribution Group 1 Group 2 Vitamin K antagonists Direct-acting oral anticoagulants (Sociodemographic variables, (Sociodemographic variables, Comorbidities) Comorbidities) Cessaon of follow-up: • Move away • Death 2017 Follow-up • Pregnancy • AF diagnosis withdrawn 2018 • Appearance of valvulopathy Evaluation of costs Evaluation of costs ---------------- ---------------- Appearance of health events Appearance of health events (ischaemic or haemorrhagic) (ischaemic or haemorrhagic) Data analysis (Comparison of pharmacological groups) Figure 1. Flow diagram. during the year 2017, 126,702 cases of NVAF had been (2) Patient undergoing treatment with VKA or DOAC. registered in the ICS, 61,002 of which were receiving anticoagu- Exclusion criteria: lant treatment (41,430 VKA and 19,548 DOAC). Inclusion criteria: (1) Patient with indications for anticoagulation other than NVAF. (1) Patient with a diagnosis of NVAF at least 1 year old. (2) Pregnancy. 3 Hernández Rojas et al. Medicine (2020) 99:36 Medicine Table 1 Variables, sampling period, and source of information. Variables Baseline 1 year 2 years Source of information Source of costs Personal history Sociodemographic characteristics • • • eCAP Primary care centre of origin • eCAP Comorbidities • • • eCAP CHA2DS2-VASc score • • • eCAP HAS–BLED score • • • eCAP Date of NVFA diagnosis • eCAP Charlson comorbidity index • • • eCAP Pharmacy variables Anticoagulant • • • Pharmacy Follow-up variables Biannual TTR • • • eCAP Estimated GF • • • eCAP DOAC dosage • • • eCAP Cost variables Primary care visits • • eCAP DOGC Hospital admissions • • CMBD-AH DOGC Institutionalisations • • eCAP DOGC Visits to specialists • • eCAP DOGC Temporary disability • • eCAP INE Laboratory tests • • Laboratory DOGC Complementary tests • • CMBD-AH DOGC Pharmaceutical expenses • • Pharmacy RPI, DH Effectiveness variables Ischemic stroke • • eCAP/CMBD-AH Other thromboembolic episodes • • eCAP/CMBD-AH Intracranial haemorrhage • • eCAP/CMBD-AH Digestive haemorrhage • • eCAP/CMBD-AH Other haemorrhages • • eCAP/CMBD-AH Mortality • • DH CMBD-AH = Conjunt Minim Bàsic de Dades - dates, diagnoses and procedures (ICD-9 Codes) linked to admissions in any of the hospitals in Catalonia, DH = Department of Health, DOAC = direct-acting oral anticoagulant, DOGC = Diari Oficial de la Generalitat de Catalunya [Official Journal of the Government of Catalonia], eCAP = Estación Clínica de Atención Primaria [Clinical History of Primary Care], GF = glomerular filtration, INE = Instituto Nacional de Estadística [National Statistics Institute], NVFA = nonvalvular atrial fibrillation, RPI = retail price index, TTR = time in therapeutic range. Data collection will end on 31 December 2018 or sooner if any (1) data from the eCAP program: demographic data, PC visits, of the following occurs: health events, clinical variables, referrals, deaths, prescrip- tions and sick leave; (1) Anticoagulation prescription suspended for more than 180 (2) laboratory results: these will be extracted directly from the days. laboratory database, rather than depending on manual (2) Patient moves to another Autonomous Community. records, thereby guaranteeing data quality; (3) Patient becomes pregnant. (3) medication dispensed by pharmacy offices: this information (4) Withdrawal of the diagnosis of NVAF. will be obtained directly from those offices; (5) Appearance of a diagnosis of valvular AF. (4) other, external sources of information: (4.1.) Conjunt Minim (6) Death of the patient. Bàsic de Dades (CMBD), which is a population registry that collects ICD-9 pathology data linked to hospitalization at all of the Hospitals in Catalonia,[31] (4.2.) Mortality: data 2.3. Data source supplied by the Department of Health, including cause and At ICS, more than 9175 professionals working in PC use the same date of all deaths of residents of Catalonia.[28] computerized medical history program, called eCAP. eCAP data will be obtained through SIDIAP, which is a unique database, previously validated, and highly representative of the Catalan 2.4. Variables population.[29,30] In this way, SIDIAP provides, from each of the 5.8 million The variables considered in the study will be divided into 5 citizens assigned to the various PCCs of the ICS, information groups: cost, follow-up, effectiveness, pharmacy, and those linked to a unique, anonymised identifier. The following corresponding to personal history (summarised in Table 1). information will be obtained from all NVAF patients receiving All oral anticoagulants that are marketed in Spain will be anticoagulants on January 1, 2017: included in the study: 4 Hernández Rojas et al. Medicine (2020) 99:36 www.md-journal.com (1) VKAs: Acenocoumarol and warfarin. developed to determine which factors may have the most (2) DOACs: Apixaban, dabigatran, edoxaban and rivaroxaban. substantial influence on determining costs. A generalized linear model (GLM) will be derived, which, given the asymmetry of the cost variable, will be based on the gamma family of distributions, 2.5. Cost estimation with a log link function,[38] because these give a best fit to the data Cost estimates will take into account direct expenses (from when estimating the impact of the factors associated with the pharmacy, PC, emergencies and hospitalisation) and indirect composition of patient costs. expenses (loss of productivity), from a social perspective, in patients diagnosed with NVAF and treatment with oral 2.8. Sensitivity analysis anticoagulants. Two groups will be established: patients receiving A univariate (1-way) deterministic sensitivity analysis will be VKAs and those receiving DOACs. We will use a 3% discount carried out, in the items which the greatest impact on the rate, which will be applied in the second year. composition of the cost, or in which there may be a uncertainty We will use the charges set out in the Diari Oficial de la with the real price.[39] This will allow us to test the robustness of Generalitat de Catalunya (DOGC)[32] closest to the date of our results. analysis (2014) to estimate the costs. The euro will be used as the reference currency (which will not require conversion). Prices will be adjusted to those of 2018 to take into account inflation in the 3. Ethics and dissemination health sector.[33] The costs of hospital admissions will be This study will be carried out in accordance with the norms and obtained from the Grups Relacionats amb el Diagnòstic principles of the Declaration of Helsinki. The protocol was (GRD) patient classification system, published in the approved by the Clinical Research Ethics Committee of the DOGC.[34,35] For drug prices, the retail price taken from Fundació Institut Universitari per a la recerca a l’ Atenció pharmacy billing data will be used. The final costs of the 2 Primària de salut Jordi Gol i Gurina (IDIAPJGol) on May 30, study groups (patients treated with VKAs and those treated with 2018 with code P18/080, and has also been authorized by the DOACs) will be calculated by quantifying the resources Primary Healthcare Directorate of the ICS. The SIDIAP database consumed by each individual with respect to each variable and provides anonymised data, identified with an internal code, then multiplying by the cost of each resource. This procedure will which makes it impossible, even for the research team, to identify yield the total cost of the resource per individual, as well as the any subject. For the same reason, informed consent will not be total cost for each study group. obtained. By these means, the confidentiality of the data of the individuals included in the study will be guaranteed, in 2.6. Measurement of effectiveness accordance with the Organic Law on Protection of Personal The primary health event in this project will be “ischaemic Data (03/2018 of December 5, and in accordance with the stroke” and the indicator of cost-effectiveness will be “avoided provisions of the Regulation 2016/679 of the European strokes”. To estimate avoided strokes we will determine the total Parliament and of the Council of 27 April 2016 on data economic cost for the 2 pharmacological groups, as well as the protection, and all applicable national regulations. If any number of strokes registered during the 2 years of the study modification is required, it will be presented to the IDIAPJGol period. Finally, we will compare the groups and thereby Clinical Research Ethics Committee for approval. determine the economic cost of avoiding a stroke. The results and conclusions of this study will be disseminated Secondarily, we will determine the appearance of the following in Spain, as well as in international scientific and professional health events during the 2 years of the study: forums, with the results published in indexed scientific journals.  Ischaemic events: (1) Ischemic stroke. 4. Discussion (2) Other thromboembolic events (transient ischaemic attack The aim of the present study is to carry out an economic (TIA), angina, acute myocardial infarction, and peripheral evaluation of all the anticoagulants indicated for the prevention embolism). of thromboembolic complications in patients with NVAF. • Haemorrhagic events: VKAs are the most used anticoagulants worldwide.[40] (1) Intracranial haemorrhage. However, the administration criteria differs in each country. (2) Gastrointestinal bleeding. For example, while in Continental Europe, the VKA most widely (3) Other haemorrhages. used is the acenocoumarol, in the United States and the rest of • All-cause mortality. Europe, it is the warfarin.[41] Several studies have been carried out with warfarin. [15,16,19,20,25,42] Though, there is scarce literature on acenocoumarol, [21,22,24,27] which is the most popular VKA in 2.7. Economic analysis Spain. In the clinical practice, the results of the studies including The cost of DOACs will be compared with that of VKAs, warfarin are assumed to be generalizable.[7] Nevertheless, and following the methodological guidelines for this type of study from the best of our knowledge, there is no evidence to support proposed by the International Society for Pharmacoeconomics this assumption. and Outcomes Research (ISPOR)[36] and the International The main strength of our study is the analysis of all marketed Network of Agencies for Health Technology Assessment anticoagulants from Spain, which will provide strong evidence (INAHTA).[37] about their differences, including both VKAs (warfarin and After comparing the average cost of both groups of drugs and acenocoumarol). Another strength includes the quantification of estimating their confidence intervals, a regression model will be indirect costs, such as the loss of productivity, associated with the 5 Hernández Rojas et al. Medicine (2020) 99:36 Medicine 2 types of treatments. Since it is a population-based study, with Formal analysis: Marc Casajuana, José Fernández-Sáez. RWD, more accurate and representative information will be Investigation: Zojaina Hernández Rojas, Maria Rosa Dalmau available for health decision-makers.[43,44] Moreover, most of the Llorca, Carina Aguilar Martín, Marc Casajuana, Dolores published research on this topic is based on hospital data, Rodríguez Cumplido, Emma Forcadell Drago, Noèlia Car- whereas our project will include variables and expenses related to rasco-Querol, Josep Maria Pepio Vilaubí, Alessandra PC settings, where most VKA controls are performed in our Queiroga Gonçalves, Josep M Alegret. milieu. Methodology: Carina Aguilar Martín, Marc Casajuana, Ales- sandra Queiroga Gonçalves, Dolores Rodríguez. Project administration: Zojaina Hernández Rojas, Maria Rosa 4.1. Limitations Dalmau Llorca. Our study has certain limitations. First, the database includes data Supervision: Maria Rosa Dalmau Llorca, Josep M Alegret. that are dependent on the professional registry, which means that Validation: José Fernández-Sáez. there may be some under-registration of some variables of interest. Visualization: Dolores Rodríguez Cumplido, Emma Forcadell In addition, there are expenses that we cannot determine, such as Drago. those of medical transport or the journeys undertaken by patients Writing – original draft: Zojaina Hernández Rojas. themselves, which could therefore lead to the costs of VKA being Writing – review and editing: Maria Rosa Dalmau Llorca, Carina underestimated. Also, we will not be able to include intangible Aguilar Martín, Marc Casajuana, Alessandra Queiroga costs (as patient suffering due to complications or caregiver Gonçalves, José FernándezSáez, Dolores Rodríguez Cum- overload) and operating costs, (such as overheads: water, plido, Emma Forcadell Drago, Noèlia Carrasco-Querol, Josep furniture, building maintenance and input costs). However, these Maria Pepio Vilaubí, Josep M Alegret. limitations should affect both groups of anticoagulants equally, so it is reasonable to assume that they will not have a significant References impact on the result. Finally, a time horizon of 2 years could be too short a time for a health event to occur. [1] Chugh SS, Havmoeller R, Narayanan K, et al. Worldwide epidemiology of atrial fibrillation: a global burden of disease 2010 study. Circulation 2014;129:837–47. 4.2. Future directions [2] Gómez-Doblas JJ, Muñiz J, Martin JJA, et al. Prevalencia de fibrilación auricular en España. Resultados del estudio OFRECE. Rev Esp Cardiol The evaluation of drugs with real-world data, based on cost- 2014;67:259–69. effectiveness analyses, provides a valuable insight for other [3] Jovic´evic´ M, Divjak I, Slankamenac P, et al. The most frequent causes of European countries, where the conditions of use of these drugs ischemic stroke in young adults. Med Pregl 2011;64:331–5. [4] Arboix A, Solà E, Castillo M, et al. Comparación del perfil de factores de are usually similar. riesgo cerebrovascular entre accidentes isquémicos transitorios e infartos This study could be of interest throughout Spain, as well as to cerebrales. Med Clin (Barc) 2003;121:292–4. other countries with similar guidelines for treating NVAF, as it [5] Camm AJ, Lip GY, De Caterina R, et al. 2012 focused update of the ESC will provide relevant information about direct and indirect costs, Guidelines for the management of atrial fibrillation: an update of the 010 ESC Guidelines for the management of atrial fibrillation. Developed with which, together with aspects of effectiveness and safety, could the special contribution of the European Heart Rhythm Association. Eur help shape the consensual decision-making of evaluating bodies, Heart J 2012;33:2719–47. and thereby provide a stimulus to update current consensus [6] Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy documents. On the other hand, we consider that it would be to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann interesting for future research to include the opinions of patients Intern Med 2007;146:857–67. [7] Generalitat de Catalunya. Servei Català de la Salut. Pautes per a about their preferences for 1 type anticoagulant over another, l’harmonització de l’us d’anticoagulants orals per a la prevenció de l’ictus since this is a factor that could influence adherence to treatment i l’embòlia sistèmica en pacients amb fibril·lació auricular; 2018. and, indirectly, its effectiveness and economic cost. Available at: https://catsalut.gencat.cat/web/.content/minisite/catsalut/ proveidors_professionals/medicaments_farmacia/harmonitzacio/pautes/ anticoagulants-orals-fibrilacio-auricular/pautes-harmonitzacio-anticoa Acknowledgments gulants-orals-en-fibrilacio-auricular.pdf [Accessed July 21, 2020]. [8] Vidal Group . Vademecum: Su fuente de conocimiento farmacológico. The authors acknowledge IDIAPJGol for having awarded the Madrid: Barcelona; 2015. Available at: https://www.vademecum.es/. database scholarships and for enabling the intensification of the Accessed February 29, 2020 research activity, both of which are essential for the successful [9] Lin PJ. Reviewing the reality: why we need to change. Eur Hear Journal (Suppl ) 2005;7(E):15–20. execution of the present study. We also thank the Fundació Dr. [10] Schulman S. Care of patients receiving long-term anticoagulant therapy. Ferran for the scholarship and their support for the Generalitat de N Engl J Med 2003;349:675–83. Catalunya’s Industrial Doctorate training programme. We [11] Cheng JW, Barillari G. Non-vitamin K antagonist oral anticoagulants in greatly appreciate the valuable support of Dr Maria Ferré and cardiovascular disease management: evidence and unanswered ques- Nurse Nuria Beguer. tions. J Clin Pharm Ther 2014;39:118–35. [12] Yeh CH, Hogg K, Weitz JI. Overview of the new oral anticoagulants: opportunities and challenges. Arterioscler Thromb Vasc Biol 2015; Author contributions 35:1056–65. [13] Solla-Ruiz I, Villanueva-Benito I, Paredes-Galán E, et al. Differences Conceptualization: Zojaina Hernández Rojas, Maria Rosa between patient-driven adherence to vitamin K antagonists and direct Dalmau Llorca, Carina Aguilar Martín. oral anticoagulants. Do few missed doses matter? ACO-MEMS Study Thromb Res 2019;20–7. Data curation: Carina Aguilar Martín, Marc Casajuana, José [14] Steffel J, Verhamme P, Potpara TS, et al. The 2018 European Heart Fernández-Sáez. Rhythm Association Practical Guide on the use of non-Vitamin K Funding acquisition: Zojaina Hernández Rojas, Maria Rosa antagonist oral anticoagulants in patients with atrial fibrillation. Eur Dalmau Llorca. Heart J 2018;39:1330–93. 6 Hernández Rojas et al. Medicine (2020) 99:36 www.md-journal.com [15] Kirchhof P, Benussi S, Kotecha D, et al. Guía ESC 2016 sobre el www.sidiap.org/index.php/es/component/content/article?id=33:base- diagnóstico y tratamiento de la fibrilación auricular, desarrollada en de-dades. Published 2014 [Accessed February 29, 2020]. colaboración con la EACTS. Rev Esp Cardiol 2017;70:50.e1–e84. [29] Bolíbar B, Fina Avilés F, Morros R, et al. Base de datos SIDIAP: la [16] January CT, Wann LS, Calkins H, et al. 2019 AHA/ACC/HRS focused historia clínica informatizada de Atención Primaria como fuente de update of the 2014 AHA/ACC/HRS guideline for the management of información para la investigación epidemiológica. Med Clin (Barc) patients with atrial fibrillation: a report of the American College of 2012;138:617–21. Cardiology/American Heart Association Task Force on Clinical Practice [30] Del Mar García-Gil M, Hermosilla E, Prieto-Alhambra D, et al. Guidelines and the Heart Rhythm Society in collaboration with the Construction and validation of a scoring system for the selection of high- Society of Thoracic Surgeons. Circulation 2019;140:e125–51. quality data in a Spanish population primary care database (SIDIAP). [17] Ministerio de Sanidad, Servicios Sociales e Igualdad. Agencia Española Inform Prim Care 2012;19:135–45. de Medicamentos y Productos Sanitarios. Informe de Posicionamiento [31] Generalitat de Catalunya. Departament de Salut. CatSalut. Servei Català Terapéutico UT_ACOD/V5/21112016. Criterios y recomendaciones de la Salut. Conjunt mínim bàsic de dades (CMBD); 2017. Available at: generales para el uso de nuevos anticoagulantes orales directos (ACOD) https://catsalut.gencat.cat/ca/proveidors-professionals/registres-catalegs/ en la prevención del ictus y la embolia sistémica en pacientes con registres/cmbd/[Accessed February 29, 2020]. fibrilación auricular no valvular. November 2016. Available at: https:// [32] Generalitat de Catalunya. DOGC (Diari Oficial de la Generalitat de www.aemps.gob.es/medicamentosUsoHumano/informesPublicos/docs/ Catalunya); 2016. Available at: https://dogc.gencat.cat/es/pdogc_so criterios-anticoagulantes-orales.pdf [Accessed November 15, 2019]. bre_el_dogc/[Accessed February 29, 2020]. [18] Escobar C, Borrás X, Bover Freire R, et al. A Delphi consensus on the [33] Generalitat de Catalunya. Institut d’Estadística de Catalunya. Idescat. management of oral anticoagulation in patients with non-valvular atrial Indicadors de conjuntura econòmica. Índex de preus de consum (IPC); fibrillation in Spain: ACOPREFERENCE study. PLoS One 2020;15: 2002. Available at: https://www.idescat.cat/indicadors/?id=conj&n= e0231565. 10261 [Accessed February 29, 2020]. [19] Reynolds SL, Ghate SR, Sheer R, et al. Healthcare utilization and costs [34] Generalitat de Catalunya. Portal Jurídic de Catalunya. ORDRE SLT/42/ for patients initiating Dabigatran or Warfarin. Health Qual Life 2012, de 24 de febrer, per la qual es regulen els supòsits i conceptes Outcomes 2017;15:1–9. facturables i s’aproven els preus p ublics corresponents als serveis que [20] Kansal AR, Zheng Y, Pokora T, et al. Cost-effectiveness of new oral presta l’Institut Català de la Salut. DOCG; 2012. Available at: https:// anticoagulants in the prevention of stroke in patients with atrial portaljuridic.gencat.cat/ca/pjur_ocults/pjur_resultats_fitxa/?documen fibrillation. Best Pract Res Clin Haematol 2013;26:225–37. tId=601235&action=fitxa [Accessed February 29, 2020]. [21] Barón Esquivias G, Escolar Albaladejo G, Zamorano JL, et al. Cost- [35] Fetter RB, Shin Y, Freeman JL, et al. Case mix definition by diagnosis- effectiveness analysis comparing apixaban and acenocoumarol in the related groups. Med Care 1980;18(Suppl 2):1–53. prevention of stroke in patients with nonvalvular atrial fibrillation in [36] Husereau D, Drummond M, Petrou S, et al. Consolidated health Spain. Rev Española Cardiol 2015;68:680–90. economic evaluation reporting standards (CHEERS)-explanation and [22] Restovic G, Carcedo D, McLeod EJ, et al. PCV72 cost-effectiveness of elaboration: a report of the ISPOR health economic evaluation rivaroxaban versus acenocumarol in the stroke prevention in patients publication guidelines good reporting practices task force. Value Health with non-valvular atrial fibrilation in the Spanish setting. Value Health 2013;16:231–50. 2012;15:A375. [37] INAHTA. The International Network of Agencies for Health Technolo- [23] Bowrin K, Briere J-B, Fauchier L, et al. Real-world cost-effectiveness of gy Assessment. Available at: http://www.inahta.org/[Accessed February rivaroxaban compared with vitamin K antagonists in the context of 29, 2020]. stroke prevention in atrial fibrillation in France. PLoS One 2020;15: [38] Dodd S, Bassi A, Bodger K, et al. A comparison of multivariable e0225301. regression models to analyse cost data. J Eval Clin Pract 2006;12:76–86. [24] Monreal-Bosch M, Soulard S, Crespo C, et al. Comparison of the cost- [39] Walker D. Allowing for uncertainty in economic evaluations: qualitative utility of direct oral anticoagulants for the prevention of stroke in patients sensitivity analysis. Health Policy Plan 2001;16:435–43. with atrial fibrillation in Spain. Rev Neurol 2017;64:247–56. [40] Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus Warfarin [25] González-Juanatey JR, Álvarez-Sabin J, Lobos JM, et al. Cost- in patients with atrial fibrillation. N Engl J Med 2009;361:1139–51. effectiveness of dabigatran for stroke prevention in non-valvular atrial [41] Hirsh J, Dalen JE, Anderson DR, et al. Oral anticoagulants: mechanism fibrillation in Spain. Rev Española Cardiol 2012;65:901–10. of action, clinical effectiveness, and optimal therapeutic range. CHEST [26] Giner-Soriano M, Casajuana M, Roso-Llorach A, et al. Effectiveness, 2001;119:8S–21S. safety and costs of stroke prevention in non-valvular auricular [42] Escobar C, Martí-Almor J, Pérez Cabeza A, et al. Direct oral fibrillation. Study of cohorts matched by Propensity score. Aten Primaria anticoagulants versus vitamin K antagonists in real-life patients with 2020;52:176–84. atrial fibrillation. a systematic review and meta-analysis. Rev Española [27] Lekuona I, Anguita M, Zamorano JL, et al. Would the use of edoxaban Cardiol 2019;72:305–16. be cost-effective for the prevention of stroke and systemic embolism in [43] Garrison LP, Neumann PJ, Erickson P, et al. Using real-world data for patients with nonvalvular atrial fibrillation in Spain? Rev Española coverage and payment decisions: the ISPOR real-world data task force Cardiol 2019;72:398–406. report. Value Health 2007;10:326–35. [28] IDIAP Jordi Gol. SIDIAP - Sistema de Información para el Desarrollo de [44] Mahajan R. Real world data: additional source for making clinical la Investigación en Atención Primaria. Barcelona. Available at: https:// decisions. Int J Appl Basic Med Res 2015;5:82. 7