Program Using a Mobile Application Versus Telephone Advice on Patients at Risk of Coronary Heart Disease : a Pilot RCT

Aims and Hypotheses to be Tested:

This study compares the effects of a nurse-led support programme - an app support programme (ASP) - with telephone support (TS) for patients discharged from the ED who are at risk of developing CHD. We hypothesize that:

Within 3 months, when compared with the TS group clients, clients who participate in the ASP will have:

Primary outcome: Better self-efficacy and self-management behaviour

Secondary outcomes:

• Less ED attendance and hospitalization frequency;
• A better physiological health profile (e.g., blood pressure, body mass index, blood cholesterol level, blood triglyceride level, and blood glucose level) and better cardiovascular functional endurance;
• An increase in the total amount of exercise;
• A lower perceived stress level; and
• Better health literacy regarding cardiac care.

Plan of Investigation:

(i)Methods: A prospective multi-center, parallel, randomized controlled trial with two arms - an app support programme (ASP) group vs a telephone support (TS) group - will be adopted. 1. Settings: Data will be collected from an EDs at a regional Hong Kong hospital, namely Tuen Mun Hospital and the community centres. The demographic profile of the residents and the setting and care protocol will be similar, to minimize the cluster effect. Each ED includes a short stay unit and an emergency medical unit (EMW) with about 26 beds. Patients with cardiac disorders account for approximately 25% of all EMW admissions (n=1900 per year) and generally have hypertension or hyperlipidemia and angina 4.

1. Recruitment of subjects: The admission records of all eligible clients with a medical diagnosis of at-risk CHD in the study ED will be screened. Identified patients will be approached by the research assistant (RA1) for further screening and those deemed to be eligible will be invited to join the study. Appendix 2 presents the CONSORT flow diagram of the study.

About 80 potential patients will be approached. Eligible patients who consent to participate in the study and then who submit to the collection of their baseline data will be randomly allocated to either the ASP group or the TS group according to the pre-set randomization sequence. Participants will attend an appointment with a research nurse (A) who has experience in cardiac nursing and specific training in the intervention. Follow-ups will be conducted at 1 month, and 3 months by another RA (2), who will be blinded to the intervention allocation.

Regarding patient recruitment, we estimate that 75% of eligible CHD patients (n=5-7 per week) will agree to join the study at an ED and that at least 8-10 patients will be recruited from both EDs per week. These estimations are based on local departmental statistics and on a previous study that we conducted in a regional ED. 4 We further estimate that 10 weeks will be required to recruit all of the patients required for the study (60 patients). Due to the Covid- 19 situation that subject recruitment is not allowed, we also extend the subject recruitment to community centres..Subject inclusion criteria are the same. Ethical approval has been obtained from the university and the study venues.

1. Follow-up data collection: During the 1-, and 3-month follow-ups, the trained RA (2) and student research assistants (who are master's nursing students), who will be blinded to group allocation, will collect data and perform physical assessments of the participants at the university's research laboratory office. This office has sufficient space and equipment to perform health assessments and laboratory tests. "Item e" shows the facilities in detail. All participants will be asked to complete the questionnaires, place them in an envelope, and hand the envelope to the research assistants.

To ensure the quality of patient recruitment, intervention, and data collection processes, the research team will ensure that the RAs and research nurses are capable of carrying out the proper procedures. All of the research nurses are qualified and experienced in cardiac nursing. They will be provided with specific training workshops by the PI and Co-Is prior to the commencement of the study. One research nurse (A) who is skilled in cardiac nursing and counseling will deliver all of the educational interventions to ensure consistency. Audit procedures will be conducted to ensure the quality and uniformity of the educational intervention. Another research nurse (B) will provide the telephone advice intervention to the TSG. The team will visit the EDs at least once per week during the initial recruitment period. Weekly meetings with research nurses and research assistants will be held to ensure good quality in data collection and to solve any logistical problems.

Data processing and analysis To minimize subject contamination, the different groups will be assigned different follow-up dates. The research assistants who are responsible for collecting data will be blinded to the group allocations. They will receive a briefing and training session on collecting data, taking blood using the finger-stick method, and using the auto blood analyzers. All questionnaires will be distributed to the individual participants, who will be asked to place their own completed questionnaire in a locked box.

Data will be analyzed using SPSS. The intention-to-treat principle will be applied. The baseline characteristics of the groups will be compared using a Chi-square or Fisher's exact test for categorical variables, and a t-test or Mann-Whitney test for continuous variables, to examine the comparability of the groups by randomization. Confounding variables will be accounted for in subsequent analyses. The effectiveness of an e-health programme using the app vs telephone support will be evaluated by comparing the study outcomes using Generalized Estimating Equations (GEE) models. For the primary outcomes of self-efficacy and self-management behavior, two GEE models with an identity link function will be performed. For the two secondary outcome variables that are measured in counts (i.e., ED attendance and the number of hospitalizations), GEE models with a log link function and a Poisson distribution will be performed. For the other secondary outcomes, which are continuous, GEE models with an identity link will be fitted. Significant time by group interactions in the GEE models will support our hypotheses on the effects of the e-health intervention using app support. Descriptive statistics will be used to summarize the level of satisfaction with the use of the app at T1 to T3 in the intervention group, and a GEE model will be used to assess changes in satisfaction over time. All of the tests will be two-sided and a p-value of <0.05 will be considered statistically significant.

This pilot study will examine the feasibility of the main RCT study by using the app,identify any recruitment, implementation of the intervention and logistic issues arising from the recruitment, intervention, and follow-ups. It can also provide the preliminary finding for sample size calculation for the main RCT later on.