Weight Status Among Minnesota Hispanic or Latino/a Youth: An Exploration of Protective Factors

Abstract

Purpose

Pediatric obesity disproportionately impacts Hispanic or Latino/a adolescents. Culturally appropriate family-based behavioral initiatives to improve weight status are warranted. The purpose of this research was to determine prevalence rates and identify protective factors associated with having overweight/obesity (body mass index ≥ 85^th percentile) to inform Hispanic or Latino/a-targeted behavioral intervention development.

Design

Secondary data analyses of a population-based statewide survey.

Setting

Minnesota public high schools.

Participants

Male (n = 2,644) and female (n = 2,798) Hispanic or Latino/a 9^th and 11^th graders (N = 5,442).

Measures

Obesity-related behaviors (meeting fruit and vegetable [F&V] and physical activity [PA] recommendations), family caring, family country/region of origin, and weight status.

Analysis

Stepwise logistic regression models (F&V, PA), stratified by biological sex, were used to identify protective factors of overweight/obesity.

Results

The overall prevalence of meeting F&V and PA recommendations was 11.0% and 11.8%, respectively. Meeting F&V recommendations was not protective against overweight/obesity in either sex. Yet, males and females who met PA recommendations had significantly lower odds of having overweight/obesity (p < .05). In F&V and PA models, family caring was protective against overweight/obesity in females (p < .05), and family country/region of origin was protective against overweight/obesity in both sexes (p < .05).

Conclusion

Findings illustrate a need for obesity prevention initiatives for Hispanic or Latino/a youth. More research is needed to understand the protective nature of family caring and country/region of origin.

Keywords

weight control adolescents health policy social support racial minority groups community

Purpose

Pediatric obesity¹ is a public health problem in the United States (US).^2,3 Having overweight/obesity and related weight stigma puts youth at risk for poor physical and mental health sequela that can persist into adulthood.^4,5 According to US 2019 prevalence estimates, nearly 1 in 3 of all high schoolers had overweight/obesity.⁶

High schoolers who self-identify as Hispanic or Latino/a have the second-highest rate (38.8%) of overweight/obesity.⁶ Isasi et al.⁷ documented that Hispanic or Latino/a youth have higher rates of severe obesity and faster increases in body mass index (BMI) during adolescence than non-Hispanic or Latino/a whites. Albrecht et al.⁸ found that BMI increases are most prominent in those who self-identify as Mexican American or Puerto Rican. Overall rates of overweight/obesity differ between Hispanic or Latino males (41.3%) and Latina females (36.3%).⁶ Identifying factors that might reduce the risk of overweight/obesity among Hispanic or Latino/a youth by ethnic subgroup and biological sex is necessary for developing culturally relevant prevention and intervention strategies and health policies.

Assets-oriented prevention and intervention strategies often explore the role of protective factors (ie, competencies of individuals, families, and communities that promote the health and well-being of youth and families).⁹ There are several protective factors associated with weight status, including self-efficacy,¹⁰ social support,¹¹ high socioeconomic status, familial and cultural norms, and healthy lifestyle behaviors.¹² An abundance of pediatric obesity prevention literature investigates these latter factors, specifically the interplay between physical activity (PA) and dietary behaviors and weight status,¹³ given that rates of PA and fruit and vegetable (F&V) consumption are low among all youth.¹⁴ Rates of meeting the Centers for Disease Control and Prevention (CDC) guidelines of 60 minutes of moderate-to-vigorous PA daily are the second lowest in Hispanic or Latino/a (20.9%) youth.⁶

Evidence indicates that high levels of family functioning, parent-mediated practices (eg, family meals), and positive parenting serve as weight-related protective factors for many youth.^15-17 Within Hispanic or Latino/a communities, familismo (familism) is an essential cultural construct that includes a variety of components—parent and family caring, family connectedness, and respeto a los padres (respecting one’s parents)—that promote the health of youth.^18-20 Parents and extended family play an integral role in promoting adolescents’ healthy weight status.^21,22 Adolescents’ perceptions of family caring (the degree to which they feel cared for/supported by their families) influence obesity-related health behaviors.^15,23 Despite the breadth of literature on parent and family caring and connectedness,^24,25 little research has explored how family caring influences weight status in Hispanic or Latino/a youth.

Hispanic or Latino/a youth, particularly adolescents, are disproportionately affected by obesity yet remain underrepresented in obesity prevention interventions.^3,26,27 Although 95% of US Hispanic or Latino/a children are US-born,²⁸ Hispanic or Latino/a adults and their ancestors have emigrated from 20+ countries; thus, their families’ cultural heritages, among many other factors, are diverse.²⁹ Previous research has often relied on broad racial categorizations, which has facilitated high-level comparison but has limited understanding of unique characteristics or cultural nuances. In order to be well-positioned to positively intervene and create tailored intervention strategies, researchers must emphasize weight status and weight-related behaviors by ethnic subgroup identity or country of origin.^8,30

To this end, our secondary analysis study explored individual (ie, meeting F&V and PA recommendations) and familial (ie, family caring) protective factors, using ethnic subgroup identity (ie, family country/region of origin) as a proxy for cultural influences on weight status among Hispanic or Latino/a adolescents.

More specifically, the study aimed to answer the following questions:

(1) What are the prevalence rates of meeting F&V recommendations, meeting PA recommendations, and overweight/obesity?

(2) What are the associations between (a) meeting F&V recommendations and having overweight/obesity and (b) meeting PA recommendations and having overweight/obesity (controlling for potential confounders)?

(3) What are the direct effects of perceived family caring on having overweight/obesity?

(4) What are the direct effects of family country/region of origin on having overweight/obesity?

Methods

Design

An adapted version of Elder and colleagues²⁰ socioecological framework created for Hispanic or Latino/a populations informed our study. We utilized data from the 2019 Minnesota Student Survey (MSS). The Minnesota Departments of Education, Health, Human Services, and Public Safety coordinates the survey, which is administered every 3 years to students in 4 grades: 5^th, 8^th, 9^th, and 11^th. The MSS asks approximately 90 questions about health behaviors and is voluntary for school districts and students.³¹

Sample

In 2019, 81% of Minnesota public school districts participated in the MSS, totaling 170,128 students. To focus on the adolescent high school population, we limited our sample to 9^th or 11^th graders (N = 80,456), of whom 7,266 students self-identified as “Hispanic or Latino/Latina” only or “Hispanic or Latino/Latina” and any other additional racial-ethnic category. Our final analytic sample included 75% of the 7,266 students who met our eligibility criteria (N = 5,442). See the analytic section for more detail. Since MSS data were publicly available and de-identified, the University of Minnesota Institutional Review Board (IRB) approved the study as exempt from full IRB review.

Measures

Descriptions of all analyzed items can be found in Table 1.

Table 1.

Items Used for the Analyses.

#	Variable	Question(s)	Answer
1	Biological sex	What is your biological sex?	Male	Female
2	Meeting F&V recommendations^a	A. During the last 7 days, how many times did you eat fruit? (Do not count fruit juice)	I did NOT eat or drink this
			1 to 3 times in the last 7 days
			4 to 6 times in the last 7 days
		B. During the last 7 days, how many times did you eat green salad, potatoes, carrots or other vegetables? (Do not count French fries, fried potatoes, or potato chips)	1 time per day
			2 times per day
			3 times per day
			4 or more times per day
3	Meeting PA recommendations^b	During the last 7 days, on how many days were you physically active for a total of at least 60 minutes per day? (Add up all the time you spent in any kind of physical activity that increased your heart rate and made you breathe hard some of the time)	0 days	1 day
			2 days	3 days
			4 days	5 days
			6 days	7 days
4	Family caring^c	A. How much do you feel your parents care about you?	Not at all
			A little
			Some
		B. How much do you feel other adult relatives care about you?	Some
			Quite a bit
			Very much
5	Family country of origin^d	If you are Hispanic or Latino/Latina, which group best describes you? (If more than one describes you, mark ALL that apply)	Colombian
			Ecuadoran
			Guatemalan
			Mexican
			Puerto Rican
			Salvadoran
			Spanish/Spanish-American
			Other Hispanic or Latino/Latina
6	Weight status	Self-reported height and weight^e	BMI-for-age percentiles (%)^f
		How old are you?^e	Healthy/underweight: < 85^th
		What is your biological sex?^e	Overweight/obese: ≥ 85^th
7	Grade	What is your grade in school right now?	9^th	11^th
8	Skipped meals	During the last 30 days, have you had to skip meals because your family did not have enough money to buy food?	Yes
8	Skipped meals		No

Note: Items are from the Minnesota Student Survey Interagency Team.

Abbreviations: BMI, body mass index; F&V, fruit and vegetable; PA, physical activity.

^aQuestions 2A and 2B totaled and sum dichotomized as meeting the F&V recommendation of 5 or more total servings per day; or not meeting recommendation.

^bQuestion 3 dichotomized as meeting PA recommendation of 7 days; or not meeting recommendation.

^cQuestions 4A and 4B categorized as both parents and other adult relatives cared “very much”; either parents or other adult relatives cared “very much”; and neither parents nor other adult relatives cared “very much.”

^dQuestion 5 combined as Mexican only; Guatemalan/Salvadoran only; Colombian/Ecuadoran only; Puerto Rican only; (Spanish/Spanish-American)/(Other Hispanic or Latino/Latina), including any non-listed combination.

^eVariables used by the Minnesota Student Survey Interagency Team to calculate BMI-for-age percentiles.

^fCalculated from self-reported height, weight, age, and biological sex.

Independent variables

We used 4 independent variables in this study: (1) meeting F&V recommendations, (2) meeting PA recommendations, (3) family caring, and (4) family country of origin. To create the meeting F&V recommendations variable, we combined responses to 1 item about fruit intake with a second item about vegetable intake. We dichotomized this 2-item composite variable as “met F&V recommendations” (≥ 5 times per day) or “did not meet F&V recommendations” (< 5 times per day) based on the daily frequency of F&V intake and the CDC “5 a Day” recommendation.³² Given the disparity in units across answer choices (“times in the last 7 days” vs “times per day”), we created the variable using identical code to that used by the MSS Interagency Team to transform “times in the last 7 days” to “times per day.”³¹ We made an a priori decision to exclude “100% fruit juice” since literature regarding the obesogenic effects of fruit juice in adolescence is inconclusive.^33,34 Second, for the meeting PA recommendations variable, we used an item that asked about PA levels in the last 7 days. To align with the CDC recommendation of 60 minutes of moderate-to-vigorous PA daily,³⁵ we dichotomized this independent variable as “met PA recommendations” (7 days per week) or “did not meet PA recommendations” (< 7 days per week).

The perceived family caring variable combined responses to the 2 items shown in Table 1 into 3 categories: (1) perceptions that both parents and other adult relatives cared “very much”; (2) perceptions that either parents or other adult relatives cared “very much”; and (3) all other ratings (any combination of “not at all” to “quite a bit” for both parents and other adult relatives).

The 2019 MSS was the first version of the survey to ask about group identification based on family country of origin directly following a race-ethnicity question. For our analyses, we collapsed the family country of origin categories into 6 family country/region of origin groups by geographic location, as seen in Table 1. Since “Spanish/Spanish-American” often refers to those who have descendants from Spain and “Hispanic” often is used to describe someone who is “Spanish speaking,” we combined these 2 groups as “Other” since they did not explicitly represent a geographic region in Latin America.

Dependent variable

Using CDC guidelines, the MSS Interagency Team created a weight status variable (healthy/underweight, overweight, and obese) based on self-reported height, weight, age, and biological sex.^31,36 We combined the overweight and obese groups for our analyses, creating a dichotomous weight status dependent variable to compare those categorized as “overweight/obese” (≥ 85^th percentile) with those categorized as “healthy/underweight” (< 85^th percentile).

Covariates

We included grade in school and whether a student had ever skipped meals because their family did not have enough money for food as covariates in our analyses. We used grade (“9^th” or “11^th”) in school as a proxy to account for developmental differences between students. We used skipped meals (“Yes” or “No”) as a proxy for socioeconomic status since skipped meals can be associated with dietary quality and weight status.³⁷

Analysis

Our analyses were informed by the STrengthening the Reporting of OBservational Studies in Epidemiology (STROBE) Statement.³⁸ We used listwise deletion to remove respondents with missing data. Missingness for each independent variable ranged from 7.7% to 9.8%. We excluded approximately 20% of the 7,266 eligible students because these students skipped pertinent questions (height, weight, age, or biological sex) comprising the primary dependent variable, weight status. Our final analytic sample (N = 5,422) included 2,644 males and 2,798 females.

As stated above, obesity-related outcomes often differ between male and female Hispanic or Latino/a adolescents;⁶ thus, we stratified all models by biological sex. We used Chi-square tests of association to identify prevalence rates and to detect differences in the prevalence of all study variables between males and females (Research Question [RQ] 1). We used logistic regression models to test the association between meeting F&V recommendations and having overweight/obesity and between meeting PA recommendations and having overweight/obesity. We used multivariable logistic regression models to test these associations in adjusted models, controlling for grade in school and skipped meals (RQ 2).

To test the direct effect of the perceptions of family caring on having overweight/obesity (RQ 3), we added family caring to the logistic regression models. To explore the direct effect of family country/region of origin on having overweight/obesity (RQ 4), we added family country/region of origin to the models. We used students who self-identified as Mexican as the reference group for all logistic regression models that included family country/region of origin, as they were the largest group (62.6%) in our sample. Due to our moderately large sample size, we used an alpha level of .05 to determine statistical significance. We conducted our analyses using SAS OnDemand for Academics.³⁹

Results

Demographics and Sample Characteristics

Table 2 shows the characteristics of the overall analytic sample and by biological sex.

Table 2.

Sample Characteristics of 9^th and 11^th Grade Hispanic or Latina/o Students and by Biological Sex.

Parameter	Overall (N = 5,442)	Male n = 2,644 (48.6%)	Female n = 2,798 (51.4%)	χ² (DF)^a	p-value
Grade, n (%)				1.98 (1)	.16
9^th	3,121 (57.4)	1,542 (58.3)	1,579 (56.4)
11^th	2,321 (42.6)	1,102 (41.7)	1,219 (43.6)
Family country/region of origin, n (%)				9.33 (5)	.10
Mexican	3,409 (62.6)	1,653 (62.5)	1,756 (62.8)
Central American	420 (7.7)	227 (8.6)	193 (6.9)
South American	266 (4.9)	121 (4.6)	145 (5.2)
Puerto Rican	267 (4.9)	127 (4.8)	140 (5.0)
Multiple regions	229 (4.2)	98 (3.7)	131 (4.7)
Other	851 (15.6)	418 (15.8)	433 (15.5)
Skipped meals, n (%)				8.13 (1)	< .01
No	5,022 (92.3)	2,468 (93.3)	2,554 (91.3)
Yes	420 (7.7)	176 (6.7)	244 (8.7)
Weight status, n (%)				5.64 (1)	.02
Normal or underweight	3,303 (60.7)	1,562 (59.1)	1,741 (62.2)
Overweight/obese	2,139 (39.3)	1,082 (40.9)	1,057 (37.8)
Overweight	1,092 (20.1)	479 (18.1)	613 (21.9)
Obese	1,047 (19.2)	603 (22.8)	444 (15.9)
Met F&V recommendations, n (%)				.27 (1)	.60
Did not meet	4,843 (89.0)	2,359 (89.2)	2,484 (88.8)
Met	599 (11.0)	285 (10.8)	314 (11.2)
Met PA recommendations, n (%)
Did not meet	4,801 (88.2)	2,201 (83.3)	2,600 (92.9)	122.54(1)	< .01
Met	641 (11.8)	443 (16.8)	198 (7.1)
Family caring (“very much”), n (%)				58.36 (2)	< .01
Neither^b parents nor other relatives^c	1,710 (31.4)	702 (26.6)	1,008 (36.0)
Either parents or other relatives^c	1,533 (28.2)	778 (29.4)	755 (27.0)
Both parents and other relatives^c	2,199 (40.4)	1,164 (44.0)	1,035 (37.0)

Abbreviations: DF, degrees of freedom; F&V, fruit and vegetable; PA, physical activity; SD, standard deviation.

^aChi-square tests calculated to determine sex differences among categorical variables.

^bIncludes all other combinations of ratings, other than “very much.”

^cOther adult relatives.

Prevalence rates

As shown in Table 2, 11.0% of the sample met F&V recommendations, 11.8% met PA recommendations, and 39.3% reported having overweight/obesity. Many Hispanic or Latina/o students (40.4%) reported that both their parents and other adult relatives cared “very much” about them.

Direct Effects

Obesity-related behaviors and having overweight/obesity

As shown in Table 3 (males) and Table 4 (females), the association between meeting F&V recommendations and having overweight/obesity was not statistically significant for either Hispanic or Latino/a male or female students in unadjusted models (Model 1[m], Model 1[f]) nor adjusted models (Model 2[m], Model 2[f]). However, as seen in Table 5 (males) and Table 6 (females), meeting PA recommendations was significantly associated with lower odds of having overweight/obesity for males and females in both the unadjusted models (Model 5[m], Model 5[f]) and adjusted models (Model 6[m], Model 6[f]). After adjusting for covariates (Model 6[m]), compared to males who did not meet PA recommendations, males who met PA recommendations had significantly lower odds of having overweight/obesity. After adjusting for covariates (Model 6[f]), compared to females who did not meet PA recommendations, females who met PA recommendations had significantly lower odds of having overweight/obesity.

Table 3.

Odds Ratios and 95% Confidence Intervals for Associations Between the Independent Variables (Meeting Fruit and Vegetable Recommendations, Covariates, Family Caring, and Family Country/Region of Origin) and the Dependent Variable (Weight Status) for Male (m) Hispanic or Latino Students.

Independent variables	Dependent variable: Weight status (overweight/obese)
	Male students
	Model 1(m)	Model 2(m)	Model 3(m)	Model 4(m)
Meeting F&V recommendations
Did not meet (Ref)	1.00	1.00	1.00	1.00
Met	1.18 (.92 - 1.51)	1.18 (.92 - 1.51)	1.19 (.93 - 1.53)	1.22 (.95 - 1.56)
Grade
9^th (Ref)	NA	1.00	1.00	1.00
11^th	NA	.91 (.78 - 1.07)	.91 (.78 - 1.06)	.91 (.77 - 1.06)
Skipped meals
No (Ref)	NA	1.00	1.00	1.00
Yes	NA	1.28 (.94 - 1.74)	1.28 (.94 - 1.74)	1.29 (.95 - 1.77)
Family caring (“very much”)
Neither^a parents nor other relatives^b (Ref)	NA	NA	1.00	1.00
Either parents or other relatives^b	NA	NA	1.11 (.90 - 1.37)	1.10 (.90 - 1.36)
Both parents and other relatives^b	NA	NA	.94 (.77 - 1.14)	.95 (.78 - 1.15)
Family country/region of origin
Mexican (Ref)	NA	NA	NA	1.00
Central American	NA	NA	NA	.79 (.59 - 1.05)
South American	NA	NA	NA	.63 (.43 - .94)*
Puerto Rican	NA	NA	NA	.62 (.42 - .91)*
Multiple regions	NA	NA	NA	1.12 (.74 - 1.68)
Other	NA	NA	NA	.77 (.62 - .96)*

Abbreviations: F&V, fruit and vegetable; NA, not applicable; Ref, reference value.

*p < .05.

^aIncludes all other combinations of ratings, other than “very much.”

^bOther adult relatives.

Table 4.

Independent variables	Dependent variable: Weight status (overweight/obese)
	Female students
	Model 1(f)	Model 2(f)	Model 3(f)	Model 4(f)
Meeting F&V recommendations
Did not meet (Ref)	1.00	1.00	1.00	1.00
Met	.78 (.61 - 1.00)	.80 (.62 - 1.03)	.83 (.65 - 1.07)	.83 (.65 - 1.07)
Grade
9^th (Ref)	NA	1.00	1.00	1.00
11^th	NA	1.17 (1.00 - 1.37)	1.17 (1.00 - 1.36)	1.15 (.99 - 1.35)
Skipped meals
No (Ref)	NA	1.00	1.00	1.00
Yes	NA	1.37 (1.05 - 1.79)*	1.24 (.95 - 1.62)	1.22 (.93 - 1.60)
Family caring (“very much”)
Neither^a parents nor other relatives^b (Ref)	NA	NA	1.00	1.00
Either parents or other relatives^b	NA	NA	.91 (.75 - 1.10)	.92 (.76 - 1.12)
Both parents and other relatives^b	NA	NA	.60 (.50 - .72)*	.62 (.51 - .74)*
Family country/region of origin
Mexican (Ref)	NA	NA	NA	1.00
Central American	NA	NA	NA	.87 (.64 - 1.18)
South American	NA	NA	NA	.44 (.30 - .67)*
Puerto Rican	NA	NA	NA	.79 (.55 - 1.14)
Multiple regions	NA	NA	NA	.93 (.65 - 1.35)
Other	NA	NA	NA	.66 (.52 - .83)*

Abbreviations: F&V, fruit and vegetable; NA, not applicable; Ref, reference value.

*p < .05.

^aIncludes all other combinations of ratings, other than “very much.”

^bOther adult relatives.

Table 5.

Odds Ratios and 95% Confidence Intervals for Associations Between the Independent Variables (Meeting Physical Activity Recommendations, Covariates, Family Caring, and Family Country/Region of Origin) and the Dependent Variable (Weight Status) for Male (m) Hispanic or Latino Students.

Independent variables	Dependent variable: Weight status (overweight/obese)
	Male students
	Model 5(m)	Model 6(m)	Model 7(m)	Model 8(m)
Meeting PA recommendations
Did not meet (Ref)	1.00	1.00	1.00	1.00
Met	.66 (.54 - .82)*	.66 (.53 - .82)*	.67 (.54 - .83)*	.67 (.54 - .83)*
Grade
9^th (Ref)	NA	1.00	1.00	1.00
11^th	NA	.90 (.77 - 1.05)	.89 (.76 - 1.05)	.89 (.76 - 1.04)
Skipped meals
No (Ref)	NA	1.00	1.00	1.00
Yes	NA	1.28 (.94 - 1.74)	1.28 (.94 - 1.75)	1.30 (.95 - 1.77)
Family caring (“very much”)
Neither^a parents nor other relatives^b (Ref)	NA	NA	1.00	1.00
Either parents or other relatives^b	NA	NA	1.11 (.90 - 1.37)	1.10 (.89 - 1.36)
Both parents and other relatives^b	NA	NA	.96 (.79 - 1.17)	.97 (.80 - 1.18)
Family country/region of origin
Mexican (Ref)	NA	NA	NA	1.00
Central American	NA	NA	NA	.79 (.60 - 1.06)
South American	NA	NA	NA	.65 (.43 - .96)*
Puerto Rican	NA	NA	NA	.63 (.43 - .92)*
Multiple regions	NA	NA	NA	1.14 (.76 - 1.72)
Other	NA	NA	NA	.78 (.62 - .97)*

Abbreviations: NA, not applicable; PA, physical activity; Ref, reference value.

*p < .05.

^aIncludes all other combinations of ratings, other than “very much.”

^bOther adult relatives.

Table 6.

Independent variables	Dependent variable: Weight status (overweight/obese)
	Female students
	Model 5(f)	Model 6(f)	Model 7(f)	Model 8(f)
Meeting PA recommendations
Did not meet (Ref)	1.00	1.00	1.00	1.00
Met	.68 (.50 - .93)*	.69 (.51 - .95)*	.70 (.51 - .96)*	.71 (.51 - .97)*
Grade
9^th (Ref)	NA	1.00	1.00	1.00
11^th	NA	1.16 (1.00 - 1.36)	1.16 (.99 - 1.35)	1.15 (.98 - 1.34)
Skipped meals
No (Ref)	NA	1.00	1.00	1.00
Yes	NA	1.38 (1.06 - 1.81)*	1.24 (.95 - 1.63)	1.22 (.93 - 1.61)
Family caring (“very much”)
Neither^a parents nor other relatives^b (Ref)	NA	NA	1.00	1.00
Either parents or other relatives^b	NA	NA	.90 (.74 - 1.09)	.91 (.75 - 1.11)
Both parents and other relatives^b	NA	NA	.59 (.49 - .71)*	.61 (.51 - .74)*
Family country/region of origin
Mexican (Ref)	NA	NA	NA	1.00
Central American	NA	NA	NA	.86 (.63 - 1.18)
South American	NA	NA	NA	.44 (.30 - .67)*
Puerto Rican	NA	NA	NA	.81 (.56 - 1.16)
Multiple regions	NA	NA	NA	.92 (.64 - 1.33)
Other	NA	NA	NA	.66 (.53 - .83)*

Abbreviations: NA, not applicable; PA, physical activity; Ref, reference value.

*p < .05.

^aIncludes all other combinations of ratings, other than “very much.”

^bOther adult relatives.

Family caring and having overweight/obesity

As shown in Table 3 (males) and Table 5 (males), the association between family caring and having overweight/obesity was not statistically significant for Hispanic or Latino male students in the adjusted models accounting for F&V recommendations (Model 3[m]), PA recommendations (Model 7[m]), and covariates. However, as shown in Table 4 (females) and Table 6 (females), for Hispanic or Latina females, the association between family caring and having overweight/obesity was statistically significant in adjusted models accounting for F&V recommendations (Model 3[f]), PA recommendations (Model 7[f]), and covariates. Compared to females who perceived that neither their parents nor other adult relatives cared “very much” about them, females who perceived that both their parents and other adult relatives cared “very much” had significantly lower odds of having overweight/obesity in the model accounting for F&V recommendations and covariates (Model 3[f]) and significantly lower odds of having overweight/obesity in the model accounting for PA recommendations and covariates (Model 7[f]).

Family country/region of origin and having overweight/obesity

Several significant differences in overweight/obesity emerged across family country/region of origin as seen in Tables 3 and 5 (males) and Tables 4 and 6 (females), which display models accounting for F&V recommendations (Model 4), PA recommendations (Model 8), and covariates, including family caring. As seen in Table 3, compared to male students who identified with the group Mexican, males categorized as South American, Puerto Rican, or “Other” had significantly lower odds of having overweight/obesity after accounting for meeting F&V recommendations and covariates (Model 4[m]). Similarly, as seen in Table 5, compared to male students who identified with the group Mexican, males categorized as South American, Puerto Rican, or “Other” had significantly lower odds of having overweight/obesity after accounting for meeting PA recommendations and covariates (Model 8[m]). As seen in Table 4, compared to females who identified as Mexican, females categorized as South American or “Other” had significantly lower odds of having overweight/obesity after accounting for meeting F&V recommendations and covariates (Model 4[f]). Likewise, as seen in Table 6, compared to females who identified as Mexican, females categorized as South American or “Other” had significantly lower odds of having overweight/obesity after accounting for meeting PA recommendations and covariates (Model 8[f]).

Discussion

The present study sought to advance the pediatric obesity literature by examining protective factors associated with healthy weight status among Hispanic or Latino/a youth who may be at high risk for having overweight/obesity. More specifically, we aimed to examine the associations between individual (ie, meeting F&V and PA recommendations), familial (ie, family caring), and cultural (ie, family country/region of origin) protective factors and weight status in 9^th- and 11^th-grade male and female students.

Rates of overweight/obesity in our sample of Minnesota Hispanic or Latino/a students were comparable to national rates of all American Hispanic or Latino/a youth.⁶ The differences in overweight/obesity across Hispanic or Latino/a subgroups by biological sex in our sample also aligned with existing literature.⁷ Concerningly, overall rates of overweight/obesity have increased steadily over the last decade in Minnesota among Hispanic or Latina females and remained high among Hispanic or Latino males. In MSS samples, rates of overweight/obesity among Hispanic or Latina females were 28.2% in 2013, 32.8% in 2016, and 37.8% in 2019.³¹ Similarly, rates of overweight/obesity among Hispanic or Latino males were 36.6% in 2013, 35.2% in 2016, and 40.9% in 2019.³¹ Given these trends, it is vital to continue Hispanic or Latino/a-targeted obesity prevention research.

Rates of meeting F&V recommendations among our sample of Hispanic or Latina/o youth are low. While our F&V data are not directly comparable to existing Hispanic or Latino/a-targeted research with different cut points or continuous (not categorical) scales, existing evidence documents generally low F&V intake among Hispanic or Latino/a adolescents.⁶ When exploring meeting F&V recommendations as a protective factor, we did not detect significant associations with having overweight/obesity. Despite this, it is vital to consider youth’s dietary habits when designing obesity prevention interventions.

Among all Hispanic or Latino/a youth in our sample, the rate of meeting PA recommendations was almost 10% lower than the national average of all American youth who identify as Hispanic or Latino/a.⁶ This low rate is noteworthy given the significant variability in meeting PA recommendations by biological sex and the steady secular decline in meeting PA recommendations among male and female youth in Minnesota who self-identified as Hispanic or Latino/a of any race over the past decade. Minnesota Student Survey data showed that, among Hispanic or Latino males, rates of meeting PA recommendations were 21.1% in 2013, 18.0% in 2016, and 16.8% in 2019.³¹ Among Minnesota Hispanic or Latina females, rates of meeting PA recommendations were 8.8% in 2013, 8.0% in 2016, and 7.1% in 2019.³¹ The higher rates of meeting PA recommendations in Hispanic or Latino males compared to females align with substantial evidence that Hispanic and Latino male youth are more likely to meet PA recommendations.⁴⁰

Most of the sample felt cared for “very much,” either by (1) both their parents and other adult relatives or by (2) either their parents or other adult relatives in both the F&V and PA models; feeling cared for by both parents and other adult relatives protected Latina girls against having overweight/obesity. Substantial literature shows the protective nature of family caring, connection, and communication on various obesity-related health behaviors.^41,42 Furthermore, evidence indicates that supportive parents and youth’s satisfaction with family relationships are associated with positive physical and psychological health outcomes among Hispanic or Latino/a adolescents.^43,44 However, many existing studies only included youth’s perceptions of their parents, not of extended family, and most studies tested family factors within the context of acculturation.^45,46 Thus, these findings depended on additional factors, such as immigrant status or length of time in the US. Very few studies have explored the role of extended family in pediatric obesity. More research is needed to determine if and how different levels of parental and/or other adult relative support, as well as acculturation factors, may impact weight status among Hispanic or Latino/a youth.

Our findings add to the increasing body of literature exploring Hispanic or Latino/a subgroups by family country/region of origin and serve as a foundation for further exploration of heterogeneity in weight outcomes across ethnic subgroups.^8,47 Given that there were lower odds of having overweight/obesity in both Hispanic or Latino/a males and females who identified with the group South American, further investigation of specific cultural factors that may impact weight status among this particular geographic region is needed. Additional research may illustrate nuances responsible for disparities in weight status, such as acculturation, distinct ethnic or cultural dietary staples, educational diversity, or economic resources.⁴⁸ For example, socioeconomic status differs among Mexican, Ecuadoran, and Columbian subgroups in Minnesota,⁴⁹ which could affect food access and participation levels in school and community sports.

Additional qualitative research with respondents of diverse Hispanic or Latino/a backgrounds may also prove valuable in understanding the lived experience of this population of adolescents within the context of weight status, especially among those categorized as “Other.” For example, researchers could interview youth to better understand the role of ethnic heritage among 3^rd and 4^th generation Hispanic or Latino/a youth, who tend to have higher odds of having overweight/obesity than family members of earlier generations.⁴⁸ Finally, it might be helpful to use mixed methods to explore associations among family country/region of origin, family caring, and meeting F&V and PA recommendations, given these variables’ inherent ethnic/cultural compositions. Understanding family dynamics around home-based meals, dietary preferences, and family-based PA levels of ethnic subgroups may provide practical insights to family-based obesity interventionists working with culturally diverse Hispanic or Latino/a youth and their families.

Limitations and Strengths

There are several limitations of our study. First, we could not determine causation since the MSS data are cross-sectional. Second, we used only 2 items to create our family caring variable, yet family caring is a complex concept that may be better measured using more items or primary data. Third, BMI-for-age calculations were based on self-reported height and weight, which are less accurate than objective measures. Fourth, the relatively large amount of missing data on the weight status variable (∼20%) may have biased the sample towards those who were more aware of their height and weight or comfortable providing this information. Finally, our interpretation of the findings regarding our family country/region of origin variable within the context of weight outcomes is inherently limited, as we did not have access to other ethnic/cultural (eg, acculturation) variables.

Despite these limitations, our study adds to the adolescent obesity prevention literature. Our analyses included unique family country/region of origin data to explore weight status by ethnic subgroup, and we used an assets-oriented approach by exploring potential protective factors. Given the moderately large sample size, we distinguished differences by biological sex that may help groups design and implement tailored programming to meet the distinct needs of Hispanic or Latino/a youth. It is critical to advance Hispanic or Latino/a-targeted obesity prevention research given the low rates of meeting F&V and PA recommendations and the high percentages of overweight/obesity among Hispanic or Latino/a males and females. Given our significant findings of family caring among Hispanic or Latina females, obesity initiatives may also consider engaging family members in activity-based interventions to increase PA among Latina females. Interventions to increase PA among Latino males may also be a promising avenue to address overweight/obesity.

Conclusions and Next Steps

Further research is needed to understand the interplay between dietary intake, PA behaviors, and weight status among Hispanic or Latino/a youth by ethnic subgroup. If premised on the socioecological framework,²⁰ this research should explore additional contextual factors, including other interpersonal influences (eg, peers, siblings) or organizational and community influences.⁵⁰ A comprehensive research approach may elucidate other important predictors (eg, access to affordable F&V, availability of school- or community-based sports, self-efficacy) of weight status. Obesity prevention researchers should also consider investigating the interplay between cultural influences on adolescents’ dietary and PA behaviors. To improve weight-related behaviors and outcomes among this population, researchers should collaborate directly with Hispanic or Latino/a communities and experts in psychology, anthropology, sociology, and cultural studies to design and implement culturally appropriate public health initiatives. Collaborative research efforts would advance the science of pediatric obesity prevention among this understudied population and inform public health interventionists, healthcare providers, school districts, policymakers, and Hispanic or Latino/a youth and families alike.

So What …

What Is Already Known on This Topic?

Overweight/obesity rates are disproportionately high among Hispanic or Latino/a adolescents. Pediatric obesity prevention researchers often promote healthy dietary and physical activity behaviors among youth, and culturally appropriate research with Hispanic or Latino/a families often includes the family unit, as parents serve as supportive partners of behavior change. Health equity research supports collecting country-specific categories for ethnicity to identify areas of need across Hispanic or Latino/a subgroups.

What Does This Article Add?

Our findings demonstrated that meeting physical activity recommendations was associated with lower odds of having overweight/obesity in Hispanic or Latino/a male and female adolescents; family caring was protective against overweight/obesity in Latina females; and family country/region of origin was protective against overweight/obesity among Latino males and Latina females.

What Are the Implications for Health Promotion Practice or Research?

Findings should serve as an impetus for continued interdisciplinary research using population-based, data-driven approaches along with qualitative methods to further elucidate individual, familial, and cultural nuances associated with weight status among Hispanic or Latino/a youth.

Footnotes

Author Contributions

Christie L. Martin conceptualized the study in partnership with an academic advisor (Fulkerson); designed analyses in collaboration with dissertation committee members (Drs. Fulkerson, McMorris, Eisenberg, Sieving, and Porta) and statistician (Mathiason); led all aspects of the study; completed quantitative analyses; drafted manuscript; revised manuscript.

Barbara J. McMorris designed the analyses in partnership with Martin and dissertation committee members, reviewed the manuscript for accuracy, and substantially contributed to manuscript revision.

Marla E. Eisenberg designed the analyses in partnership with Martin and dissertation committee members, reviewed the manuscript for accuracy, and substantially contributed to manuscript revision.

Renee E. Sieving designed the analyses in partnership with Martin and dissertation committee members, reviewed the manuscript for accuracy, and substantially contributed to manuscript revision.

Carolyn M. Porta designed the analyses in partnership with Martin and dissertation committee members, reviewed the manuscript for accuracy, and substantially contributed to manuscript revision.

Michelle A. Mathiason designed the analyses in partnership with Martin and dissertation committee members; assisted with data analyses; reviewed the manuscript for accuracy; substantially contributed to manuscript revision.

Sarah M. Espinoza reviewed the manuscript for accuracy and substantially contributed to manuscript revision, providing a cultural lens and expertise in physical activity.

Yazmin A. Cespedes reviewed the manuscript for accuracy and substantially contributed to manuscript revision, providing a cultural lens.

Jayne A. Fulkerson assisted with the project’s conceptualization with Martin; designed the analyses in partnership with Martin and dissertation committee members; reviewed manuscript for accuracy; substantially contributed to manuscript revision.

All authors approved the submitted manuscript and have sufficiently participated in the work to take public responsibility for its content.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethics Approval

University of Minnesota IRB ID STUDY00008173 (11/18/2019); Not Human Research; Investigator: Dr. Jayne Fulkerson

ORCID iDs

Christie L. Martin

Barbara J. McMorris

Michelle A. Mathiason

References

Centers for Disease Control and Prevention . Childhood Obesity Facts | Overweight & Obesity. https://www.cdc.gov/obesity/data/childhood.html. Published May 17, 2022. Accessed July 20, 2022.

Sanyaolu

Okorie

Locke

Rehman

. Childhood and adolescent obesity in the United States: A public health concern. Glob Pediatr Health. 2019;6. 2333794X1989130, doi:10.1177/2333794X19891305.

Rafei

Elliott

Jones

Riosmena

Cunningham

Mehta

. Obesity incidence in U.S. children and young adults: A pooled analysis. Am J Prev Med. 2022;63(1):51-59. doi:10.1016/j.amepre.2021.12.021.

Chu

Nguyet

NTM

Nga

, et al. An update on obesity: Mental consequences and psychological interventions. Diabetes Metab Syndr. 2019;13(1):155-160. doi:10.1016/j.dsx.2018.07.015.

Sommer

Twig

. The impact of childhood and adolescent obesity on cardiovascular risk in adulthood: A systematic review. Curr Diab Rep. 2018;18, 91(10). doi:10.1007/s11892-018-1062-9.

Centers for Disease Control and Prevention . 1991-2019 High School Youth Risk Behavior Survey Data. nccd.cdc.gov/youthonline/. Published 2019. Accessed July 20, 2022.

Isasi

Rastogi

Molina

. Health issues in Hispanic/Latino youth. Physiol Behav. 2018;176(5):139-148. doi:10.1037/lat0000054.

Albrecht

Gordon-Larsen

. Ethnic differences in body mass index trajectories from adolescence to adulthood: A focus on Hispanic and Asian subgroups in the United States. PLoS One. 2013;8(9):e72983. doi:10.1371/journal.pone.0072983.

Deković

. Risk and protective factors in the development of problem behavior during adolescence. J Youth Adolesc. 1999;28(6):667-685. doi:10.1023/A:1021635516758.

10.

Schroeder

Kubik

Lee

Sirard

Fulkerson

. Self-efficacy, not peer or parent support, is associated with more physical activity and less sedentary time among 8- to 12-year-old youth with elevated body mass index. J Phys Act Health. 2020;17(1):74-79. doi:10.1123/jpah.2019-0108.

11.

Wang

Druker

Gapinski

, et al. The role of social support vs. modeling on adolescents’ diet and physical activity: Findings from a school-based weight management trial. J Child Adolesc Behav. 2014;2(2):1-10. doi:10.4172/2375-4494.1000132.

12.

Kiraly

Turk

Kalarchian

Shaffer

. Applying ecological frameworks in obesity intervention studies in Hispanic/Latino youth: A systematic review. Hisp Health Care Int. 2017;15(3):130-142. doi:10.1177/1540415317731069.

13.

Anderson

. A review of the prevention and medical management of childhood obesity. Child Adolesc Psychiatr Clin N Am. 2018;27(1):63-76. doi:10.1016/j.chc.2017.08.003.

14.

Lange

Moore

Harris

, et al. Percentage of adolescents meeting federal fruit and vegetable intake recommendations-youth risk behavior surveillance system, United States, 2017. MMWR Morb Mort Wkly Rep. 2021; 70:69-74. doi:10.15585/mmwr.mm7003a1.

15.

Haines

Rifas-Shiman

Horton

, et al. Family functioning and quality of parent-adolescent relationship: Cross-sectional associations with adolescent weight-related behaviors and weight status. Int J Behav Nutr Phys Act. 2016;13(1):1-12. doi:10.1186/s12966-016-0393-7.

16.

Pratt

Skelton

. Family functioning and childhood obesity treatment: A family systems theory-informed approach. Acad Pediatr. 2018;18(6):620-627. doi:10.1016/j.acap.2018.04.001.

17.

Ochoa

Berge

. Home environmental influences on childhood obesity in the Latino population: A decade review of literature. J Immigr Minor Health. 2017;19(2):430-447. doi:10.1007/s10903-016-0539-3.

18.

Campos

Kim

. Culture, relationships and health: Incorporating the cultural diversity of family and close relationships into our understanding of health. Am Psychol. 2017;72(6):543-554.

19.

Wheeler

Zeiders

Updegraff

Umaña-Taylor

Rodríguez de Jesús

Perez-Brena

. Mexican-origin youth’s risk behavior from adolescence to young adulthood: The role of familism values. Dev Psychol. 2017;53(1):126-137. doi:10.1037/dev0000251.

20.

Elder

Ayala

Parra-Medina

Talavera

. Health communication in the Latino community: Issues and approaches. Annu Rev Public Health. 2009;30(1):227-251. doi:10.1146/annurev.publhealth.031308.100300.

21.

Salvy

Miles

JNV

Shih

Tucker

D’Amico

. Neighborhood, family and peer-level predictors of obesity-related health behaviors among young adolescents. J Pediatr Psychol. 2017;42(2):153-161. doi:10.1093/jpepsy/jsw035.

22.

Ash

Agaronov

Young

Aftosmes-Tobio

Davison

. Family-based childhood obesity prevention interventions: A systematic review and quantitative content analysis. Int J Behav Nutr Phys Act. 2017;14(1):113. doi:10.1186/s12966-017-0571-2.

23.

Kitzman-Ulrich

Wilson

St George

Lawman

Segal

Fairchild

. The integration of a family systems approach for understanding youth obesity, physical activity, and dietary programs. Clin Child Fam Psychol Rev. 2010;13(3):231-253. doi:10.1007/s10567-010-0073-0.

24.

Areba

Eisenberg

McMorris

. Relationships between family structure, adolescent health status and substance use: Does ethnicity matter? J Community Psychol. 2018;46(1):44-57. doi:10.1002/jcop.21915.

25.

Carver

Elliott

Kennedy

Hanley

. Parent–child connectedness and communication in relation to alcohol, tobacco and drug use in adolescence: An integrative review of the literature. Drugs. 2017;24(2):119-133. doi:10.1080/09687637.2016.1221060.

26.

Branscum

Sharma

. A systematic analysis of childhood obesity prevention interventions targeting Hispanic children: Lessons learned from the previous decade. Obes Rev. 2011;12(501):151-158. doi:10.1111/j.1467-789X.2010.00809.x.

27.

Ash

Agaronov

Young

Aftosmes-Tobio

Davison

. Family-based childhood obesity prevention interventions: A systematic review and quantitative content analysis. Int J Behav Nutr Phys Act. 2017;14(1):113. doi:10.1186/s12966-017-0571-2.

28.

Mather

. Trends and challenges facing America’s Latino children | Population Reference Bureau. https://www.prb.org/trends-and-challenges-facing-americas-latino-children/. Published September 28, 2016. Accessed July 20, 2022.

29.

Aragones

Hayes

Chen

González

Gany

. Characterization of the Hispanic or Latino population in health research: A systematic review. J Immigr Minor Health. 2014;16(3):429-439. doi:10.1007/s10903-013-9773-0.

30.

Fenelon

Chinn

Anderson

. A comprehensive analysis of the mortality experience of Hispanic subgroups in the United States: Variation by age, country of origin, and nativity. SSM Popul Health. 2017;3:245-254.

31.

Minnesota Department of Health. Minnesota

Student Survey. https://www.health.state.mn.us/data/mchs/surveys/mss/index.html. Accessed July 20, 2022.

32.

Centers for Disease Control and Prevention . 5 a Day Works!; 2005. https://www.cdc.gov/nccdphp/dnpa/nutrition/health_professionals/programs/5aday_works.pdf. Accessed July 20, 2022.

33.

Sharma

Chung

Kim

Hong

. Paradoxical effects of fruit on obesity. Nutrients. 2016;8(10):1-16. doi:10.3390/nu8100633.

34.

Auerbach

Wolf

Hikida

, et al. Fruit juice and change in BMI: A meta-analysis. Pediatrics. 2017;139(4):e20162454. doi:10.1542/peds.2016-2454.

35.

U.S. Department of Health and Human Services . Physical Activity Guidelines for Americans. 2nd ed. Accessed July 20, 2022. https://health.gov/sites/default/files/2019-09/Physical_Activity_Guidelines_2nd_edition.pdf (2018).

36.

Centers for Disease Control and Prevention . Recommended BMI-for-age Cutoffs - BMI for age training course. https://www.cdc.gov/nccdphp/dnpao/growthcharts/training/bmiage/page4.html. Accessed July 20, 2022.

37.

Monzani

Ricotti

Caputo

, et al.

A systematic review of the association of skipping breakfast with weight and cardiometabolic risk factors in children and adolescents. What should we better investigate in the future?

Nutrients. 2019;11(2):1-23. doi:10.3390/nu11020387.

38.

von Elm

Altman

Egger

Pocock

Gøtzsche

Vandenbroucke

. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495-1499. doi:10.1016/j.ijsu.2014.07.013.

39.

SAS OnDemand for Academics | SAS . SAS Institute Inc. https://www.sas.com/en_us/software/on-demand-for-academics.html. Published 2021.

40.

Armstrong

Wong

Perrin

Page

Sibley

Skinner

. Association of physical activity with income, race/ethnicity, and sex among adolescents and young adults in the United States findings from the national health and nutrition examination survey, 2007-2016. JAMA Pediatr. 2018;172(8):732-740. doi:10.1001/jamapediatrics.2018.1273.

41.

Crossman

Sullivan

Benin

. The family environment and American adolescents’ risk of obesity as young adults. Soc Sci Med. 2006;63(9):2255-2267. doi:10.1016/j.socscimed.2006.05.027.

42.

Ornelas

Perreira

Ayala

. Parental influences on adolescent physical activity: A longitudinal study. Int J Behav Nutr Phys Act. 2007;4:1-10. doi:10.1186/1479-5868-4-3.

43.

Almeida

Molnar

Kawachi

Subramanian

. Ethnicity and nativity status as determinants of perceived social support: Testing the concept of familism. Soc Sci Med. 2009;68(10):1852-1858. doi:10.1016/j.socscimed.2009.02.029.

44.

Isasi

Hua

Jung

, et al. The association of parental/caregiver chronic stress with youth Obesity: Findings from the study of latino youth and the hispanic community health study/study of latinos sociocultural ancillary study. Child Obes. 2017;13(4):251-258. doi:10.1089/chi.2016.0205.

45.

Cupito

Stein

Gonzalez

Supple

. Familism and Latino adolescent depressive symptoms: The role of maternal warmth and support and school support. Cultur Divers Ethnic Minor Psychol. 2016;22(4):517-523. doi:10.1037/cdp0000097.

46.

Hardway

Fuligni

. Dimensions of family connectedness among adolescents with Mexican, Chinese, and European backgrounds. Dev Psychol. 2006;42(6):1246-1258. doi:10.1037/0012-1649.42.6.1246.

47.

Neighbors

Marquez

Marcus

. Leisure-time physical activity disparities among Hispanic subgroups in the United States. Am J Public Health. 2008;98(8):1460-1464. doi:10.2105/AJPH.2006.096982.

48.

Frisco

Quiros

Van Hook

. One size may not fit all: How obesity among Mexican-origin youth varies by generation, gender, and age. Demography. 2016;53(6):2031-2043. doi:10.1007/s13524-016-0525-3.

49.

Minnesota Compass . Cultural communities | MN Compass. https://www.mncompass.org/cultural-communities. Accessed July 20, 2022.

50.

Berrigan

Arteaga

Colón-Ramos

, et al. Measurement challenges for childhood obesity research within and between Latin America and the United States. Obes Rev. 2021;22(S3):1-15. doi:10.1111/obr.13242.