Background and design of the empowering cardiovascular health in grandfamilies (ECHG) study: A mixed methods study research protocol

Abstract

A steady increase in the number of grandparent caregivers - accelerated by recent trends such as children orphaned by the COVID-19 pandemic or the opioid crisis - magnifies the importance of addressing cardiovascular disease (CVD) risk in that already at high-risk group. However, few efforts have directly addressed the specific cardiac health needs of grandparent primary caregivers (or custodial grandparents). In this paper, we describe the research protocol for the Rural Caregiver Heart Health Education (RICHH) intervention for implementation with grandparent caregivers (RICHH-GP) study funded by the National Institute of Heart, Lung, and Blood. This study will tailor an existing CVD risk reduction intervention (RICHH) for implementation with grandparent primary caregivers and then test the modified design for feasibility, acceptability, and preliminary efficacy. The expected outcome of RICHH-GP is to establish the groundwork for an expanded future clinical trial in which we will test effectiveness with a longer follow-up and with increased inclusivity of participants from marginalized populations. The specific aims are: 1) modify the RICHH intervention to target custodial grandparents from a variety of backgrounds; 2) employ the tailored intervention and assess its feasibility and acceptability via a 2-arm randomized controlled trial; 3) test the preliminary efficacy of RICHH-GP based on measures of CVD risk (physical activity, dietary intake, nicotine exposure, sleep, body mass index, lipids, blood pressure, and HbA1c), self-management behaviors, and depressive symptoms at 4- and 6-months compared to baseline and; 4) (exploratory) examine the proximal impact of the RICHH-GP intervention on CVD risk and depressive symptoms on a subsample of grandchildren of the grandparent participants. The ultimate objectives of this work are to prevent CVD-related morbidity and mortality and overall physical decline, improve psychological well-being among grandparent caregivers, and foster a more heart-healthy environment in grandfamilies.

What is already known.

• •The number of grandparent primary caregivers is increasing, and this population is at elevated risk for poor cardiovascular and psychological health.
• •To date there has been little effort to address the specific heart health needs of this under investigated population (grandparent caregivers and their grandchildren).

What this paper adds.

• •In this study we present a tailored heart health intervention for grandparent primary caregivers, focusing on feasibility, acceptability, and preliminary efficacy.
• •We use a mixed methods design to inform intervention refinement and preparation for a larger-scale clinical trial.

1. Introduction

1.1. Custodial grandparents and their key challenges

As of 2024, approximately 2.12 million Americans were primary caregivers for their grandchildren, (U.S. Census Bureau 2026) with a steady increase in recent decades (Paw Research Center 2026). Many have gained custody of grandchildren due to domestic violence in the parents’ homes (Boserup et al., Dec 2020, Piquero et al., 2021), parents’ incarceration, (Sawyer and Bertram, 2026) or parents’ death due to the COVID-19 pandemic (Hillis et al., 2022, Global orphanhood estimates real time calculator 2026) or the opioid crisis (Centers for Disease Control and Prevention 2026). These grandparents face ongoing stressors as they navigate legal custody arrangements, cope with the loss of their own child(ren), (Sampson and Hertlein, 2015) or face social isolation (Whitley et al., 2001). They are also negatively affected by a combination of social determinants of health (SDoH) such as lower levels of education (Paw Research Center 2026), higher rates of poverty (Wu, 2018) and higher rates of single-caregiver households (U.S. Census Bureau 2026, Provenzano et al., 2021). Additionally, custodial grandparents appear to have higher rates of adverse childhood experiences (ACEs) when compared to other caregivers and the general US population (Smith et al., 2022, Song et al., 2024). Finally, custodial grandparents often prioritize well-being of their dependent grandchildren over their own health (Lee and Jang, 2019). This unique mixture of stressors negatively affects grandparent caregivers’ health when compared to their non-caregiving peers, as evidenced by higher rates of depression (Musil et al., 2017) and health decline, (Musil et al., 2011) and most directly relevant to the proposed study, a higher risk of cardiovascular disease (CVD) (Lee et al., 2003, Smalls et al., 2020).

1.2. Importance of addressing CVD risk in custodial grandparents

Addressing custodial grandparents’ CVD risk is a critical public health concern. Reducing CVD risk and development of CVD promotes healthy aging. To date, interventions have focused primarily on addressing custodial grandparents’ depressive symptoms (Smith et al., 2018) (highly prevalent among custodial grandparents (Lee and Jang, 2019, Musil et al., 2017)), overall self-care (Fruhauf et al., 2022), health (Hayslip and Fruhauf, 2019), resourcefulness (Zauszniewski et al., 2014, Musil et al., 2015), and parenting skills (Smith et al., 2018, Kirby and Sanders, 2014). Investigators have not explored CVD risk in this group, even though CVD and depressive symptoms are often linked (Rafiei et al., 2023, De Hert et al., 2018). In the absence of increased understanding of their cardiovascular (CV) health, addressing their CVD risk and related health decline will remain unlikely.

1.2.1. Custodial grandparents’ CVH has intergenerational implications for child health

Critically, custodial grandparents’ CV health has an intergenerational impact. Minimizing CVD risk in custodial grandparents further minimizes disruption in their grandchildren’s lives (2.6 million children in US live in grandfamilies (The Annie E. Casey Foundation 2026)). Loss or disability of a grandparent from CVD can leave grandchildren without a caregiver, setting the stage for childhood trauma. There is an established positive association between primary caregivers’ CV health behaviors and the CV health of children under their care (Muchira et al., 2020, Ma et al., 2022) as well as between the CV health scores of custodial grandparents and those of their grandchildren (Song et al., 2024). Similarly, caregiver depression increases the risk for depression among children under their care (Dachew et al., 2023, Chithiramohan and Eslick, 2023), and depressive symptoms are elevated among grandparent caregivers (Musil et al., 2011, Whitley et al., 2016). This underscores the importance viewing the health and well-being children living in grandfamilies as interconnected with addressing depressive symptoms in custodial grandparents. These considerations make it clear that there is a critical need to identify interventions that are accessible and address CVD and mental health risk among custodial grandparents and their grandchildren.

1.3. Improving CVD risk of custodial grandparents – using the modified RICHH intervention

The Rural Caregiver Heart Health Education [RICHH] is an evidence-based, educational-behavioral and counseling intervention that targets adult caregivers of adult family members with a chronic illness, designed by Moser et al (Moser et al., 2023, Smith et al., 2022) (a co-investigator of this study). The goal of the intervention is to promote adult caregivers’ self-care and enhancing their skills and motivation to engage in CVD risk reduction (Moser et al., 2023, Smith et al., 2022). Moser’s team has previously demonstrated that the RICHH CVD risk reduction intervention is effective in reducing CVD risk. The intervention alleviates depressive symptoms and promotes engagement in self-management behaviors, both of which are linked to CVD risk in caregivers (Haley et al., 2010, Xu et al., 2020, Capistrant et al., 2012) and the general population (Rafiei et al., 2023, De Hert et al., 2018) (Fig. 1). Considering the disproportionately high prevalence of depressive symptoms among custodial grandparents when compared to their non-caregiving peers, (Musil et al., 2011, Whitley et al., 2016) the RICHH intervention’s target of reducing depressive symptoms appears particularly well-suited for custodial grandparents.

Fig. 1.

Conceptual framework guiding the proposed study.

The RICHH intervention leverages video-conferencing technology, a cost-effective and efficient means for providing individualized interventions for caregivers who have limited access to care. Custodial grandparents, similarly to rural caregivers who have participated in previous RICHH interventions, often encounter limited access to care due to various barriers (e.g., logistical [transportation and childcare] challenges) (Dolbin-MacNab and O'Connell, 2021). Thus, we anticipate that the RICHH intervention, in being virtually delivered, will likely be particularly viable for custodial grandparents.

1.4. Identifying needed modifications to the RICHH intervention to target custodial grandparents

Below, we highlight compelling reasons in support of our efforts to examine the feasibility, acceptability, and explore the preliminary efficacy of modifying the RICHH intervention for grandparent caregivers (RICHH-GP).

Custodial grandparents report a high number of adverse childhood experiences (ACEs).Approximately 90% of custodial grandparents have ≥1 ACE (Smith et al., 2022, Song et al., 2024) whereas only a small percentage of RICHH participants (<4.5%) reported concerns about someone in the home using drugs/alcohol, feeling unsafe/threatened in daily life, or fearing abuse. Given this potential disparity in experiences with trauma between the two groups, modifying the RICHH intervention to integrate a trauma-informed approach (Edelman, 2023) becomes critical for application with custodial grandparents. Towards that end, this study will focus on preventing re-traumatization, fostering resilience and addressing the unique needs of this population through these modification processes (Edelman, 2023). Such efforts may positively impact engagement in self-care behaviors during- and post-intervention. Building on our prior work (Song et al., 2024), we also plan to solicit feedback from community health workers (CHWs) who work with grandfamilies and understand grandparents’ unique needs and priorities to help us create a trauma-informed CVD reduction intervention.

Custodial grandparents are influenced by a complex interplay of various SDoH. Unfavorable SDoH adversely affect CV health and disease prevention, (Javed et al., 2023, Jilani et al., 2021, Teshale et al., 2023) placing custodial grandparents at higher risk. These same challenges may also limit participation in research. To date, RICHH participants were less exposed to SDoH challenges, making it critical to consider SDoH challenges for participants with greater exposure:

Race/ethnicity. Almost all (96.0%) of earlier RICHH participants were White, which is not representative of demographics of custodial grandparents in the US (59.4% are White; 19.9% Black/African American) (U.S. Census Bureau 2026). Upstream evaluation of the appropriateness and acceptability of study materials for custodial grandparents from racial/ethnic minorities is crucial. For instance, the intervention modules should reflect food preferences and choices in line with relevant considerations, such as addressing the challenges of accessing healthy foods in environments where many racial/ethnic minority populations reside (Kris-Etherton et al., 2020).

Poverty/food insecurity. While less than 17.4% of RICHH participants struggled to make ends meet, over one-third of grandfamilies live in poverty (Wu, 2018). Additionally, 25% of grandfamilies experience food insecurity, which is more than twice the national rate (Generations United, 2022). Thus, it is critical that intervention materials address the affordability and accessibility of heart-healthy food in the local area.

Health literacy. The RICHH intervention was designed with materials written at a 6^th-grade reading level and included pictures and graphics. The RICHH study reported that improvement in diet quality score from the RICHH intervention was greater in the low health literacy group than in the high health literacy group (Kang et al., 2021). This suggests that the intervention may be effective for custodial grandparents, among whom approximately one third have less than a 12^th grade education with no high school diploma (Rapoport et al., 2020). However, it is important to note that ∼90% of RICHH participants to date had more than a 12^th grade education (Mean education 14.0 ± 2.7 years), and the average health literacy score was considered as “adequate” (4.6 ± 1.5, with a score of ≥ 4 indicating adequate health literacy (Weiss et al., 2005)). Given the lower education levels among custodial grandparents and the absence of current information about health literacy in this population, it is worth noting the positive association between education attainment and health literacy (Magnani et al., 2018). Also, lower health literacy is more prevalent among older adults, under-represented populations, and those with less education (Kutner et al., 2006). It will be important to modify content to ensure it aligns with their health literacy levels.

2. Study specific aims

To optimize healthy aging and CV health among this growing, at-risk population for which there is no dedicated current support, the goal of study is to modify an established intervention for implementation with custodial grandparents, and then test the feasibility, acceptability, and preliminary efficacy of that modified RICHH-GP protocol.

Specific Aim #1: Modify the RICHH intervention to target custodial grandparents from varied backgrounds (e.g., SDoH groups). We will conduct trauma-informed focus groups with (N=∼12-16) grandparents and CHWs who work with grandfamilies, to identify needed modifications to a) recruitment and retention strategies, b) intervention/study materials, and c) data collection/management procedures.

Specific Aim #2: Employ the re-designed RICHH intervention (RICHH-GP) and determine its feasibility and acceptability with custodial grandparents at 4-months post intervention, based on:

• a)Feasibility: (Surendran et al., 2023) Adequate recruitment (target N=70), attrition (10-15%) and adherence (80%) rates
• b)Acceptability: (Sekhon et al., 2022) Affective attitude (80% satisfaction with the intervention), burden (80% satisfaction with the intervention structure/content), opportunity costs and perceived effectiveness, and self-efficacy (80% confidence in adhering to the target behaviors). This will be assessed through (1) grandparents’ written evaluations and (2) focus-group interviews (target N= ∼16).

Specific Aim #3: Evaluate and refine the RICHH-GP protocol (i.e., data collection procedures, measures, data management plans) among our team members and explore the preliminary efficacy of the RICHH-GP intervention based on changes in the following measures in custodial grandparents at 4-months and 6-months compared to baseline on:

• a)Primary outcomes of physical activity, dietary intake, and body mass index [BMI]
• b)Secondary outcomes of lipids, blood pressure, HbA1c, nicotine exposure, and sleep
• c)Secondary outcome of depressive symptoms

Specific Aim #4 (Exploratory): Evaluate the proximal impact of the RICHH-GP intervention on CVH and depressive symptoms among a subsample of grandchildren of the grandparent participants

3. Research design and method

3.1. Aims and overarching research design (Fig. 2)

Fig. 2.

Overall research design.

Our study will utilize a mixed methods approach. In the pre-intervention stage, we will use a qual design to address Aim 1 (Modifying the RICHH intervention for custodial grandparents). Next, we will conduct a 2-arm randomized-controlled trial (RCT) to implement the modified RICHH-GP intervention. In the post-intervention stage, we will use a mixed method design, using both QUANTITATIVE (QUAN) and qualitative (qual) methods to address Aim 2 (determining the feasibility and acceptability of the RICHH-GP intervention). The QUAN method will also be used to address Aim 3a (exploring the preliminary efficacy of the RICHH-GP intervention). This approach was chosen to leverage the strengths of both QUAN and qual data to develop a tailored intervention for custodial grandparents. Throughout the intervention and post-intervention stages, we will continuously evaluate and refine the RICHH-GP protocol (aim 3b).

3.2. Pre-intervention (QUAL): tailoring RICHH for custodial grandparents (aim 1)

Research Design. We will employ a qualitative design and adhere to the Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines for data collected (Tong et al., 2007).

Sampling. We will use purposive sampling to recruit ∼12-16 participants into 4-5 focus-groups from the Portland Metropolitan area or Southern Washington.

Data Collection. We will conduct audio-recorded focus group interviews in English (approximately 90 minutes). We plan to conduct 4-5 in-person focus groups with custodial grandparents and 1 additional group with CHWs, either in-person or virtually. We will discontinue recruiting when saturation of the data occurs or when inductive coding from focus groups with custodial grandparents yields only 5% new information (Guest et al., 2020). The focus groups will be recruited from: 1) custodial grandparents, pooled from organizations serving underrepresented communities in the Portland Metropolitan Area or Southern Washington (e.g., an organization that serves Black/African American families); 2) community organizations that serve grandparents raising grandchildren (e.g., the Oregon Kinship Navigator program); 3) custodial grandparents who participated in our previous study or who are referred by those participants; 4) advertisement via posts on Facebook and other social media; 5) Distribution of study brochures to local sites (e.g., senior community centers, local family medicine offices); and 6) referrals from community health workers (CHWs) from networks that extend across the Portland Metropolitan Area or Southern Washington.

A facilitator will use a focus group interview guide informed by the Theory of Planned Behavior (TPB) (Ajzen and Madden, 1986). Participants will be invited for our team’s presentation covering the following topics: 1) recruitment/retention strategies our team has planned; 2) RICHH-GP intervention and study materials; and 3) data collection, and management procedures. After the presentation, participants will be invited to share their perceptions of facilitators/barriers to implementing the RICHH-GP intervention, resources needed for implementing the intervention, and feedback on our study protocol/materials. Each focus group grandparent participant will receive $45, and each CHW participant will receive $75.

Data Management and Analysis. Qual data will be managed by Ms. Rosenkranz, with oversight by PI Song. Field notes, with attention to interaction data, will be prepared within 24 hours after each focus group, and used to further explain or verify results from the thematic analysis. Focus group audio-recordings will be transcribed verbatim, verified for accuracy, and de-identified by assigning designators to participant names randomly (such as A, B, C). All data will be entered into NVivo 15 (Lumivero) to help organize/manage the data and codes. Data will be kept on a secure, password-protected network hosted at OHSU. Data will be analyzed using thematic analysis (Braun and Clarke, 2006) which includes: (1) familiarization with the data, (2) initial coding and codebook development, (3) searching for themes based on initial coding and a recurrent process of coding, (4) review of the themes with the group, (5) theme definition and labelling, and (6) report writing. Our team has used this approach successfully previously (Dolbin-MacNab et al., 2022). The familiarization step will begin with the coding team members (the coding team: the PI, Dr. Hansen, and Ms. Rosenkranz) reading and rereading the responses multiple times, and applying a priori codes derived from the TPB (Hsieh and Shannon, 2005). New codes will be developed from the content and applied. After the data familiarization phase, the same coding team members will collaboratively develop an initial codebook of all agreed upon codes. Then at least two members of the team will code the interviews independently. The initial codes and associated emerging themes/subthemes will be discussed among the coding team until a consensus is reached, resulting in a final codebook to be used by the coding team for subsequent coding of the interviews. All coding discrepancies will be resolved among the coding team members, who will work collaboratively to collapse and combine the initial codes into the themes. These themes will be further defined and refined within the context of raw data, so that their scope and content or “essence” are clear (Braun and Clarke, 2006). We will examine the extent to which the themes identified conform to the TPB model, or whether new themes have emerge (Hsieh and Shannon, 2005).

To accomplish these steps, the coding team will meet at least bi-weekly and prepare summary reports. At team meetings, preliminary data summaries will be reviewed to inform the final analyses of the focus group data. Initial themes will be shared with volunteer participants for “member checking,” a key qual data verification step, asking key informants to confirm findings (Ghanbarzadeh et al., 2014, Hoffart, 1991).

Expected Outcomes. When analysis is completed, the whole research team will meet to discuss the results, address identified problems on the design of the RICHH-GP intervention and/or the study protocol (e.g., recruitment, retention strategies, data collection and management procedures), and adapt accordingly. Currently, Phase 1 of this study has been completed; its results will be reported in a separate manuscript.

3.3. RCT and post-intervention evaluation (QUAN+ QUAL): feasibility, acceptability, and preliminary efficacy (aims 2–3)

The objective for Aim 2: to employ the modified RICHH-GP intervention and assess its feasibility and acceptability with custodial grandparents at 4-months post intervention. We will obtain QUAN data on feasibility (recruitment numbers/rate, attrition, adherence, and attendance rates) (Surendran et al., 2023) and acceptability (affective attitude, burden, opportunity costs, perceived effectiveness, and self-efficacy) of the intervention (Sekhon et al., 2022). We will also obtain qual data on the same acceptability domains through grandparents’ written evaluation and focus group interviews to complement the QUAN data.

The objective for Aim 3: During implementation of the RICHH-GP intervention, our team will evaluate and refine the modifications to the RICHH-GP protocol through early and regular discussions. In the post-intervention phase, we will assess changes in CVD risk, self-management behaviors, and depressive symptoms among custodial grandparents at 4- and 6-months compared to baseline. We expect that the intervention group will have: a) higher physical activity levels, better diet quality, and lower BMI; b) better lipid profile, decreased blood pressure, lowered HbA1c, less nicotine exposure, worse sleep; and c) lower depressive symptoms, compared to the usual care control group.

Research Design. We will use a mixed-methods design as it offers more comprehensive evidence than what could be learned from using either method alone (Creswell et al., 2011). The qual data will be used to provide further insight into the views of custodial grandparents regarding whether and how the RICHH-GP intervention is feasible and acceptable, (and secondarily the impact of their participation in the RICHH-GP protocol on their grandchild’s health). QUAN data will be obtained from questionnaires and objective measures. Qual data will be obtained through written responses to open-ended questions using Qualtrics^XM and focus-group interviews. The two sets of data will be analyzed separately, and then the results from both sets will be merged using an interactive approach. Integrating these results will achieve complementarity and be an essential step for interpreting and communicating our findings (Creswell and Plano Clark, 2018).

Study Sample/Sampling. We will use a two-stage sampling strategy (Palinkas et al., 2015).

Stage 1: QUAN strand. We will use sampling to enroll 70 participants (to reach our analytic goal of N=60) over 15 months of active recruitment. We will aim to enroll a 50%/50% mix of White and non-White (primarily, Black or African American) participants. Previous literature show differences in CV health based on racial/ethnicity status (O'Hearn et al., 2022, Lopez-Neyman et al., 2022) and a mix of participants will allow us to explore variations in effects by race/ethnicity. Based on previous grandparent caregivers’ studies including ours, (Song et al., 2024, Smalls et al., 2020, Bessette et al., 2021) we expect ∼90% of our recruits will be female. We will target recruiting at least 10% of the participants to be male). Based on our experience from previous studies (Tracy et al., 2022, Wallace et al., 2021) we expect our recruitment goals to be very feasible.

Stage 2: Qual strand. We will recruit participants via purposive sampling of a cross section of participants from stage 1, enabling us to collect feedback from a diverse representation to address key variations in the target population. Over 3 months of recruitment in this stage we will conduct ∼5 focus groups with custodial grandparents (N=∼16-20): (1) White; (2) non-White (primarily Black/African American); (3) low socioeconomic status (< federal poverty level 200%); (4) high ACEs (≥ 4); and (5) low health literacy.

Inclusion criteria for custodial grandparents: (1) grandparent who provides full-time care to at least one grandchild (< 18 years old) for at least 6 months; (2) grandparent who lives in the same household as the grandchild(ren) with either no biological parent residing in the household, or with a residing parent living in the household who provides less than 25% of the caregiving for the grandchild(ren); (2) living in Oregon or Washington; (3) English proficiency sufficient for informed consent and completion of study measures (whenever possible, we will include bilingual participants); and (4) informed consent. If primary caregiving is shared among grandparents and other adult(s) in the same household, we will ask the duo to decide which one will participate in the study (based on criteria such as percentage of time spent with the child, responsibility for meals, or level of responsibility for other specific caregiving tasks). Further, we will collect additional caregiver status information that we will control for in our analysis: marital status, and answers to the question “Are there any adults (other than the grandparent participants) who engage(s) in caregiving in the same household, and if so what caregiving activities do they engage in (e.g., homework support)?”

Exclusion criteria for custodial grandparents: (1) have plans for transfer of care to other caregivers or setting for the next 12 months; (2) reported significant physical or emotional impairment, or psychiatric illness that might interfere with engagement in their own self-management or that is likely to result in their needing a caregiver in the next 12 months. Further, caregivers will be ineligible if they have any of the following apply: (3) chronic drug abuse; (4) current active cancer (i.e., undergoing active treatment for cancer) other than isolated skin cancer; These individuals are excluded as they are not likely to be able to attend the intervention, and likely to have multiple factors that interfere with caregiving.

Intervention vs. Control (Usual Care). The intervention is 12 weeks long. A video-conferencing program on a digital device (i.e., Amazon Fire Tablet) will be used for the interventions. Post-intervention effects will be assessed in both groups at 4-months after initiation of the intervention to identify effects, and again at 6-months to estimate sustained effects. Video-conferencing technology is used to enhance the feasibility and decrease the burden of participating in the intervention. The 12-week duration was chosen based on variations of individual goal accomplishment due to caregiving responsibility, allowing participants sufficient time to adopt healthy behaviors, and reduce burden.

Intervention Group. Informed by the TPB, (Ajzen and Madden, 1986) the RICHH-GP intervention is an educational-behavioral and counseling intervention that promotes caregivers’ knowledge, skills and motivation to engage in CVD risk reduction. The intervention is delivered individually to caregivers in their homes using video-conferencing technology on mini-iPads. The program consists of 12 weekly sessions [30-45 minutes] that will be held at the caregivers’ preferred times. A nurse or educator who has a background in cardiovascular health and is trained in motivational interviewing techniques and cognitive behavioral therapy will deliver the RICHH intervention over 12 weekly sessions (Table 1). Based on the expertise and experience, the interventionist will assist in individual goal settings and reinforce strategies using motivational interviewing strategies in each session. In contrast with most approaches, where a single risk factor is targeted with risk factor-specific intervention, we target multiple CVD risk factors. This “whole-health” approach, which is grounded in the TPB, (Ajzen and Madden, 1986) improves the CVD risk profile by promoting self-management of multiple CVD risk factors, lifelong health, and reduction of risk from all chronic diseases (Kottke et al., 2009). Of note, the TPB is commonly used successfully to organize lifestyle modification interventions that require behavior change (Conn et al., 2003, Conner et al., 2002, Deskins et al., 2006, Lien et al., 2002, Masalu and Astrom, 2003, Povey et al., 2000, Bentley et al., 2009) and our team developed and tested three successful self-care interventions using the TPB (Wu et al., 2012, Welsh et al., 2012, Dekker et al., 2012). The TPB states the determinants of behavior change are attitudes, subjective norm, and perceived behavioral control. Attitudes are determined by the individual’s beliefs about outcomes of performing the behavior. An individual who holds strong beliefs that positively valued outcomes will result from a behavior will have a positive attitude toward that behavior. An individual’s subjective norm is determined by normative beliefs—whether important significant others approve or disapprove of the behavior. Perceived behavioral control is determined by beliefs about the presence or absence of resources for and impediments to behavioral performance.

Table 1.

Content of the rural caregiver heart health education.

Modules	Content/focus
1. Self-management for heart health [use booklet and self-management logs; week1]	Self-management principles; feedback of data collected at baseline (i.e., blood pressure, lipids, BMI, eating pattern and physical activity level) and implications for the caregivers’ CV health; mutual identification of targets for intervention and participants’ goals; identification of caregiver specific barriers to achieving goals
2. Depressive symptom management [use booklet; weeks 2-12]	What is depression related to caregiving; Identify caregivers’ stressors and source for depressive symptoms; strategies for managing stressors including socialization, verbalization, cognitive restructuring; management of depressive symptoms.
3. Heart healthy eating [use booklet and self-management logs; weeks 3-12]	The whole food approach to easy heart healthy eating principles overview – what and how to eat to achieve heart health given individual circumstances – how to choose healthier foods when eating out; how to choose from menus; how to ask for healthier alternatives; reading food labels; portion control (first regardless of other choices); facts about fats revealed; low- sodium.
4. Physical activity [use booklet weeks 3-12]	How much and what types of physical activity are adequate for heart health; how to incorporate enough physical activity into a day
5. Co-morbidity(ies) management [use booklet; weeks 4-5]	Management of hypertension, diabetes, obesity
6. Adherence to medication and smoking cessation [use booklet; weeks 4-12]	Medication review; methods to enhance adherence to medication. Linking medications taken with target and outcomes; smoking cessation; avoid second-hand smoke.

The modules are written at a 6^th grade reading level. Each module includes skill building and knowledge provision and comes with supplementary materials to accommodate those who desire more complex materials. Because of the importance of depression management and prevention, and of nutrition, content from both modules is integrated throughout each weekly session. Examples of how we achieve the principles of individualization, promotion of self-management, and reducing barriers to following CVD risk reduction practices are as follows.

Individualization. At the first session, we discuss results from baseline data collection, indicate what they mean for the individual, and set up individual, realistic goals for CVD risk reduction. We also discuss individual barriers to risk reduction and use motivational interviewing to reduce some of these barriers.

Self-management. We concentrate in each module on providing skills and using demonstration of these skills via video (e.g., healthy cooking). Self-management principles are incorporated into each module.

Reducing barriers. With the caregiver, we identify and discuss ways to overcome, their personal and environmental barriers to risk reduction. All sessions include barrier reducing skills, and problem solving. For example, we map, for each individual, a walking route that is accessible to them; we provide tips about how to shop and cook using locally bought groceries and how to use what is available to cook a heart healthy meal; we provide information on eating out; we take a recipe and modify it using heart healthy principles.

Control Group. The Usual Care (control) group will receive Amazon Fire Tablets loaded with: 1) “Caregiver” (What is caregiver burnout? How should I care for myself? What are the caregivers’ rights? How should I communicate with my loved one and with others? How can I support my loved one?), and 2) CVD risk reduction pamphlets (Life’s Essential 8) in PDF format along with the associated links from the American Heart Association. In addition, we will email them using the Amazon Fire Tablet monthly in order to keep them engaged in the study. Those randomized to usual care will be offered the RICHH-GP intervention content at the end as an “opportunity.”

Data Collection.

Aim #2. We will calculate recruitment, attrition, adherence, and attendance rates at the end of the 1:1 parallel group RCT. Custodial grandparents will complete the post-intervention survey measuring feasibility (e.g., perceptions of the ease of the blood collection procedures, or perceptions of the logistical requirements of the intervention such as the timing of data collection) and acceptability of the RICHH-GP intervention (e.g., most/least helpful elements at 4 months post-intervention and Likert Scale questions). We will conduct audio-recorded interviews (∼60 minutes) with 4 grandparents in approximately 4-5 focus groups in English to assess feasibility and acceptability. We will also assess grandparent’s perceptions of the impact of the intervention on the grandchild’s health. Recruitment for the focus groups will end when “saturation” of the data occurs, or when inductive coding yields only 5% new information (Guest et al., 2020). A facilitator will use a semi-structured interview guide directed at exploring grandparent perspectives of the RICHH-GP intervention. Grandparents will be invited to share their opinions on acceptability of the intervention, and their perceptions of barriers and enhancements for participating in the intervention. The sequencing (QUAN followed by qual) is purposeful to avoid biased answers on QUAN. We will ask participants if they think QUAN data collection influenced their qual responses. Focus group participants will receive $45.

Aim #3. Data will be collected at the baseline, 4-month, and 6-month visits for both groups. The home study visits will be from 2-2.5 hours long, with blood sample collection in the last portion of visits so that potential stress response from the finger stick will not influence other measures. Each participant in the RCT will receive up to $300 ($100/visit). Table 3 outlines study measures.

Table 3.

Measures.

Domain	Measure
Grandparent's Cardiovascular Disease Risk Factors
Body mass index	Weight (scale) and height (stadiometer)
Blood pressure/heart rate	Blood pressure cuff with monitor
Blood lipids	Collecting non-fasting blood samples via finger stick
Blood sugar (Hemoglobin A1c %)
Waist circumference	Non-extensible tape measurement
Grandparent's Adherence to Self-management Behaviors
Physical activity	Accelerometer and questionnaire
Sleep health	Accelerometer and Pitssburg Sleep Quality Index questionnaire
Dietary intake	Viocare food-frequency questionnaire
Nicotine exposure	Questions
Depressive Symptoms	Patient Health Questionnaire-9
Other Variables of Interest
Caregiving burden	Zarit Burden Interview questionnaire
Other caregiving characteristics	Questions on duration of caregiving, the duration of living together with their grandchildren, number of children in their care
Medical conditions	Charlson Comorbidity Index
Medication(s)	Questions
Resilience	Connor-Davidson Resilience Scale and Post-traumatic growth inventory questionnaire
Health care utilization	Questionnaire
Adverse chilhood experiences	Questionnaire
Grandchild's Cardiovascular Health Behaviors and Depressive Symptoms
Physical activity	Accelerometer and questionnaire
Sleep health	Accelerometer and Pitssburg Sleep Quality Index questionnaire
Dietary intake	Viocare food-frequency questionnaire
Nicotine exposure	Questionnaire
Adverse chilhood experiences	Questionnaire (report by grandparents)
Depressive symptoms	Questionnaire
Grandchild's Cardiovascular Disease Risk Factors
Body mass index	Weight (scale) and height (stadiometer)
Blood pressure/heart rate	Blood pressure cuff with monitor
Blood lipids	Collecting non-fasting blood samples via finger stick
Blood sugar (Hemoglobin A1c %)

CVD Risk Factors

Body Mass Index (BMI) and Waist Circumference. BMI (kg/m²) will be calculated from height and body weight measured. Measurements will be taken without shoes and with all over-garments removed (Heyward and Wangner, 2004). To obtain weight, we will measure to the nearest 0.1 kg using a Seca 813, and standing height to the nearest 0.1 cm using a Seca 213 portable stadiometer. Waist circumference will be measured to the nearest 0.1 cm using a non-extensible tape at the narrowest portion of the torso between the iliac crest and inferior rib (Garvey et al., 2016, Després et al., 2021). At least two independent measurements will be taken and averaged (Katzmarzyk et al., 2013).

Blood pressure. Our trained staff will measure seated blood pressure using a manual sphygmomanometer with an appropriately fitted cuff and best practices (Muntner et al., 2019, Cheung et al., 2023). Per a standard protocol, (Centers for Disease Control and Prevention 2026) three consecutive blood pressure readings will be obtained and averaged. All staff are trained to accurately and reliably measure blood pressure and are tested every 3 months in the following areas: 1) choosing proper cuff size and placing the cuff correctly; 2) proper patient positioning; and 3) appropriate measurement protocol.

Blood lipids and HbA1c. Non-fasting blood samples (as lipids do not vary substantially after eating (Darras et al., 2018)) will be collected via finger stick. We will use CardioChek® Plus Analyzer (Polymer Systems Technology Diagnostics, Whitestown, IN) (Polymer Systems Technology Diagnostics, 2026) to analyze lipids and the A1CNow®⁺ (Polymer Systems Technology Diagnostics, Whitestown, IN) (Polymer Systems Technology Diagnostics, 2026) to analyze HbA1c (%). Both analyzers were used in our earlier pilot with grandfamilies (Song et al., 2024, Song et al., 2022).

Adherence to Self-management Behaviors

Physical activity. Physical activity will be measured using ActiGraph accelerometers (wGT3X-BT, ActiGraph LLC, Pensacola, FL; ActiGraph LLC, 2026). Participants will be instructed to wear the accelerometers during all hours for 7 consecutive days, to achieve a minimum of 4 valid days of monitoring (including one weekend day) (Vetrovsky et al., 2020, Migueles et al., 2017). A valid day will be defined as a minimum of 10 hours of wearing time. Our team have several years of experience using ActiGraphs in studies with (older) adults (Bessette et al., 2021, Song et al., 2022, Song and Bonds, 2020, Song et al., 2018, Song et al., 2017). To supplement ActiGraph data and to account for the possibility of incomplete data from some participants (e.g., due to illness, travel, loss of monitors), we will administer the International Physical Activity Questionnaire for self-report of physical activity (Craig et al., 2003). We will calculate moderate-to-vigorous and light activity time, and sedentary time from these measures.

Sleep. Sleep will be assessed via self-report and actigraphy. Participants will complete the Pittsburg Sleep Quality Index (PSQI) (Buysse et al., 1989, Carpenter and Andrykowski, 1998), a validated measure of subjective sleep quality and sleep duration. Participants will also wear a wrist-worn activity monitor continuously for 7 consecutive days to acquire objective measures of sleep duration, bedtime, wake time, and sleep-wake patterns. The use of both subjective and objective measures will enhance the validity and comprehensiveness of sleep assessment.

Food intake. We will administer a self-report food-frequency questionnaire Viocare (Viocare, Inc., Princeton, NJ, USA) (Deierlein et al., 2019, Cullen et al., 2008) that we used successfully in our previous study (Song et al., 2022). It is a user-friendly and cost- and time-effective method of collecting dietary patterns. Following the AHA diet scoring criteria (Lloyd-Jones et al., 2022), Viocare data will be used to generate Healthy Eating Index-2020 (USDA Food and Nutrition Administration 2026). It requires about 30 minutes to complete. The trained staff will be available to clarify questions or assist with completion of Viocare. An earlier form was validated against the 24-hour dietary recall method (Boucher et al., 2006).

Nicotine exposure. Questions regarding smoking status (18 items) (Centers for Disease Control and Prevention 2026) and secondhand smoke exposure (15 items) (Ksinan et al., 2021) will be asked to assess their nicotine exposure.

Depressive Symptoms. Depressive symptoms will be assessed using the Patient Health Questionnaire-9 (PHQ-9) (Huang et al., 2006, Kroenke et al., 2001, Spitzer et al., 1999). Nine items of the PHQ-9 corresponds to the nine symptoms of the major depressive disorder criteria of the Diagnostic and Statistical Manual of Mental Disorders-IV. Participants rate items based on how often they experience these symptoms over two weeks on a 4-point Liker scale ranging from 0 (not at all) to 3 (nearly every day). The reliability and validity of the PHQ-9 have been demonstrated extensively in geriatric populations (McManus et al., 2005). Its brevity and sensitivity make it useful in research and clinical settings (Kroenke et al., 2001).

Measurement of Other Variables of Interest. To characterize participants, the following variables will be assessed. Caregiving Burden will be measured with the Zarit Burden Interview (ZBI) (Bedard et al., 2001). The ZBI has 22-items and each item is rated on a 5-point Likert scale from 0 (never) to 4 (nearly always). A total score of 17 or higher (out of a total of 88) is considered to be severe burden. Racial/ethnic minorities are overrepresented among grandfamilies and pre-existing racial disparities and systemic racism may contribute to negative outcomes (Dolbin-MacNab and Few-Demo, 2018). thus, grandparents’ race/ethnicity will be collected. Since sex and marital status may influence CV health (among grandparent-headed households [no parents residing], 78.4% were with only one caregiver, (Keim et al., 2022) and among grandparent-headed households with only one caregiver, 92.8% were headed by a grandmother (Keim et al., 2022, Demographic, 2015), we will include grandparent sex/gender and marital status. It is important to note that the findings of the RICHH intervention reported that sex did not moderate intervention effects on outcomes of BMI, depressive symptoms, lipid profile, blood pressure, or diet quality; but it moderated the effects on adherence scale scores (Moser et al., 2023). Grandparents will be also asked about duration of their caregiving, the duration of living together with their grandchildren, and the number of children in their care, since these variables may also play an important role when considering outcomes with custodial grandparents (Infurna et al., 2022). Data about grandparents’ education level, comorbidities, medications, history of ACEs, and health care utilization will be collected using a standardized questionnaire. Medical Conditions will be measured using the Charlson Comorbidity Index (Charlson et al., 1987, Katz et al., 1996). Grandparents will be queried about preexisting diseases at the time of enrollment. Most conditions are scored with 1 point, although some (e.g., hemiplegia, cirrhosis) are assigned >1 point. Scores can range from 0 to 34. Medication will be assessed by a brief interview. History of ACEs. Grandparents will complete the ACES aware questionnaire for adults (Aces aware 2026) Grandparents will complete the Connor-Davidson Resilience Scale (10 items) (Connor and Davidson, 2003, Campbell-Sills and Stein, 2007) and Posttraumatic Growth Inventory (21 items) (Tedeschi and Calhoun, 1996) to test their role as moderators on the impact of the RICHH intervention.

Recruitment Plan. We plan an approximately 15-month recruitment period to recruit 70 participants, about 4-6 grandparents/month. The total accounts for an estimated 10-15% attrition – on average found in our similar previous studies (90% retention in a study with grandparent-caregivers (Song and Bonds, 2020) and 89% retention in a RICHH intervention at 4 months follow-up visits and 85% at 12 months follow-up visits (Moser et al., 2023) – to reach the study’s sample size of 60.

We have determined, through interviews and community partner guidance, that the most successful recruitment approach will be a multi-pronged, involving (1) community organizations serving underrepresented communities in the Portland Metropolitan Area (e.g., N/NE AARP) or Washington; (2) working in a newly established collaboration with Greater Oregon Behavioral Health, Inc., which operates Oregon Kinship Navigator program; (3) recruitment through Facebook and other social media; (4) Distribution of study brochures to local sites (e.g., senior community centers, local family medicine offices); (5) referrals from CHWs from network that exists across the Portland metropolitan area or Washington and who serve grandparents raising their grandchildren in the community.

During interviews, (Song et al., 2024) we learned that many CHWs who work with grandfamilies are hired by community-based agencies. We will attend regularly scheduled meetings at those agencies, provide refreshments, share study information, and seek feedback on our recruitment approach, and tailoring study procedures with cultural perspectives in mind. Every 6 months, or upon reaching a predefined recruitment milestone (e.g., 25% of the target sample enrolled), we will report back to the agencies on our study’s progression. We will follow similar engagement procedures with other collaborating community partners.

We have connected with two organizations which serve Black/African American families in the Portland Metropolitan Area. We will work with CHWs with whom we have discussed collaboration on research. Finally, we will advertise on a study Facebook page (recruiting grandparent caregivers via Facebook has a track record of success (Wallace et al., 2021).

Potential participants can learn by calling our study phone number or by providing contact information by email, phone, or on “sign-up” sheets during meetings. Our team will then follow up with those who express interest. Potential participants will be screened via email, phone, or Qualtrics^XM survey to determine eligibility.

Baseline data collection: A research staff member will arrange a study visit to collect baseline data from participants. Questionnaires will be completed using the Qualtrics^XM system (more information on this system is detailed in the data analysis section). CVD risk will be assessed through physical examination including a finger stick for lipids and HbA1c.

Randomization and Intervention: After baseline data collection, participants will be randomly assigned to the intervention or control group (using block size=4) to ensure equal allocation (1:1) between the two groups across month (assuming roughly n=4 enrolled per month) using a randomization schedule created by the study biostatistician with the blockrand package in R (R Core Team n.d). Following baseline assessment, the PI will present each participant with an envelope with their study group assignment. This randomization process will be replicated with each wave until the target sample (N=70) is reached. Only those assigned to the treatment condition will receive the intervention. A trained staff member will deliver education materials and a Amazon Fire Tablet to the caregivers in the intervention group and instruct them how to use the video conference program (i.e., WebEx) on aTablet by running a trial video-conference. We provide a phone number for caregivers to call at any time for assistance using the Tablet. An appointment will be made for the first session. The interventionist will deliver the intervention sessions using the video-conferencing program to caregivers in their homes at their preferred time. Caregivers in usual care group will participate in the same data collection procedures as caregivers in the intervention group.

Follow up data collection/Retention plan. Caregivers in both groups will follow the same data collection procedures as the baseline assessment at 4- and 6- months to evaluate post-intervention effects. At follow-up, all participants will complete the same questionnaires and physical examination as completed at baseline. Successful longitudinal retention of participants and follow-up requires careful use of a combination of encouragement, establishing a caring professional relationship, and using regular contact methods (i.e., telephone calls, emails, using a Fire Tablet [for the intervention only]) to ensure retention. Lastly, as with recruitment, we will rely on our study partners (e.g., organizations) to assist us should we have difficulty with follow up contacts. Thus, we do not expect difficulty with retention, given the limited time requirements and incentives for study participations.

Procedures. Baseline and follow-up data will be collected in English during a single 8-day cycle at a place where participants feel the most comfortable with and may vary by site: (1) participant’s home or (2) a community-site, such as CHW’s office, neighborhood center, library, or county office. Data collection also includes home use of an actigraph. We have established a safety protocol for home visits and will follow it for all study visits. Data will be managed primarily using Qualtrics^XM, which provides user-friendly web-based case report forms, real-time data entry validation (e.g., range checks) to reduce the risk of missing data, audit trials, and a de-identified data export mechanism to statistical packages (Qualtrics, 2026). Most measures we will use in Qualtrics^XM, were used previously or are in current use by our team.

Protocol Evaluation/refinement (Among team members). During the trial, we will evaluate data collection procedures, the delivery of measures, and data management plans/procedures. To ensure a comprehensive record of our progress, the study staff will maintain a structured log, documenting any encountered barriers, challenges, and solutions to implementation of the RCT. We will create summaries from findings and present to our regular team meetings to solicit feedback to further refine/finalize the protocol and study procedures.

Data Management and Analysis.

Aim #2 Analysis Plan. 1) QUAN data: As per the pilot and feasibility studies guidelines, QUAN data analysis will be descriptive. This analysis will provide descriptive statistics (e.g., N, percentages with 95% confidence intervals [CIs]) for the number of participants recruited (target N=70), recruitment rate (N per month), % of participants who completed the intervention (target 80%) and 4-month (and 6-month) follow-up visits (target 85-90%), the demographic characteristics of participants, and the number of intervention sessions attended by each participant (target 80%). Acceptability of the intervention will be estimated by the percentage of participants who indicate high acceptability across five different domains (e.g., respond “strongly like” or “like” the intervention), with a target of 80% for each domain. These results will be used to guide future intervention development. All analyses will be performed using R v4.4.2 (R Core Team n.d).

Aim #2 Power and Sample Size. The proposed sample size was chosen balancing realistic recruitment goals and the statistical precision of the feasibility and acceptability estimates, as reflected in our “go” or “no-go” criteria. For the formal power analyses, we assumed power=.80, a one-sample test of proportion and 95% CIs (i.e., alpha=.05) (R Core Team n.d). With our proposed sample size of 60, 95% CIs will span ±10-15 percentage points around observed feasibility and acceptability point estimates.

2) Qual data (from written evaluation [e.g., most/least helpful elements] and focus group interviews): All data will be entered into NVivo 15 (Lumivero) to help organize/manage and code the data. Written evaluation and data from focus groups will be analyzed and verified as described below. After all the focus groups are completed, we will offer web-based participant meetings (approximately 1 hour) to explain how attendees could help researchers improve and evaluate study results and share summarized study findings (de Loyola González-Salgado et al., 2022).

Triangulation and Interpretation of Mixed Methods Study: Data findings from each set of (QUAN and qual) data will be merged, compared (using the triangulation method (Creswell et al., 2011, Sandelowski, 2000)), and then interpreted (Creswell and Plano Clark, 2018). The qual strand will add more extensive and descriptive explanations that are not captured by the QUAN strand. An informational matrix will be created to compare and contrast the qual data findings with the QUAN data findings (Creswell and Plano Clark, 2018). This joint display allows for direct comparison to determine if the two sets of results are in full concordance, in partial concordance, missing in one set of results, in discordance or complementary. Differences in the data findings will be examined in relation to the nature of the data (its type - QUAN vs. qual, coverage, and depth), described, interpreted, and documented (Creswell et al., 2011, Patton, 1999). Results from this comparison is intended to situate the QUAN results in the appropriate context regarding the extent to which participants accepted the intervention and how they evaluate the influence of the intervention on their health. If some of the feasibility and acceptability criteria are not met, we will investigate the potential causes of those failures. This information will allow modifications to our intervention prior to the development of a future clinical trial. The interpretation of the combined results will be used to inform development of interventions that are tailored to custodial grandparents. Trustworthiness: Several strategies will ensure the trustworthiness of the qual analysis. First, the use of four independent coders, and the process of reaching full consensus with the coding and themes, will help mitigate potential researcher biases and enhance the credibility and confirmability of the analysis (Creswell and Poth, 2018, Lincoln and Guba, 1985). We will enhance the trustworthiness of the analysis by conducting peer debriefings with Co-I Musil. Co-I Musil will provide her extensive expertise in working with grandmother caregivers over decades. We will maintain an audit trail of all steps and decisions in the analysis to enhance dependability (Creswell and Poth, 2018, Lincoln and Guba, 1985).

Aim #3 Analysis Plan. Psychometric analysis (e.g., Cronbach’s alpha) will be used to confirm the reliability of all self-report measures. We will examine patterns of missing data (e.g., drop outs) to determine mechanisms (missing completely at random, missing at random, or nonignorable) (Little and Rubin, 2022), and may use full information maximum likelihood estimation depending on observed mechanisms of missingness.

For the preliminary efficacy estimates, we will estimate between-group differences in rates of CVD risk, self-management behaviors, and depressive symptoms from baseline to 4 months and the sustained effects of the intervention at 6 months using a mixed-effects modeling framework (Pinheiro and Bates, 2000). This model will include fixed effects for time and intervention group and random effects as appropriate given the model and fit statistics (e.g., random slopes across time). Parameter estimates from these models will be used to calculate effect sizes (e.g. Cohen’s d and Odds ratios, as appropriate) for the magnitude of change across time in each group and between groups. We also plan to conduct exploratory analyses to investigate potential differences by race/ethnicity and exploring race/ethnicity, age, and/or resilience as moderators in the mixed-effects models. All analyses will be performed using R v4.4.2 (R Core Team n.d).

Aim #3 Power and Sample Size. Formal power analyses were conducted for Aim 2. Findings from the preliminary efficacy analyses of exploratory Aim 3 will provide variability and effect size estimates (with 95% CIs) to inform statistical power analyses and eventual sample size requirements for future trials.

3.4. (Exploratory) evaluate the proximal impact of the RICHH-GP intervention on grandchildren of the grandparent participants (Aim 4)

The objective for Aim 4: to evaluate the proximal impact of the RICHH-GP intervention on CVH and depressive symptoms among a subsample of grandchildren of the grandparent participants

Research Design/Study Sample. We will employ an observational (pre- and post-intervention) design. We will recruit grandchildren (n=20) of grandparent participants assigned to the RICHH-GP intervention.

Inclusion criteria and Exclusion criteria (for grandchildren): Grandchild will be eligible for participation if they: (1) are aged 7 to 14 years; (2) reside with the participating grandparent caregiver; (3) live in Oregon or Washington; and (4) have English proficiency sufficient for informed assent and completion of study measures. If more than one child in the household meets these criteria, we will select the child whose birthday is closest to the date of the eligibility screen. Children will be excluded if they report a significant acute general medical or psychiatric illness that might interfere with completion of the study. Table 2

Table 2.

Data Collection and Evaluation Plan for Assessing Acceptability.

Objective	Indicators	Measure of success
Acceptability (Guest et al., 2020)	Affective attitude	At least 80% of participants report “strongly like” or “like” in answer to “Did you like or dislike the intervention?”
	Burden	At least 80% of participants report “no effort,” “little effort”, or “no opinion” in answer to: “How much effort did it take to complete the intervention?”
	Opportunity costs and perceived effectiveness	At least 80% of participants respond “strongly disagree” or “disagree” to the question, “Intervention interfered with my other priorities.” At least 80% of participants respond “strongly agree” or “agree” to the item: “The intervention has improved my cardiovascular (heart) health.”
	Self-efficacy	At least 80% of participants respond “strongly agree” or “agree” to the question, “How confident did you feel about engaging with the intervention?”

Data collection. Data will be collected from grandchildren over 8-days at baseline and again at the end of the intervention (at 4-months). Each data collection period will begin with an in-person scheduled home visit with the grandparents enrolled in RICHH-GP. During this visit, grandchildren will complete a questionnaire (via Qualtrics ^XM) assessing demographics, the grandchild’s physical activity, sleep, food intake, nicotine smoke exposure as well as depressive symptoms, and undergo assessment of CVH physiological indicators (BMI, blood pressure/heart rate, lipids and Hemoglobin A1c). We will collect blood samples at the end of the study visits so that any potential stress response from finger stick will not influence other study measures. Following the visit, the grandchildren will be asked to wear wrist-worn activity monitors for 7 consecutive days. Prepaid, stamped envelopes will be provided to each grandparent-grandchild dyad for returning the sleep monitors and daily activity/sleep logs 1 week after the study visit. Each in-person visit with grandchildren will last approximately 45 to 60 minutes. Table 3 provides an overview of study measures. Each grandchild participant will receive up to $150 ($75/visit), along with a one-time insulated water bottle at the baseline visit as a token of appreciation (Afkinich and Blachman-Demner, 2020).

Recruitment and Retention Plan. Eligible grandchildren will be recruited from grandparent participants assigned to the RICHH-GP intervention arm. Recruitment and retention procedures are described under the “Recruitment Plan” and Retention plan” for Aims 2 and 3. After obtaining legal guardian permission, research staff will explain the study to eligible grandchildren and obtain their assent using an IRB-approved assent form, either through email or during an in-person meeting.

Data Management and Analysis.

Aim #4 Analysis Plan. We will describe the sample using standard descriptive statistics at each time point. Changes in grandchildren’s CVH indicators and depressive symptoms from baseline to 4 months will be estimated with paired t-tests (or equivalent tests for categorical variables) and appropriate measures of effect size (e.g., Cohen’s d and Odds Ratios). All results will be interpreted as exploratory given the pilot nature of this study. All analyses will be performed using R v4.4.2 (R Core Team n.d).

Aim #4 Power and Sample Size. Findings from the preliminary efficacy analyses for this exploratory Aim 4 will provide variability and effect size estimates (with 95% CIs) to inform statistical power analyses and sample size requirements for future trials.

4. Discussion

This study will be the first to use a trauma-informed lens to adapt a well-established CVD risk reduction intervention for an at-risk target population (grandparent primary caregivers and their grandchildren). We expect to reveal and implement strategies for implementation with custodial grandparents that are not addressed by applying non-modified, existing intervention paradigms previously designed and tested for adult caregivers of adult patients (e.g., by spouses, children), such as the RICHH intervention. This paper outlines the significance of and methods for addressing the three study aims. Upon completion, results of our analysis are expected to provide scientifically essential data to support a larger clinical trial where we modify the RICHH-GP intervention to promote healthy aging and optimal CV health among custodial grandparents.

4.1. Anticipated results and implications

We expect to complete the scientifically necessary groundwork for an expanded future clinical trial that will test the effectiveness of the RICHH-GP intervention with a longer follow-up and with an increased inclusivity of participants from historically marginalized populations. To achieve our goal, we will employ a mixed methods design. The QUAN strand will track the initial effect of the RICHH-GP intervention, while the qual strand will explore grandparents’ perceptions regarding feasibility, acceptability, and the downstream effect of their participation in the RICHH-GP program on their grandchild’s CVH indicators and depressive symptoms.

Considering that custodial grandparents are influenced by a complex interplay of SDoH and higher rates of risk factors (e.g., ACEs) than their non-caregiving peers, our findings will provide invaluable guidance to health professionals and policy makers who work with this fast-growing, high-risk group and other groups of caregivers who face SDoH-related obstacles. If the RICHH-GP intervention is effective with custodial grandparents, it will support the framework outlined in Fig. 1 and demonstrate the effectiveness of the TPB to guide intervention to this under-studied caregiver population. The component of RICHH-GP which explicitly incorporates the perspective and experience of custodial grandparents is critical for modifying interventions that promote intergenerational health, equity, and strengthen existing community programs and policies (Substance Abuse and Mental Health Services Administration, 2014).

5. Conclusions

This study will tailor an established CVD risk reduction intervention for implementation with custodial grandparent caregivers and evaluate the feasibility, acceptability, and explore the initial effects of the modified intervention (RICHH-GP) on grandparent caregivers and their grandchildren. Data from this study will guide refinement of the RICHH-GP intervention, study procedures, and future trial design, including potential enhancements to strengthen intergenerational pathways to CV health. Findings will provide critical preliminary evidence to support a future fully powered clinical trial of the RICHH-GP intervention, with the long-term goal of improving CV health, supporting healthy aging among custodial grandparents, and promoting cardiovascular health across generations.

Ethics approval

The study protocol was approved by the Institutional Review Board at Oregon Health & Science University (IRB# 28858).

Ethics declaration

The authors have NOT obtained informed consent from participants or their legal representatives. This is a study protocol and does not report the results of the study.

This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. This study was approved by the Oregon Health & Science University. (Approval No. 28858)

The results of this clinical trial and any associated work have been posted in a registry. This clinical trial was registered with number NCT07083895 (clinicalgov).

Trial status

This trial is registered at clinicalgov (registration # NCT07083895); We are currently preparing to recruit participants for Phase 2 of this study. Recruitment will begin in October 2026 and the study is expected to be completed in July 2028. This manuscript describes version 2.0 of the study protocol, dated July 28, 2026.

CRediT authorship contribution statement

MinKyoung Song: Writing – review & editing, Writing – original draft, Visualization, Project administration, Funding acquisition, Conceptualization. Lissi Hansen: Writing – review & editing, Funding acquisition, Conceptualization. Carol M. Musil: Writing – review & editing, Funding acquisition, Conceptualization. Nathan F. Dieckmann: Writing – review & editing, Funding acquisition, Conceptualization. Susan J. Rosenkranz: Writing – review & editing, Data curation. Debra K. Moser: Writing – review & editing, Methodology, Investigation, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

No data were used for the research described in the article.