Adherence to Nutritional Treatment Following MI Using Telemedicine Treatment (ADNUT)

Background: Cardiovascular diseases (CVDs) are the leading cause of death and disability globally (1). It has been suggested that among all modifiable risk factors for CVD, an unhealthy diet has a major contributor to morbidity and mortality from CVD (2). An extensive body of evidence has led to the development of dietary guidelines for CVD prevention (3). The Mediterranean diet has been established as a major dietary pattern for the primary prevention of CVD (4). Evolving evidence support the contribution of the Mediterranean diet to secondary CVD (4, 5). Successful cardiac rehabilitation (CR) plans are comprehensive and extensive and include medical supervision, physical activity, physiotherapy, and nutrition support. Despite being an evidence-based clinical standard, and demonstrating prevention of readmission and higher health-related quality of life (QoL)(6) referral and participation in CR are low, especially among women and older adults (7). Accessibility difficulties, group setting, non-flexible hours, long commute times, and travel costs have been identified as barriers to participating in CR programs. Remote CR programs may address accessibility barriers and may represent an alternative for certain populations, and the need to develop additional effective modalities made apparent during the COVID-19 outbreak (8). Modalities for delivering remote physical activity and cardiac monitoring have shown promising results (9) and have been implemented successfully in Israel (10). However, it remains unclear whether remotely delivered nutrition therapy is effective.

In this proposal, we suggest comparing modalities of nutrition treatment for patients in a CR program in a randomized clinical trial. We will compare existing, in-person nutrition counseling (IPNC) program to a remote, web-based nutrition counseling (WBNC) program for three months post-discharge due to Myocardial infarction (MI) and examine its effectiveness for the first six months post discharge. The primary hypothesis is that there will be greater adherence to the Mediterranean diet in participants randomly assigned to WBNC than in those randomly assigned to IPNC.

1. Study Design, including:
2. Detailed Plan of the Study; the research plan should be outlined for the whole research period requested. A single-blind, randomized controlled, single-center study
3. The study population will include patients within a month within one month from hospital discharge for MI, with of cardiac risk 1-2, capable of conducting a conversation using Zoom software, in either Hebrew or English. We will exclude patients with a prognosis of one year or less due to comorbidity, patients with renal failure, or patients with hemodynamic instability, those who were already participating in a remote cardiac rehabilitation program, patients with hearing or vision impairments prevented from reasonable participation in an online call (zoom) or patients, who do not have access to a computer/smartphone.
4. Baseline data (Time 0 - T0) Socioeconomic and lifestyle information on smoking and physical activity as well as the patient's willingness to undergo a lifestyle change will be obtained during a personal interview. Food intake will be measured using 24 H food recall, adherence to MedDiet will be measured using the I-MEDAS (11), QoL will be assessed using the SF-36 (12) . Anthropometric measures: height, weight, waist circumference and Handgrip strength (HGS) (13) will be ascertained. Information recorded from the patient chart will include blood pressure, fasting blood glucose, HbA1C, cholesterol, and smoking status (current, past, never). These measures will be recorded again 3 months at completion of the intervention (T2). Blood tests will not be performed as part of the study procedures, but records will be extracted regarding blood tests performed in routine care in the patient's HMO.
5. Length of intervention- 3 months. Follow-up will take place immediately after the intervention (T1) and at 6 months after the intervention (T2). The main outcome variable will be adherence to MedDiet at T1 as measured by I-MEDAS (validated in the Israeli population, scores range from 1-17) (11), and the change in I-MEDAS from T0 to T1, T0 to T2, T1 to T2, and T1 to T2.
6. The interventions will take place in Cardiac Rehabilitation Unit, Sheba Medical Center, Ramat Gan. Individuals fulfilling the eligibility criteria will receive an explanation on the study procedures. After providing written informed consent, patients will be randomly assigned to either intervention or control groups. Control group participants will receive usual medical, physical activity and standard in-person nutrition counseling (IPNC) on site. Intervention group participants will receive remote, web-based nutrition counseling (WBNC) in addition to the standard medical counseling and on-site physical activity.
7. Methods (Sample size and its justification should be included \& available equipment / means)
8. Randomization: Individuals will be randomized to intervention or control groups by stratified randomization where the stratification will be by cardiac risk level and sex. All eligible patients fulfilling the eligibility criteria will be randomly sampled from the clinic lists. These patients will be contacted by study personnel after approval from the treating physician and asked to come for a baseline meeting and explanation of study procedures. Those meeting all inclusion criteria and willing to provide informed consent will be enrolled. A log will be kept of refusals and reasons for refusal. We expect a refusal rate of 20-35%.
9. Sample size: Assuming baseline I-MEDAS score of median (IQR) of 7(2), the study would require 35 participants per group to achieve 80% power to detect a change in mead I-MEDAS score of 2 units (e.g. to 9 (3) (assuming a larger variance in T1 and T2, T3 results than at baseline) Assuming 30% lost to follow-up, each arm will be increased to 45 per group.
10. Statistical analysis: Due to the multilevel nature of the sampling and allocation, and presumed dependence between participants in the same clinic - multilevel analytic tools will be used. Baseline parameters including age, glucose, BP, and cholesterol measures will be compared between and among groups using multi-level linear models, while distribution of smoking and other categorical variables will be assessed using multi-level models for comparing proportions. Outcome analysis will examine median I-MEDAS across the strata and among treatment arms using linear multilevel models. We will also examine within-subject differences using repeated measures and evaluating outcomes at the three time points (T0, T1, T2 \& T3).