ABC | Volume 110, Nº1, January 2018

Original Article Stephan et al Mobile health and atrial fibrillation Arq Bras Cardiol. 2018; 110(1):7-15 These interventions are based on the premise that the healthcare professional is responsible for the provision of essential information to the patient and for stimulating the patient to search for knowledge. In this context, technology shows up as an allied, by improving information access, organization, transmission and retention. In particular, mobile technology introduces a new era of health care, by bringing care closer to the patient and allowing a better doctor-patient interaction. In this rapidly expanding market, in 2015, there were 45,000 mHealth publishers and more than 3 billion mHealth app downloads. 29 Current evaluations are, in general, favorable. A recent analysis of the American Heart Association on mHealth and cardiovascular disease prevention included 69 apps for weight loss, increase in physical activity, smoking cessation, glycemic control, hypertension and dyslipidemia. Despite heterogeneous, positive results were found for the proposed behavioral changes, and future studies using more rigorous methodology, more diversified samples and a long-term follow-up were suggested to evaluate the duration of the effects. 30 With respect to the target populations, the literature highlights the necessity of these technologies to encompass other specific populations – older subjects with age-related changes (e.g. reduced vision or mobility), minorities in need of culturally sensitive contents and interventions, and low-income adults with inconsistent access to mobile communication. 30-32 AF is a largely explored subject in mHealth. Most studies have reported the use of home monitoring devices for heart rate. With regards to patients’ education, the American Heart Association and the European Society of Cardiology have high‑quality applications and web materials in English that help in the shared decision-making process. 33,34 There are also many risk calculation methods available for the clinical practice. However, neither the development process nor the evaluation of these apps is described in the literature. Also, we have not found any support instrument for shared decision-making in AF, be it in mHealth or in other media. A strength of our study was the development of the app based on evidence, taking into account many factors mentioned in guidelines of shared decision making and care of anticoagulated patients. 25,35 The level of patients’ previous knowledge was analyzed, and the learning style was adequate to their preferences of terminology and navigability. The amount and detail of information was adjusted, and could be increased or reduced, according to each individual’s understanding. Another advantage was the fact that patients’ evaluation could be saved for further analyses by other professionals, indicating the role of the instrument as a bridge in the multidisciplinary care. In an integrated outpatient service, for example, the patient could watch the video and have their risk factor evaluated during the screening process and focus on treatment during the medical visit. In addition, the selected population was appropriate for implementing a shared decision strategy. Most patients had a SAMe-TT2R2 score equal to or greater than two, suggesting a lower probability to maintain anticoagulation at acceptable levels with coumarins and hence a greater necessity for strategies for an adequate control. Results of the analysis of patients’ risk perception showed how this understanding is inappropriate and requires attention. Most patients believed they had a stroke risk lower than the calculated and one third of patients believed they had a bleeding risk with the use of OAC higher than the calculated. Other studies showed similar results on awareness of the risk of stroke. 36,37 Such inadequate understanding may lead to poor treatment adherence, since patients do not perceive themselves to be at risk for thromboembolic events and also believe they have a high risk of bleeding using the medication. After interacting with the app, no significant change in risk perception was observed. In attempt to improve such perception, the following observation was added to the second version of the app, currently under test: “This risk is considered LOW/INTERMEDIATE/HIGH”, with a color code to each level of risk (green/yellow/red), together with the percentages exhibited on the screen “Understanding risks and benefits”. Several limitations are inherent to the development of an instrument that utilizes a relatively new technology for our population. Although the screen size, the visual communication methods and the terminology had been carefully considered, they still can be inadequate for some patients. Besides, even though the information provided to the patients had been adapted to the patients, the fact that it had been excessive in some cases and not maintained after some months cannot be ruled out. It is expected that the continuous provision of information by SMS compensate part of this issue. Besides, the interaction with the app may be repeated in other visits whenever necessary. The small number of patients studied may also be questioned. Nevertheless, in studies evaluating the usability of apps, the number of subjects involved is usually small and shown sufficient. 38 Another current limitation is the necessity of a long-term evaluation of the outcomes, such as the TTR, adherence and occurrence of thromboembolic events and bleeding. This limitation is expected to be eliminated with a randomized intervention study, by using the app in the care of our patients attending the anticoagulation outpatient service and comparing the results with the care currently provided. Conclusions The use of the mHealth app during the medical visit about anticoagulation in AF improves disease knowledge and the treatment of low-income patients with low educational level, enabling a shared decision with low decisional conflict. Further studies are needed to confirm whether such improvement can be translated into hard outcomes. Author contributions Conception and design of the research, Writing of the manuscript and Critical revision of the manuscript for intellectual content: Stephan LS, Almeida ED, Guimarães RB, Ley AG, Mathias RG, Assis MV, Leiria TLL; Acquisition of data: 12