Intergenerational Determinants of Weight Status Concordance and Discordance in Parent/Adolescent Dyads from the Family Life,Activity,Sun,Health,and Eating Study

Abstract

Background:

Studies have shown that obesity (OB) has strong intergenerational linkages and tends to cluster in families, but there is a dearth of research examining the socioecological factors predictive of weight status concordance and discordance among parents and adolescents.

Methods:

We ran a stepwise multinomial logistic regression to assess for sociodemographic, individual-, and family-level predictors of four dyadic weight status groups using data from 1516 parent/adolescent dyads from the National Cancer Institute's Family Life, Activity, Sun, Health, and Eating Study. We categorized parent/adolescent dyads into one of four groups based on their BMI: (1) Healthy Weight Concordance (i.e., both parent and adolescent in the normal weight range); (2) overweight (OW)/OB Concordance (i.e., both parent and adolescent with OW or OB); (3) Discordance-Parent OW/OB (i.e., parent with OW/OB and adolescent in the normal weight range); or (4) Discordance-Adolescent OW/OB (i.e., adolescent with OW/OB and parent in the normal weight range).

Results:

There were 475 parent/adolescent dyads (31.3%) in the Healthy Weight Concordance group, 351 (23.2%) in the OW/OB Concordance group, 604 (39.8%) in the Discordance-Parent OW/OB group, and 86 (5.7%) in the Discordance-Adolescent OW/OB group. Being from a low socioeconomic family, identifying as a minority, and identifying as a male parent were associated with an OW/OB dyadic BMI. Higher levels of adolescent and parent emotional eating were significantly associated with parent/adolescent dyads being in the OW/OB Concordance group. Parent emotional eating was also associated with Discordance-Parent OW/OB. In contrast, parents' and adolescents' physical activity self-efficacy was associated with a Healthy Weight dyadic BMI vs. OW/OB Concordance or Discordance-Parent OW/OB.

Conclusions:

Our findings highlight the strength of OW/OB concordance in families, especially among lower socioeconomic and ethnic minority families and the significance of emotional eating and physical activity self-efficacy in contributing to BMI patterns among parents and adolescents.

Introduction

Adult and adolescent obesity (OB) has reached epidemic levels in the United States, where an estimated 73.6% of adults (20+ years old) and 21% of adolescents (12–19 years old) are classified as having overweight (OW) or OB.^1,2 Individuals with OB are at increased risk for chronic diseases, including, but not limited to, hypertension, dyslipidemia, type 2 diabetes, coronary heart disease, stroke, and some cancers.³

Research has shown that adolescents in the OW category tend to become adults with OB, which in turn leads to a higher incidence, prevalence, and mortality rate of coronary heart disease or other causes attributable to OB.⁴ OW and OB have strong intergenerational linkages and tend to cluster in families due to genetic predisposition and a shared environment, including diet and activity patterns.^5,6 To reduce disease burden and promote health in adolescents, it is important to understand weight status associations in families.

Previous research consistently demonstrates the intergenerational transmission of OW and OB from parents, particularly mothers, to children, especially older children and adolescents.^6,7 For example, James et al. found that adolescents of mothers with OB had an increased risk of OB themselves.⁷ Notably, the intergenerational transmission of OB in this study followed a socioeconomic gradient, such that transmission was stronger among disadvantaged households vs. more affluent households.⁷

Similarly, Naess et al. observed a positive association between parents' BMI and their offspring's BMI, which was further strengthened when parental education level was lower.⁶ The authors found that having parents with OW or OB was associated with higher BMI in children. Although the literature shows that weight status is generally concordant among parent/adolescent dyads (i.e., both parent and adolescent with healthy weight, or OW/OB), there is less information regarding discordant parent/adolescent dyads (i.e., parent with OW/OB and healthy weight adolescent, or vice versa) as well as factors differentially associated with weight status concordance or discordance.

From a socioecological perspective, researchers have identified determinants of adolescent OW and OB across multiple levels, including sociodemographic (e.g., socioeconomic status [SES]), individual (e.g., self-efficacy), and family characteristics (e.g., parenting practices).^8,9 Beyond parental weight status, individual-level factors associated with weight status include lifestyle behaviors (e.g., physical activity, sedentary behavior, diet) and psychological constructs (e.g., emotional regulation, self-efficacy).^10,11 Lifestyle behaviors of family members often correspond to one another.^12,13

The literature on parental dietary intake, for example, has been found to be more predictive of adolescent dietary intake than youth's own food preferences or knowledge.^12,13 Similarly, parental increases in objectively measured physical activity have been associated with increased youths' physical activity levels.¹⁴ Studies have also consistently shown associations between parent and adolescent emotional regulation,¹⁵ and how maladaptive emotional regulation contributes to weight gain, OW, or OB through lifestyle behaviors, specifically by increasing emotional eating, sedentary habits, and physical inactivity.¹⁰

Konttinen et al. identified emotional eating and physical activity self-efficacy as significant predictors of BMI among adults.¹¹ In fact, the strongest and most consistent psychological correlate of engaging in exercise is self-efficacy,¹⁶ which has been shown to predict higher levels of success in increasing activity levels and in weight control among adults.¹⁷ Adolescents with vs. without OB also report lower levels of physical activity self-efficacy.¹⁸

Family factors, including the home environment, have also been linked to diet, physical activity, and sedentary habits in both parents and adolescents.⁸ Positive parenting, including setting healthy rules and limits, has been associated with an overall healthier diet, more consumption of fruits and vegetables, and lower intake of junk food and sugar-sweetened beverages among adolescents.^19,20 Similarly, parental modeling healthy food intake, buying more fruits and vegetables and less junk food, and creating healthy meal-time routines, such as family meals together without TV, have shown protective effects on dietary intake and quality among adolescents.²⁰

For example, Van Lippevelde et al. found that children and adolescents' physical activity levels revolve around parental involvement, specifically in the form of joining them in physical activities (e.g., modeling) and providing logistical support.²¹ Parents have been shown to impact their children's and adolescents' weight-related behaviors by providing access and availability to foods (e.g., fruits and vegetables, junk food, and sugar and sweetened beverages),^22,23 equipment for physical activity and play,²⁴ and television in their children's bedroom.⁸

Overall, although we have a general understanding of socioecological factors that place parents and adolescents at greater risk for OB, it is less clear how these may be associated with weight status concordance (both healthy weight or OW/OB BMI) or discordance (one OW/OB and the other healthy weight BMI) among parent/adolescent dyads. As such, the aims of the study were to (1) describe distinct concordance and discordance in parent/adolescent weight status; and (2) examine socioecological predictors of dyadic weight status, specifically sociodemographic, individual-level (i.e., lifestyle behaviors, psychological), and family-level factors (i.e., parenting, home environment).

Materials and Methods

Sample

The current study was a secondary data analysis using the National Cancer Institute's Family Life, Activity, Sun, Health, and Eating (FLASHE) Study (https://cancercontrol.cancer.gov/brp/hbrb/flashe.html), which collected cross-sectional information on individual, interpersonal, and environmental correlates of cancer-preventive lifestyle behaviors in a national sample of 1859 parent/adolescent dyads.²⁵ Based on a nonprobability sampling method, participants from across the United States were recruited through the Ipsos Consumer Opinion Panel. A total of 5027 families were screened for eligibility. Families were eligible if the adolescent was 12–17 years old, and if the parent and adolescent lived together for at least 50% of the time.

The screening strategy entailed efforts to balance the web-based sample as much as possible to reflect the national population based on key demographic variables of census division, gender, household income, household size, and race/ethnicity. The nationwide FLASHE sample was similar to the US population on variables of age and census division and adolescents were relatively even for age and gender, but the majority of the parents were females. Parent and adolescent participants completed two surveys each, one diet focused and the other on physical activity (along with other cancer preventive health behaviors).

The FLASHE study received approval from the National Institutes of Health Institutional Review Board (IRB), the Westat IRB, and the Office of Management and Budget. The deidentified data sets were released and made publicly available. More detailed information on FLASHE design and methodology is reported elsewhere.²⁵ Of the 1859 dyads in the sample, 1516 dyads completed all the measures included in our analyses, and thus were included in our study.

Measures

We selected self-report measures from the available demographic, physical activity, and diet FLASHE surveys.

Dyadic weight status

Adolescents' and parents' self-reported height and weight: Using the CDC criteria for adult BMI and adolescent age- and sex-specific BMI percentiles, we created three weight status groups for both adults and adolescents: normal weight, OW, and OB. We then categorized parent/adolescent dyads into one of the following dyadic weight status groups: (1) Healthy Weight Concordance (i.e., both parent and adolescent in the normal weight range); (2) OW/OB Concordance (i.e., both parent and adolescent with OW or OB); (3) Discordance-Parent OW/OB (i.e., parent with OW/OB and adolescent in the normal weight range); or (4) Discordance-Adolescent OW/OB (i.e., adolescent with OW/OB and parent in the normal weight range).

Sociodemographic factors

The data set included demographic information for adolescents on age, sex, race, and weight status. Demographics for parents include age, sex, marital status, education level, race medical insurance information, income, medical insurance, and weight status. We created a composite variable as a proxy for SES, such that any parent receiving Medicaid insurance and/or having less than a high school degree or General Educational Development (GED) was categorized as lower SES (29.0%), all others as higher SES (71.0%).

Individual-level factors

Emotional regulation

We calculated a summed score for parents and adolescents using the Emotional Regulation Questionnaire-Suppression, a 4-item measure of skills and processes associated with suppression of physiological, behavioral, and emotional tendencies. Response options ranged from 1 = “Strongly Disagree” to 5 = “Strongly Agree.” Cronbach's α for parents and adolescents was 0.80 and 0.81, respectively.

Emotional eating

We calculated a summed score for parents and adolescents using two items from the Eating in the Absence of Hunger Questionnaire,²⁶ which assesses whether an individual's eating behaviors are driven by emotions. Response options ranged from 1 = “Never” to 5 = “Always.” Cronbach's α for both parents and adolescents was 0.86.

Physical activity self-efficacy

We used a single item for parents and adolescents from the Perceived Competence Scale,²⁷ which assesses if individuals feel confident in their ability to exercise regularly. Response options ranged from 1 = “Strongly Disagree” to 5 = “Strongly Agree.”

Dietary intake

We calculated summed scores for parent and adolescent fruit and vegetable intake and junk food and sugar-sweetened beverage intake using the 27-item Dietary Screener Questionnaire from the 2010 National Youth Physical Activity and Nutrition Survey.²⁸

Physical activity and sedentary behavior

We used the 15-item Youth Activity Profile (YAP) questionnaire²⁹ to calculate a summed score for adolescents' total weekly minutes of moderate-to-vigorous physical activity by capturing in and out of school weekday minutes and weekend minutes. Using the YAP questionnaire, adolescents' sedentary behavior was calculated based on adolescents' total weekly minutes of sedentary behavior. We used the International Physical Activity Questionnaire (IPAQ) Short Form³⁰ to calculate total parent physical activity by adding the number of moderate and vigorous physical activity minutes. For parents' sedentary behavior, we used the same IPAQ Short Form creating a variable capturing the total number of sitting minutes during the last 7 days.

Family-level factors

Home availability

We assessed parent- and adolescent-reported home availability of fruits and vegetables, junk food and sugar-sweetened beverages, and the presence of TV in adolescents' bedroom using items sourced from the National Youth Physical Activity and Nutrition Survey.²⁸ Response options ranged from 1 = “Never” to 5 = “Always.” We additionally assessed the availability and use of sports and exercise equipment among adolescents using an item from the “Active Where? Survey.”³¹

Domain-specific parenting practices

We calculated summed scores for parent- and adolescent-reported domain-specific parenting practices for fruit and vegetable intake, junk food and sugar intake, physical activity, and use of electronic devices using items sourced using the Child Feeding Questionnaire,³² Comprehensive Feeding Practices Questionnaire,³³ Parenting Feeding Style Questionnaire,³⁴ and the Legitimacy of Parental Authority instrument.³⁵ Response options ranged from 1 = “Strongly Disagree” to 5 = “Strongly Agree.” Cronbach's α ranged from 0.74 to 0.86.

Statistical Analyses

The dependent variable for this study was dyadic weight status. Sociodemographic, individual-, and family-level factors were the independent variables. We further categorized independent variables as being either diet or activity related and included them in separate models: (1) diet-related variables and (2) activity-related variables. The statistical plan consisted of several steps. First, we examined descriptive statistics (including means, standard deviations, or frequencies) across all study variables. We assessed all predictors for collinearity by evaluating the acceptable variance inflation factor and tolerance values. We did not find any collinearity issues, with tolerance values ranging from 0.40 to 0.96 and variance inflation factor (VIF) ranging from 1.04 to 2.5. Second, we created the dyadic weight status variable using parent and adolescent BMI categories. Third, we used multiple logistic regression models to examine associations between dyadic weight status and multiple predictors.

To examine the additive effects of all sociodemographic, individual-, and family-level factors on dyadic weight status, we ran a series of stepwise multinomial logistic regression models. The reference category for all analyses was the Healthy Weight Concordance group. First, we entered all the sociodemographic variables of interest (i.e., parents' and adolescents' sex, age, marital status, SES, and race) into Model 1. Second, we entered individual-level variables into the model with demographic variables (Models 2 and 4). For the dietary model, individual-level variables included adolescents' and parents' emotional regulation, emotional eating, fruit and vegetable intake, and junk food/sugar-sweetened beverage (SSB) intake. For the activity model, individual-level variables included adolescents' and parents' emotional regulation, self-efficacy for physical activity, self-reported physical activity levels, and sedentary behavior.

Third, we added family-level factors (Models 3 and 5). For the dietary model, family-level variables included fruit and vegetable availability, SSB availability, and domain-specific parenting practices for fruit and vegetable intake and junk food/SSB intake. For the activity model, family-level variables included adolescents' TV in bedroom availability, adolescents' and parents' sports equipment activity, and domain-specific parenting practices for physical activity and electronic devices. In efforts to maintain clarity and simplicity when reporting our results, we only report results from the final models (Model 3-Diet and Model 5-Activity).

We estimated unstandardized logistic regression coefficients (b), odds ratios (ORs), and 95% confidence intervals (CIs). We applied Bonferroni corrections to all test p-values with a significance level of 0.05 and further dividing by the total number of predictors in the model to adjust for multiple testing and comparisons. We handled missing data using full-information maximum likelihood,³⁶ and ran all logistic analyses using Mplus version 8.0.

Results

We included a total of 1516 parent/adolescent dyads in this study (Table 1; Supplementary Table S1). Adolescents ranged from 12 to 17 years old (12–13 years old, 32.8%; 13–14 years old, 34.8%; 16–17 years old, 32.4%), about half were female (50.7%), mostly white (63.4%), and were in the normal weight range (71.2%). Most parents were between 25 to 55 years old (86%), female (73.7%), married (71.5%), employed (66.6%), not covered by Medicaid insurance (66.7%), and more than half of parents had either OW (30.8%) or OB (32.2%).

Table 1.

Family Life, Activity, Sun, Health, and Eating Participant's Sociodemographic Characteristics (n = 1516)

	Mean (SD) or n (%)
Adolescents
Age	14.48 (1.61)
12–13	498 (32.8%)
14–15	527 (34.8%)
16–17	491 (32.4%)
Sex
Male	748 (49.3%)
Female	768 (50.7%)
Race
Hispanic	153 (10.1%)
Non-Hispanic black	257 (17.1%)
White	951 (63.4%)
Other	140 (9.3%)
Adolescent weight status
Normal	1079 (71.2%)
Overweight	239 (15.8%)
Obese	198 (13.1%)
Parents
Age
18–34	173 (11.4%)
35–44	656 (43.3%)
45–59	647 (42.7%)
60+	39 (2.57%)
Sex
Male	399 (26.3%)
Female	1116 (73.7%)
Marital status
Married	1074 (71.5%)
Divorced, widowed, separated	192 (12.8%)
Never married	150 (9.9%)
In relationship (unmarried)	87 (5.8%)
Education
<High school	18 (1.2%)
HS GED	246 (16.3%)
Some college	544 (36%)
4-Year college degree+	703 (46.5%)
Employment
Employed or self-employed	1005 (66.6%)
Unemployed, homemaker, student, retired	505 (33.4%)
Income
$0–99,999	1190 (79.3%)
$100,000+	310 (20.7%)
Medicaid insurance
Not applicable	102 (6.8%)
Yes	396 (26.5%)
No	999 (66.7%)
Parent weight status
Normal	561 (37.0%)
Overweight	467 (30.8%)
Obese	488 (32.2%)

GED, General Educational Development; HS, high school; SD, standard deviation.

We categorized parents and adolescents into the following dyadic BMI status groups: 475 (31.3%) in the Healthy Weight Concordance group, 351 (23.2%) in the OW/OB Concordance group, 604 (39.8%) in the Discordance-Parent OW/OB group, and 86 (5.7%) in the Discordance-Adolescent OW/OB group. The following sections describe the observed results from the dietary model (Table 2) and activity model (Table 3).

Table 2.

Multinomial Logistic Regression (Odds Ratio and 95% Confidence Intervals) for Concordance and Discordance of BMI Patterns—Diet Model (N = 1516)

	Model 1			Model 2			Model 3
	Discordance-Parent OW/OB	Discordance-Adolescent OW/OB	OW/OB Concordance	Discordance-Parent OW/OB	Discordance-Adolescent OW/OB	OW/OB Concordance	Discordance-Parent OW/OB	Discordance-Adolescent OW/OB	OW/OB Concordance
Reference group	Healthy Weight Concordance			Healthy Weight Concordance			Healthy Weight Concordance
Sociodemographic
Adolescent age	0.99 (0.91–1.06)	0.92 (0.80–1.07)	0.92 (0.84–1.01)
Adolescent sex	0.78 (0.61–0.99)	1.28 (0.80–2.04)	1.03 (0.78–1.37)
Parent age	1.21 (1.01–1.46)	1.29 (0.90–1.85)	0.99 (0.80–1.23)
Parent sex (female, male)	1.93 (1.44–2.59)	1.27 (0.72–2.23)	1.58 (1.12–2.43) ^a	2.31 (1.69–3.16) ^a	1.58 (0.87–2.85)	1.90 (1.31–2.76) ^a	2.24 (1.63–3.07) ^a	1.52 (0.84–2.74)	1.89 (1.30–2.75) ^a
Marital status	1.08 (0.92–1.26)	1.06 (0.80–1.39)	1.17 (0.99–1.40)
SES (higher, lower)	1.78 (1.30–2.46) ^a	1.66 (0.92–2.99)	2.56 (1.81–3.63) ^a	1.71 (1.23–2.36) ^a	1.67 (0.93–2.99)	2.65 (1.86–3.79) ^a	1.75 (1.25–2.44) ^a	1.59 (0.87–2.90)	2.64 (1.84–3.77) ^a
Race (white, nonwhite)	1.66 (1.24–2.20) ^a	1.45 (0.85–2.47)	1.72 (1.25–2.37) ^a	1.83 (1.36–2.46) ^a	1.60 (0.92–2.77)	2.26 (1.62–3.16) ^a	1.98 (1.46–2.68) ^a	1.62 (0.91–2.88)	2.27 (1.63–3.18) ^a
Individual-level
Adolescent emotional regulation				1.02 (0.98–1.06)	1.06 (0.99–1.14)	1.02 (0.98–1.07)	1.01 (0.98–1.05)	1.06 (0.99–1.14)	1.02 (0.97–1.06)
Parent emotional regulation				0.99 (0.96–1.03)	0.99 (0.92–1.05)	1.03 (0.99–1.08)	1.00 (0.96–1.04)	0.98 (0.92–1.05)	1.03 (0.98–1.07)
Adolescent emotional eating				1.05 (0.98–1.13)	1.14 (1.01–1.29)	1.15 (1.07–1.25) ^a	1.04 (0.97–1.12)	1.11 (0.98–1.26)	1.14 (1.05–1.24) ^a
Parent emotional eating				1.19 (1.10–1.28) ^a	1.08 (0.95–1.23)	1.21 (1.11–1.32) ^a	1.21 (1.12–1.30) ^a	1.10 (0.97–1.25)	1.22 (1.11–1.34) ^a
Adolescent fruit and vegetable intake				0.97 (0.91–1.05)	0.97 (0.83–1.13)	0.87 (0.79–0.96) ^a	1.01 (0.94–1.09)	1.00 (0.86–1.17)	0.90 (0.81–1.00)
Parent fruit and vegetable intake				0.92 (0.85–0.99)	0.90 (0.77–1.06)	0.92 (0.83–1.01)	0.95 (0.88–1.04)	0.94 (0.79–1.13)	0.96 (0.86–1.06)
Adolescent junk food/SSB intake				0.95 (0.88–1.02)	1.00 (0.85–1.17)	1.00 (0.92–1.09)	0.92 (0.85–1.00)	0.97 (0.83–1.14)	0.96 (0.88–1.06)
Parent junk food/SSB intake				1.04 (0.95–1.13)	0.96 (0.80–1.15)	1.00 (0.90–1.11)	1.02 (0.93–1.13)	0.97 (0.80–1.17)	0.99 (0.88–1.11)
Family- and environmental level
Adolescent fruit and vegetable availability							1.03 (0.84–1.25)	0.94 (0.68–1.30)	0.94 (0.75–1.16)
Parent fruit and vegetable availability							0.93 (0.76–1.15)	0.86 (0.59–1.25)	0.86 (0.68–1.09)
Adolescent SSBs availability							1.15 (0.98–1.34)	1.07 (0.83–1.37)	1.12 (0.95–1.32)
Parent SSBs availability							0.93 (0.80–1.07)	0.91 (0.71–1.16)	1.01 (0.85–1.19)
Adolescent parenting practices fruit and vegetable							0.98 (0.95–1.02)	0.93 (0.88–0.98)	0.98 (0.94–1.02)
Parent parenting practices fruit and vegetable							0.97 (0.93–1.00)	1.00 (0.94–1.06)	1.00 (0.96–1.05)
Adolescent parenting practices junk food/SSBs							0.99 (0.95–1.03)	1.04 (0.97–1.10)	0.99 (0.95–1.03)
Parent parenting practices junk food/SSBs							1.00 (0.96–1.04)	1.01 (0.94–1.09)	1.03 (0.98–1.08)

Significant after adjustment for multiple comparisons using Bonferroni correction.

SES, socioeconomic status; SSB, sugar-sweetened beverage.

Table 3.

Multinomial Logistic Regression (Odds Ratio and 95% confidence intervals) for Concordance and Discordance of BMI Patterns—Activity Model (N = 1516)

	Model 1 (repeat)			Model 4			Model 5
	Discordance-Parent OW/OB	Discordance-Adolescent OW/OB	OW/OB Concordance	Discordance-Parent OW/OB	Discordance-Adolescent OW/OB	OW/OB Concordance	Discordance-Parent OW/OB	Discordance-Adolescent OW/OB	OW/OB Concordance
Reference group	Healthy Weight Concordance			Healthy Weight Concordance			Healthy Weight Concordance
Sociodemographic
Adolescent age	0.99 (0.91–1.06)	0.92 (0.80–1.07)	0.92 (0.84–1.01)
Adolescent sex	0.78 (0.61–0.99)	1.28 (0.80–2.04)	1.03 (0.78–1.37)
Parent age	1.21 (1.01–1.46)	1.29 (0.90–1.85)	0.99 (0.80–1.23)
Parent sex (female, male)	1.93 (1.44–2.59)	1.27 (0.72–2.23)	1.58 (1.12–2.43) ^a	2.27 (1.67–3.01) ^a	1.45 (0.81–2.60)	1.72 (1.19–2.48) ^a	2.33(1.70, 3.21) ^a	1.51(0.85, 2.68)	1.80(1.24, 2.61) ^a
Marital status	1.08 (0.92–1.26)	1.06 (0.80–1.39)	1.17 (0.99–1.40)
SES (higher, lower)	1.78 (1.30–2.46) ^a	1.66 (0.92–2.99)	2.56 (1.81–3.63) ^a	1.61 (1.16–2.23) ^a	1.60 (0.90–2.82)	2.41 (1.70–3.42) ^a	1.50(1.08, 2.08) ^a	1.55(0.87, 2.77)	2.31(1.62, 3.31) ^a
Race (white, nonwhite)	1.66 (1.24–2.20) ^a	1.45 (0.85–2.47)	1.72 (1.25–2.37) ^a	1.78 (1.33–2.39) ^a	1.46 (0.85–2.49)	2.13 (1.53–2.97) ^a	1.74(1.28, 2.35) ^a	1.47(0.83, 2.61)	2.08(1.49, 2.92) ^a
Individual-level
Adolescent emotional regulation				1.03 (1.00–1.07)	1.08 (1.01–1.15)	1.05 (1.00–1.09)	1.03 (0.99–1.07)	1.08 (1.01–1.15)	1.04 (1.00–1.09)
Parent emotional regulation				1.01 (0.97–1.05)	0.99 (0.93–1.06)	1.05 (1.00–1.09)	1.01 (0.97–1.05)	0.99 (0.93–1.06)	1.05 (1.00–1.09)
Adolescent self-efficacy for PA				0.90 (0.79–1.03)	0.71 (0.57–0.89) ^a	0.71 (0.62–0.82) ^a	0.94 (0.82–1.07)	0.72 (0.56–0.91)	0.74 (0.64–0.85) ^a
Parent self-efficacy for PA				0.76 (0.68–0.84) ^a	0.90 (0.73–1.10)	0.73 (0.65–0.83) ^a	0.78 (0.70–0.87) ^a	0.88 (0.72–1.07)	0.72 (0.64–0.82) ^a
Parent physical activity				1.00 (1.00–1.00)	1.00 (1.00–1.00)	1.00 (1.00–1.00)	1.00 (1.00–1.00)	1.00 (1.00–1.01)	1.00 (1.00–1.00)
Adolescent physical activity				1.00 (1.00–1.00)	1.00 (1.00–1.01)	1.00 (1.00–1.00)	1.00 (1.00–1.00)	1.00 (1.00–1.00)	1.00 (1.00–1.00)
Adolescent sedentary behavior				1.00 (1.00–1.00)	1.00 (1.00–1.01)	1.00 (1.00–1.00)	1.00 (1.00–1.00)	1.00 (1.00–1.01)	1.00 (1.00–1.00)
Parent sedentary behavior				1.02 (0.99–1.05)	1.03 (0.97–1.09)	1.03 (0.99–1.06)	1.02 (0.99–1.05)	1.02 (0.96–1.09)	1.03 (0.99–1.06)
Family- and environmental level
Adolescent TV in bedroom							0.82 (0.62–1.09)	0.68 (0.39–1.20)	0.77 (0.55–1.07)
Adolescent sports equipment							0.92 (0.84–1.01)	0.94 (0.79–1.12)	0.93 (0.83–1.03)
Adolescent parenting practices physical activity							0.96 (0.92–0.99)	1.011 (0.94–1.08)	0.98 (0.94–1.03)
Parent parenting practices physical activity							0.98 (0.94–1.02)	1.09 (1.01–1.18)	1.04 (0.99–1.08)
Adolescent parenting practices screen time							1.03 (1.00–1.07)	0.96 (0.90–1.02)	0.98 (0.94–1.02)
Parent parenting practices screen time							0.99 (0.95–1.03)	0.93 (0.86–1.00)	0.99 (0.94–1.04)

Significant after adjustment for multiple comparisons using Bonferroni correction.

PA, physical activity.

Dietary Model

In the final diet model (Table 2, Model 3), sociodemographic variables including parental sex, SES, and race, as well as the individual-level predictor of emotional eating, all remained significant predictors of dyadic weight status. Male parental sex was significantly associated with parent/adolescent dyads being in either the Discordance-Parent OW/OB (OR = 2.24; 95% CI = 1.63–3.07) or OW/OB Concordance groups (OR = 1.89; 95% CI = 1.30–2.75) vs. the Healthy Weight Concordance group.

Lower family SES was significantly associated with parent/adolescent dyads being in either the Discordance-Parent OW/OB (OR = 1.75; 95% CI = 1.25–2.44) or OW/OB Concordance groups (OR = 2.64; 95% CI = 1.84–3.77) vs. the Healthy Weight Concordance group. Identifying as a minority (nonwhite) was significantly associated with parent/adolescent dyads being in either the Discordance-Parent OW/OB (OR = 1.98; 95% CI = 1.46–2.68) or OW/OB Concordance groups (OR = 2.27; 95% CI = 1.63–3.18) vs. the Healthy Weight Concordance group.

Among the individual-level factors, higher levels of adolescent emotional eating were significantly associated with parent/adolescent dyads being in the OW/OB Concordance group (OR = 1.14; 95% CI = 1.05–1.24) vs. the Healthy Weight Concordance group. Higher levels of parent emotional eating were also still associated with parent/adolescent dyads being in either the Discordance-Parent OW/OB (OR = 1.21; 95% CI = 1.12–1.30) or OW/OB Concordance groups (OR = 1.22; 95% CI = 1.11–1.34) vs. the Healthy Weight Concordance group.

None of the family-level factors in this model was significantly associated with dyadic weight status after applying the Bonferroni correction.

Physical Activity Model

In the final physical activity model (Table 3, Model 5), sociodemographic variables, including parental sex, SES, and race, as well as individual-level predictor of physical activity self-efficacy for both parents and adolescents, all remained significant predictors of dyadic weight status. Male parental sex was significantly associated with parent/adolescent dyads being in either the Discordance-Parent OW/OB (OR = 2.33; 95% CI = 1.70–3.21) or OW/OB Concordance groups (OR = 1.80; 95% CI = 1.24–2.61) vs. Healthy Weight Concordance group.

Lower family SES was significantly associated with parent/adolescent dyads being in either the OW/OB Concordance groups (OR = 2.31; 95% CI = 1.62–3.31) vs. the Healthy Weight Concordance group. Identifying as a minority (nonwhite) was significantly associated with parent/adolescent dyads being either in the Discordance-Parent OW/OB (OR = 1.74; 95% CI = 1.28–2.35) or OW/OB Concordance groups (OR = 2.08; 95% CI = 1.49–2.92) vs. the Healthy Weight Concordance group.

Among the individual-level factors, higher levels of adolescent self-efficacy for physical activity (PA) were associated with parent/adolescent dyads being in the Healthy Weight Concordance group vs. the OW/OB Concordance group (OR = 0.74; 95% CI = 0.64–0.85). Higher levels of parent physical activity self-efficacy were associated with parent/adolescent dyads being in the Healthy Weight Concordance group, vs. being in either the Discordance-Parent OW/OB (OR = 0.78; 95% CI = 0.70–0.87) or OW/OB Concordance groups (OR = 0.72; 95% CI = 0.64–0.82).

Discussion

Prior literature has shown that weight status is generally concordant among parent/adolescent dyads (i.e., both parent and adolescent with healthy weight, or OW/OB), but there is a dearth of studies examining discordance in families, and more specifically, parent/adolescent dyads (i.e., OW/OB weight parent and healthy weight adolescent, or vice versa) as well as factors predictive of weight status concordance or discordance.

The current study sought to evaluate discordance in parent/adolescent dyads using a national study and to explore socioecological factors that place parents and adolescents at greater risk for OB by examining associations between sociodemographic, individual-, and family-level factors and weight status concordance and discordance in parent/adolescent dyads. First, we examined the discordance in families and found that the majority of families were concordant in their BMI; there were a very small proportion of adolescents who were in the OW/OB range if they had a normal weight parent. Next, after assessing socioecological predictors, our main findings from our study showed that fathers (vs. mothers), lower SES, and identifying as minority were significantly associated with parent/adolescent dyads being more likely to have an OW/OB dyadic weight status (either Discordance-Parent OW/OB or OW/OB Concordance) than a healthy one (Healthy Weight Concordance).

At the individual-level, adolescent and parental emotional eating was associated with dyadic weight status, such that parental emotional eating was associated with Discordance-Parent OW/OB, and both adolescent and parental emotional eating was associated with OW/OB Concordance BMI. Regarding physical activity, parental physical activity self-efficacy was associated with Discordance-Parent OW/OB, and both adolescent and parental physical activity self-efficacy was associated with OW/OB Concordance BMI. None of our family-level predictors was significant.

Our sample included 13% of adolescents and 32% of adults with OB, compared with the general US population of 21% of adolescents (12–19 years old) and 43% of adults (20+ years old) in the OB range.^1,2 Thus, our sample had lower rates of OB compared with the average American adolescents or adults. More than half (55%) of the parent/adolescent dyads in our sample had a concordant (healthy weight or OW/OB) weight status, which is consistent with the literature on intergenerational transmission of parent and child BMI.⁷

Approximately 40% of dyads were categorized as Discordance-Parent OW/OB, suggesting an increased risk for the adolescent becoming OW or OB in the future. The Discordant-Adolescent OW/OB group comprised the smallest number of dyads (6%), which suggests that the combination of a normal weight parent with an OW/OB weight child is rare; this finding supports the notion that parents pass on contributing lifestyle behaviors that support normal weight status in their children.

In our study, female parental sex was significantly associated with being more likely to be in Healthy Weight Concordance vs. OW/OB Concordance or the Discordance-Parent OW/OB groups. We believe this finding may be because mothers, who made up approximately three-quarters of the sample, had slightly lower levels of OW or OB (60%) compared with fathers (70%).

Consistent with our findings, Freeman et al., conducted a longitudinal study involving 3285 families and found that having an OW/OB weight father and healthy weight mother (vs. healthy weight parents) led to a 15 times higher likelihood of childhood OB.³⁷ The authors also reported that families with an OW/OB weight mother and healthy weight father did not significantly increase the likelihood of OW/OB weight in youth. While the FLASHE study did not include both parents as respondents, our results show that fathers may be just as influential, if not more so, on their adolescent's weight status and OW/OB concordance within their families.

Parental sociodemographic variables held across both dietary and activity models, which highlights their influence on weight status within families. As in our study, high SES is consistently linked to better health outcomes within families.^38,39 Using data from NHANES and the National Survey of Children's Health, Frederick et al. found that the prevalence of OB among high-SES adolescents decreased in recent years, whereas the prevalence among low-SES peers has continued to increase over time.³⁸ Additional analyses suggested that these differences were due to differential levels of physical activity and caloric intake.

Other studies also observe this widening disparity with lower prevalence of OB among youth from wealthier families and higher prevalence in poorer families.³⁹ Overall, the intergenerational transmission literature consistently shows that families from low SES often have strong familial clustering of OW/OB.^6,40

We found that race/ethnicity was associated with dyadic weight status, with those identifying as white having a higher likelihood of being in healthy weight concordant families vs. those identifying as nonwhite more likely to be in OW/OB or Discordance-Parent OW/OB groups. These results align with observed disparities for both adults and adolescents among Hispanic and non-Hispanic black populations.

Parents' and adolescents' emotional eating and low physical activity self-efficacy were associated with OW/OB dyadic weight status. Emotional eating may lead parents and adolescents to consume more energy-dense, nutrient-poor foods leading to higher BMI within families or OW/OB concordance.⁴¹ This finding is consistent with a review conducted by Puder and Munsch that showed that familial patterns of eating behavior, particularly those with high levels of familial stress, may contribute to the development of childhood OB.⁴² Similarly, previous research has shown that adolescents with OB often report lower levels of PA self-efficacy when compared with their healthy weight peers.¹⁸

There are several limitations that should be noted. First, this is a cross-sectional study, so causality and temporal sequence could not be established. Future longitudinal research could explore these associations further as well as reasons for no direct effects of family-level variables and parent/adolescent BMI status patterns. It could be that family-level predictors are not directly associated with the outcome of dyadic BMI status but rather that these family-level variables may be indirectly associated through individual-level variables.

Second, measures included in our analyses were self-reported and thus may have been subject to recall- or social-desirability bias. The literature has shown that self-reported anthropometric data (i.e., height, weight, BMI) could be prone to an underestimation of weight, especially in females, and an overestimation in height.^43,44

As noted above, another limitation is the study's web-based sampling strategy, which can sometimes result in nonrepresentativeness. Balancing techniques were used, and the participants matched the US population on most key demographic variables (i.e., age, census division, and adolescents' gender), but the FLASHE sample included more mothers than fathers and tended to be highly educated. Future research should be conducted to replicate our findings with balanced samples.

A strength of this study was the use of the FLASHE data set, which is a large national data set. This study also represents one of the first to examine weight status discordance and concordance among parents and adolescents.

Conclusions

In conclusion, our findings highlight the strength of OW/OB concordance in families, especially among low-SES and ethnic minority populations and the significance of proximal individual-level factors for both adolescents and parents, namely, emotional eating and physical activity self-efficacy in contributing to dyadic weight status. More research is needed to understand these associations over time.

Footnotes

Authors' Contributions

The authors' responsibilities were as follows: M.A.K. and S.M.S. conceptualized and designed the study. M.A.K. and T.K.L. analyzed the data. M.A.K. and R.M.B. drafted the article. T.K.L. and S.M.S. critically reviewed the article for intellectual content. All authors approved the final article as submitted and agree to be accountable for all aspects of the work.

Funding Information

No funding source to declare.

Author Disclosure Statement

The authors have no conflicts of interest relevant to this article to disclose.

Supplementary Material

References

Fryar

, Carroll

, Afful

. Prevalence of overweight, obesity, and severe obesity among adults aged 20 and over: United

States

, 1960–1962 through 2017–2018. NCHS Health

E-Stats

. 2020. https://www.cdc.gov/nchs/data/hestat/obesity-adult-17-18/obesity-adult.htm (Last accessed January 10, 2022).

Fryar

, Carroll

, Afful

. Prevalence of overweight, obesity, and severe obesity among children and adolescents aged 2–19 years: United

States

, 1963–1965 through 2017–2018. NCHS Health

E-Stats 2020

. https://www.cdc.gov/nchs/data/hestat/obesity-child-17-18/overweight-obesity-child-H.pdf (Last accessed January 10, 2022).

Centers for Disease Control and Prevention. Adult Obesity Causes & Consequences. 2021. Available at https://www.cdc.gov/obesity/adult/causes.html (Last accessed January 10, 2022).

Bibbins-Domingo

, Coxson

, Pletcher

, et al. Adolescent overweight and future adult coronary heart disease. N Engl J Med, 2007; 357:2371–2379.

Heslehurst

, Vieira

, Akhter

, et al. The association between maternal body mass index and child obesity: A systematic review and meta-analysis. PLoS Med, 2019; 16:e1002817.

Naess

, Sund

, Holmen

, Kvaløy

. Implications of parental lifestyle changes and education level on adolescent offspring weight: A population based cohort study-The HUNT Study, Norway. BMJ Open, 2018; 8:e023406.

James

, Mendolia

, Paloyo

. Intergenerational transmission of body mass and obesity status in Australia. Econom Record, 2020; 96:1–18.

Sahoo

, Sahoo

, Choudhury

, et al. Childhood obesity: Causes and consequences. J Fam Med Prim Care, 2015; 4:187–192.

Klein

, Lytle

, Chen

. Social ecological predictors of the transition to overweight in youth: Results from the Teens Eating for Energy and Nutrition at Schools (TEENS) study. J Am Dietet Assoc, 2008; 108:1163–1169.

10.

Aparicio

, Canals

, Arija

, et al. The role of emotion regulation in childhood obesity: Implications for prevention and treatment. Nutr Res Rev, 2016; 29:17–29.

11.

Konttinen

, Silventoinen

, Sarlio-Lähteenkorva

, et al. Emotional eating and physical activity self-efficacy as pathways in the association between depressive symptoms and adiposity indicators. Am J Clin Nutr, 2010; 92:1031–1039.

12.

Lundeen

, Park

, Onufrak

, et al. Adolescent sugar-sweetened beverage intake is associated with parent intake, not knowledge of health risks. Am J Health Promot, 2018; 32:1661–1670.

13.

Raynor

, Van Walleghen

, Osterholt

, et al. The relationship between child and parent food hedonics and parent and child food group intake in children with overweight/obesity. J Am Dietet Assoc, 2011; 111:425–430.

14.

Craig

, Cameron

, Tudor-Locke

. Relationship between parent and child pedometer-determined physical activity: A sub-study of the CANPLAY surveillance study. Int J Behav Nutr Phys Activity, 2013; 10:1–8.

15.

Herd

, King-Casas

, Kim-Spoon

. Developmental changes in emotion regulation during adolescence: Associations with socioeconomic risk and family emotional context. J Youth Adolesc, 2020; 49:1545–1557.

16.

Trost

, Owen

, Bauman

, et al. Correlates of adults' participation in physical activity: Review and update. Med Sci Sports Exerc, 2002; 34:1996–2001.

17.

Nezami

, Lang

, Jakicic

, et al. The effect of self-efficacy on behavior and weight in a behavioral weight-loss intervention. Health Psychol, 2016; 35:714–722.

18.

Trost

, Kerr

, Ward

, Pate

. Physical activity and determinants of physical activity in obese and non-obese children. Int J Obes, 2001; 25:822–829.

19.

Verzeletti

, Maes

, Santinello

, et al. Food-related family lifestyle associated with fruit and vegetable consumption among young adolescents in Belgium Flanders and the Veneto Region of Italy. Appetite, 2010; 54:394–397.

20.

Vaughn

, Ward

, Fisher

, et al. Fundamental constructs in food parenting practices: A content map to guide future research. Nutr Rev, 2016; 74:98–117.

21.

Van Lippevelde

, Verloigne

, De Bourdeaudhuij

, et al. Does parental involvement make a difference in school-based nutrition and physical activity interventions? A systematic review of randomized controlled trials. Int J Public Health, 2012; 57:673–678.

22.

Nepper

, Chai

. Associations of the home food environment with eating behaviors and weight status among children and adolescents. J Nutr Food Sci, 2015; S12:1.

23.

Amuta

, Jacobs

, Idoko

, et al. Influence of the home food environment on children's fruit and vegetable consumption: A study of rural low-income families. Health Promot Pract, 2015; 16:689–698.

24.

Tandon

, Grow

, Couch

, et al. Physical and social home environment in relation to children's overall and home-based physical activity and sedentary time. Prevent Med, 2014; 66:39–44.

25.

Nebeling

, Hennessy

, Oh

, et al. The FLASHE study: Survey development, dyadic perspectives, and participant characteristics. Am J Prevent Med, 2017; 52:839–848.

26.

Tanofsky-Kraff

, Ranzenhofer

, Yanovski

, et al. Psychometric properties of a new questionnaire to assess eating in the absence of hunger in children and adolescents. Appetite, 2008; 51:148–155.

27.

Harter

. The perceived competence scale for children. Child Dev, 1982; 53:87–97.

28.

Song

, Carroll

, Lee

, Fulton

. Physical activities of US high school students—2010 National Youth Physical Activity and Nutrition Survey. J Phys Activity Health, 2015; 12(s1):S11–S17.

29.

Saint-Maurice

, Welk

. Validity and calibration of the youth activity profile. PLoS One, 2015; 10:e0143949.

30.

Craig

, Marshall

, Sjöström

, et al. And the IPAQ consensus group and the IPAQ reliability and validity study group. international physical activity questionnaire (IPAQ): 12-country reliability and validity. Med Sci Sports Exerc, 2003; 35:81–95.

31.

Kerr

, Sallis

, Rosenberg

, et al. Active where. In: Surveys: Active Living Research. 2008. https://drjimsallis.org/measure_activewhere.html (Last accessed January 10, 2022).

32.

Birch

, Fisher

, Grimm-Thomas

, et al. Confirmatory factor analysis of the Child Feeding Questionnaire: A measure of parental attitudes, beliefs and practices about child feeding and obesity proneness. Appetite, 2001; 36:201–210.

33.

Musher-Eizenman

, Holub

. Comprehensive feeding practices questionnaire: Validation of a new measure of parental feeding practices. J Pediatr Psychol, 2007; 32:960–972.

34.

Wardle

, Sanderson

, Guthrie

, et al. Parental feeding style and the inter-generational transmission of obesity risk. Obes Res, 2002; 10:453–462.

35.

Darling

, Cumsille

, Martínez

. Individual differences in adolescents' beliefs about the legitimacy of parental authority and their own obligation to obey: A longitudinal investigation. Child Dev, 2008; 79:1103–1118.

36.

Enders

. Applied Missing Data Analysis. New York, NY: Guilford Press, 2010.

37.

Freeman

, Fletcher

, Collins

, et al. Preventing and treating childhood obesity: Time to target fathers. Int J Obes (2005), 2012; 36:12–15.

38.

Frederick

, Snellman

, Putnam

. Increasing socioeconomic disparities in adolescent obesity. Proc Natl Acad Sci USA, 2014; 111:1338–1342.

39.

Chung

, Backholer

, Wong

, et al. Trends in child and adolescent obesity prevalence in economically advanced countries according to socioeconomic position: A systematic review. Obes Rev, 2016; 17:276–295.

40.

Umer

, Kelley

, Cottrell

, et al. Childhood obesity and adult cardiovascular disease risk factors: A systematic review with meta-analysis. BMC Public Health, 2017; 17:683.

41.

van Strien

, Cebolla

, Etchemendy

, et al. Emotional eating and food intake after sadness and joy. Appetite, 2013; 66:20–25.

42.

Puder

, Munsch

. Psychological correlates of childhood obesity. Int J Obes, 2010; 34:S37–S43.

43.

McAdams

, Van Dam

, Hu

. Comparison of self-reported and measured BMI as correlates of disease markers in US adults. Obesity, 2007; 15:188.

44.

Gorber

, Tremblay

, Moher

, Gorber

. A comparison of direct vs. self-report measures for assessing height, weight and body mass index: A systematic review. Obes Rev, 2007; 8:307–326.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB