Family-Level Factors That Influence Physical Activity among Hispanic Adolescents at Risk for Type 2 Diabetes

Abstract

Objective

Family-based health promotion and disease prevention strategies are recommended as best practice; however, there is limited knowledge regarding the family-level factors that influence modifiable behavior risk factors like physical activity (PA) among Latinx adolescents. This study addressed this knowledge gap by using qualitative methods to identify perceptions of family-level factors that influence PA among Hispanic youth.

Methods

We conducted semistructured, open-ended interviews with 20 Latinx adolescents (14-16 years) with obesity (body mass index ≥95th percentile) to identify their perceptions of how family influences PA. Content analysis was used to identify emergent themes, which were then compared across demographic factors, other identified themes, and participant-identified personal values.

Results

Fourteen adolescents (70%) perceived family factors that facilitated PA. They described family support received as doing PA together, modeling PA, and providing motivational or financial support. Ten adolescents (50%) viewed family as a barrier, more often citing active barriers such as family responsibilities. Across demographics, youth with divorced parents and parents who reported more working hours (≥40 hours per week) perceived their family as less supportive of PA. Additionally, perceptions of family support shaped self-identified health values (eg, being healthy and fit).

Conclusions

Family-based obesity prevention strategies should leverage factors that facilitate PA and should be designed to consider factors that serve as barriers to PA among this age group and population. Given that there is little guidance on the development and implementation of family-based obesity-prevention strategies, findings from this study will inform the development of future family-based prevention opportunities among high-risk youth and families.

Introduction

Obesity rates among adolescents in the United States continue to increase, and Hispanic adolescents experience higher rates of obesity compared with the general population (27% vs 21%), placing them at increased risk for obesity-related diseases like Type 2 diabetes.1,2 Family-based interventions are considered best practices for effectively addressing obesity disparities among ethnic minority populations and are recommended in obesity prevention guidelines published by the US Preventive Services Task Force, the Centers for Disease Control and Prevention, and the Centers for Medicaid and Medicare Services.3-5 Higher family functioning has been linked to increased physical activity (PA), sufficient sleep, and decreased body mass index (BMI) among general populations of adolescents in the US.6 As a result, it is widely acknowledged that family plays a central role in the development of lifestyle.

Family-based interventions may be particularly relevant for engaging Hispanic adolescents given that familism, or familismo, is a strong cultural value in the Latinx community.7 Familismo highlights the importance of maintaining close relationships with family and of fulfilling familial obligations.7 Among Latinx adolescents, behavioral manifestations of familismo include assuming additional responsibilities such as interpreting for parents, caring for siblings, performing household chores, and striving to reflect positively on their families through academic excellence and respect for parental authority.7

Research conducted among Latinx adolescents suggests that a strong sense of familismo is a protective factor against various adverse outcomes such as depression, suicide, substance use, and externalizing behaviors.7 However, the impact of familismo varies with context and outcome of interest. For example, among populations under stress, familismo may be detrimental or insufficient in mitigating the adverse effects of external factors such as discrimination. Furthermore, adolescent Latinas with a previous suicide attempt often attribute their attempt to “sacrificing” themselves because of their inability to fulfill family responsibilities.7 Similarly, obesity may present a distinct context in which the effects of familismo vary. Children with obesity are subject to social exclusion and weight-based discrimination, subsequently contributing to depression, anxiety, and stress.8-10 Therefore, familismo alone might not sufficiently buffer these negative impacts for Latinx adolescents with obesity. This could serve as an explanation for the mixed results observed in family-based interventions targeting obesity in Latinx youth, with some studies reporting no significant changes in BMI following intervention.11

One prevalent approach used in family-based interventions is fostering family social support (FSS) for improving obesity-related behaviors like PA.12 Previous research has shown that FSS for PA is associated with increased activity among Latinx youth.13-15 In the context of the self-determination theory (SDT), receiving social support from one’s family can build a sense of relatedness, which can help internalize and integrate behaviors like PA into one’s sense of self.16

Additionally, the SDT holds that when a behavior like PA is aligned with self-identified personal values, that behavior is more likely to become initially adopted and maintained over time.17,18 Previous research has shown that having self-endorsed values related to activity or having an active lifestyle (such as being healthy and fit or avoiding disease) is associated with increased PA behaviors.17,18 Personal values, and their links with behavior, can be developed on the individual level as well as the interpersonal level in the context of family. For example, having shared values is a core characteristic of families that are “health promoting” to their children.19 Thus, receiving FSS can shape health-related values that can significantly impact health behaviors and outcomes.

Although family-based approaches are recognized as critical for addressing obesity disparities, recommendations for implementing family-based strategies do not provide specific guidelines for how to engage families in interventions, and there is a significant gap in understanding how to foster FSS. Our limited knowledge on how to operationalize FSS, and the contextual factors that influence FSS in this high-risk population, limits our ability to effectively intervene to address obesity disparities. This is particularly true regarding family-based interventions for adolescents with obesity, as they are in a unique developmental period when behaviors and values are being established. Adolescence is characterized by growing autonomy, individualizing oneself from parental figures, and developing individualized cultural, familial, and health-related values. However, this developmental period may contrast with the expectations and obligations of familismo, suggesting that this life stage may have an impact on the desire, needs, and preferences for FSS among Latinx adolescents with obesity.7,20

Given the lack of consensus on the effects of familismo on exercise behaviors and the approach to designing family-based interventions among Latinx youth with obesity, it is essential to explore the perspectives of family factors on PA in this high-risk, yet understudied, population. The primary objective of this qualitative study was to identify themes regarding perceptions of FSS for PA in a sample of Latinx adolescents with obesity. The secondary objective of this study was to examine differences in perceptions of FSS across family-level contextual factors. Lastly, the tertiary aim was to examine the relationship between perceived FSS for PA and self-endorsed values. The results of this study will aid in identifying family-based facilitators and barriers that can be leveraged or addressed in family-based health promotion and disease prevention opportunities among Latinx adolescents with obesity.

METHODS

Recruitment and Participants

This qualitative study builds on existing work from a National Institutes of Health funded project aimed at identifying barriers and facilitators to health behaviors among Hispanic adolescents with obesity to inform the development of a digital diabetes prevention program among this population. Participants were recruited from the greater Houston metropolitan area through collaborations with local pediatric clinics (n=5), the US Department of Agriculture Children’s Nutrition Research Center volunteer research database (n=3), community organizations (n=21), and word of mouth (n=7). Interested participants were screened for inclusion/exclusion criteria via phone. Participants met inclusion criteria if they self-identified as Hispanic or Latino origin, met obesity cutoffs (BMI ≥95th percentile and <120% of 95th percentile), between the ages of 14 and 16 years, and owned his or her own phone. Exclusion criteria included taking a medication (eg, steroids) or previous diagnosis of a condition (eg, sleep apnea) that influences activity, sleep, and/or cognition; recent hospitalization or injury that interfered with normal PA; pregnant; currently enrolled in an exercise program; or currently using a personal activity monitoring device.

Interview Procedures

Semistructured, in-depth interviews were conducted by 2 trained interviewers using private Baylor College of Medicine Zoom meeting rooms between November 2021 and January 2022. All interviews were scheduled at the convenience of the participant, conducted in English, and audio recorded through Zoom. Participants were compensated $25 in appreciation of their participation. All study procedures and materials were approved by the Institutional Review Board at Baylor College of Medicine (H-49195) and participants provided written informed parental consent and youth assent prior to their participation.

Interview Guide

A semistructured interview guide was developed by the investigative team. Interview questions were informed by the SDT, the guiding theoretical framework for the parent study. The final interview guide included 9 open-ended interview questions about encouragement of PA, barriers to PA, personal values, and autonomy. These questions were designed to elicit information on perceived facilitators and barriers to PA, participants’ personal values related to health, and perceptions of constructs associated with the SDT: autonomy, competence, and relatedness.

Qualitative Data Analysis

Audio recordings were 35 minutes long on average and were professionally transcribed using a transcription service. Transcripts were coded in the qualitative analysis software program NVivo (version 12.5; QSR International). Coders independently read and coded each transcript using content analysis21 to identify repeated ideas and important quotes regarding FSS for PA and family-level social factors that influence health behaviors. This step generated 2 initial lists of codes and definitions within NVivo. Coders then met to compare and contrast codes and either confirmed, modified, or rejected each code to yield one guiding codebook. Coders then independently read the transcripts for another cycle, met to compare codes, and made additional edits. These cycles were repeated until no further themes were identified. Codes were then organized into groups of related family-level social factors that informed emerging themes and associated subthemes.

Descriptive analyses on participant demographics were calculated. Youth were categorized into 2 groups: (1) youth who reported receiving FSS for PA and (2) youth who discussed barriers to FSS of PA. Some youth (n=5) were categorized into both groups as they discussed instances in which family was supportive and instances in which there were barriers to family support. Cramer V nonparametric correlations and χ² tests of association were used to examine associations among individual and demographic variables across these 2 groups. Independent t tests were used to identify differences in demographic variables across both groups.

Results

Demographic data and characteristics are presented in Table 1. In total, 20 Hispanic adolescent girls (n=11) and boys (n=9) participated in 40- to 70-minute semistructured interviews. The parents of the adolescent participants self-reported the adolescents’ ethnicity, parent relationship status (separated, female-headed, male-headed, or other), parental education level, parental employment, and household income. The sample consisted primarily of adolescents from lower socioeconomic backgrounds, and the majority (85%) identified as Mexican American.

Table 1.

Demographic characteristics

Variable	Girls (n=11)	Boys (n=9)
Adolescent age, y	15.1	15.2
Adolescent mean BMI, kg/m2	37.3	35.7
Adolescent Hispanic group, %
Mexican	91	78
Salvadoran	0	22
Honduran	9	0
Parental relationship status, %
Two-parent household	63.6	33.3
Female-headed household	18.1	33.3
Separated parents	18.1	22.2
Other	0	11.1
Parental highest level of education, %
Some high school or less	54	56
High school or equivalent	27	11
Some college/technical school	18	22
Postgraduate study	0	11
Parental employment, %
Yes	37	89
No	63	11
Household income, %
Less than $10,000	9	0
$10,000-19,999	18	11
$20,000-29,999	27	22
$30,000-39,999	27	22
$40,000-49,999	9	22
$50,000-59,999	9	11
Greater than $60,000	0	11

BMI, body mass index.

Qualitative Insights

This section provides an overview of qualitative insights regarding youth’s perceptions of FSS for PA.

Perceptions of FSS for PA

A total of 14 participants reported that they received FSS for PA. The types of social support discussed by youth were categorized using a framework developed by Beets et al,22 who identified 4 major categories of social support for PA-related behaviors in youth: informational, conditional, instrumental, and motivational. Informational support includes providing knowledge, advice, or guidance that encourages more engagement in PA (eg, transportation, gym membership). Conditional support is defined by physical presence or the direct involvement of the family in the PA. Motivational support is defined as verbal and nonverbal support for PA. Based on readings of transcripts, we identified 2 additional categories: observational and functional support. Observational support includes being encouraged by observing family members engage in health behaviors. Functional support was defined as performing jobs or tasks with family members that include support on nontraditional, leisure, or occupational physical activities like gardening, construction, or operating machinery. Conditional and observational, followed by motivational, were the most commonly reported types of FSS. Example quotes representing each type of support are included in Table 2. Conditional and observational support were the most commonly cited supports for PA, which is in line with previous literature suggesting that parental exercise behaviors and modeling of PA are associated with increased PA in children.23,24 A significant portion of participants also cited motivational support, which is supported by evidence suggesting that verbal praise of PA by parents is also associated with increased PA in the child.23,25-27

Table 2.

Exemplar quotes

Category	Quotes
Encouragement categories
Informational (n=1)	“... [my aunt] watches workout videos …She buys the little workout outfits just like, ‘Hey, look, this looks cute. Let’s go show it off and stuff.’ So … She motivates me a lot to be active.”
Conditional (n=6)	“Like when my sister says she’ll take me I know I’m going to work out with her and she’s going to be there. I think just having like, someone there is like, really good for me.”
Motivational (n=4)	“My family [encourages me], ’cause then I know that they believe in me, and they know that I won’t…like I’m not the type of person that will give up …they cheer me on. And I have more confidence for that.”
Observational (n=6)	“[My mom]’s always about taking care of her health... And you know, at some point, it crossed my mind that, “Hey, I need to do something.” So I don’t end up bad with my health or something.”
Functional (n=2)	“Since I was young, I would work with my dad, I mean my grandpa, in heavy machinery. So that’s what like, made me strong. And like, I could carry like, a lot of weight”
Barrier categories
Active barriers
Family responsibilities (n=5)	“Most days I get home from school and I have to watch my niece to pretty late until maybe 6 where—and usually is already dark… but it’s always I have to babysit or I have to help my mom.”
Protection (n=2)	“My mom doesn’t really like me going for a walk in this neighborhood, because it’s a little rough.” “My brother lifts real heavy weights. And my mom told me not to do that because I’m gonna hurt myself.”
Passive barriers
Lack of conditional support (n=1)	“[My family] arrives at home around 7:00-7:30. …the sun’s already down. So it makes it difficult to be physically active with them.”
Lack of instrumental support (n=2)	“...since my parents don’t really have like the money to help pay for it, or gas often to drive around, so I can find a place to exercise.”
Lack of motivational support (n=1)	“Since I’ve been big my entire life and [my family] is more I would say ‘skinnier.’ They tend to make fun of me. They call me names and stuff.”

Perceptions of Barriers to FSS

Overall, 10 participants mentioned barriers to FSS for PA (Figure 1). Barriers were categorized as active or passive. Active barriers included actions that limited PA because of the presence of an action, belief, or expectation. Examples of active barriers included family responsibilities such as household chores and parental safety concerns that limited outdoor play. Passive barriers included the adolescents’ perceived absence of support categories listed above including the absence of conditional, instrumental, and motivational support. Exemplar quotes for barriers are also included in Table 2.

Figure 1.

Thematic network

In regard to family-level factors that impacted perceptions of FSS, independent t tests showed a difference between youth who reported family as supportive and youth who reported barriers to family support with regard to age (t=−1.52, df=8.14, P=.08), and this difference approached significance. Youth who reported family as supportive were slightly younger (mean=14.75 years) compared with youth who reported more barriers to family support for PA (mean=15.25 years). Additionally, parents of youth who discussed barriers to family support reported that they worked an average of 42.35 hours per week, compared with youth who reported family as supportive, who worked an average of 19.75 hours per week (t=2.12, df=18, P=.05). We also found that youth discussing more barriers to FSS for PA was positively correlated with parent marital status (Cramér V=0.66, P=.07) and this finding approached significance (Cramér V=.656, P=.072, Figure 2). About 70% of youth with married parents and 75% of youth with separated parents discussed receiving FSS for PA. In contrast, 100% of youth with separated parents discussed barriers to FSS for PA.

Figure 2.

Associations between adolescent perceptions and family as an encouragement or barrier to physical activity (PA). LR, likelihood ratio

When examining the impact of group membership and self-endorsed values, a χ² test of associations revealed an association among youth who discussed family as supportive of PA and valuing success (likelihood ratio, 4.74; df=1, P=.029), being good at something (likelihood ratio, 3.44; df=1, P=.06), and being healthy and fit (likelihood ratio, 3.29; df=1, P=.07) (Figure 2). No other significant associations were observed.

Discussion

Few studies and guidelines provide guidance on how to incorporate FSS into family-based disease prevention interventions, and even fewer have explored the needs and preferences for FSS for behavior change during adolescence among a high-risk pediatric population. This study provides novel information regarding perceptions of FSS for PA among Hispanic adolescents with obesity and identifies additional family-level factors that influence perceptions of FSS. Information gleaned from this study can be used to guide and strengthen existing family-based strategies among this growing high-risk population.

In our study, the majority of participants who perceived family as supportive of PA mentioned receiving conditional, observational, and motivational support for PA, signifying that most adolescents perceive higher support when family members engage in PA with them, model PA behaviors, or encourage PA. This finding is consistent with previous research, which shows that parental modeling, participation in PA, and providing reinforcement increase PA in children.15 Therefore, interventions that leverage family support to increase PA should focus on creating opportunities for family members to be active together, educating parents on the importance of modeling healthy behaviors, and emphasizing the importance of parents motivating youth towards behavior change through verbal and nonverbal praise and encouragement. Interestingly, instrumental support was not mentioned by as many adolescents, although it has been significantly correlated with PA in other studies.15,22,28

Family responsibilities were one of the most commonly mentioned barriers to FSS for PA. Youth from lower-socioeconomic-status backgrounds typically have a larger role in assisting with family responsibilities that are vital to maintaining stability in the family unit, such as cooking or caring for other siblings.29 Future interventions should focus on ways that families can work together to fulfill family responsibilities while engaging in PA. For example, for youth who carry the responsibility of babysitting younger siblings, a potential strategy for increasing PA may be to suggest games and outdoor play that can be done safely with siblings. In the clinical setting, motivational interviewing can serve as an approach to help family members uncover active and passive barriers to social support. This method can also guide families toward feasible solutions that are tailored to their specific context and needs.

There were no differences in whether participants perceived their family as being supportive or a barrier to PA between sexes. This finding is not consistent with previous research that found girls often perceive less family support and/or limited opportunities for support.28,30,31 This can possibly be explained by our small sample size or by other factors not accounted for in our study, such as birth order. For example, firstborn Latina daughters are frequently expected to take on caregiver roles in their families and are often seen as a third parental figure.32 There was also no difference in the perception of family support across income categories in our participants. Previous research describes decreased support for PA in children and adolescents from lower-income households.25,33 We may not have detected differences across income given that this was a predominantly low-income population, and as such there was little variation in income. There was a minimal difference detected across ages, with younger participants more likely to report family as encouraging. Prior research found that the association between family support and PA decreases as adolescents age, while the influence of peer support increases.34 Our findings support that younger adolescents may perceive and require different types of support compared with older adolescents. This could indicate the need for extrafamilial support in later adolescence. However, future research is needed to understand these age-specific differences in familial support in order to better tailor family-based interventions.

In this study, 100% of youth with separated parents described aspects of their family as barriers to PA. This suggests that family-based strategies should consider family structure and may need to specifically address support needs for youth whose parents are separated. Additionally, adolescents whose parents worked greater than 40 hours per week were more likely to report barriers to family support. Prior studies that involved parent interviewing revealed that Latino parents similarly identified long work schedules as barriers to engaging in PA, both on their own and with their children.35,36 Contextual factors such as marital status and work cannot easily be altered and arguably should not be intervened in. However, the findings from this study highlight the significant need for family-based strategies that are adaptable to fit the needs of different family environments. These findings suggest that family-based approaches that identify strategies for providing FSS in a manner that is feasible despite potential contextual barriers may be critical for reaching this population.

Study findings also suggest that FSS may influence self-endorsed values among youth. Although it remains unclear whether having a family that encourages PA directs adolescents to hold specific values (e.g., valuing success, being good at something, and being healthy and fit), or whether families that hold and teach these values also tend to be more encouraging of PA, it is clear that perceived familial support and adolescent values are interconnected. Given that personal values developed during adolescence are predictive of adult behaviors, PA interventions that foster health values in adolescents could enhance the sustainability of healthy behaviors. When a behavior aligns with a self-endorsed personal value, the behavior is more likely to become internalized and sustained.12 Future interventions should consider the use of evidence-based strategies to develop personal and shared values that support healthy lifestyle behaviors.

Although family-based interventions are the gold standard for obesity prevention among pediatric populations, engaging family members in family-based interventions remains challenging.35,37 Similar to previous studies, this study highlights how contextual complexities may influence key intervention components like FSS for PA, making it difficult for families to fully engage in family-based strategies. An intervention’s efficacy is influenced by family structure, routines, culture, and community resources, which all interplay to create a unique context for each family and their health behaviors.38 This study and the findings of others point to the importance of perceiving the family as a complex social system and underscore the need for a deeper understanding of the structural and social mechanisms that influence parent-child relationships, and therefore impact health behaviors.32,33 There is a need for more specific family-based intervention guidelines and recommendations that are considerate of the complexities within the family units of various racial and ethnic subgroups and can foster FSS in a responsible manner that does not lead to unintended outcomes or lay blame for the actions of any one member of the family system.

Strengths and Limitations of This Study

Although many studies have explored the importance of family-centered interventions in Latinx populations, most have focused on adult or early childhood participants. This study addresses a gap in the literature by eliciting qualitative perspectives from Latinx adolescents with obesity and from low-income backgrounds, a vulnerable population that is often left out of studies. However, the limitations of this study can also include the narrow scope of the focus population. Participants were recruited at community partnership events for resource distribution, such as food drives; thus, many were experiencing food insecurity or other socioeconomic barriers to health. Thus, perspectives elicited in this study are not representative of all Latinx adolescents across income levels and socioeconomic statuses. However, insights gleaned from this study may be particularly helpful for reaching high-risk, low-income populations that are often underrepresented in obesity research. Furthermore, a great majority of adolescents were of Mexican origin; therefore, it is likely that the themes identified are not representative of Latinx individuals from different countries of origin.

Conclusions

Latinx adolescents in this study perceived conditional support, motivational support, and modeling as common types of support provided by family to promote PA. These support constructs should serve as critical inputs in future family-based interventions among this population to encourage the development of healthy PA behaviors. Our findings indicate that FSS is influential on health behaviors even throughout the developmental stage of adolescence and impacts the formation of self-endorsed values during this life stage. Our study also highlights the importance of considering additional family-level factors, such as family structure or parent work hours, and the role that these factors play in a family’s ability to support their adolescent in behavior change. This study provides useful information that can inform the design and development of a framework for involving families in community behavioral health initiatives, to address the call for family-based interventions as the gold standard in obesity prevention. This includes specific strategies for creating a supportive family environment to promote health behaviors and address growing diabetes disparities in the key population of low-income Hispanic adolescents.