Different Lifestyle Interventions in Adults From Underserved Communities

Journal Pre-proof Different Lifestyle Interventions in Adults from Underserved Communities: The FAMILIA Trial Rodrigo Fernandez-Jimenez, MD, PhD, Risa Jaslow, MS, RDN, Sameer Bansilal, MD, MS, Raquel Diaz-Munoz, RN, MS, Monali Fatterpekar, PhD, Maribel Santana, BS, Andrea Clarke-Littman, BS, Jacqueline Latina, MD, MS, Ana V. Soto, MD, Christopher A. Hill, BA, Mohamed Al-Kazaz, MD, Rajeev Samtani, MD, Rajesh Vedanthan, MD, MPH, Chiara Giannarelli, MD, PhD, Jason C. Kovacic, MD, PhD, Emilia Bagiella, PhD, Andrew Kasarskis, PhD, Zahi A. Fayad, PhD, Valentin Fuster, MD, PhD PII:

S0735-1097(19)38083-0

DOI:

https://doi.org/10.1016/j.jacc.2019.10.021

Reference:

JAC 26843

To appear in:

Journal of the American College of Cardiology

Received Date: 28 August 2019 Revised Date:

17 October 2019

Accepted Date: 22 October 2019

Please cite this article as: Fernandez-Jimenez R, Jaslow R, Bansilal S, Diaz-Munoz R, Fatterpekar M, Santana M, Clarke-Littman A, Latina J, Soto AV, Hill CA, Al-Kazaz M, Samtani R, Vedanthan R, Giannarelli C, Kovacic JC, Bagiella E, Kasarskis A, Fayad ZA, Fuster V, Different Lifestyle Interventions in Adults from Underserved Communities: The FAMILIA Trial Journal of the American College of Cardiology (2019), doi: https://doi.org/10.1016/j.jacc.2019.10.021. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 The Author(s). Published by Elsevier on behalf of the American College of Cardiology Foundation.

Different Lifestyle Interventions in Adults from Underserved Communities: The FAMILIA Trial Rodrigo Fernandez-Jimenez, MD, PhDa,b,c; Risa Jaslow, MS, RDNa; Sameer Bansilal, MD, MSa; Raquel Diaz-Munoz, RN, MSa; Monali Fatterpekar, PhDa; Maribel Santana, BSa; Andrea ClarkeLittman, BSa; Jacqueline Latina, MD, MSa; Ana V Soto, MDd; Christopher A Hill, BAa; Mohamed Al-Kazaz, MDa; Rajeev Samtani, MDa; Rajesh Vedanthan, MD, MPHe; Chiara Giannarelli, MD, PhDa; Jason C Kovacic, MD, PhDa; Emilia Bagiella, PhDf; Andrew Kasarskis, PhDa; Zahi A Fayad, PhDa,g; Valentin Fuster, MD, PhDa,b a

The Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai (ISMMS), New York, USA; bCentro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain; cCIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain; dMount Sinai Medical Center, Miami Beach, USA; eDepartment of Population Health, NYU School of Medicine, New York, USA; fCenter for Biostatistics, ISMMS, New York, USA; g Translational and Molecular Imaging Institute, ISMMS, New York, USA. Running title: Family-Based Health Promotion in Underserved Communities Funding: The American Heart Association, under grant No 14SFRN20490315, funded this study. R.F-J is a recipient of funding from the European Union Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 707642. The CNIC is supported by the Instituto de Salud Carlos III (ISCIII), the Ministerio de Ciencia, Innovación y Universidades (MCNU) and the Pro CNIC Foundation, and is a Severo Ochoa Center of Excellence (SEV-2015-0505). Disclosures: Authors declare no relationship with industry or other relevant entities that might pose a conflict of interest in connection with the submitted article. Acknowledgments: The authors are indebted to the children, parents/caregivers, family members, and school staff for their invaluable participation in the study. We are also indebted to all volunteers who collaborated in the study. The authors would also like to thank Amaya de CosGandoy for statistical support, Gloria Santos-Beneit for assistance on the design of the intervention, and Dr Daniella Kadian-Dodov for supervision of vascular ultrasound protocol and procedures. Tweet: The @American_Heart funded FAMILIA trial @MountSinaiHeart highlights that high adherence to lifestyle interventions may improve health outcomes *Corresponding author: Valentin Fuster, MD, PhD Zena and Michael A. Wiener Cardiovascular Institute, Icahn School of Medicine at Mount Sinai 1 Gustave L Levy Place New York, NY 10029, USA Centro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain Telephone: 212-241-7911 Fax: 212-423-9488 E-mail: [email protected] Twitter: @rodrigo_fjez | @MountSinaiHeart | @CNIC_Cardio 1

ABSTRACT Background: The current trends of unhealthy lifestyle behaviors in underserved communities are disturbing. Thus, effective health promotion strategies constitute an unmet need. Objective: To assess the impact of two different lifestyle interventions on parents/caregivers of children attending preschools in a socioeconomically disadvantaged community. Methods: The FAMILIA study is a cluster-randomized trial involving 15 Head Start preschools in Harlem, New York. Schools, and their children’s parents/caregivers, were randomized to receive either an “individual-focused” or “peer-to-peer based” lifestyle intervention program for 12 months or control. The primary outcome was the change from baseline to 12 months in a composite health score related to Blood pressure, Exercise, Weight, Alimentation and Tobacco (Fuster-BEWAT Score, FBS), ranging from 0 to 15 (ideal health=15). To assess the sustainability of the intervention, we evaluated the change of FBS at 24 months. Main prespecified secondary outcomes included changes in FBS subcomponents and the impact of the knowledge of presence of atherosclerosis as assessed by bilateral carotid/femoral vascular ultrasound. Mixed-effects models were used to test for intervention effects. Results: We enrolled 635 parents/caregivers with a mean age of 38±11 years, 83% female, 57% Hispanic/Latino and 31% African American, and a baseline FBS of 9.3±2.4 points. The mean within-group change in FBS from baseline to 12 months was ~0.20 points in all groups, with no overall between-group differences. However, high-adherence participants to the intervention exhibited a greater change in FBS than their low-adherence counterparts: 0.30 points (95% CI: 0.03 to 0.57; p-value = 0.025) vs. 0.00 points (95% CI: -0.43 to 0.43; p-value = 1.0), respectively. Furthermore, the knowledge by the participant of the presence of atherosclerosis significantly boosted the intervention effects. Similar results were sustained at 24 months. Conclusions: Although we did not observe overall significant differences between intervention and control groups, the FAMILIA trial highlights that high adherence rates to lifestyle interventions may improve health outcomes. It also suggests a potential contributory role of the presentation of atherosclerosis pictures, providing helpful information to improve future lifestyle interventions in adults. Condensed abstract: We enrolled 635 parents/caregivers of children attending preschools from a socioeconomically disadvantaged multi-ethnic community in New York City. Participants were randomized to intervention (individual-focused or peer-to-peer based lifestyle programs) for 12 months or control. The study did not show differences in the change of the non-laboratory based Fuster-BEWAT score between control and intervention arms neither at 12 months (immediate post-intervention) nor at 24 months (sustainability). However, a dose-response relationship was observed with high-adherence participants exhibiting a significant improvement. Furthermore, a potential contributory role of the presentation of atherosclerosis pictures was suggested. The study provides helpful information to improve lifestyle interventions. Key words: Health promotion, Vulnerable populations, Parents, Caregivers, Lifestyle Abbreviations IIIP = Intensive individual intervention program PPPI = Peer-to-peer program intervention FBS = Fuster-BEWAT Score 2

Introduction Cardiovascular (CV) health disparities among different racial, ethnic and socioeconomic groups persist in the United States (1). Differences in CV disease burden are broadly attributable to modifiable risk exposures such as hypertension, high cholesterol levels, smoking, dietary risks, obesity, and low physical activity (2). The Committee of the National Academies of Sciences, Engineering, and Medicine recently identified promoting cardiovascular health as one of the 4 priorities areas for action (3). Primary prevention guidelines also emphasize that the most important way to prevent disease is to promote a healthy lifestyle throughout life (4); however, changing our lifestyle behaviors as adults is challenging. Therefore, there is a need for creating approaches that may result in positive, measurable changes in prevalent poor health metrics (5). In particular, the sustainability of health promotion interventions remain largely unknown. Proven success strategies could be potentially implemented at a wider scale to fulfill the goal of reducing the burden of modifiable risk factors significantly at a population level (6). With the aim of promoting health in a socioeconomically disadvantaged community using a family-centered approach, we launched in 2015 the “Family-Based Approach in a Minority Community Integrating Systems-Biology for Promotion of Health (FAMILIA)” trial (7). Children attending Harlem public preschools and their parents/caregivers were recruited and randomized to parallel health promotion intervention programs or to control. We recently demonstrated that children aged 3 to 5 years of age receiving a 4-month preschool-based educational intervention to promote health improved their knowledge, attitudes and habits scores toward a healthy lifestyle (8). The present manuscript reports the peak effects (primary endpoint) and sustained effects (sustainability) of the health promotion interventions delivered among the parents/caregivers participating in the FAMILIA trial. 3

Methods Study design, setting and participants The design and rationale of the FAMILIA study has been previously published (7). Briefly, the study is a parallel-group cluster-randomized controlled trial targeting children aged 3 to 5 years, and their parents/caregivers from 15 public preschools in Harlem, New York City (NYC). Participating schools are part of The Head Start program (U.S. Department of Health and Human Services) that provides comprehensive services to low-income children and their families. School staff were also considered as “caregivers” because they are with the children for <50 hours weekly. During the course of the study, schools participating in FAMILIA agreed not to take part in any other major structured health intervention program aside from the usual curriculum. Schools and parents/caregivers were recruited between October 2015 through May 2017, and were randomized in a 3:2 ratio (3 intervention/2 control) with their assignment aligned to the assignment for the child (8). Those in the “intervention” group were then assigned to one of two interventions in a 1:2 ratio, thus accounting for disparate cluster sizes. The first intervention was the “Intensive Individual Intervention Program (IIIP)” consisting of 8 to 12 individual counseling sessions and a personal activity monitoring device as a motivational agent. The second intervention was the “Peer-To-Peer Program Intervention (PPPI),” consisting of monthly group meetings of participants to help everyone in the group collectively improve their cardiovascular risk factor profile. The intervention program continued for 12 months and participants were advised to participate in a minimum of 8 sessions. As main follow-up assessments, parents/caregivers were scheduled to be evaluated at baseline (prior to the intervention), at approximately 12 months (immediate post intervention, 4

peak-effect) and at approximately 24 months (12 months post intervention, sustainability). Informed written consent was required from all participants. The Icahn School of Medicine at Mount Sinai Institutional Review Board approved the study (HS#:14-01054), which was conducted in accordance with institutional and federal guidelines involving human subjects research. The study is registered in ClinicalTrials.gov, identifier number NCT02481401. Intervention The description of the intervention adheres to the TIDieR (Template for Intervention Description and Replication) guidelines (9). More details about the intervention can be found in the Online Appendix and Online Tables 1 to 3. Intensive Individual Intervention Program (IIIP) The IIIP comprised of one-on-one counseling sessions with a trained lifestyle coach. Lifestyles coaches were assigned to each of the participants and met them at an event at their children’s school where they gave the Lifestyle Coaching Manual to the participants. The participants used the Lifestyle Coaching Manual to discuss the session topics pertaining to cardiovascular health over the course of the intervention. A personal activity monitoring device (Garmin Vivofit 2 and 3) was given to the participants to motivate them and to improve their activity level. The 8 coaching sessions were held once every 3-4 weeks, lasting approximately 45 minutes during the first 8 months. An additional 4 complimentary sessions were offered to the participants over the following 4 months for a total intervention duration of approximately 12 months. Peer-To-Peer Program Intervention (PPPI) The PPPI group meetings occurred at the participant children’s preschools lasting approximately 12 months. After the two first sessions, the participants selected peer leaders and co-leaders who accepted their role on a voluntary basis. These peer leaders and co-leaders received a full day (65

hour long) leadership training provided by the scientific team member of FAMILIA. The coleaders were trained so they could step-up and carry out the responsibility of a leader if the leader was not available to lead the scheduled session during the intervention. At the peer leader led group meetings the participants shared their own experiences, problems, knowledge associated with health habits, evaluated changes and offered mutual support. 45-minute sessions were held every 3-4 weeks and lasted for approximately 12 months. The FAMILIA research team trained and provided regular support to the lifestyle coaches of IIIP and the peer leaders and co-leaders of the PPPI during the course of the intervention. An incentive structure was established for the participants based on their session attendance with their lifestyle coaches and the group meetings. Control Group Parents and caregivers randomized to the control arm did not receive any structured program for the first 4 months of the study. This coincided with the children not receiving any educational intervention program. After 4 months, coinciding with the crossover of the children’s intervention, the control adult participants received relevant components of the health-based curriculum activities the children completed in the classroom, including newsletters with poems, arts and crafts activities or home routines for the children related to healthy lifestyles. A detailed description of the child intervention was published elsewhere (8). Additionally, we provided informational and educational sessions on summer safety, dental hygiene, foot health, financial management and tax advice on a quarterly basis. Materials provided during the assessments When participants came to the study assessments, they all received an individualized counseling session to discuss their results (e.g., weight, waist and hip circumferences, blood 6

pressure, body-mass index, lipid profile, vascular ultrasound) irrespective of their study arm allocation. They also received a representative printed picture of their carotid and/or femoral vessels, and handouts with information about weight control, guidance on interpreting results for their levels of glucose, cholesterol, and blood pressure, and tips about portion control (Online Appendix). All questionnaires and brochures were available in English and Spanish, and health counselors were fluent in both languages to accommodate participant language preferences. Primary outcome: change in overall Fuster-BEWAT score The primary outcome was the difference between groups in the change from baseline of a composite health score related to Blood pressure, Exercise, Weight (body-mass index), Alimentation (fruit and vegetable consumption) and Tobacco (smoking habit) (Fuster-BEWAT Score, FBS) at the completion of the 12-month intervention. A final 24-month assessment was performed to evaluate the sustainability of the impact of the 2 lifestyle interventions on FBS approximately 12 months after the intervention program ended. The overall score ranges from 0 (poor health) to 15 (ideal cardiovascular health), and it is derived from the sum of individual components of the score (0-3 points each). The details for the calculation of the FBS at baseline and follow-up assessments, alongside the inverse association of the FBS with the presence and extent of subclinical atherosclerosis, have been previously published (10,11). Overall FBS was calculated for all participants with data available for measuring at least 4 out of the 5 individual components of the score. For those participants with one individual component missing (<10%), the corresponding variable was imputed using the calculated average of the remaining four components of the given subject. Secondary outcomes: changes in FBS subcomponents and subgroup analysis

7

As secondary outcomes, the changes in the individual components of the FBS and the impact of the intervention on the overall FBS change in different subgroups were evaluated. Variables for subgroup analysis included the following: age (<40 years old vs. 40 years or older), sex, self-identified race/ethnicity (non-Hispanic black, Hispanic/Latino, other/multiracial), baseline Fuster-BEWAT score (tertiles), place of birth (United States vs. outside United States), school staff (yes/no), self-reported annual household income (<$25,000; $25,000-$50,000; >$50,000), self-reported highest level of education (secondary school or lower, high school, college/university/postgraduate), and the results of the baseline three-dimensional carotid/femoral vascular ultrasound performed (evidence of atherosclerosis at any location evaluated vs. absence of atherosclerosis). The methodology used for data collection alongside the ultrasound imaging protocol are described in the Online Appendix. Statistical analysis All study data were first collected on paper, and then entered into a REDCap (Research Electronic Data Capture) database hosted at the Icahn School of Medicine at Mount Sinai, New York. Continuous variables are reported as mean and standard deviation (SD) while discrete variables are reported as frequencies and percentages, unless otherwise specified. Multilevel linear mixed-effects models that account for the hierarchical cluster randomized design were used to test for the adjusted intervention effect (change in FBS). Fixed effects were the corresponding baseline FBS (as a continuous variable), and treatment group. Schools and families were handled as random effects. The same linear mixed models were applied for the analysis of the change in the subcomponents of the FBS. Interaction models were also fitted to identify possible age-, sex-, race/ethnicity- baseline FBS-, place of birth-, school staff-,

8

household income-, education level-, and vascular ultrasound results-by-treatment effects for the main outcome variable (overall FBS). To assess a potential dose-response effect of the intervention, differences in FBS between participants receiving less than 50% of the program sessions (<4 sessions, low adherence group) versus those receiving 50 or more of the program sessions (≥4 sessions, high adherence group) were explored by the use of similar linear mixed-effects models (16). Data about adherence to the intervention was based on number of sessions attended and was collected at the individual level. Every attempt was made to follow all enrolled participants irrespective of allocation or withdrawal from treatment. All participants were included in the analysis in the groups to which they were randomized. A complete-case intention-to-treat analysis was performed as the main analysis. Under the assumption of missing at random, multiple imputation using multivariate normal distribution was performed to include all randomized enrolled participants as sensitivity analysis. For subgroup analyses, missing values (if any) on variables used to create subgroups were not imputed; instead, pairwise deletion was performed. Details on sample size and power calculations and multiple imputation procedures carried out can be found in Supplemental Methods. Statistical significance was set at a p-value <0.05, except in the case of the interaction analyses in which statistical significance was set at a p-value <0.10 (17). Because of the a priori defined primary hypothesis and the exploratory nature of secondary analyses, results were not adjusted for multiplicity (18). All analyses were performed using STATA version 15 (StataCorp, College Station, Texas).

9

Results Participant flow diagram and baseline characteristics The study enrolled 15 schools of which 9 were randomized to intervention groups (3 schools to the Intensive Individual Intervention Program and 6 schools to Peer-To-Peer Program Intervention) and 6 to the control condition, totaling 635 adults enrolled and assessed at baseline. After a median follow-up of 17 and 27 months, ~30% and ~33% of participants were lost to follow-up or had incomplete data, respectively. Therefore, 446 and 424 adults were included in the main analysis (complete-case intention-to-treat analysis) of the primary outcome (Figure 1) and sustainability outcome (Online Figure 1) of the study, respectively. No school withdrew from the trial during the study period, and no adverse events were reported. Table 1 contains a summary of the collected baseline information at the school and individual levels. In summary, no significant differences were found between control and intervention groups at baseline, with the exception of a higher proportion of Hispanic/Latino and of participants born outside the United States in the intensive individual intervention group. Baseline information for those participants lost to follow-up or with incomplete data and for remaining participants included in the primary outcome analysis, either in the intervention groups or control, are presented in Online Table 4. Primary outcome: change in Fuster-BEWAT score at 12 months (peak effect) Baseline overall FBS were 9.3 (SD 2.6), 9.4 (SD 2.3) and 9.2 (SD 2.4) in the control, IIIP-Individual and PPPI-Peer groups, respectively. The mean within-group change from baseline at 12-month assessment in the overall FBS was approximately 0.20 points in all groups (Table 2). The average absolute differences in overall FBS were -0.01 points (95% CI: -0.53 to 0.52; p = 0.978) between control and IIIP-Individual groups, and 0.00 points (95% CI: -0.50 to 10

0.49; p = 0.988) between control and PPPI-Peer groups. No significant differences were observed in any of the FBS components between groups. Overall results were similar when including all randomized enrolled participants (n = 635) after multiple imputation (data not shown). In the IIIP-Individual group, 69% of participants (n = 149 out of all 217 randomized) and 85% of participants (n = 119 out of all 157 included in the complete-case intention-to-treat analysis at 12 months) attended ≥ 4 of scheduled sessions (high adherence group). In the PPPIPeer group, 50% of participants (n = 117 out of all 236 randomized) and 65% (n = 111 out of all 170 included in the complete-case intention-to-treat analysis at 12 months) attended ≥ 4 of scheduled sessions. Overall, high adherence participants exhibited a greater change in the overall FBS at 12 months than their low adherence counterparts: 0.00 points (95% CI: -0.43 to 0.43; pvalue = 1.0) in the low adherence groups vs. 0.30 points (95% CI: 0.03 to 0.57; p-value = 0.025) in the high adherence groups; however, this difference between groups was not statistically significant (p = 0.225). Online Table 5 details the changes and differential changes in the low and high adherence groups in each of the intervention arms at 12 months. Sustainability: change in Fuster-BEWAT score at 24 months The mean within-group changes from baseline at 24-month assessment in the overall FBS were 0.11, 0.22 and -0.19 points in control, IIIP-Individual and PPPI-Peer groups, respectively, with no overall between-group differences (Table 3). Overall results were similar when including all randomized enrolled participants (n = 635) after multiple imputation (data not shown). In the IIIP-Individual group, 81% of participants (n = 111 out of all 161 included in the complete-case intention-to-treat analysis at 24 months) attended ≥ 4 of scheduled sessions (high 11

adherence group). In the PPPI-Peer group, 66% (n = 101 out of all 152 included in the completecase intention-to-treat analysis at 24 months) attended ≥ 4 of scheduled sessions. Overall, high adherence participants exhibited a greater change in the overall FBS at 24 months than their low adherence counterparts: -0.40 points (95% CI: -0.82 to 0.02; p-value = 0.065) in the low adherence groups vs. 0.16 points (95% CI: -0.09 to 0.42; p-value = 0.208) in the high adherence groups; and this difference was statistically significant (p = 0.025). Online Table 6 details the changes and differential changes in the low- and high adherence groups in each of the intervention arms at 24 months. Determinants of the intervention effects A breakdown of the mean differences (intervention vs. control) in the change of overall FBS at 12 months according to variables of interest is shown in Figure 2, while mean changes in each of the intervention groups are shown in Online Figure 2. This stratified analysis revealed significant interactions of the effect of the intervention with baseline overall FBS, place of birth and the results of vascular ultrasound. Participants starting from a lower baseline score, who were born outside US and with evidence of atherosclerosis benefited more from the intervention. A trend suggesting larger benefits among school staff participants and higher self-reported annual household income (>$50,000) was also observed; however these interactions did not reach statistical significance. A breakdown of the mean differences (intervention vs. control) in the change of overall FBS at 24 months according to variables of interest is shown in Figure 3, while mean changes in each of the intervention groups are shown in Online Figure 3. This stratified analysis revealed significant interactions of the effect of both intervention groups with the results of vascular ultrasound and the condition of the individual in regards to be part of the school staff or not. 12

Participants with evidence of atherosclerosis and school staff benefited more from the intervention. A trend suggesting larger benefits among participants 40 years of age or older, those who were born outside United States, and with a higher self-reported annual household income (>$50,000) and education level was observed; however these interactions were not statistically significant. Discussion Among preschool children’s parents/caregivers from an urban multi-ethnic community, neither an intensive individual intervention program nor a peer-to-peer program intervention delivered over the course of 12 months, demonstrated a significant impact on simple health metrics as compared to controls. The overall lack of intervention effects was seen both at immediate post-intervention and at approximately 1 year after the end of the intervention, suggesting that there were no delayed results of the intervention. However, a dose-response intervention effect was observed; those individuals attending 50% or more of the intervention program demonstrated healthier changes than their low adherence counterparts at both timepoints evaluated. The identification and understanding of ultrasound-based images of atherosclerosis by the participant consistently boosted the intervention effects. Furthermore, we identified subgroups of participants in which lifestyle interventions could have a greater impact. All this information may be useful to tailor future health promotion programs in adults (Central Illustration). Reasons for a lack of overall intervention effects in the FAMILIA trial Although evidence on hard, longer-term CV outcomes resulting from lifestyle interventions is needed (19), health promotion programs using counseling and education aimed at behavior change might be effective when targeted to individuals with CV risk factors, such as 13

type 2 diabetes or hypertension. In contrast, they may have limited use in general populations (20). The FAMILIA trial stems from prior comprehensive school-based (the SI! Program) (2124) and peer group-based (the Fifty-Fifty Program) (16) lifestyle interventions conducted in Spain and Colombia. The Fifty-Fifty Program in Spain targeted adults 25 to 50 years of age with at least 1 CV risk factor and showed a significant improvement in the same composite FBS using a similar peer-to-peer intervention strategy, as compared to controls. Beyond the complexity of changing health-related behaviors (25), several factors may have decreased the ability to generate larger intervention effects in the FAMILIA trial. Firstly, the present study enrolled a relatively young and healthy population. Thus, the mean initial FBS was ~20% higher in the FAMILIA adult population than in the Fifty-Fifty trial (higher score indicating better health), making it more challenging to demonstrate a significant change. This finding was mostly due to healthier scores observed in the subcomponents of blood pressure and exercise among participants enrolled in the FAMILIA trial. Secondly, the control group received relevant health promotion messages over the course of the study. This included up to 5 counseling sessions to discuss their health and vascular ultrasound results that were offered at the end of study assessments. In addition, control parents/caregivers received relevant components of the health-based curriculum activities the children completed in the classroom. The fact that FBS in the control group increased similarly as in the intervention groups at 12 months (by ~0.20 points) supports the notion that health status knowledge and awareness might serve as triggers for actual behavior change (26). Thirdly, the overall intervention adherence was modest; and this could have decreased the overall intervention effects. A dose-response relationship was observed, consistently with prior studies (8,16). Factors that affect intervention fidelity in health promotion programs warrants further research. Fourthly, contamination in the 14

form of shared intervention messages or products between individuals in the community cannot be excluded because the relatively small area in which the trial was conducted (i.e. Harlem neighborhood). Finally, social determinants may play an important role on intervention effects. The FAMILIA trial enrolled relatively homogenous low-income families. Assuming intervention effectively reached the parents/caregivers participating in the trial, low-income populations still frequently face food insecurity issues (27). Furthermore, environmental variables such as availability of affordable sport facilities and neighborhood safety may also impact the participation in physical activities (28). Determinants of intervention effects A breakdown of the effect of the intervention suggested larger effects in some specific subgroups. The most consistent finding was the association of the knowledge and awareness of the presence of atherosclerosis as assessed by three-dimensional vascular ultrasound with a higher intervention effect at both time-points evaluated. This is consistent with the results of a recent pragmatic randomized trial that enrolled 3,532 individuals aged 40 to 60 years with one or more conventional CV risk factors, in which the contributory role of pictorial presentation of silent atherosclerosis for the control of such risk factors was demonstrated (29). Although exploratory, the fact that the FAMILIA trial recruited a younger population at low risk for CV disease and with a low prevalence of (very early) atherosclerosis offers the potential for real impact on CV health promotion at the population level. Yet speculative, the mechanisms of the higher intervention effect observed at 12 months in migrants to the US (as compared to participants born in the US) might include positive health selection of migrants from originating countries; cultural differences in health-related behaviors, and supportive familial and social networks among migrants (30). We observed a greater impact 15

of the intervention at 24 months in school staff as compared to parents or other family-related caregivers of the children. School staff are one of the cornerstones in children’s behavior development given that children spend a significant amount of time in school interacting with them (31). The particular impact of school staff motivation and health promotion programs targeted to school staff as means of positively affecting the lifestyle-related behavior of the children warrants future research. Study limitations and strengths Despite implementing intensive retention strategies over the course of the study, ~ 30% of participants were lost to follow-up at the primary endpoint assessment. This could have made the study underpowered to detect significant differences. Those completing the study were slightly older, more frequently Hispanic/Latino and school staff, and with a relatively higher prevalence of atherosclerosis. Although we cannot completely exclude a significant impact of lost to follow up participants on our results, complete-case intention-to-treat analyses were complemented with sensitivity analysis using multiple imputation procedures obtaining similar findings. A discussion on the acceptability of intervention programs and lessons learned is presented in the Online Appendix. Because recruitment was performed through schools and there were more female parents/caregivers present with their children at schools, the proportion of males enrolled in the FAMILIA trial was low. This is consistent with prior findings indicating that males, especially ethnic males, are underrepresented in lifestyle weight loss trials (32). One of the main strengths of the FAMILIA study is the cluster-randomized controlled design that allows isolation of the intervention effects. The low number of clusters in this study might lead to imbalances in effect estimates. As sensitivity analysis, we performed unadjusted cluster-level analysis obtaining similar results (no overall differences between groups in the 16

change of FBS from baseline; data not shown). The main outcome measure was a non-laboratory health score, i.e. FBS. Despite its simplicity, FBS performs similarly to other laboratory-based scores (33,34). These characteristics make the FBS particularly suited as a reliable low-cost indicator of CV health. Because of the low prevalence of atherosclerosis in the FAMILIA study, the possibility of evaluating the impact of lifestyle interventions on the progression of atherosclerosis was limited. Fidelity of the intervention was measured through individual attendance records which allowed us to evaluate a potential dose-response relationship. The fact that participants starting from a lower baseline score (among other subgroups) benefited more from the intervention could reflect a phenomenon of regression to the mean to a certain extent. Several strategies were applied to alleviate this effect at the overall design (random allocation to comparison groups) and analysis (statistical models included adjustment for the baseline value of the outcome score) stages of the study. Nevertheless, results from subgroup analyses should be considered as exploratory. Despite the overall neutral effects, the study included the collection of adult blood material pre- and post-intervention. Samples will be analyzed in an effort to understand the complex relationships between lifestyle behavioral changes, atherosclerosis and genomics (35). Conclusions Among parents/caregivers of children attending preschools from an urban multi-ethnic community enrolled in the FAMILIA trial, neither an intensive individual intervention program nor a peer-to-peer program intervention delivered over the course of 12 months, demonstrated a significant impact on simple health metrics as compared to controls. A dose-response relationship was observed, indicating that intervention adherence is critical on achieving a potential beneficial impact. Furthermore, a contributory role of the presentation of 17

atherosclerosis pictures to the participant was suggested. This study provides helpful information to tailor and improve lifestyle interventions. A wider adoption of effective health promotion programs at the population level may have a meaningful impact on the reduction of health disparities and CV disease burden.

18

CLINICAL PERSPECTIVES Competency in Medical Knowledge: Effective health promotion constitutes a global unmet challenge. The FAMILIA trial failed to demonstrate an overall beneficial impact of a 12-month lifestyle intervention (individual-focused or peer-to-peer based) on health scores in an urban minority community. Translational Outlook 1: High adherence participants to the intervention exhibited a significantly greater change in health metrics, indicating that intervention adherence is critical for efficacy. Factors that may affect fidelity to the intervention warrants further research. Translational Outlook 2: Consistent with previous studies, the presentation of the own participant’s atherosclerosis pictures significantly boosted the intervention effects. This information may be helpful to improve future lifestyle interventions.

19

References 1.

Mensah GA, Cooper RS, Siega-Riz AM et al. Reducing Cardiovascular Disparities

Through Community-Engaged Implementation Research: A National Heart, Lung, and Blood Institute Workshop Report. Circulation research 2018;122:213-230. 2.

Global Burden of Cardiovascular Diseases C, Roth GA, Johnson CO et al. The Burden of

Cardiovascular Diseases Among US States, 1990-2016. JAMA cardiology 2018;3:375-389. 3.

Fuster V, Frazer J, Snair M et al. The Future Role of the United States in Global Health:

Emphasis on Cardiovascular Disease. J Am Coll Cardiol 2017;70:3140-3156. 4.

Arnett DK, Blumenthal RS, Albert MA et al. 2019 ACC/AHA Guideline on the Primary

Prevention of Cardiovascular Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol 2019;74:e177-e232. 5.

Arena R, Guazzi M, Lianov L et al. Healthy lifestyle interventions to combat

noncommunicable disease-a novel nonhierarchical connectivity model for key stakeholders: a policy statement from the American Heart Association, European Society of Cardiology, European Association for Cardiovascular Prevention and Rehabilitation, and American College of Preventive Medicine. Eur Heart J 2015;36:2097-2109. 6.

Benjamin EJ, Muntner P, Alonso A et al. Heart Disease and Stroke Statistics-2019

Update: A Report From the American Heart Association. Circulation 2019;139:e56-e528. 7.

Bansilal S, Vedanthan R, Kovacic JC et al. Rationale and Design of Family-Based

Approach in a Minority Community Integrating Systems-Biology for Promotion of Health (FAMILIA). Am Heart J 2017;187:170-181.

20

8.

Fernandez-Jimenez R, Jaslow R, Bansilal S et al. Child Health Promotion in Underserved

Communities: The FAMILIA Trial. J Am Coll Cardiol 2019;73:2011-2021. 9.

Hoffmann TC, Glasziou PP, Boutron I et al. Better reporting of interventions: template

for intervention description and replication (TIDieR) checklist and guide. BMJ 2014;348:g1687. 10.

Gómez-Pardo E, Fernández-Alvira JM, Vilanova M et al. A Comprehensive Lifestyle

Peer Group–Based Intervention on Cardiovascular Risk Factors: The Randomized Controlled Fifty-Fifty Program. J Am Coll Cardiol. 2016;67:476-485. 11.

Fernández-Alvira JM, Fuster V, Pocock S et al. Predicting Subclinical Atherosclerosis in

Low-Risk Individuals: Ideal Cardiovascular Health Score and Fuster-BEWAT Score. J Am Coll Cardiol. 2017;70:2463-2473. 12.

Craig CL, Marshall AL, Sjostrom M et al. International physical activity questionnaire:

12-country reliability and validity. Medicine and science sports exercise 2003;35:1381-95. 13.

Banna JC, Vera Becerra LE, Kaiser LL, Townsend MS. Using qualitative methods to

improve questionnaires for Spanish speakers: assessing face validity of a food behavior checklist. J Am Diet Assoc 2010;110:80-90. 14.

Riley L, Guthold R, Cowan M et al. The World Health Organization STEPwise Approach

to Noncommunicable Disease Risk-Factor Surveillance: Methods, Challenges, and Opportunities. Am J Public Health 2016;106:74-8. 15.

Lopez-Melgar B, Fernandez-Friera L, Oliva B et al. Subclinical Atherosclerosis Burden

by 3D Ultrasound in Mid-Life: The PESA Study. J Am Coll Cardiol 2017;70:301-313. 16.

Gomez-Pardo E, Fernandez-Alvira JM, Vilanova M et al. A Comprehensive Lifestyle

Peer Group-Based Intervention on Cardiovascular Risk Factors: The Randomized Controlled Fifty-Fifty Program. J Am Coll Cardiol 2016;67:476-85. 21

17.

Richardson M, Garner P, Donegan S. Interpretation of subgroup analyses in systematic

reviews: A tutorial. Clinical Epidemiology and Global Health 2019;7:192-198. 18.

Althouse AD. Adjust for Multiple Comparisons? It's Not That Simple. Ann Thorac Surg

2016;101:1644-5. 19.

Schwarz PEH, Timpel P, Harst L et al. Blood Sugar Regulation for Cardiovascular Health

Promotion and Disease Prevention: JACC Health Promotion Series. J Am Coll Cardiol 2018;72:1829-1844. 20.

Ebrahim S, Taylor F, Ward K, Beswick A, Burke M, Davey Smith G. Multiple risk factor

interventions for primary prevention of coronary heart disease. Cochrane database systematic reviews 2011:CD001561. 21.

Cespedes J, Briceno G, Farkouh ME et al. Targeting preschool children to promote

cardiovascular health: cluster randomized trial. Am J Med. 2013;126:27-35 e3. 22.

Santos-Beneit G, Bodega P, de Miguel M et al. Rationale and design of the SI! Program

for health promotion in elementary students aged 6 to 11 years: A cluster randomized trial. Am Heart J 2019;210:9-17. 23.

Fernandez-Jimenez R, Santos-Beneit G, Tresserra-Rimbau A et al. Rationale and design

of the school-based SI! Program to face obesity and promote health among Spanish adolescents: A cluster-randomized controlled trial. Am Heart J 2019;215:27-40. 24.

Penalvo JL, Santos-Beneit G, Sotos-Prieto M et al. The SI! Program for Cardiovascular

Health Promotion in Early Childhood: A Cluster-Randomized Trial. J Am Coll Cardiol 2015;66:1525-1534. 25.

Kelly MP, Barker M. Why is changing health-related behaviour so difficult? Public

Health 2016;136:109-16. 22

26.

Stonerock GL, Blumenthal JA. Role of Counseling to Promote Adherence in Healthy

Lifestyle Medicine: Strategies to Improve Exercise Adherence and Enhance Physical Activity. Prog Cardiovasc Dis 2017;59:455-462. 27.

Yu E, Malik VS, Hu FB. Cardiovascular Disease Prevention by Diet Modification: JACC

Health Promotion Series. J Am Coll Cardiol 2018;72:914-926. 28.

Fletcher GF, Landolfo C, Niebauer J, Ozemek C, Arena R, Lavie CJ. Promoting Physical

Activity and Exercise: JACC Health Promotion Series. J Am Coll Cardiol 2018;72:1622-1639. 29.

Naslund U, Ng N, Lundgren A et al. Visualization of asymptomatic atherosclerotic

disease for optimum cardiovascular prevention (VIPVIZA): a pragmatic, open-label, randomised controlled trial. Lancet 2019;393:133-142. 30.

Le-Scherban F, Albrecht SS, Bertoni A, Kandula N, Mehta N, Diez Roux AV. Immigrant

status and cardiovascular risk over time: results from the Multi-Ethnic Study of Atherosclerosis. Ann epidemiol. 2016;26:429-435 e1. 31.

Fernandez-Jimenez R, Al-Kazaz M, Jaslow R, Carvajal I, Fuster V. Children Present a

Window of Opportunity for Promoting Health: JACC Review Topic of the Week. J Am Coll Cardiol 2018;72:3310-3319. 32.

Pagoto SL, Schneider KL, Oleski JL, Luciani JM, Bodenlos JS, Whited MC. Male

inclusion in randomized controlled trials of lifestyle weight loss interventions. Obesity 2012;20:1234-9. 33.

Wang H, Sun G, Li Z, Guo X, Sun Y. Ideal Cardiovascular Health Score and Fuster-

BEWAT Score as Predictors of Left Ventricular Hypertrophy Classification. J Am Coll Cardiol 2018;72:581-582.

23

34.

Fernandez-Alvira JM, Fuster V, Pocock S et al. Predicting Subclinical Atherosclerosis in

Low-Risk Individuals: Ideal Cardiovascular Health Score and Fuster-BEWAT Score. J Am Coll Cardiol 2017;70:2463-2473. 35.

Khera AV, Emdin CA, Drake I et al. Genetic Risk, Adherence to a Healthy Lifestyle, and

Coronary Disease. N Engl J med. 2016;375:2349-2358.

24

Figure Legends Figure 1. The FAMILIA study flow diagram. Recruitment of schools and parents/caregivers, and completeness of baseline and 12-month follow-up measures, according to the guidelines in the CONSORT 2010 statement for the reporting of cluster randomized trials. Four participants are not reflected in the consort diagram: two participants in the PPPI-Peer group withdrew their consent and requested all their data to be eliminated; signed consent documents were not found for one participant in the PPPI-Peer group and one participant in the control group. Figure 2. Subgroup analysis of the differential changes in overall FBS at 12 months between control and intervention groups. Forest plot representing mean differences (95% confidence interval) in the overall Fuster-BEWAT Score (FBS) changes at 12-month assessment between parents/caregivers in the intervention groups (IIIP-Individual, PPPI-Peer) and control groups, after stratified linear mixed-effects models by selected variables. Fixed effects were the corresponding baseline score and treatment group, while schools and families were handled as random effects. For baseline FBS subgroup analysis, the score was categorized in tertiles (low, intermediate, high) and continuous baseline score was not included in the model. C: control; I: intervention. Figure 3. Subgroup analysis of the differential changes in overall FBS at 24 months between control and intervention groups. Forest plot representing mean differences (95% (95% confidence interval) in the overall Fuster-BEWAT Score (FBS) changes at 24-month assessment between parents/caregivers in the intervention groups (IIIP-Individual, PPPI-Peer) and control groups, after stratified linear mixed-effects models by selected variables (as in Figure 2).

25

Central Illustration. Effect of 2 different lifestyle interventions in underserved communities. The FAMILIA trial enrolled 635 parents/caregivers of children attending preschools from an urban socioeconomically disadvantaged multi-ethnic community in New York City. Participants were randomized to intervention (individual-focused or peer-to-peer based lifestyle programs) for 12 months or control. The study did not show differences in the change of the non-laboratory-based health Fuster-BEWAT score between control and intervention arms neither at 12 months (immediate post-intervention) nor at 24 months (sustainability). However, a dose-response relationship was observed with high-adherence participants exhibiting a significantly greater change. Furthermore, a potential contributory role of the presentation of atherosclerosis pictures was suggested.

26

Table 1. Baseline characteristics of enrolled schools and adults in the FAMILIA study. Overall

Control

Intervention

Intervention

(IIIP-Individ.) (PPPI-Peer) Schools No. of schools

15

6

3

6

No. of adults/school

42.3 ± 32.3

30.3 ± 16.7

72.3 ± 52.5

39.3 ± 28.7

No. of adults

635

182

217

236

Age, years

37.8 ± 11.3

38.9 ± 11.4

37.7 ± 11.4

37.1 ± 11.0

Female

524 (82.5)

152 (83.5)

175 (80.7)

197 (83.5)

Non-Hispanic black

197 (31.0)

70 (38.5)

39 (18.0)

88 (37.3)

Hispanic/Latino

359 (56.5)

80 (44.0)

163 (75.1)

116 (49.2)

Other/multiracial

79 (12.5)

32 (17.5)

15 (6.9)

32 (13.6)

Born outside United States

305 (48.0)

83 (45.6)

128 (59.0)

94 (39.8)

School staff

159 (25.0)

44 (24.2)

52 (24.0)

63 (26.7)

<$25,000

281 (44.3)

78 (42.9)

106 (48.9)

97 (41.1)

$25,000 - $50,000

164 (25.8)

51 (28.0)

45 (20.7)

68 (28.8)

>$50,000

63 (9.9)

20 (11.0)

13 (6.0)

30 (12.7)

Unknown

127 (20.0)

33 (18.1)

53 (24.4)

41 (17.4)

Secondary school or lower

183 (28.8)

41 (22.5)

85 (39.2)

57 (24.2)

High school

299 (47.1)

85 (46.7)

96 (44.2)

118 (50.0)

College/University/Postgraduate

140 (22.1)

49 (26.9)

34 (15.7)

57 (24.2)

Parents/Caregivers

Race/ethnicity

Annual household income

Highest level of education completed

27

Unknown

13 (2.1)

7 (3.9)

2 (0.9)

4 (1.7)

FBS overall (range 0-15)

9.3 ± 2.4

9.3 ± 2.6

9.4 ± 2.3

9.2 ± 2.4

B- Blood pressure (range 0-3)

2.2 ± 1.1

2.2 ± 1.1

2.3 ± 1.0

2.1 ± 1.1

E- Exercise (range 0-3)

1.9 ± 1.2

1.9 ± 1.2

2.0 ± 1.2

1.9 ± 1.1

W- Weight (range 0-3)

1.0 ± 1.1

1.1 ± 1.2

0.9 ± 1.1

1.0 ± 1.2

A- Alimentation (range 0-3)

1.4 ± 0.7

1.4 ± 0.8

1.5 ± 0.7

1.4 ± 0.7

T- Tobacco (range 0-3)

2.7 ± 0.7

2.7 ± 0.7

2.8 ± 0.6

2.7 ± 0.7

Absence of atherosclerosis

468 (73.7)

123 (67.6)

178 (82.0)

167 (70.8)

Evidence of atherosclerosis

47 (7.4)

19 (10.4)

11 (5.1)

17 (7.2)

Not performed/not analyzable

120 (18.9)

40 (22.0)

28 (12.9)

52 (22.0)

Fuster-BEWAT score

Vascular ultrasound

Values are mean ± standard deviation, or n (%). Adults’ race/ethnicity was self-identified. FBS: Fuster-BEWAT score.

28

Table 2. Change from baseline in overall Fuster-BEWAT score and components among parents/caregivers in the FAMILIA trial at 12-months (primary endpoint).

Within group differences Score

Control

range

Between group difference

Intervention

Intervention

Difference (95% CI)

Difference (95% CI)

(IIIP-Individual)

(PPPI-Peer)

Control vs. Individual Control vs. Peer

0-15

0.20 (-0.17 to 0.58)

0.20 (-0.17 to 0.57)

0.20 (-0.12 to 0.53)

-0.01 (-0.53 to 0.52)

0.00 (-0.50 to 0.49)

B- Blood pressure

0-3

0.09 (-0.08 to 0.26)

0.12 (-0.06 to 0.29)

0.11 (-0.04 to 0.26)

0.02 (-0.22 to 0.27)

0.02 (-0.21 to 0.25)

E- Exercise

0-3

0.17 (-0.02 to 0.36)

0.21 (0.05 to 0.37)

0.07 (-0.08 to 0.23)

0.04 (-0.21 to 0.29)

-0.09 (-0.34 to 0.15)

W- Weight

0-3

-0.06 (-0.19 to 0.07)

-0.14 (-0.26 to -0.03)

-0.01 (-0.12 to 0.10)

-0.08 (-0.26 to 0.09)

0.04 (-0.13 to 0.22)

A- Alimentation

0-3

0.01 (-0.15 to 0.16)

0.01 (-0.16 to 0.19)

0.01 (-0.14 to 0.15)

0.01 (-0.23 to 0.24)

0.00 (-0.21 to 0.21)

T- Tobacco

0-3

0.01 (-0.09 to 0.11)

0.03 (-0.06 to 0.12)

0.03 (-0.06 to 0.11)

0.02 (-0.11 to 0.15)

0.02 (-0.11 to 0.14)

FBS overall

Data are presented as marginal mean differences and 95% confidence intervals (CI) as derived from linear mixed-effects models. Fixed effects were the corresponding baseline score and treatment group, while schools and families were handled as random effects. FBS: Fuster-BEWAT score.

29

Table 3. Change from baseline in overall Fuster-BEWAT score and components among parents/caregivers in the FAMILIA trial at 24-months (sustainability).

Within group differences Score

Control

range

Between group difference

Intervention

Intervention

Difference (95% CI)

Difference (95% CI)

(IIIP-Individual)

(PPPI-Peer)

Control vs. Individual Control vs. Peer

0-15

0.11 (-0.26 to 0.48)

0.22 (-0.08 to 0.53)

-0.19 (-0.50 to 0.13)

0.11 (-0.36 to 0.59)

-0.30 (-0.78 to 0.19)

B- Blood pressure

0-3

0.14 (0.00 to 0.28)

0.19 (0.07 to 0.31)

0.02 (-0.10 to 0.14)

0.05 (-0.13 to 0.24)

-0.12 (-0.31 to 0.07)

E- Exercise

0-3

0.07 (-0.14 to 0.29)

0.10 (-0.08 to 0.27)

0.02 (-0.16 to 0.20)

0.03 (-0.25 to 0.30)

-0.05 (-0.33 to 0.23)

W- Weight

0-3

0.02 (-0.13 to 0.16)

-0.10 (-0.22 to 0.03)

-0.10 (-0.23 to 0.03)

-0.11 (-0.31 to 0.08)

-0.12 (-0.31 to 0.08)

A- Alimentation

0-3

-0.08 (-0.20 to 0.04)

0.00 (-0.10 to 0.10)

-0.06 (-0.16 to 0.05)

0.08 (-0.08 to 0.23)

0.02 (-0.14 to 0.18)

T- Tobacco

0-3

0.03 (-0.06 to 0.12)

0.03 (-0.05 to 0.10)

-0.03 (-0.11 to 0.04)

-0.01 (-0.12 to 0.11)

-0.07 (-0.18 to 0.05)

FBS overall

Data are presented as marginal mean differences and 95% confidence intervals (CI) as derived from linear mixed-effects models. Fixed effects were the corresponding baseline score and treatment group, while schools and families were handled as random effects. FBS: Fuster-BEWAT score

30

ONLINE APPENDIX

Contents SUPPLEMENTAL METHODS ............................................................................................................. 2 Description of the intervention programs ....................................................................................... 2 Materials and procedures ............................................................................................................... 2 Intervention provider ...................................................................................................................... 3 Modes of delivery ............................................................................................................................ 4 Study setting .................................................................................................................................... 5 Duration and intensity .................................................................................................................... 5 Tailoring ......................................................................................................................................... 5 Modifications .................................................................................................................................. 6 Intervention adherence ................................................................................................................... 6 Data collection ................................................................................................................................... 8 Vascular ultrasound imaging protocol............................................................................................ 8 Sample size and power calculations .............................................................................................. 10 Multiple imputation procedures .................................................................................................... 11 SUPPLEMENTAL DISCUSSION ....................................................................................................... 12 Acceptability of the intervention programs .................................................................................. 12 Lessons learned ............................................................................................................................... 14 SUPPLEMENTAL TABLES ............................................................................................................... 15 SUPPLEMENTAL FIGURES AND FIGURE LEGENDS .................................................................. 26 HANDOUTS PROVIDED DURING THE ASSESSMENTS ............................................................. 29 SUPPLEMENTAL REFERENCES ..................................................................................................... 35

1

SUPPLEMENTAL METHODS Description of the intervention programs Materials and procedures The adult intervention program was developed to bring a change in the lifestyle and behavior of the participants to promote cardiovascular health and to reduce the risk of cardiovascular disease. There were two types of intervention: Intensive Individual Intervention Program (IIIP), and Peer-to-Peer Program Intervention (PPPI). The IIIP consisted on one-on-one counseling sessions with a trained lifestyle coach, and the PPPI consisted on groups meetings to support each other on their goals achievement with the help of a trained leader. The first session for both the interventions was getting to know the participants and an introduction to the FAMILIA study. For the IIIP group the scientific team of FAMILIA assigned a trained lifestyle coach to each of the participants. They were introduced to the participants at an event where the coaches gave the Lifestyle Coaching Participant Manual to the participants and scheduled their first in-person session. The participants used the Lifestyle Coaching Participant Manual to discuss the session topics with their lifestyle coaches over the phone. After the first in-person session, a personal activity monitoring device (Garmin Vivofit 2 and 3) was given to the participants to motivate and improve their activity level. After the second session meeting of the PPPI, the participants selected the peer leaders and co-leaders who accepted their role on a voluntary basis. The subsequent sessions for both interventions were focused on the different factors of cardiovascular health: healthy diet, weight management, physical activity, stress and behavior change. The lifestyle coaches and peer leaders set the goals with the participants and they encouraged them to achieve their goals at each session. The lifestyle coaches of IIIP also motivated the participants to wear the

2

physical activity device. Details of both intervention materials are provided in the Online Table 1 for IIIP and in the Online Table 2 for PPPI. The FAMILIA research team provided regular support to the lifestyle coaches, the peer leaders and the co-leaders during the course of the intervention. There was a mid-point in-person check-in with the lifestyle coaches and the relevant research team members at Icahn School of Medicine at Mount Sinai. There was also a midpoint check-in dinner meeting with the Principle Investigator of the FAMILIA project, relevant research team members, school directors/leaderships, parent ambassadors and peer leaders and co-leaders. These meetings were held to discuss the adherence of the intervention and to encourage them of the respective interventions to continue with their efforts to keep the participants engaged. A point-based incentive structure was established in which participants in the individual or peer-to-peer intervention earned points based on their attendance/participation in the intervention sessions, educational and informational meetings. Mid-point gifts were provided to the participants based on the number of sessions they completed, and at the end of intervention, the gift cards were provided based on the total number of sessions completed by the participants. Intervention provider The first two meetings of the PPPI were facilitated by a qualified research team member. The 23 selected peer leaders and co-leaders led the subsequent group sessions. These peer leaders and co-leaders received a full day (6 hours long) leadership training provided by the scientific team member of FAMILIA. The co-leaders were trained so they could step-up and carry out the responsibility of a leader if the leader was not available to lead the scheduled session during the intervention. The goals of the leadership training were to (i) Raise the awareness of other members of the importance of learning to listen and encourage mutual support; (ii) Mark the tone of the discussion; (iii) Addressing issues that 3

arise during the meeting; (iv) Monitor the evolution of the lifestyle and behavior changes discussed at the meetings; and (v) Encourage and maintain peer involvement. Each of the 23 trained peer leaders conducted the group meetings. For the IIIP, a total of 10 lifestyle coaches with a degree in nutrition or health sciences were recruited for leading the sessions. The lifestyle coaches were given a full day lifestyle coaching training (6 hours long) at Mount Sinai hospital. The goal of training was to familiarize the lifestyle coaches with: i) FAMILIA Lifestyle Intervention study, and ii) Core concepts of counseling including motivational interviewing strategies to be used in the study. All the sessions for IIIP were conducted by the lifestyle coaches. Parent ambassadors were identified by the school and were parents/caregivers of children that attended the participating preschool. They were the liaisons between the research team, peer leaders, and the lifestyle coaches and they helped in the participant engagement during the adult intervention. Modes of delivery The lifestyle coaches for the IIIP were given a workload of 20-30 participants. These lifestyle coaches first met with the participants at their children’s’ school and scheduled the first in-person session. At the end of their first session the lifestyle coaches scheduled the second session with the participants. The sessions 2 to 8 and the complimentary session were held over the phone. The lifestyle coaches called their participants a week before their scheduled session to remind the participants of their upcoming session. The participants of PPPI were asked to choose a group. There were 14 groups in the study with approximately 17 participants in each group. During the first 2 meetings the groups selected the peer leaders, and co-leaders who accepted on a voluntary basis. The peer group leaders with the help of parent ambassadors, school staff and key research team

4

members scheduled the group meetings. When the group sessions were held by phone, the trained peer leaders conducted the calls. Study setting The in-person group meetings for PPPI were held at the participant’s children’s preschool. The first in-person session with the lifestyle coaches of IIIP took place at the participant’s children’s school, and the subsequent sessions were conducted over the phone. The activity devices for the IIIP were distributed to the participants after they completed first session. A presentation about the working of the device was given to the participants at their children’s school. The leader training sessions for PPPI and lifestyle coaches training for IIIP were held at Icahn School of Mount Sinai, New York. Duration and intensity The participants in the IIIP were scheduled to attend 8 lifestyle coaching sessions during the first 8 months and 4 additional complimentary sessions with their lifestyle coach for duration of approximately 12 months for this intervention. These coaching sessions were held once every 3-4 weeks, lasting approximately for 45 minutes. The participants in the PPPI were scheduled to initially attend 12 group sessions. Each group meeting lasted for approximately 45 minutes. The peer leaders conducted 10 sessions with their group of participants after their training. These 10 group meetings after the leaders training were each held once every 3-4 weeks. Thus, the peer-to-peer intervention was designed with a total of 12 sessions held 3-4 weeks apart and to be completed within 12 months period from the start of the intervention. Tailoring Six out of ten lifestyle coaches of IIIP were bilingual (Spanish/English). These trained lifestyle coaches were assigned based on the language preference of the participants. The Lifestyle Coaching Participant Manual that was given to the participants had English and 5

Spanish versions. The PPPI in-person or phone sessions were also held based on the language of preference (Spanish/English) of the participants. Modifications At midway through the peer-to-peer intervention, it was realized that due to time constraints some peer group participants preferred to continue the meetings over the phone. These participants were offered a selection of scheduled days and times (i.e. 8PM 2nd Thursday of the month) and language preference. Day and time selections were based on feedback received from participants that had not been able to make it to the group sessions. Once the participant identified their preferred session reminder they signed up and a flier with the details of the call was provided. A research team member was present on the calls, announced themselves during the call and provided limited input such as relevant announcements regarding the study (i.e. upcoming assessment events). A conference call service (UberConference) was used to hold the calls. The peer leaders/parent ambassadors made reminder calls/text a few days prior to the scheduled call. In person peer groups sessions continued to be held for those participants that were able to attend. The number of sessions to be completed for peer-to-peer groups was reduced from 10 to 8 sessions to keep consistency with the individual counseling intervention and to accommodate the limitations of time the groups could convene. Intervention adherence The lifestyle coaches of the IIIP kept records of the sessions they completed with the participants. They reported the session updates every week and attended biweekly conference calls with FAMILIA research team members to discuss the progress of the sessions and challenges they faced in scheduling the sessions with the participants. Those participants not attending sessions were categorized as no shows, hard to schedule, phone not working or not interested in continuing with the sessions. With the help of the parent ambassadors the hard to 6

schedule participants and the participants whose phones were not working were contacted at their children’s school during the drop-off or pick-up times. The information about the change in the contact numbers was passed on to the lifestyle coaches. The coaches then contacted the participants and encouraged them to continue with the sessions. The peer leaders of the PPPI kept records of the sessions they completed with the participants. They reported the session updates and discussed the progress of the sessions and challenges they faced in scheduling the sessions with the participants with the FAMILIA research team member every 3-4 weeks. Also, with the help of the parent ambassadors, participants not attending sessions were categorized and contacted using similar strategies as with the IIIP group and were encouraged to continue with their sessions.

7

Data collection Blood pressure was recorded with the subject in a sitting position, resting quietly for five minutes with legs uncrossed by the use of a digital automatic blood pressure monitor (Omron M2). Three blood pressure readings were taken at approximately 1-minute intervals, and the mean used for analysis. Physical activity was evaluated with the self-administered short International Physical Activity Questionnaire (IPAQ) form “last 7 days recall” (1). A stadiometer was used to measure height in centimeters/inches. A balance beam scale calibrated to zero was used to measure weight in kilograms/pounds. Body-mass-index (BMI) was calculated as body weight divided by height squared (kg/m2). Fruits and vegetables consumption was assessed with the use of the visually enhanced Food Behaviors Checklist (2) and the WHO STEPwise approach to non-communicable disease risk factor Surveillance (Step 1: Behavioral Measurements; core questions) (3). The latter was also used to assess smoking habits. Vascular ultrasound imaging protocol Vascular ultrasound imaging was performed with the Philips EPIQ-7G ultrasonography system (Philips Healthcare, Andover, MA, USA). A 9L-D linear array transducer (9-3.1MHz), C1-6 curved array transducer (6-1 MHz) and VL 13-5 linear volume array transducer (13-5 MHz) were used for this study. The imaging protocol was adapted from the one used in the Progression and Early detection of Subclinical Atherosclerosis (PESA) study (4-6). The procedure was performed by experienced registered vascular technologists that completed additional specialized training with the Philips ultrasonography system. The scanning protocol included standard imaging of the left and right carotid artery bifurcation and its branches (internal and external carotid arteries), as well as the left and right common femoral artery bifurcation and its branches (superficial femoral and profunda femoral arteries). These four areas were scanned in cross-section with a 2D linear or curved 8

array transducer to evaluate for the presence of plaques and to summarize the plaque burden. The presence of plaque was defined as a focal protrusion into the arterial lumen of thickness >0.5 mm or >50% of the surrounding intima-media thickness, or a diffuse thickness ≥ 1.5 mm (4,7-9). Longitudinal views were also obtained. The same vessels were evaluated with 3D ultrasound using a VL 13-5 linear array volume transducer, which performs a mechanical automated sweep in cross-section allowing for the assessment of plaque volume and estimation of total atherosclerotic burden. All ultrasound recordings and digital images were analyzed in the Zena and Michael A. Wiener Cardiovascular Institute at Icahn School of Medicine at Mount Sinai. Dedicated software specially engineered by Philips (QStation-VPQ [vascular plaque quantification] 3.5) was used to analyze the 3D images. Each carotid, femoral 2D cross-sectional image was evaluated for the presence of plaque. When a plaque was identified, QStation was used to evaluate the 3D dataset, which was displayed as multiple transverse slices, and summarized the sum of plaque volumes in each vascular bed (carotid or femoral) on each side (left or right), as well as the total plaque volume of all vascular beds (bilateral carotid and femoral). The maximum percent stenosis was also estimated.

9

Sample size and power calculations Details on the original sample size calculation made in the FAMILIA study were previously published (10). Briefly, effect size anticipated was based on mean differences between control and intervention arms actually observed in the Fifty-Fifty study (0.7) for the Fuster-BEWAT score and a common standard deviation of 2.3 (11). Because the cluster design of the study, a range of number of clusters (schools) depending on the cluster size were computed. With a 2:1 randomization, an intraclass correlation of 0.01 and a drop-out rate of 10%, it was estimated that 12-15 clusters with a mean of 40-50 participants each (total n = 600) would need to be recruited. This would provide a power of ~82% to detect significant differences between control and the aggregate intervention group. The study finally enrolled 15 schools of which 9 were randomized to intervention groups (3 schools to the Intensive Individual Intervention Program and 6 schools to Peer-ToPeer Program Intervention) and 6 to the control condition, totaling 635 adults enrolled and assessed at baseline. At primary outcome evaluation, approximately 30% of participants were lost to follow-up or had incomplete data. Therefore, the power of the study to detect a 0.7 point-difference in the Fuster-BEWAT scores between control and the aggregate intervention group using a complete-case intention-to-treat analysis approach (15 clusters, n = 446 participants) was ~73%.

10

Multiple imputation procedures Multiple imputation using multivariate normal distribution (Markov Chain Monte Carlo procedures), implemented by the command mi, was performed to include all randomized enrolled participants as sensitivity analysis. The change from baseline in the overall Fuster-BEWAT score (FBS) of the participant was set as imputed dependent variable. Complete cases represented ~70% (446 and 424 participants, out of 635 randomized enrolled participants, for the primary outcome and sustainability, respectively). Missing data were assumed to be missing at random. The following variables were included as auxiliary variables: baseline FBS (continuous variable), age (continuous variable), sex (binary variable), race/ethnicity (categorical variable), and treatment (categorical variable). Complete covariates information was available for most randomized enrolled participants (n = 610, 96%); and auxiliary variables were imputed as needed. Number of imputations was set at 50. A ridge prior distribution with 10 degrees of freedom was used to stabilize inferences (12). A random-number seed was set to ensure reproducibility of the imputed values. Estimations on the imputed data were run with the “mi estimate” command, which adjusts coefficients and standard errors for the variability between imputations according to the combination rules by Rubin (13). Multilevel linear mixed-effects models that account for the hierarchical cluster randomized design were used to test for the adjusted intervention effect. Fixed effects were the corresponding baseline score (as a continuous variable), and treatment group. Schools and families were handled as random effects. Diagnostic checks of the imputation model were obtained using the vartable and dftable options of the “mi estimate” command. All analyses were performed using STATA version 15 (StataCorp, College Station, Texas).

11

SUPPLEMENTAL DISCUSSION Acceptability of the intervention programs We assessed the acceptability of the intervention programs by conducting focus group interviews of the PPPI participants, Peer Leaders, IIIP participants, and parent ambassadors. Overall, the participants felt very positive about the program as a whole, but they expressed a desire for more personal interactions with health experts. Some participants expressed initially being suspicious of the overall motive of research-based projects but slowly they recognized the benefit of a free family-based health program. Their initial uncertainty abated and evolved into interest in the project. Many participants liked the health assessments and medical counselling where they could talk to medical professionals about their test results. The participants expressed an interest in learning more about health issues that related to diet and nutrition and wanted a monthly in-person lectures by a physician to help alleviate some of their concerns about their health, and to provide information such as what to do when they go to the hospital etc. Participants also identified barriers, such as having competing priorities, which compromised their ability to participate in the project. PPPI participants appreciated the FAMILIA program as it gave them an opportunity to meet and talk to other parents and to work together, support and encourage each other to acquire healthy eating habits, and learn few tricks about getting their children to eat more vegetables. They felt that group sessions for the parents and the FAMILIA program for their children at the school helped them and their children to improve their habits and knowledge about health. However, they expressed for more structure in the overall program with a focus on better communication and more hands on activities about how to make changes in their habits and behaviours. They preferred more showing and doing rather than telling and implicitly judging 12

Peer Leaders recognized that FAMILIA program helped many participants become more aware of their health and ways to improve it, from exercise to better eating habits. They were satisfied in the way the peer group meetings were held and in the overall organization of the project. However, the peer leaders suggested less paperwork at the onset of the program, better communication and fewer reminder calls to the participants, and incorporating cooking and exercise workshops for the participants. IIIP participants were grateful to their lifestyle coaches as they helped them to make better nutrition choices. They liked having a session with the coach once a month. They felt that setting up goals and challenges with their coach at the session and following up on these goals at their subsequent sessions kept them in check to eat well and exercise. They acknowledged that they would have had a different experience if the program consisted only of meetings. Overall, IIIP participants felt very positively about the program, but they also expressed a desire for more personal interactions with health experts. Parent Ambassadors were a very enthusiastic and engaged group. They described themselves as “the voice” of FAMILIA. They liked the family focussed, research-based and knowledge sharing FAMILIA program. In their role to engage the participants, they learned about the initial barriers faced by the participants and their hesitation to join the program. They also realized how stressed and closed off many of the families were, and they worked with these families by sharing their own experiences. They encouraged the participants to join the program by explaining to them the benefits the program. Once they become engaged, they began to feel the benefit of opening up, gaining knowledge and group support. They also helped the FAMLIA research team to schedule the sessions and accompanied the participants to their meetings. The parent ambassadors suggested that addressing stress and mental health, building trust, giving a listening ear to the participants’ stories so they could open up and be

13

more receptive to making changing in their lives and incorporating yoga and exercise classes would be more helpful. Finally, even though there were many suggestions made by the participants to improve the FAMILIA program, several participants expressed an intention to continue practicing what they learned after the conclusion of the program. Lessons learned The study population was a combination of parents/caregivers (75%) and school staff (25%). The rationale for including school staff in the study was twofold. During school hours, they have the opportunity to influence children through the provision of family style meals, age appropriate health education, physical activity opportunities, social emotional development and through interactions with the children’s parents/caregivers. Furthermore, in the pilot phase of the study, the inclusion of school staff in the study was seen as a positive aspect by school directors and staff to engage and motivate school personnel over the course of the trial. The main reason for lost to follow-up was due to children aging out of the preschool program and going on to elementary school; as the children aged out it became increasingly challenging to remain in contact and engage their parents/caregivers. Additionally, participants perceived that the program involved too much commitment as well as having competing priorities for themselves and their children. Lessons learned included the need for improvement of our ability to remain in contact with the families as children aged out of their participating preschool. In this regard, we developed a communication system that included multiple and tailored modalities to reduce lost to follow up as well as additional strategies (rewards and incentives, special child friendly events on weekends, centralized clinic locations and working with a community event planner).

14

SUPPLEMENTAL TABLES Online Table 1. Intensive Individual Intervention Program (IIIP) – Session Summary Session

1

2

3

Content

Resources

Getting to know you • Introduction to Project FAMILIA. • Explain Lifestyle Intervention Program. • Discuss Heart health and Heart disease – refer to Heart Health and Heart Disease know the facts handouts. • Assess the -readiness and confidence to change overall lifestyle and behaviour of the participant. • Set health goals for the sessions and explain the purpose of goal setting for the sessions. • Ask the participants to note their diet 24 hours before their next session. • Schedule the next phone session. Eating for health - Part A • Explain the importance of healthy eating. • Discuss Heart health. • Follow up on the goals and discuss the 24-hour diet recall. • Ask the participant to record their diet and their mood (food-mood) before their next session. • Give recommendations for healthy eating. • Discuss the handouts. • Ask the participants to take down notes for the session. • Explain the usage of the physical activity device (Garmin Vivofit 2). • Set goals for the next session. • Schedule the next phone session. Eating for health - Part B • Check-in on the goals that were set up in the previous session. • Discuss the last few weeks’ diet the participant recorded after the previous session. • Explain the benefits of reducing sugary drinks, salt and sodium from the diet. Discuss the handouts. • Ask the participants to take down notes for the session. • Give recommendations for making additional changes to the diet. • Ask the participant to record their diet and their mood (food-mood) before their next session. • Follow up on the usage of physical activity device. • Set goals for the next session. • Schedule the next phone session.

15

•

•

•

•

• •

• •

Heart Health https://trustedpartner.azureedge.net/docs/libra ry/AreaAgningOnAgency2012/Heart%20Healt h.pdf Heart Disease https://www.cdc.gov/heartdisease/docs/Consu merEd_HeartDisease.pdf

Irrational Ideas About Eating https://www.move.va.gov/docs/NewHandouts/ BehavioralHealth/B17_IrrationalIdeasAboutEa ting.pdf Build a Healthy Meal https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ DGTipsheet7BuildaHealthyMeal_0.pdf Healthy Eating for an Active Lifestyle https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ Physical activity device information https://static.garmin.com/pumac/vivofit2_OM_ EN.pdf Salt and Sodium https://www.healthit.gov/sites/default/files/reso urces/dgtipsheet14saltandsodium.pdf Make Better Beverage Choices https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ DGTipsheet19MakeBetterBeverageChoices.p df

4

5

6

7

Weight management • Check-in on the goals that were set up in the previous session. • Discuss the last few weeks’ diet the participant recorded after the last session. • Explain the importance of weight management. Discuss the weight management handout. • Ask the participants to take down notes for the session. • Give recommendations for losing weight and maintaining a healthy weight. • Ask the participant to record their diet and their mood (food-mood) before their next session, and write a self-reflection diary. • Follow up on the usage of physical activity device. • Set goals for the next session. • Schedule the next phone session. Moving for fun! • Check-in on the goals that were set up in the previous session. • Discuss the last few weeks’ diet the participant recorded after the last session. • Check-in on the self-reflection diary. • Ask the participants to take down notes for the session. Give recommendations for being more physical active. • Refer to the session handouts for additional suggestions. • Follow up on the usage of physical activity device. • Set goals for the next session. • Schedule the next phone session. Habits and behaviors • Check-in on the goals that were set up in the previous session. • Discuss the last few weeks’ diet the participant recorded after the previous session. • Ask the participant to list the habits that they want to work on and note down their progress on changing the specific habit or habits (e.g. eating junk food, alcohol, smoking etc.). • Ask the participants to take down notes for the session. • Give recommendations for changing the habits. • Discuss the session handouts. • Follow up on the usage of physical activity device. • Set goals for the next session. • Schedule the next phone session. De-stressing your life • Check-in on the goals that were set up in the previous session. • Understand the participant’s stress level and explain the effects of physical and mental stress on health. • Give recommendations for managing stress and discuss different ways to destress (e.g. breathing exercises) • Ask the participant to note down the techniques they used for destressing before their next session. • Follow up on the usage of physical activity device.

16

•

• •

•

•

•

• •

Getting Ready To Lose Some Weight https://www.move.va.gov/docs/NewHandouts/ BehavioralHealth/B04_GettingReadyToLoseS omeWeight.pdf The Basics Of Weight Control https://www.move.va.gov/docs/NewHandouts/ Standard/S01_TheBasicsOfWeightControl.pdf Walk https://www.move.va.gov/docs/NewHandouts/ PhysicalActivity/P15_Walk.pdf

Be Active Adults https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ DGTipsheet30BeActiveAdults.pdf Benefits of Regular Physical activity https://www.move.va.gov/docs/NewHandouts/ PhysicalActivity/P02_BenefitsOfRegularExerci se.pdf Barriers to Physical Activity https://www.move.va.gov/docs/NewHandouts/ PhysicalActivity/P01_BarriersToPhysicalActivi ty.pdf

Change Your Behavior https://www.move.va.gov/docs/NewHandouts/ Standard/S09_ChangeYourBehavior.pdf Changing My Eating Habits https://www.move.va.gov/docs/NewHandouts/ Standard/S04_ChangingMyEatingHabits.pdf

8

9*

• Set goals for the next session. • Schedule the next phone session. Beyond FAMILIA-Continuing the change • Recap of all the sessions (1-7). • Discuss the session handouts • Follow up on the usage of physical activity device. • Discuss life Beyond FAMILIA. Offer complimentary sessions to the participant. • Schedule the next phone session (if the participant wishes to continue with the sessions).

Healthy cooking, smart shopping on a budget and reading labels • Welcome the participant to the complimentary session. • Explain the significance of healthy cooking, smart shopping and reading labels. • Ask the participant to keep a weekly smart shopping record and make a note of healthy and unhealthy foods they shop weekly at the grocery store. • Recommend healthy recipes • Give tips on shopping on a budget and on reading nutrition labels. Discuss the handouts • Follow up on the usage of physical activity device. • Schedule the next phone session (if the participant wishes to continue with the sessions).

Eating as a family, screen time, being active as a family, media • Check-in on the goals that were set up in the previous session. • Follow up on Weekly Smart Shopping Record 10* • Discuss using the handouts on reducing screen time, eating as a family, and being active as a family. • Follow up on the usage of physical activity device. • Schedule the next phone session (if the participant wishes to continue with the sessions). Individual choice • Discuss the topic that the participant has chosen for this session 11* • Follow up on the usage of physical activity device. • Schedule the next phone session (if the participant wishes to continue with the sessions). Individual choice • Discuss the topic that the participant has chosen for 12* this session • Recap of all the sessions. Note: * Complimentary Sessions

17

• • •

•

•

•

•

• •

U R What U Eat – We Can! https://www.nhlbi.nih.gov/health/educational/w ecan/downloads/urwhateat.pdf Keep Your Health https://www.move.va.gov/docs/NewHandouts/ Miscellaneous/M08_KeepYourHealth.pdf Take Control Of Your Thoughts Feeling And Behavior https://www.move.va.gov/docs/NewHandouts/ BehavioralHealth/B29_TakeControlOfYourTh oughtsFeelingsAndBehavior.pdf Smart Shopping for Vegetables and Fruits https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ DGTipsheet9SmartShopping.pdf Eating Better On A Budget https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ DGTipsheet16EatingBetterOnABudget.pdf Understanding Food Nutrition Labels https://www.accessdata.fda.gov/scripts/Intera ctiveNutritionFactsLabel/factsheets/Whats_O n_The_Nutrition_Facts_Label.pdf

Be a Healthy Role Model https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ DGTipsheet12BeAHealthyRoleModel.pdf Making Healthier Food Choices (We Can!) https://www.nhlbi.nih.gov/health/educational/w ecan/downloads/tip_choices.pdf Screen Time Chart (We Can!) https://www.nhlbi.nih.gov/health/educational/w ecan/downloads/screen-time-log.pdf

Online Table 2. Peer-to-Peer Program Intervention (PPPI) – Session Summary Session

1

2

3

Content

Resources

Welcome to FAMILIA • Introduction • Peer to peer groups • Topic guide • Leaders/co-leaders feedback form • Monthly reflection/action form • Sign-in sheet • Group rules Group leadership • Group dynamics • Leader characteristics • Leader selection • Roles within groups

•

Community health workers https://www.nhlbi.nih.gov/files/docs/resources/ heart/lat_mnl_en.pdf

•

Leadership skills https://eclkc.ohs.acf.hhs.gov/publication/enha nce-parents-advocacy-leadership-skills-guidehead-start-early-head-start-staff Effective Communication https://extension.umaine.edu/publications/610 3e/

Eating for health – Part A • Heart healthy foods • Sodium and fat • Food label from a food commonly used by group members • Behavior change contract

•

Eating for health – Part B • Balanced meals • Sugar in foods

•

•

4

•

•

Habits and behavior • Habit/behavior formation • Behavior/habit change

•

5

6

7

8

•

Healthy Eating http://www.heart.org/HEARTORG/GettingHeal thy/NutritionCenter/NutritionCenter_UCM_001188_SubHomePage.jsp Salt and Sodium https://www.healthit.gov/sites/default/files/reso urces/dgtipsheet14saltandsodium.pdf Build a Healthy Meal https://choosemyplateprod.azureedge.net/sites/default/files/tentips/ DGTipsheet7BuildaHealthyMeal_0.pdf Cut Down on Added Sugars https://health.gov/dietaryguidelines/2015/reso urces/DGA_Cut-Down-On-Added-Sugars.pdf Motivation: the drive to change https://www.psychologytoday.com/us/basics/ motivation Drugs and addiction https://easyread.drugabuse.gov/content/preve ntion

Weight management • Weight gain- why and how • Maintaining a healthy weight

•

Weight Management https://health.gov/dietaryguidelines/dga2005/d ocument/html/chapter3.htm

Moving for fun! • Benefits of physical activity • Types of exercise • Movement in daily life • Be an active family • Barriers to physical activity

•

Active families https://www.nhlbi.nih.gov/health/educational/w ecan/get-active/index.htm http://www.nhlbi.nih.gov/health/educational/w ecan/

De-stressing your life • Stress in our lives • Coping with stress • Dealing with stress, anxiety and depression

•

Managing Stress http://www.helpguide.org/articles/stress/stress -management.htm http://www.webmd.com/balance/guide/tips-to-

18

control-stress •

9

Food shopping and cooking • Food shopping tips • Healthy cooking tips • Smart shopping for vegetables and fruits • Understanding food nutrition labels • Eating better on a budget

How to understand and use the nutrition facts label https://www.fda.gov/food/nutrition-educationresources-materials/guide-older-adults-usingnutrition-facts-label

•

10

The media • The media and our health • Thinking critically about media

Quality entertainment and tech https://www.commonsensemedia.org

Eating as a family • Benefits of eating together • “How to” of family meals • Importance of the Family Meal

•

Tips to improve family dinners http://thefamilydinnerproject.org

Beyond FAMILIA • What was learned? • Moving forward • Certificate of participation • Keep your health • Take control of your thoughts, feelings and behavior

•

Balance thoughts for long term maintenance http://www.cdc.gov/diabetes/prevention/pdf/po stcurriculum_session9.pdf

11

12

19

Online Table 3. The TIDieR (Template for Intervention Description and Replication) Checklist. Item

Item

Where located

number

Primary paper

Other (details)

(page or appendix number)

1.

BRIEF NAME Provide the name or a phrase that describes the intervention.

Page 6

https://clinicaltrials.gov/ct 2/show/NCT02481401

WHY 2.

Describe any rationale, theory, or goal of the elements essential to the intervention.

Pages 5 and 6

WHAT 3.

Materials: Describe any physical or informational materials used in the intervention, including those provided to participants

Pages 5 to 9

Online Table 1-2,

or used in intervention delivery or in training of intervention providers. Provide information on where the materials can be

Supplemental

accessed (e.g. online appendix, URL).

pages 2 and 3, Handouts

4.

Procedures: Describe each of the procedures, activities, and/or processes used in the intervention, including any enabling

Pages 7 and 8

or support activities.

Supplemental pages 2 and 3

WHO PROVIDED 5.

For each category of intervention provider (e.g. psychologist, nursing assistant), describe their expertise, background and any specific training given.

Pages 7 and 8

Supplemental pages 3 and 4

HOW 20

6.

Describe the modes of delivery (e.g. face-to-face or by some other mechanism, such as internet or telephone) of the

Pages 7 and 8

intervention and whether it was provided individually or in a group.

Supplemental pages 4 and 5

WHERE 7.

Describe the type(s) of location(s) where the intervention occurred, including any necessary infrastructure or relevant

Pages 7 and 8

features.

Supplemental page 5

WHEN and HOW MUCH 8.

Describe the number of times the intervention was delivered and over what period of time including the number of

Pages 7 and 8

sessions, their schedule, and their duration, intensity or dose.

Supplemental page 5

TAILORING 9.

If the intervention was planned to be personalised, titrated or adapted, then describe what, why, when, and how.

Supplemental pages 5 and 6

MODIFICATIONS 10.

If the intervention was modified during the course of the study, describe the changes (what, why, when, and how).

Supplemental page 6

HOW WELL 11. 12.

Planned: If intervention adherence or fidelity was assessed, describe how and by whom, and if any strategies were used to

Supplemental

Supplemental page

maintain or improve fidelity, describe them.

pages 6 and 7

14

Actual: If intervention adherence or fidelity was assessed, describe the extent to which the intervention was delivered as planned.

21

Pages 14 and 15

Online Table 4. Baseline characteristics of adults included and excluded from the primary analysis (complete-case intention-to-treat analysis) in the FAMILIA study. Included in the analysis Overall

Control

Excluded from the analysis

Intervention

Intervention

(IIIP-

(PPPI-Peer)

Overall

Control

Individual)

Intervention

Intervention

(IIIP-

(PPPI-Peer)

Individual)

Adults No. of adults

446

119

157

170

189

63

60

66

Age, years

38.8 ± 11.6

40.3 ± 11.9

38.5 ± 11.6

38.1 ± 11.3

35.3 ± 10.1

36.1 ± 9.6

35.7 ± 10.7

34.3 ± 9.9

Female

378 (84.8)

102 (85.7)

129 (82.2)

147 (86.5)

146 (77.3)

50 (79.4)

46 (76.7)

50 (75.8)

Non-Hispanic black

136 (30.5)

42 (35.3)

30 (19.1)

64 (37.7)

61 (32.3)

28 (44.4)

9 (15.0)

24 (36.4)

Hispanic/Latino

263 (59.0)

58 (48.7)

120 (76.4)

85 (50.0)

96 (50.8)

22 (34.9)

43 (71.7)

31 (47.0)

Other/multiracial

47 (10.5)

19 (16.0)

7 (4.5)

21 (12.4)

32 (16.9)

13 (20.7)

8 (13.3)

11 (16.7)

Born outside United States

218 (48.9)

58 (48.7)

95 (60.5)

65 (38.2)

87 (46.0)

25 (39.7)

33 (55.0)

29 (43.9)

School staff

127 (28.5)

32 (26.9)

42 (26.8)

53 (31.2)

32 (16.9)

12 (19.1)

10 (16.7)

10 (15.2)

<$25,000

202 (45.3)

56 (47.1)

81 (51.6)

65 (38.2)

79 (41.8)

22 (34.9)

25 (41.7)

32 (48.5)

$25,000 - $50,000

125 (28.0)

37 (31.1)

33 (21.0)

55 (32.4)

39 (20.6)

14 (22.2)

12 (20.0)

13 (19.7)

>$50,000

48 (10.8)

12 (10.1)

11 (7.0)

25 (14.7)

15 (7.9)

8 (12.7)

2 (3.3)

5 (7.6)

Unknown

71 (15.9)

14 (11.8)

32 (20.4)

25 (14.7)

56 (29.6)

19 (30.2)

21 (35.0)

16 (24.2)

Race/ethnicity

Annual household income

Highest level of education completed 22

Secondary school or lower

130 (29.2)

27 (22.7)

61 (38.9)

42 (24.7)

53 (28.0)

14 (22.2)

24 (40.0)

15 (22.7)

High school

207 (46.4)

60 (50.4)

68 (43.3)

79 (46.5)

92 (48.7)

25 (39.7)

28 (46.7)

39 (59.1)

College/University/Postgraduate

104 (23.3)

29 (24.4)

28 (17.8)

47 (27.7)

36 (19.1)

20 (31.8)

6 (10.0)

10 (15.2)

Unknown

5 (1.1)

3 (2.5)

0 (0.0)

2 (1.2)

8 (4.2)

4 (6.4)

2 (3.3)

2 (3.0)

FBS overall (range 0-15)

9.2 ± 2.5

9.3 ± 2.6

9.3 ± 2.3

9.1 ± 2.4

9.4 ± 2.4

9.4 ± 2.7

9.7 ± 2.2

9.3 ± 2.4

B- Blood pressure (range 0-3)

2.2 ± 1.1

2.2 ± 1.1

2.2 ± 1.0

2.1 ± 1.1

2.3 ± 1.0

2.1 ± 1.1

2.4 ± 0.8

2.4 ± 0.9

E- Exercise (range 0-3)

1.9 ± 1.2

1.9 ± 1.2

1.9 ± 1.2

1.9 ± 1.1

2.0 ± 1.1

2.1 ± 1.2

2.1 ± 1.1

1.9 ± 1.2

W- Weight (range 0-3)

0.9 ± 1.1

1.0 ± 1.2

0.9 ± 1.0

0.9 ± 1.1

1.1 ± 1.2

1.3 ± 1.3

1.0 ± 1.1

1.1 ± 1.2

A- Alimentation (range 0-3)

1.4 ± 0.7

1.4 ± 0.8

1.5 ± 0.7

1.4 ± 0.7

1.4 ± 0.8

1.4 ± 0.8

1.5 ± 0.8

1.4 ± 0.8

T- Tobacco (range 0-3)

2.8 ± 0.6

2.8 ± 0.6

2.8 ± 0.6

2.8 ± 0.6

2.6 ± 0.8

2.5 ± 0.9

2.7 ± 0.7

2.6 ± 0.8

Absence of atherosclerosis

367 (82.3)

94 (79.0)

143 (91.1)

130 (76.5)

101 (53.4)

29 (46.0)

35 (58.3)

37 (56.1)

Evidence of atherosclerosis

37 (8.3)

16 (13.5)

9 (5.7)

12 (7.1)

10 (5.3)

3 (4.8)

2 (3.3)

5 (7.6)

Not performed/not analyzable

42 (9.4)

9 (7.6)

5 (3.2)

28 (16.5)

78 (41.3)

31 (49.2)

23 (38.3)

24 (36.4)

Fuster-BEWAT score

Vascular ultrasound

Values are mean ± standard deviation, or n (%). Adults’ race/ethnicity was self-identified. FBS: Fuster-BEWAT score.

23

Online Table 5. Change from baseline at 12-month assessment in overall Fuster-BEWAT score by adherence to the intervention among parents/caregivers in the FAMILIA trial. Within group differences Low Adherence

High Adherence

Between group differences Difference (95% CI) p-value High vs. Low

IIIP- Individual

-0.13 (-1.02 to 0.76)

0.15 (-0.34 to 0.64)

0.28 (-0.61 to 1.16)

0.539

PPPI- Peer

0.09 (-0.37 to 0.55)

0.38 (0.05 to 0.72)

0.29 (-0.28 to 0.87)

0.317

Data are presented as marginal mean differences and 95% confidence intervals (CI) as derived from marginal means after linear mixed-effects models. Fixed effects were the corresponding baseline score and the categorized adherence to the intervention: completion of less than 50% of the program sessions (low adherence group) vs. 50% or more (high adherence group) of sessions. Schools and families were handled as random effects.

24

Online Table 6. Change from baseline at 24-month assessment in overall Fuster-BEWAT score by adherence to the intervention among parents/caregivers in the FAMILIA trial.

Within group differences Low Adherence

High Adherence

Between group differences Difference (95% CI) p-value High vs. Low

IIIP- Individual

-0.26 (-0.96 to 0.43)

0.27 (-0.07 to 0.62)

0.53 (-0.24 to 1.30)

0.173

PPPI- Peer

-0.43 (-0.95 to 0.08)

0.00 (-0.37 to 0.36)

0.43 (-0.20 to 1.06)

0.317

Data are presented as marginal mean differences and 95% confidence intervals (CI) as derived from marginal means after linear mixed-effects models. Fixed effects were the corresponding baseline score and the categorized adherence to the intervention: completion of less than 50% of the program sessions (low adherence group) vs. 50% or more (high adherence group) of sessions. Schools and families were handled as random effects.

25

SUPPLEMENTAL FIGURES AND FIGURE LEGENDS Online Figure 1. Sustainability (24-month) FAMILIA study flow diagram. Recruitment of schools and parents/caregivers, and completeness of baseline and 24-month follow-up measures, according to the guidelines in the CONSORT 2010 statement for the reporting of cluster randomized trials (14).

26

Online Figure 2. Subgroup analysis of the change in overall FBS at 12 months (intervention groups). Forest plot representing mean differences (95% CI) in the overall Fuster-BEWAT Score (FBS) changes at 12-month assessment in parents/caregivers allocated to the intervention groups, after stratified linear mixed-effects models by selected variables. Same considerations as for Figure 2.

27

Online Figure 3. Subgroup analysis of the change in overall FBS at 24 months (intervention groups). Forest plot representing mean differences (95% CI) in the overall Fuster-BEWAT Score (FBS) changes at 24-month assessment in parents/caregivers allocated to the intervention groups, after stratified linear mixed-effects models by selected variables. Same considerations as for Figure 3.

28

HANDOUTS PROVIDED DURING THE ASSESSMENTS

29

30

31

32

33

34

SUPPLEMENTAL REFERENCES 1.

Craig CL, Marshall AL, Sjostrom M et al. International physical activity

questionnaire: 12-country reliability and validity. Medicine and science in sports and exercise 2003;35:1381-95. 2.

Banna JC, Vera Becerra LE, Kaiser LL, Townsend MS. Using qualitative methods to

improve questionnaires for Spanish speakers: assessing face validity of a food behavior checklist. J Am Diet Assoc 2010;110:80-90. 3.

Riley L, Guthold R, Cowan M et al. The World Health Organization STEPwise

Approach to Noncommunicable Disease Risk-Factor Surveillance: Methods, Challenges, and Opportunities. Am J Public Health 2016;106:74-8. 4.

Fernandez-Friera L, Penalvo JL, Fernandez-Ortiz A et al. Prevalence, Vascular

Distribution, and Multiterritorial Extent of Subclinical Atherosclerosis in a Middle-Aged Cohort: The PESA (Progression of Early Subclinical Atherosclerosis) Study. Circulation 2015;131:2104-13. 5.

Fernandez-Ortiz A, Jimenez-Borreguero LJ, Penalvo JL et al. The Progression and

Early detection of Subclinical Atherosclerosis (PESA) study: rationale and design. Am Heart J 2013;166:990-8. 6.

Lopez-Melgar B, Fernandez-Friera L, Oliva B et al. Subclinical Atherosclerosis

Burden by 3D Ultrasound in Mid-Life: The PESA Study. J Am Coll Cardiol 2017;70:301313. 7.

Baber U, Mehran R, Sartori S et al. Prevalence, impact, and predictive value of

detecting subclinical coronary and carotid atherosclerosis in asymptomatic adults: the BioImage study. J Am Coll Cardiol 2015;65:1065-74. 8.

Stein JH, Korcarz CE, Hurst RT et al. Use of carotid ultrasound to identify subclinical

vascular disease and evaluate cardiovascular disease risk: a consensus statement from the 35

American Society of Echocardiography Carotid Intima-Media Thickness Task Force. Endorsed by the Society for Vascular Medicine. J Am Soc Echocardiogr 2008;21:93-111; quiz 189-90. 9.

Touboul PJ, Hennerici MG, Meairs S et al. Mannheim carotid intima-media thickness

and plaque consensus (2004-2006-2011). An update on behalf of the advisory board of the 3rd, 4th and 5th watching the risk symposia, at the 13th, 15th and 20th European Stroke Conferences, Mannheim, Germany, 2004, Brussels, Belgium, 2006, and Hamburg, Germany, 2011. Cerebrovasc Dis 2012;34:290-6. 10.

Bansilal S, Vedanthan R, Kovacic JC et al. Rationale and Design of Family-Based

Approach in a Minority Community Integrating Systems-Biology for Promotion of Health (FAMILIA). Am Heart J 2017;187:170-181. 11.

Gomez-Pardo E, Fernandez-Alvira JM, Vilanova M et al. A Comprehensive Lifestyle

Peer Group-Based Intervention on Cardiovascular Risk Factors: The Randomized Controlled Fifty-Fifty Program. J Am Coll Cardiol 2016;67:476-85. 12.

Schafer JL. Analysis of Incomplete Multivariate Data. Boca Raton, FL: CRC Press,

1997. 13.

Rubin DB. Multiple Imputation for Nonresponse in Surveys. New York: Wiley, 1987.

14.

Campbell MK, Piaggio G, Elbourne DR, Altman DG, Group C. Consort 2010

statement: extension to cluster randomised trials. BMJ 2012;345:e5661.

36