Usability and feasibility analysis of an mHealth-tool for supporting physical activity in people with heart failure

Medical health practice using mobile devices aimed at the end user, is often referred to as mHealth, and such tools are predicted to have an important role to play in managing disease [1]. One area where mHealth could add value is in the management of heart failure (HF), a chronic and debilitating condition affecting more than 35 million people worldwide [2]. HF is a clinical syndrome characterized by breathlessness, peripheral oedema and fatigue as the most common symptoms [3]. People with HF are advised to be physically active [3], but in spite of physical activity being associated with improved prognosis [4], a majority of people with HF display low levels of physical activity [5]. Barriers to adherence to physical activity for this population, like low motivation, fear of worsening symptoms and lack of influence over what activities to perform [6], could potentially be addressed by a home-based mHealth-tool. We sought to develop a tool to support physical activity, based on the Medical Research Council (MRC) guidelines on developing and evaluating complex interventions [7]. These guidelines emphasize co-design, identification of stakeholders on all socio-ecological levels and advocate the use of a mixed-methods approach [7‐9].

While modern technology is important in realising positive behaviour modifications [10], a tool is only efficacious if it meets the needs of the user [11]. To avoid the often-high rates of attrition of 50–80% [12], focus on usability is key. Usability is defined as the extent to which a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use, or the ease of use and acceptability of a system or product for a particular class of users carrying out specific tasks in a specific environment [13].

The development process used was based on the MRC guidelines [7], the INDEX (IdentifyiNg and assessing different approaches to DEveloping compleX interventions) guidance [14], and the 6SQuID (six essential Steps for Quality Intervention Development) model [15]. These different guides were combined to create a pragmatic process, which did not take too long to complete and the process is illustrated in Fig. 1. The first two phases of the development process are described in detail in the Appendix A “Development project: Activity coach”, and in the main section of this paper we focus on the third phase, the “Evaluation- and refining phase”. The tool developed will be referred to as the “Activity coach” throughout this paper.

The aim of this study was to evaluate the usability of the Activity coach and to assess feasibility of outcome measures, for people with HF.

To address the aims of this study, a mixed-methods design was employed. Usability was assessed through analysis of system adherence, technical issues reported by the study participants, as well as through probing of user experiences. To assess feasibility of outcome measures, we defined progression criteria for the outcomes tested. Data was collected from people with HF, who were recruited to use the Activity coach for four weeks (28 days). Former research show that 8–10 people usually is enough to identify about 80% of the usability issues [16]. The ethical review board approval for this research was obtained 2020-05-11 (ID: Dnr 2020–01444, Etikprövningsmyndigheten, Box 2110, 75,002 Uppsala, etikprovning.se). The study is registered at clinicaltrials.gov with study identifier NCT05235763.

The inclusion criteria were age ≥ 18, confirmed heart failure diagnosis with functional class NYHA II-III, and signed written informed consent. Candidates who were currently enrolled in another study involving physical activity were excluded. The study participants were recruited from a primary care health centre in Stockholm, Sweden by their treating physician, randomly selected from listed people with a confirmed HF diagnosis. Those who accepted to participate in the study were contacted via telephone by the researcher. A letter was sent to all study participants including some more information and the consent form. At a scheduled home visit, the consent forms were collected, and the Activity coach was installed, and the study participants were shown how to use the Activity coach. During this visit, the researcher asked the study participant to define the two goals for the goal attainment outcome. The hardware required to use Optilogg with the Activity coach installed, i.e. a 7” tablet and a connected weight scale, were supplied to the study participants for the duration of the study.

We used an approach inspired by Hawkins et al. [26] to predefine progression criteria for the measures of usability and feasibility. In a previous study of a home-based eHealth intervention, a system adherence ≥ 60% was used as a cut-off point to describe adherence [27]. We also defined acceptable levels of accelerometer use [28]. The progression criteria are described in Table 1.

Thirteen people with HF were asked and consented to participate in the study, but three withdrew their consent prior to starting the pilot. One had been admitted to the hospital and the two others did not state any reason. Ten people with HF had the Activity coach installed. The median age was 72 (quartile range 65–76) and seven were male. Half-way through week one, one study participant chose to withdraw from the study and was therefore not included in the analyses. No reason for this withdrawal was given.

Only field-notes were used, rather than recorded interviews, in spite of recordings and subsequent verbatim transcripts providing more reliable data [35]. Although impressions, emotions, and contextual details are better recorded through field-notes since ideas and memories from interviews are likely to be lost further down in the research process [36], these benefits of field-notes should not preclude recording the interviews as well as taking the field-notes.

Three out of the original thirteen patients chose to withdraw their consent prior to even starting the study, and one immediately after the study had started. This could indicate a fear of technology, that might lead to selection bias in a larger study, and selection bias is common in studies of mHealth interventions [37, 38] and could lead to issues concerning external validity of reported findings.