OPTImal CArdiac REhabilitation (OPTICARE) Following Acute Coronary Syndromes

Rationale: Throughout the western world, healthy life style management is becoming increasingly important. Therefore, secondary prevention with an emphasis on lifestyle adjustments should be the cornerstone of modern cardiac rehabilitation (CR). This study may demonstrate the benefits of extended CR with respect to achieving optimal long term secondary prevention goals.

Objective: The objective is to investigate whether in acute coronary syndrome patients the cardiac risk profile will improve with 2 new extended cardiac rehabilitation interventions (OPTICARE and COACH) and if these patients maintain healthier life style habits and a more active life style compared to patients randomized to standard CR.

Study design: The proposed trial is a singlecenter, multi-disciplinary randomized controlled trial with a 18-month follow-up. In total 900 patients will be randomized into 3 arms:(1) Standard CR according to the Dutch guidelines; (2) OPTICARE intervention: standard CR and participation in multifactorial lifestyle sessions during rehabilitation and additional sessions at 4, 6 and 12 months and a personalized program to promote an active lifestyle; (3) COACH intervention: standard CR according to the Dutch guidelines followed by 5 telephone delivered coaching sessions starting at 6 weeks intervals (up to 6 months). The enrollment period will be 24 months, the intervention period is 12 months and the duration of the follow-up will be another 6 months. Assessments will take place at 4 time points: at baseline, 12 weeks after inclusion (end standard CR), at 12 and at 18 months.

Study population: Patients referred for standard CR will be included to participate in this trial. ACS patients treated with primary or elective PCI or CABG, with a stable medical condition and providing written informed consent are eligible to participate.

Intervention: OPTICARE intervention: Standard CR and obligation to participate in multifactorial lifestyle and risk factor sessions (medical information, dietary advises, risk factors and emotional advises). Individual sessions and a personalized home-based program to promote an active life style upon instruction of a physiotherapist and physical activity counsellor during and after completion of rehabilitation. If applicable, patients will participate in smoking cessation, dietary and stress management programs. Additional supervised multifactorial lifestyle and risk management training sessions of each 2 hours are provided at 4, 6 and 12 months. COACH intervention: standard CR according to the Dutch guidelines followed by 5 telephone based coaching sessions starting at 6 weeks intervals up to 6 months. The coaching sessions include assertiveness training, goal setting and reassessment.

Main study parameters/endpoints: The main objective is to investigate whether in patients randomized to extended CR (OPTICARE or COACH) cardiac risk profile will be improved, if these patients maintain healthier life style habits and do achieve an increase in daily activity level compared to patients randomized to standard CR.

Statistical Analysis Plan (SAP) of the OPTICARE study

Introduction The proposed trial is a single center, multi-disciplinary open-label randomized controlled trial. Between October 2011 and August 2014 922 patients were randomized into 3 arms:(1) Standard CR according to the Dutch guidelines (Opticare-Basic) (n=308); (2) OPTICARE intervention: standard CR with the addition of a personalized program to promote an active lifestyle. Besides, additional sessions were offered at 4, 6 and 12 months to improve long-term adherence to lifestyle changes. (Opticare-Capri) (n=310); (3) Heart COACH intervention: standard CR according to the Dutch guidelines followed by 5-6 telephone delivered coaching sessions starting one after finishing standard CR at 6 weeks intervals up to 12 months (Opticare-Achmea Heart Coach) (n=304). The intervention period is 12 months and the duration of the follow-up will be another 6 months. Assessments will take place at 4 time points: at baseline, 12 weeks after inclusion (end standard CR), at 12 and at 18 months.

Current status The study will be finished in January 2016. During September 2015 and January 2016 the data are checked and cleaned. In February 2016 the database will be closed and the statistical analyses can start.

Statistical analysis plan for Cardiology Primary end point and sample size Primary end point is the SCORE risk function measured at 18 months post randomization (Conroy RM et al, Eur HJ 2003; 24:987-1003). This score estimates 10-year risk of fatal cardiovascular disease. The SCORE risk function includes age, gender, systolic blood pressure, total cholesterol and smoking behaviour. The RESPONSE trial (Jorstadt HT, Heart 2013; 99:1421-1430) studied the effectiveness of a nurse-coordinated outpatient risk management program in cardiac patients. That strategy was associated with a 17 % reduction in SCORE Risk function as compared with standard care. Based on these data, and taking into account the more intensive interventions that we will perform, we expect in both the OPTICARE-CAPRI and in the OPTICARECOACH arm at least a 20 % reduction in the SCORE Risk Score at 1.5 years: from 5.40 to 4.32 points with an estimated standard deviation (SD) of 4.5. With 274 patients in each treatment arm, the study has 80 % power (beta-error=0.02) to detect this difference with an alpha-error of 0.05 (2-sided test). We will enroll a total 300 patients in each treatment arm, taking into account a 10 % drop-out rate.

N.B. In all analyses Opticare-Capri will be compared with Opticare-Basic and Opticare-Heart Coach with Opticare-Basic. The sample size computation was such that Opticare-Capri will not be compared with Opticare-Heart Coach.

Primary analysis: intention-to-treat In total, three analyses will be performed to investigate the consistency of the different analyses and to investigate whether differences between the methods can be explained.

Although according to the protocol, no comparisons will be made on cardiac events, these events (mortality, myocardial infarctions, coronary revascularizations and rehospitalizations for cardiac reasons) will be collected.

To be investigated: is the RISC score normally distributed? If necessary, a (ln-) transfrmation will be applied.

Analysis 1 (main analysis): SCORE risk function score at 18 months Opticare-Capri vs Opticare-Basic (student's t-test) and the same for Opticare-Capri vs Opticare-Basic.

Analyse 2:Mixed models will be used. These models takes into account missing data that is missing at random (Twisk, 2004). This analysis will include all patients with at least one measurement. All 4 time points will be included in these models (repeated measurements). At this way we can optimally analyze changes over time.

Analyse 3: Delta score 18 months vs baseline. All patients with a score at baseline and at 18 months will be included.

Secondary analyses: per protocol Patients have to attend at least 75% of the rehabilitation sessions to be included in the per protocol analysis, this means 18 training sessions. Besides, patients included in the Opticare-Achmea Heart Coach have to complete at least 3 out of 5 telephonic coaching sessions. Patients in the OPTICARE-Capri arm were invited to 3 active lifestyle intervention sessions and 3 multifactorial lifestyle and cardiovascular risk factor group intervention sessions. Only those Opticare patients who attended at least 4 of the 6 possible group sessions, will be included in this per-protocol analysis. It will be evaluated whether these results coincide with the intention-to-treat results.

The same analyses as described under the intention-to-treat analyses will be performed for the per protocol analysis and results will be compared to results of the intention-to-treat analyses.

In the intention-to-treat and in the per protocol analyses also the individual risks factors systolic blood pressure, total cholesterol and smoking behaviour will be analyzed separately as dependent variables.

Imputation No imputation will be performed on the end point parameters If necessary, multiple imputation on missing data of the baseline characteristics will be performed.

Statistical analysis plan for Rehabilitation Medicine Primary end point Primary end point is daily amount of moderate-to-vigorous physical activity (MVPA) measured by Actigraph (Bluetooth ® Smart wGT3X-BT).

Sample size Expected amount of minutes in (moderate-intensity) daily activity at 18 months post randomization Opticare-Basic program: 16 minutes/day (SD 13) Based on amount of minutes at start of rehabilitation(Ref. Oliveira and Stevenson et al) Opticare-Achmea Heart Coach: 25 minutes/day (SD 20) Based on amount of minutes at completion of standard CR (Ref. Stevenson et al), it is expected that patients in Opticare-Achmea Heart Coach reach (?) this level of activity Opticare-Capri: 32 minutes/day (SD 23) It is expected that patients in Opticare-Capri reach the recommended level of physical activity. At 2 year follow-up, Oliveira et al found that patients were active for a mean of 63.7 minutes at the end of a counselling session, long term effects are unknown. SD is based on highest SD measured by Oliveira.

With 202 patients in each treatment arm, the study has a power of 80% (beta-error=0.02) to detect this difference with an alpha-error of 0.05, taking into account the ANOVA Bonferroni correction (2-sided test) Taking into account missing values, failure of measurement system and drop outs (20%), a total of 245 patients in each treatment arm will be enrolled.

Analyses: per protocol Patients that dropped out of the study or CR program did not visit the center anymore, making it impossible to install the Actigraph device. Therefore, by design a mix of the intention-to-treat-analysis identical and a per-protocol analysis will be applied. In the CONSORT diagram this design will be explained in more detail.

Per-protocol analysis:

To be investigated: is MVPA normally distributed? If needed, a (ln-) transformation will be performed.

Analysis 1: MVPA at 18 months Opticare-Capri vs Opticare COACH vs Opticare-Basic (ANOVA) .

Analysis 2: Mixed models. These models takes into account missing data that is missing at random (Twisk, 2004).

This analysis will include all patients with at least one measurements. All 4 time points will be included in these models (repeated measurements). In this way we can optimally analyze changes over time.

Analysis 3: Delta minutes of MVPA 18 months vs baseline. All patients with a valid measurement at baseline and at 18 months will be included.

N.B. In the analyses also the minutes of sedentary behavior and the distribution of MVPA and sedentary behavior will be analyzed separately,