Assessing the Acceptability of a Culturally Adapted Group-Based Pediatric Intervention, Kokajjiriri, for Marshallese Mothers and Infants to Improve Nutrition and Prevent Childhood Obesity

Abstract

Background:

Marshallese Pacific Islanders experience higher rates of obesity than other racial and/or ethnic communities. Despite the obesity rates experienced in this community, there are currently no childhood obesity prevention interventions designed for Marshallese Pacific Islanders in the United States. The purpose of this study is to assess the acceptability and feasibility of a culturally adapted group-based pediatric intervention, Kokajjiriri, with Marshallese mothers to improve nutrition and reduce childhood obesity.

Methods:

A multi-methods design was used to culturally adapt the Kokajjiriri intervention for Marshallese mothers in Arkansas (n=17). In phase one, we conducted 24-hr dietary recalls with 20 Marshallese mothers to inform the cultural adaptation of the group based pediatric intervention, and then in phase two, we culturally adapted and piloted three sessions of the intervention to determine acceptability and feasibility of the intervention.

Results:

Participants found the adapted intervention to be acceptable and feasible, found the location to be convenient, and found the facilitator to be knowledgeable. Four themes emerged from the qualitative data: 1) Lactation Support; 2) Introducing Healthy Solids; 3) Rice Portion Control; and 4) Finding Resources.

Conclusions:

This is the first study to assess the acceptability and feasibility of a culturally adapted group-based pediatric intervention, Kokajjiriri, with Marshallese mothers to improve nutrition and reduce childhood obesity. The results from this culturally adapted group-based pediatric intervention, Kokajjiriri, will be used to inform future adaptations and implementation of the full intervention for Marshallese women and children.

INTRODUCTION

The United States (US) ranks as the 12^th most obese country worldwide and is considered number one among high-income countries.¹ Childhood and adolescent obesity in the US have reached epidemic levels. In the US, nearly 10% of infants and toddlers and 17% of children and adolescents are obese.² Overweight and obese children are likely to stay obese into adulthood and more likely to develop non-communicable diseases like type 2 diabetes, high blood pressure, hypertension, high cholesterol, orthopedic problems, and liver disease.³ The prevalence of childhood obesity is significantly higher among racial and/or ethnic minority children in the US. In 2021–2022, obesity rates were significantly higher for Hispanic (22.7%), non-Hispanic Black (22%), and non-Hispanic American Indian/Alaska Native children (21.4%). While information is limited, available data shows Pacific Islander children in the continental US have obesity prevalence ranging from 17.2% to 48.6%.⁴

The Center for Disease Control and Prevention (CDC) recommends exclusive breastfeeding with appropriate complementary foods introduction after six months.¹ Exclusive breastfeeding reduces the risk of childhood and life-long obesity. The introduction of solid foods prior to four months is associated with obesity later in life, while exclusive breastfeeding for at least six months can reduce the odds of becoming overweight by more than 30%, as breast milk’s protective effect increases with duration.^5,6 Further, for the mother, exclusive breastfeeding can reduce the risk of metabolic syndrome and type 2 diabetes, which can help reduce obesity in subsequent children.⁷

Arkansas currently has the largest population of Marshallese Pacific Islanders living in the continental US (~16,000 people).^8,9 Our study in 2016, with Marshallese adults in Arkansas, revealed that 41.2% of participants had measures indicating hypertension, 16.4% had prehypertension, and 90% were overweight or obese (n=401).¹⁰ Exclusive breastfeeding rates among Marshallese in the US are well below the CDC’s reccomendations.¹¹ A national survey of Marshallese mothers found that at four months of age, only 31% were exclusively breastfeeding,¹ and preliminary data of Marshallese in Arkansas demonstrates less than 2% of Marshallese women were exclusively breastfeeding at six months.¹² Qualitative data with Marshallese women demonstrate numerous barriers to exclusive breastfeeding, including verbal and non-verbal public shaming, perceived low milk production and quality, maternal employment, lack of female familial support, and institutional influences (e.g., hospital, WIC).^12–14 Further, Marshallese infants traditionally receive early introduction (before six months) of high-simple carbohydrate or sugar complementary foods, typically including rice, ramen, juice, and pureed fruit with sweetened canned milk.^15–17 Despite the obesity rates experienced in this community, there are currently no childhood obesity prevention interventions designed for Marshallese Pacific Islanders in the US.

Early-life interventions targeting eating patterns during the first 1000 days of life are crucial to promote proper nutrition and growth to reduce childhood obesity.⁸ Innovative models of pediatric care are needed to improve the health and well-being of women and infants, particularly for minoritized groups. Group pediatric care is a promising strategy to improve clinical and social outcomes for both mother and infant.^18,19 Previous outcomes associated with group pediatric care, compared to traditional pediatric care, include greater satisfaction, higher exclusive breastfeeding, higher attendance of well-child visits (65% vs. 37%) to be fully immunized (98% vs. 82%), and substantially lower rates of overweight and obesity in children at 15 months (8% vs. 24%)²⁰ and 24 months (2.1% vs. 15%).²¹

Importantly, health interventions not aligned with the cultural values, perspectives, and preferred modes of living of the target population are presumed to be less effective than culturally responsive interventions that account for these factors.^22,23 To design culturally appropriate interventions to promote healthy growth and development of children under the age of 12 months, it is necessary to better understand the current diets within the families of new Marshallese mothers.

In the first phase of this study, we first conducted 24-hour dietary recall interviews with 20 Marshallese mothers with children under the age of 12 months to characterize their dietary practices. Key findings included a median Healthy Eating Index-2020 score of 46.4 out of 100, lower than a nationally representative sample of breastfeeding women in the US,²⁴ and lower than prior studies of children living in the Republic of the Marshall Islands.^25,26 Lower diet quality scores were driven by an intake of fruits and vegetables, whole grains, and dairy below recommended amounts, and high intake of refined grains (e.g., white rice). Positive findings included a high consumption of seafood, evidenced by high scores within seafood/plant proteins and fatty acid diet quality components. In phase two, we then conducted cultural adaptations and piloted an abbreviated version of a group pediatric care intervention, CenteringParenting, with 17 (n=17) Marshallese mothers. CenteringParenting is a group pediatric intervention that has demonstrated feasible and acceptable by providers and with low-income women, but has not been tested for acceptability among racial and/or ethnic minoritized communities, and never with the Marshallese.²⁷ As part of our cultural adaptations, our Marshallese staff encouraged the use of a culturally appropriate name for the intervention, Kokajjiriri (Kokajjiriri hereafter). The word Kokajjiriri means to raise a child.

To design culturally appropriate interventions among Marshallese and other Pacific Islanders in the US, it is necessary to understand Marshallese culture and the historical relationship between the Republic of the Marshall Islands (RMI) and the US. The RMI are a group of small low-lying islands located in the Pacific. Following World War II, over a 20-year period, the US conducted nuclear testing in the Marshall Islands, detonating more than 65 nuclear bombs around the Marshall Islands. ²⁸ As a result of the nuclear fallout, dangerous conditions contaminated fish and soil, thus contaminating the food supplies. Historically, the Marshall Islanders were agriculturally self-sufficient living on a diet comprising local plants and fish.²⁹ Disruption of cultural practices and changes in food supplies were also due to islanders being relocated, with families separated by the need to move from contaminated areas. There were major restrictions placed on consumption of native foods due to ground contamination from nuclear fallout.²⁸ The inability of Pacific Islanders to be self-sufficient in food production led to significant diet changes which included canned meats (e.g., SPAM^®), ramen, white rice, and other shelf-stable products that were shipped from the US government.^30,31 Recent research with Marshallese communities in Arkansas continues to support the preferences for and consumption of these foods, which must be considered in developing culturally appropriate nutrition recommendations for this group.^16,17 The purpose of this study is to assess the acceptability and feasibility of a culturally adapted group-based pediatric intervention, Kokajjiriri, with Marshallese mothers to improve nutrition and reduce childhood obesity.

METHODS

Research Design

A multi-methods design was used to culturally adapt the Kokajjiriri intervention for Marshallese mothers in Arkansas. This approach was chosen to ensure a more comprehensive understanding of participants’ perceived acceptability and feasibility of Kokajjiriri. In the first phase, we collected pre-intervention 24-hour dietary recall data from Marshallese mothers of young children, providing a rich characterization of the dietary patterns of the target population.^32,33 Phase 2 used these data and a culturally grounded theory approach^34,35 to adapt Kokajjiriri for Marshallese mothers to assess the acceptability and feasibility of the culturally adapted curriculum.

This study used a culturally grounded approach to culturally adapt the Kokajjiriri intervention using the dietary recall results from n=20 Marshallese mothers with children under the age of 12 months and Marshallese stakeholder input.^34,35 These cultural adaptations included: 1) using words, images, and idioms that represent Marshallese culture; 2) using food models of traditional Marshallese foods; and 3) using a Marshallese facilitator in a centralized location to their community. Culturally grounded approaches utilize methods that place the culture and social context of the targeted population at the center of the intervention. ^34,35 Methods are used in which curriculum components evolve from the ground up (i.e., from the worldviews, values, beliefs, and behaviors of the population that the program is intended to serve).^34,35 In using a culturally grounded approach, we used the guidance of Marshallese staff in every step of the research process from conceptualization to dissemination. It is essential to include community members for a successful adaptation and implementation. While time and cost investments of the adaptation process are high, they may be necessary for unique populations in which available evidence-based practices have limited applicability or generalizability. Culturally grounded approaches to adapt obesity prevention evidence-based interventions have been demonstrated as highly effective with Native Hawaiian and Pacific Islander adult populations.^23,36,37

Kokajjiriri Intervention Overview

The Kokajjiriri intervention was designed to occur over nine group sessions from six weeks through 12 months of the newborn’s life. Group visits are 90–120 minutes each and follow a unique structured curriculum that incorporates standards of care around parenting, infant feeding practices, and nutrition. In the sessions, credentialed providers offer: 1) brief one-on-one health exams; 2) facilitated discussion around parenting topics; and 3) a 10–15-minute closing for group questions.

The adapted Kokajjiriri intervention was then tested in three sessions focused on nutrition topics to understand acceptability of the culturally adapted program. These three sessions focus on infant feeding, the introduction of complementary foods that are high in nutrients important for child growth and development, and general healthy eating for breastfeeding mothers. These sessions were piloted at a community-based organization by a trained bilingual Marshallese study staff.

Recruitment, Consent, and Retention

Recruitment took place via trained bilingual research staff. Eligibility criteria included being female, being an adult (≥18 years), being self-reported Marshallese, and having children under 12 months of age. Potential participants who met the eligibility criteria were asked if they would like to participate and were recruited by trained bilingual Marshallese study staff.

Potential participants who met the eligibility criteria were offered the opportunity to join the study and complete the consent process prior to participating in the intervention. Trained bilingual study staff conducted the consent process. The bilingual study staff read the consent aloud to the participants in the participant’s language of choice (English or Marshallese). Participants were given the opportunity to have their questions answered prior to consent.

The research team used an engaged approach to collaboratively develop a retention plan with Marshallese stakeholders. The retention plan specified that all study staff responsible for retention will be bilingual (Marshallese/English). Bilingual study staff obtained each participant’s contact information and preferred method of contact. Bilingual study staff also collected contact information for at least two relatives and asked participants for permission to contact their relatives if needed. Confidentiality rules were followed, and no participant information was provided to relatives. Before each data collection visit, bilingual study staff contacted study participants about the upcoming data collection visit. Each Marshallese participant received a $50 gift card upon completion of both the survey and focus group session. This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving research study participants were approved by the University of Arkansas for Medical Sciences Institutional Review Board (Protocol #274752).

Development of Instruments

The quantitative surveys and focus group guides were developed with extensive input from Marshallese stakeholders and are specific to this study (see Supplemental File 1—Interview Guide). After qualitative and quantitative data collection instruments were confirmed by community stakeholders, bilingual study staff translated the instruments into Marshallese. After translations were complete, the community-engaged research (CER) team met with the bilingual study staff and conducted two mock data collection events over the course of three months. These mock data collection events served as training for the bilingual study staff and allowed the team to evaluate any challenges in cultural nuance, comprehension, and translations. All data documents went through three iterations. When conducting CER, it is critical to allow for time and flexibility to guarantee the study and data instruments are culturally appropriate to the target community.³⁸

Data Collection

Data for phase two of this study were collected from October to November of 2023. The survey was implemented using the web-based Research Electronic Data Capture (REDCap)³⁹ system and took approximately 10–15 minutes to complete. All focus groups were conducted by trained study staff. The qualitative focus groups were audio recorded and then transcribed verbatim. A semi-structured focus group guide with open-ended questions was used to encourage providers to speak candidly while maintaining consistent inquiries across focus groups. The focus groups took approximately 30 minutes and were conducted in person. All identifiable information was kept confidential, and all participants were provided an ID number. Seventeen (n=17) participants participated in three focus groups until saturation was achieved. Saturation in qualitative research is when, through the course of interviewing, you notice the same themes emerging repeatedly.⁴⁰

Data Analysis

Qualitative data were managed using MAXQDA12 software.⁴¹ Content analysis was used to analyze qualitative data. The authors started with initial coding that consisted of naming each data segment with short summations of emergent themes. This process helped organize the data for focused thematic coding. The focused thematic codes that emerged were used to identify and develop the most salient thematic categories of the data, which became the thematic codes.⁴² There were two primary coders and one confirmation coder. The authors coded transcripts for the emergent thematic codes and collaboratively discussed the themes to ensure scientific rigor and intercoder agreement. Codes were organized into a codebook. Quantitative survey data were analyzed with descriptive statistics using SPSS.⁴³

QUANTITATIVE RESULTS

Participant demographic results are presented in Table 1. Over half of the participants were either married or in an unmarried couple (n=11; 55%). Most of the participants were high school graduates or had some high school education (n=14; 70%), and over half were either working from home/taking care of their families or out of work (n=11; 55%). Almost all the participants were born in the Marshall Islands (n=16; 80%) and were enrolled in WIC (n=16; 80%). The median age was 25.5 with a median of 4 children and a median of 4.5 adults living in the home. Participants ranged from living in the US between 2–33 years with a median of 11.5.

Table 1.

Pre-Intervention Survey Demographics (N = 20)

	Frequency	%
Relationship Status
Single	7	35%
Married	7	35%
Unmarried couple	4	20%
Missing	2	10%
Education
Some high school	5	25%
High school graduate	9	45%
Some college or technical school	1	5%
College graduate	3	15%
Missing	2	10%
Employment Status
Out of work for less than 1 year	1	5%
Out of work for 1 year or more	2	10%
Taking care of your family/home	8	40%
Employed	7	35%
Missing	2	10%
Birthplace
Marshall Islands	16	80%
US-born	1	5%
Other	1	5%
Missing	2	10%
Enrolled in WIC
Yes	16	80%
No	2	10%
Missing	2	10%
Age (Years)
Range	18 - 40	---
Median	25.50	---
Lower Quartile	22.25
Upper Quartile	32.75
IQR	10.50	---
Number of Children
Range	2 - 9	---
Median	4.00	---
Lower quartile	3.00
Upper quartile	6.00
IQR	3.00	---
Number of Adults
Range	2 - 8	---
Median	4.50	---
Lower quartile	4.00
Upper quartile	6.00
IQR	2.00	---
Length in the US (Years)
Range	2 - 33	---
Median	11.50	---
Lower quartile	7.00
Upper quartile	18.25
IQR	11.25	---

Notes: Two participants did not complete the Demographic Survey. This explains the missing data represented in the table.

Post-intervention survey results are presented in Table 2. Only 17 participants completed the intervention of the initial 20 that completed the dietary recall (Phase 1) of the intervention. All the participants reported that the information was useful in helping them think about nutrition for their baby/child (n=17; 100%). Almost all the participants said that the information provided was relevant to Marshallese women like them (n=16; 94.12%), and all participants said that the information provided was accurate (n=17; 100%). All participants reported that they learned new information about nutrition for their baby/child (n=17; 100%). Almost all participants agreed that the time and length of the session was convenient (n=16; 94.12%), and all participants said they found the group parenting sessions to be suitable for Marshallese mothers (n=17; 100%). Almost all participants said the location of the session was convenient (n=15; 88.24%), and all participants said the location of the session was safe (n=17; 100%). All participants said that the facilitator was warm and understanding, competent and well-trained, and knowledgeable (n=17; 100%).

Table 2.

Post-Intervention Survey (N=17)

	Frequency	%
The information was useful in helping me think about nutrition for my baby/child.
Yes	17	100.00%
No	0	0.00%
The information provided was relevant to Marshallese women like me.
Yes	16	94.12%
No	1	5.88%
I am confident that the information provided is accurate.
Yes	17	100.00%
No	0	0.00%
I learned new information about nutrition for my baby/child.
Yes	17	100.00%
No	0	0.00%
The time and length of the session was convenient.
Yes	16	94.12%
No	1	5.88%
I found the group parenting sessions to be suitable for Marshallese mothers.
Yes	17	100.00%
No	0	0.00%
Location of the session was convenient.
Yes	15	88.24%
No	2	11.76%
Location of the session was safe.
Yes	17	100.00%
No	0	0.00%
Facilitator was warm and understanding.
Yes	17	100.00%
No	0	0.00%
Facilitator was competent and well-trained.
Yes	17	100.00%
No	0	0.00%
Facilitator was knowledgeable.
Yes	17	100.00%
No	0	0.00%

QUALITATIVE RESULTS

Four themes emerged from the qualitative data: 1) Lactation Support; 2) Introducing Healthy Solids; 3) Rice Portion Control; and 4) Finding Resources.

Lactation Support

Participants described wanting more information around lactation. For example, one participant said, “Lactation. She was talking about colostrum. Most of us didn’t know that you were supposed to give it to your baby the first couple of days.” (Participant 11) One participant said, “In Marshallese culture we call colostrum ‘dren nana’ which means dirty water that you should throw out.” (Participant 3) Another participant requested “more information about storing milk.” (Participant 10) Participants also described needing a breast pump. One participant said, “I want access to a pump. I’ve never used one.” (Participant 9) Another participant agreed and said, “Me neither.” (Participant 8)

Introducing Healthy Solids

Participants also discussed learning about when to introduce solids to infants. One participant said, “For me it’s just learning stuff. Today’s topic was about the foods. I guess knowing when to give food to the baby. I didn’t know that six months is when you start giving [solid] food to the baby.” (Participant 17) Participants also discussed an understanding of healthy foods to introduce to their infants. One participant said:

Our Marshallese diet for our babies is usually what we have in our kitchens. I know a lot of us grew up with Top Ramen because it’s easy to chew or break it up and spoon it. Today I learned about scraping of the apples. (Participant 16)

Another participant described learning methods to improve the health of the foods she is serving her infant when she said, “As for the canned foods, they are telling us to wash them first because they are salty.” (Participant 9) Importantly, the participants described learning healthy food education that could go beyond their infants and can be used for the whole family. One participant said, “Yes, of course because it’s helpful. It’s not just about showing us what food to feed our babies but what we should eat too. And it’s not just about how to feed the babies but how to take care of the babies.” (Participant 12)

Rice Portion Control

Participants described learning about portion control specifically to white rice, which is a staple in the Marshallese community. One participant said, “We learned about balance. Sometimes we may only think that our food is good when we bring the rice. Today we learned about the right amounts of rice.” (Participant 9) Participants described a change in understanding of the right proportion of rice to put on a plate for both adults and infants. One participant said:

I am a hands-on visual person so everything she had on her table [referring to food models] was perfect. I liked how it showed the portions of it. Growing up if you had your plate of rice, it’s not in a corner it’s the whole plate. Marshallese love rice. (Participant 17)

Another participant agreed and said:

Usually, you need to finish your plate of food to be strong. Growing up it wasn’t that corner; it was the whole plate. It is not as big as an adult plate, but it was more than what that corner should have been. (Participant 16)

The concept of finishing your plate was described as also a means of not wasting food. One participant said, “Maybe at someone else’s home but not at your own home. But you don’t want to waste food. You know back home food is expensive and you don’t want to waste what you have.” (Participant 17)

Local Resources

Lastly, participants discussed the need for more local resources support. One participant said:

Having more resources. When I asked if I could take my baby [referring to adopting her baby] I was kind of squirming. I didn’t know what to do. I had nothing. No childcare, no diapers, no resources. I think if there was more of us, they would say ‘oh I need this or I need that.’ When I go to ACOM [Arkansas Coalition of Marshallese] they have been helping with formula at their pantries. You know, mine and my partners budget is just about us. Now we are switching with how we spend stuff because we have added another person. That person needs a lot of things. A lot of the programs I talked with [referring to government programs] our household income is above the scale, so a lot couldn’t help. (Participant 16)

Participants also focused on the need for better childcare programs. One participant said, “Access to childcare programs. We need help because we can’t afford help with childcare.” (Participant 7)

DISCUSSION

This is the first study to assess the acceptability and feasibility of a culturally adapted group-based pediatric intervention, Kokajjiriri, with Marshallese mothers. Almost all the participants said that they found that the culturally adapted intervention provided helpful nutrition information that was relevant to the Marshallese community. Almost all the participants found the location to be convenient, and all the participants found the facilitator of the sessions (female Marshallese Community Health Worker) to be warm and understanding, culturally competent and well-trained, and knowledgeable. Similarly, previous literature on group pediatric interventions focused on postpartum and nutrition education have also demonstrated acceptability and feasibility for families with barriers including lower education levels and transportation barriers impact postpartum education support on breastfeeding and nutrition for infants.^21,44

This study adds new insights into areas of educational needs specific to the Marshallese. The participants described a lack of knowledge around the importance of giving their infants colostrum, supporting previous literature findings that education as a significant barrier in initiation and prolonged breastfeeding.⁴⁵ Previous research has shown that women who do not give their infants colostrum in the first few days of life typically supplement with formula, which can create low milk supply and create nipple confusion for the infant, potentially complicating exclusive breastfeeding. ⁴⁶ Previous studies have identified that Marshallese women desire to exclusively breastfeed their infants but experience numerous barriers, including perceived low milk production and quality. ^12–14 This finding of the participants’ lack of the knowledge about colostrum adds to the literature and provides an opportunity to better educate on the importance of colostrum for successful exclusive breastfeeding experiences for Marshallese women, which can reduce the odds of obesity in infants and children. Additionally, the participants described a lack of knowledge around the importance of waiting until six months to introduce complimentary foods. Early introduction of complimentary foods prior to six months is associated with higher odds of obesity later in life.^5,6

It has been well documented that rice is a staple in the Marshallese community and that it is typically a first food given to infants.^14,16,17,31 Our dietary recall results identified white rice as the top food category contributing to total energy intake among Marshallese mothers with infants under the age of 12 months.³³ However, it is important to provide culturally appropriate nutrition education about rice due to the historical trauma the Marshallese have experienced. In consultation with our Marshallese stakeholders, we were encouraged to educate on rice portion control rather than exchanging white for brown rice or suggesting a different food entirely. The participants described a better understanding of portion control of rice due to the food models and hands-on approach used in the sessions, suggesting this is a culturally appropriate way to provide nutrition education on this topic. Lastly, the participants described a need for more local resources specific to childcare support, diapers, and formula.

LIMITATIONS

Several limitations need to be acknowledged. The findings of this research are based on the perspectives of a small sample of Marshallese women. The study was conducted in one US state, and the results may not be generalizable to other locations. However, our rigorous in-depth qualitative process has elicited findings that may provide valuable insights into the importance of cultural adaptation of health behavior interventions to increase acceptability and feasibility prior to implementing the full intervention. Further, although this study is specific to Marshallese, evidence-based interventions designed for Pacific Islanders may inform work with other disenfranchised and indigenous populations, thus increasing the generalizability of the proposed research.^34–38

CONCLUSION

The results from this culturally adapted nutrition component of a group-based pediatric intervention, Kokajjiriri, will be used to inform future adaptations and implementation of the full intervention for Marshallese women and children to improve nutrition and prevent childhood obesity.

Key Messages.

• Health interventions not aligned with the cultural values, perspectives, and preferred modes of living of the target population are presumed to be less effective than culturally responsive interventions that account for these factors.

• To design culturally appropriate interventions to promote healthy growth and development of children under the age of 12 months, it is necessary to better understand the current diets within the families.

• This is the first study to assess the acceptability and feasibility of a culturally adapted group-based pediatric intervention, Kokajjiriri, with Marshallese mothers to improve nutrition and reduce childhood obesity.

Data availability:

The deidentified data underlying the results presented in this study may be made available upon reasonable request from the corresponding author, Dr. Britni L. Ayers, at blayers@uams.edu.