Study design

A single-centre, non-randomised observational pilot study was conducted in Máxima Medical Center in Veldhoven and Eindhoven, the Netherlands. Patients referred for CR due to coronary artery disease (stable angina pectoris, acute coronary syndrome and/or after coronary revascularisation) or atrial fibrillation who met the inclusion criteria as described in Tab. 1 were asked to participate in the study.

Table 1 Inclusion criteria

Recruitment was performed by a research nurse. After patients provided written and verbal informed consent an account for the LifeStyleScore research platform was created. The patient received a manual on how to access and complete the questionnaires in this study. Participants were instructed to complete the questions provided via the LifeStyleScore application as well as the Patient Activation Measure® (PAM®-13) [4] and the System Usability Scale (SUS) [13] at the start of their CR programme, i.e. before their first CR intake consultation with the case manager (t = 0), and after 3 months (t = 1), i.e. at the end of their CR programme. An impression of the LifeStyleScore application is displayed in Fig. 1. In order to assess patient activation tendencies prior to reporting actual lifestyle behaviours, the PAM-13 was administered first. Results of the LifeStyleScore application were used to set and evaluate lifestyle goals during the CR programme. An overview of the study design is displayed in Fig. 2.

Fig. 1Fig. 2LifeStyleScore application

The LifeStyleScore application is based on validated assessment tools and provides insight into cardiovascular risk behaviour by assessing, scoring and giving advice in six separate lifestyle behaviour domains based on the Dutch vitality guidelines (BRAVO-kompas) [14] and a systematic review of the literature [15]. This entails that the application assesses physical activity, sedentary behaviour, smoking behaviour, alcohol consumption, nutritional intake and perceived stress, supplemented with biometric characteristics (body composition) of the patient. The first version of the application examined in this project is available in Dutch.

The assessment of lifestyle behaviour domains (except for body composition) encompasses two stages:

Stage 1: a screening questionnaire used to detect cardiovascular risk behaviours;

Stage 2: in case of possible cardiovascular risk behaviour, an in-depth questionnaire is used to more thoroughly evaluate the risk behaviour.

The patients received written feedback via the application based on their score. Cardiovascular risk behaviour scores are given as a traffic light score with the following categories: ‘Low risk’ (3—green), to ‘Medium risk’ (2—yellow) and ‘High risk’ (1—red).

For this study, the LifeStyleScore application was integrated into the CR programme. The patients filled out the application before their intake with their CR case manager. During the patients CR intake consultation, the results were discussed (motivational interviewing was used to motivate the patients if necessary) and used to formulate goals according to the SMART principle (Specific, Measurable, Achievable, Relevant and Time-Bound). The set goals were registered in the electronic patient record for follow-up. The cases were shared with other HCPs in the CR team. The questionnaires implemented in the LifeStyleScore application are based on validated questionnaires and guidelines as described in Tab. 2.

Table 2 Questionnaires used in the LifeStyleScore applicationOutcome measures

The primary outcome was usability of the LifeStyleScore application. Secondary outcomes were patient self-management and activation, changes in lifestyle behaviours and acceptance of the application by the HCPs.

Usability

Usability of the LifeStyleScore application was assessed by the 10-item SUS questionnaire (Dutch version) [13]. The SUS is known as a highly robust and versatile tool for usability professionals to quickly and easily collect the user’s subjective rating of a product’s usability. The SUS scores are presented on a scale ranging from 0 to 100. A study on the SUS scores of 500 studies [23] determined that the average SUS score is 68 points. A score above 68 indicates that the design needs minor improvements or adjustments, while a score of below 68 indicates that the design might need to be re-evaluated and improved to be considered user friendly. The mean SUS score is also suitable for evaluating the usability of digital health apps [24].

Patient activation and self-management

Patient activation was measured using the PAM®-13 questionnaire (Dutch version) [4]. The PAM-13 is a validated questionnaire that measures self-reported knowledge, skills and confidence for self-management of one’s health or chronic condition. The PAM’s algorithm produces a score along an empirical, interval level scale from 0–100 that correlates to one of four progressively higher levels of patient activation: 1) believing the patient role is important, 2) having the confidence and knowledge necessary to take action, 3) actually taking action to maintain and improve one’s health and 4) staying the course even under stress.

Lifestyle behaviour

Changes in lifestyle behaviour were measured with the newly developed mobile and desktop LifeStyleScore application. The raw scores of the validated questionnaires used in the LifeStyleScore application were translated to a score from 1 (high risk) to 3 (low risk) per category based on the cut-off scores of the original questionnaires [17,18,19,20,21], Dutch Physical Guidelines [16] and Dutch Dietary Guideline [22] (see Supplement 1 of the Electronic Supplementary Material for the cut-off scores used).

Acceptance of the application by healthcare professionals

Experiences with the application as part of the CR programme were collected via a brief focus group interview with the HCPs who were involved in patient care. The participants of the focus group session used the LifeStyleScore application and were asked to share their experience with patients using the application, whether the application was beneficial to their work, and to elaborate on how the application could be integrated further in their work and the care of patients with cardiovascular disease. Thematic analysis was used to analyse the data collected in the focus group interview.

Sample size and statistical analysis

The sample size was selected based on expected compliance and on the ‘rule of 12’ [25] for continuous variables, which recommends to include at least 12 participants in pilot studies to provide valuable preliminary information for planning larger subsequent studies. Statistical analyses were performed using IBM SPSS for Mac (version 28.0; IBM Corporation). Descriptive statistics were used to describe the study population. The within-group changes in lifestyle behaviour were analysed using paired samples T‑tests. A McNemar test was used to compare the binary variables of patient activation levels.