Usability of a mobile application on diabetic foot self-care

Rev Bras Enferm. 2020;73(4): e20180862 http://dx.doi.org/10.1590/0034-7167-2018-0862 6 of ABSTRACT Objectives: to assess the usability of an app prototype for diabetic foot self-care by an end user. Methods: a descriptive study that uses heuristic assessment of a hybrid app usability. Fifteen users of an outpatient diabetes care service in a capital of Northeastern Brazil participated in the study during April 2018. The usability measurement tool called Smartphone Usability questionnaiRE was applied. Results: the lowest score was 77 and the highest was 112, with an average usability of 96.1 points. Usability was framed in the last two levels, 70 and 8o. Users now strongly agree (level 70) and fully (level 80) with the assessed items, which represents good usability of the apps prototype. Conclusions: the final product developed focuses on user needs and requirements, which can ensure usability based on effectiveness, efficiency and satisfaction triad. Descriptors: Self-Care; Diabetes Mellitus; Telemedicine; Educational Technology; Diabetic Foot.


INTRODUCTION
Diabetes Mellitus (DM) currently affects 425 million adults worldwide. DM is projected to reach 629 million by 2045, i.e., there will be an estimated 48% increase in the number of people with the disease. When not properly administered, all types of DM can result in complications affecting many parts of the body, leading to frequent hospitalization, disability and premature death (1) .
Among the complications, the diabetic foot stands out, being the most feared by patients due to one-limb loss (1) . They are common and result in considerable, often recurring suffering, and are associated with a high mortality rate and a high cost of health, which makes it a serious public health problem (2) .
The number of hospitalizations for treatment of complicated diabetic foot and amputation is increasing compared to the other complications of DM (3) .
The exponential growth of DM patients has been a persistent concern of health professionals. In addition to health interventions, public policies and health programs recommend educating patients to actively participate in self-care (4) .
The use of mobile health apps may serve this purpose. These can provide personalized knowledge about the disease and bring about positive behavioral changes for self-care development for independent and effective management of DM (5) . Thus, considering the perpetration of mobile technology in the health field, companies and healthcare professionals have developed varieties of apps with emphasis on DM prevention and management (6) .
Usability is one of the key parameters for making a mobile app used and loved by users, as well as allowing users to achieve their specified goals. Efficiency, effectiveness and satisfaction are key factors that determine the user experience when involved with an app (7) .
DM is one of the leading chronic diseases of global health and exponential growth worldwide, with the main complication being diabetic foot. This can be prevented by very simple measures such as the simple daily assessment of feet by people with DM. Based on this context, it was proposed the creation of a multimedia app for daily care promotion for people with DM. Awareness of the theme arises from the professional performance of one of the researchers in this study as a stomatherapist nurse in home care, outpatient and hospital.
Given the above, the question is: what is the usability level of an app prototype for self-care with diabetic foot by an end user?

OBJECTIVES
To assess the usability of an app prototype for diabetic foot self-care by an end user.

Ethical aspects
The study obeyed the ethical and legal precepts governing Resolution 466/12, which involves research conducted with humans from the Brazilian National Health Board (Conselho Nacional de Saúde) (8) . The study was approved by the Research

Design, place of study and period
This is a descriptive, heuristic and usability study of a hybrid app called "PedCare". We have proposed a mobile app to help DM and at-risk patients take care of the health of their feet. This is a thesis deployment with mixed methods in sequential phases: 1) methodological study and 2) randomized controlled trial. Usability was analyzed by users when the app was in prototype phase. Sequential studies on the effect of this app were performed to verify its viability. It is noteworthy that the app is in the trademark and software registration phase and the other stages of the study are being published.
In order to ensure good app building practices, usability assessments are performed. There are several techniques to do this, ranging from informal assessment to usability testing analysis with user-class representatives. It is possible to reduce errors, increase productivity and safety when humans interact with the system (9) , which makes usability testing one of the main methods from the end user's perspective.
The study was conducted at the Integrated Center for Diabetes and Hypertension (CIDH -Centro Integrado de Diabetes e Hipertensão) in the city of Fortaleza, state of Ceará, Brazil, during April 2018.

Population or sample; inclusion and exclusion criteria
Fifteen CIDH users participated in the pilot study. It is recommended that the number of evaluators be a small group (from 3 to 5) for interface accuracy and assessment with the defined heuristics (10) . A priori, it was intended to compare usability in three distinct groups, but it was not possible to obtain statistical significance. They met the established criteria: age ≥ 18 years; DM diagnosis; active registration at the health center; possession of a mobile device for personal use and familiarity; and skill with app use. It was measured by the maneuverability test with the prototype installed. People with marked functional dependence and/or cognitive impairment or with any difficulty that made the handling unfeasible, besides people with diabetic ulcer were excluded.

Study protocol
Participants were approached at the service while waiting for consultation with some multidisciplinary specialty, diagnostic tests or medication dispensation. They were briefed on the study, and those who were interested were taken to a private room where they explained the purpose of the study, presented the app's prototype feature explanations tutorial earlier, in PowerPoint format, designed on a tablet, and, then the app prototype. After consent for the study, the participants formalized their agreement by signing the Free and Informed Consent Term.
The "PedCare" mobile app prototype is a hybrid model that does not have the need for an internet connection to function, categorized as an advanced app by using features available on of Usability of a mobile application on diabetic foot self-care Marques ADB, Moreira TMM, Jorge TV, Rabelo SMS, Carvalho REFL, Felipe GF. the installed mobile device. PedCare has 11 menus, 69 screens, based on User-Centered Design and HTML, CSS and JavaScript programming language to be compiled for Android and iOS platforms.
The user was offered the app running on smartphone or tablet, according to their preference. After handling on average 20 minutes, or when the user deemed the handling time to be adequate, the usability measurement tool called SURE (Smartphone Usability questionnaiRE) (10) , version (1.0) of the questionnaire was applied. SURE has 31 items, being specific to smartphones and developed through systematic literature review and use of Item Response Theory (IRT).
For each statement, the participant selected a response using degree of agreement on each criterion, with 1 -Inadequate, 2 -Partially Adequate, 3 -Adequate, 4 -Totally Adequate or NA -Not Applicable. The total score of the tool corresponded to the sum of all the answers. The total score of up to 124 points is calculated from the sum of the scores obtained in the items. The value is interpreted by the distribution ranges: level 30 (present a possibility to totally or partially disagree); level 40 (present possibility to agree); level 50 (fail to partially agree to strongly); level 70 (strongly agree); and level 80 (totally agree) (10) .

Analysis of data, and statistics
Data are presented through tables, being contrasted and discussed in the light of the literature on the subject. The variables were analyzed descriptively, considering simple frequency, percentage, mean, standard deviation and test mean.
The values obtained in the usability test were compared using the scores adopted by the SURE tool. Because it is based on the IRT, each evaluator's response to an item represents the probability as a function of the item's parameters and ability (10) , which allows data reliability.

RESULTS
The usability test consisted of 15 end users of the specialized care service, the majority (60.0%) were women, with a mean age of 50.8 years, 40.0% single, 40.0% married, 60.0% had completed elementary school, with an average of 11.2 years of diagnosis of the disease and most (60.0%) were using mixed therapy (use of hypoglycemic drugs and insulin), according to Table 1. Table 2 presents the results obtained in the heuristic assessment of an app prototype usability by end users through SURE. Table 3 presents each user's assessment level based on the sum of the scores proposed by the SURE tool.
Based on Table 3, the lowest score was 77 and the highest was 112, with an overall usability average of 96.1 points. Thus, usability was framed in the last two levels, 70 and 80. That is, users strongly agree (level 70) and fully (level 80) with the assessed items, which represents good usability of the app prototype. have intuitive and interactive interface, texts are easy to read, have comprehension and accessibility in the language used, among other aspects. In the last level, users totally agree with all items and the help/tip provided by an app is helpful (10) . This fact allows us to infer that the app is pleasant to use and can be incorporated into everyday life.
Mobile app usability has attracted a lot of attention in the field of human-computer interaction, as well-designed apps can enhance user experiences (11) .
Thus, it is observed that the propagation of various usability assessment models has been introduced in the literature. These models are more general, which may not apply to any specific mobile app due to their complexities and lack of appropriate descriptions on how to choose usability measures such as usability dimensions, criteria and metrics (12) .
Other factors should be considered, such as exponential growth of apps, low quality availability, multiple categories, and different platforms. Therefore, in general, usability assessment models cannot measure the characteristics of these various apps because each app category has different functional and non-functional requirements. Thus, custom usability models may be required to assess these various apps (12) .
The HealthCare Information and Management Systems Society has developed guidelines to assess the usability of mobile health apps. This is a Likert scale to classify each criterion, which does not provide an indication of its quality, although the criteria were extensive and included usability engineering criteria to assess efficiency, effectiveness, user satisfaction, and platform optimization (13) .
Until the development of this study no specific tool was identified for the accuracy of mobile health app usability. A quality assessment scale called Mobile App Rating Scale -MARS (14) has been identified, but this scale is not yet translated and validated into Portuguese.
Usability assessment of mobile apps is still in its early stages. 26 usability models have been compiled for mobile apps, but these are unexplored and most of them have no proven effectiveness. This is a critical issue as existing usability guidelines are insufficient to design effective app interfaces due to peculiar characteristics and dynamic application context (12) .
Three experts conducted 66 app usability testing targeting people with DM over 50. The total usability score was calculated from all categories, which were determined using a 5-point Likert scale. Most of the ratings ranged from 3.0 to 4.0, which corresponded to a moderate to good rating of apps, especially for those offering a small range of functions. Usability was categorized into three distinct parameters: Instant and easily understandable feedback 3,3; Intuitive usability 3.5; and Simple Recognition of Click-Sensitive Areas 3.0 (15) . Usability tests become increasingly essential before making app available to the end user. Prior to actual context applicability checks, this provides a technical baseline in which users are familiar with the potential of mobile technology. This allows users to provide richer feedback on functional requirements and use cases (16) .

Score 1 2 3 4 N/A
17. The organization of menus and action commands (such as buttons and links) is logical, allowing you to easily find them on the screen.
-04 05 06 -18. I was able to successfully complete the tasks using this app.
-----20. The app provides all the information needed to complete tasks clearly and understandably.

DISCUSSION
Usability was framed at level 70 and level 80 by end users. At level 70 stratification, users strongly agree with the items. They find it easy to enter and correct data, they believe the product meets consumer needs. The features resemble available devices, of Usability of a mobile application on diabetic foot self-care Marques ADB, Moreira TMM, Jorge TV, Rabelo SMS, Carvalho REFL, Felipe GF.
One factor that can guarantee the usability of an app is its content. Based on this premise, the International Diabetes Federation has made recommendations for the development of DM app at four levels: individual, healthcare professional, politician and app developers (6) . In the individual field, the current app must meet basic needs and is in accordance with clinical practice.

Study limitations
As a limiting factor of this study, there is the restriction of the number of participants. Furthermore, it cannot be guaranteed that everyone has answered the questionnaire faithfully and with the utmost attention to each of the questions.

Contributions to nursing, health or public policy
Mobile health use can provide mechanisms to improve the efficiency and effectiveness of clinical care provided by nurses to promote foot care for people with DM. Thus, it would empower them to actively manage their health and to hold them accountable for their quality of life and health by providing information resources and reducing administrative burdens.

CONCLUSIONS
The test identified high usability of app prototype. End users have come to strongly and fully agree with the items analyzed, being a product developed focusing on user needs and requirements, which can guarantee usability, based on effectiveness, efficiency and satisfaction triad.