Family functioning mediates the relationship between child behavior problems and parent feeding practices in youth with overweight or obesity

Abstract

Parental feeding practices play a significant role in children’s health behaviors. Given the high prevalence of childhood obesity, it is important to examine factors that may influence parental feeding practices. This study examined the role of family functioning on the relationship between child behavior problems and parent feeding practices. We hypothesized that higher problematic child behavior would correlate with lower parental engagement in healthy parent feeding practices, with greater family dysfunction mediating that relationship. Participants (n = 220) were rural-dwelling parents of school-aged children with overweight or obesity. Participants completed the McMaster Family Assessment Device, Child Behavior Checklist (CBCL), and Child Feeding Questionnaire at baseline. Mediation models were used with youth behavioral problems (CBCL) as the independent variable, parental feeding practices as the dependent variable, and family functioning as the mediator. Two significant mediation models revealed (a) family functioning mediated the relationship between child behavior problems and parental perception of responsibility taking for child eating, and (b) family functioning mediated the relationship between child behavior problems and parental monitoring of their child’s eating. More child behavior problems were associated with unhealthier family functioning, which was associated with less parental monitoring of children’s unhealthy food intake and reduced responsibility taking for feeding and meal planning of their children. These findings suggest the synergistic effect of child behavior problems with poorer family functioning may be a risk factor for unhealthy parent feeding practices in children with overweight or obesity. If these relationships are supported by longitudinal research, behavior interventions for children with obesity should consider poor family functioning and child behavior as potential barriers to parental monitoring and responsibility of children’s feeding practices.

Child behavior problems and poor family functioning may be combined risk factors for unhealthy parent-feeding practices in families with school-aged children with overweight or obesity.

Implications.

Practice: The results of this study point to other potential contributors to the obesity epidemic. Clinicians should consider providing strategies to enhance family communication and behavioral management to families with children who are overweight or obese.

Policy: Policymakers who aim to reduce the obesity epidemic should consider how other modifiable factors within the family unit are associated with child eating.

Research: Future research should explore these relationships using a longitudinal design to assess causality. If shown with longitudinal research, future interventions should consider how to address behavior management and family communication strategies within a behavioral family lifestyle intervention for pediatric obesity.

INTRODUCTION

Childhood obesity is a major public health concern affecting approximately one in five youth in the USA [1]. Children with obesity are more likely to become adults with obesity, and are at greater risk for negative physical and psychological outcomes [2–5]. Behavioral family lifestyle interventions, which aim to help families make small but meaningful changes to the family unit’s dietary intake and physical activity, are currently the gold standard for pediatric obesity treatment. A recent meta-analysis shows small to moderate effect sizes for the efficacy of these behavioral interventions [6], indicating that many, but not all youth benefit from these programs. As such, more recent research efforts have focused on identifying and understanding other modifiable factors that may influence obesity risk in order to (a) find novel targets for prevention or intervention and (b) enhance intervention outcomes.

PARENT FEEDING ATTITUDES AND PRACTICES

Numerous studies show that the family plays a significant role in the development and maintenance of childhood obesity [7–9]. For example, child and adolescent dietary intake (an important obesity-relevant behavior) is shaped by the socioemotional environment created by parents in the home [10]. One such contributor to that environment includes parental feeding practices; that is, the specific goal-directed behaviors parents may use to directly influence their children’s eating. Parents typically exert substantial control over their children’s feeding environment, often determining the types and frequency of available foods inside and outside the home [10]. Parents may increase or decrease their child’s overall intake or consumption of specific foods in a variety of ways. For example, they may take responsibility for their child’s eating (perceived responsibility), restrict certain types of food (restriction), or monitor their child’s dietary intake (monitoring) [11]. Each of these behaviors plays a role in children’s dietary quality and consistency.

Overall, studies support the notion that parents may vary their feeding practices based on different child- or family-level factors. This is important, as different parental feeding practices may have differential effects depending on the weight status of the child [12]. For example, among families of 5-year olds with overweight or obesity, higher parent-perceived responsibility for their children’s feeding and greater monitoring of their children’s eating predicted a lower child body mass index (BMI) z-scores at age 7 [13]. Additionally, others have shown that greater parental monitoring of children’s food intake is associated with lower sugar, fat, and calorie content of the foods children choose to eat [14]. Thus, there is evidence that moderate levels of parental monitoring and sense of responsibility for their child’s feeding is associated with healthier eating habits and weight status. Conversely, parental restriction of child eating (another parent feeding practice) may lead to behaviors associated with weight gain, such as eating in the absence of hunger [15]. Although this literature is largely mixed, most research shows an association between increased parental restriction of child eating and increased child weight gain [16]. Unfortunately, parents who perceive their children as overweight may be the most likely to restrict their children’s food intake, leading to further weight gain [16]. Given that it is already well established that parental feeding attitudes and practices influence child weight status, a next step in this line of research is identifying other factors that may influence parental feeding practice, and thus may serve as novel targets for intervention to address unhealthy eating habits, and ultimately pediatric obesity.

CHILDHOOD BEHAVIOR PROBLEMS

Youth with obesity often are teased and bullied, which may lead to behavior problems. For example, Datar and Sturm [17] discovered that girls with obesity had more parent- and teacher-reported behavior problems compared to nonoverweight girls. However, relatively fewer studies have examined the alternative side to this relationship: how child behavioral problems may influence obesity-related behaviors. There is compelling evidence that behavior problems, in and of themselves, may portend future weight gain. Indeed, Lumeng and colleagues [18] have longitudinal data indicating that greater parent-reported behavior problems in normal weight school-aged children predicted development of overweight 2 years later. Further, there is some evidence that children with higher parent-reported behavior problems lose less weight in family behavioral weight loss treatments [19]. In the diabetes literature, Cohen and colleagues [20] explored this question by investigating how children’s behavior and family cohesion predicted glycemic control 4 years later. They found that children with type 1 diabetes who exhibited significant externalizing behaviors (i.e., aggression, misbehaving) had poorer glycemic control (but not poorer adherence to checking their blood sugar) at follow-up. While the prevalence of obesity in the pediatric type 1 diabetes population is consistent with the general child population, this discussion is relevant to obesity because the authors hypothesized that it was children’s externalizing behaviors that may interfere with other nonmonitoring adherence-related behaviors, such as diet [20]. Experimental studies have found that distressed children are more likely to eat in the absence of hunger if their parents regularly used food to soothe the child’s negative emotions at home. This practice could potentially lead to the development of unhealthy coping/eating behaviors, and, ultimately, becoming overweight [21,22]. Although not considered a direct marker of family functioning, parental use of food to soothe their children may be one reflection of how the family addresses conflict or distress.

FAMILY FUNCTIONING IN PEDIATRIC OBESITY

Family functioning is an environmental factor that may be dysregulated both in families of children with behavioral problems and children who have obesity. Family functioning refers to how family members manage daily routines, family roles, communication, and emotionally support each other [23]. It has been established that on average families of children with obesity have greater family conflict and are less cohesive compared with families of children without obesity [24,25]. Obesity, similar to a chronic illness, often adds additional stress and medical requirements to families [26,27]. In other chronic illness populations, there are data suggesting a temporal ordering, such that family dysfunction predicts poorer later disease management adherence and disease outcomes [20,28,29]. In diabetes, Cohen and colleagues determined a temporal ordering using prospective data that poor family functioning leads to reduced parental monitoring and reminders to their children regarding disease management [20]. Rhee [23] suggested that poor family functioning may contribute to childhood obesity, as it creates additional stress on children, models an environment of unhealthy lifestyles, and contributes to poor energy regulation in children.

No studies to our knowledge have addressed the temporal association between family functioning and overweight; however, there is evidence from intervention trials suggesting that improving family functioning may precede weight-relevant changes. For example, family-centered obesity interventions [19] and family-based adherence interventions in type 1 diabetes [30] suggest that improving family communication and cohesion can lead to improved treatment outcomes. Specifically, Epstein and colleagues [19] targeted both children and their parents for weight loss and behavior change in their family behavioral weight loss trials. In this way, parents and their children were encouraged to work together to meet shared goals for improved diet and weight loss. This combined parent/child weight loss treatment was effective at producing greater weight loss compared to a control group, possibly pointing to effects of improved family functioning as one driver. Amongst families of youth with poorly controlled type 1 diabetes, Wysocki and colleagues [30] demonstrated that structured Behavioral Family Systems Theory (vs. standard care) not only improved family interactions, but also lowered youth’s blood sugar levels (which are sensitive to diet-related behaviors).

PURPOSE AND AIMS

It stands to reason that amongst families with poor family functioning, child behavior problems may be more difficult to manage. Taking that a step further, parents in families with both of these factors (i.e., children with behavior problems and family dysfunction) may be less able to provide healthy parent feeding practices. Supporting this idea is literature on child temperament, which shows that child temperament mediates the relationship between parents’ personality and parenting style. That is, parents of children with more emotional dysregulation were more likely to use harsh parenting styles [31]. This evidence suggests that children’s temperaments and behaviors can affect parenting style and overall family functioning. When considered in the context of meal times (which are considered to be mildly stressful for most families [23]), the combination of problematic child behaviors and family dysfunction may result in less than helpful parent feeding practices during that meal.

Taken together, there is evidence that parental feeding practices, childhood behavior problems, and family functioning are all independently related to child weight status. However, it is unknown how they operate together, and in what order. Given that child behavior problems are related to family functioning, and poor family functioning has been shown to contribute to childhood obesity [23], it is plausible that child behavior problems and family functioning work together to influence parental feeding behaviors, a direct contributor to childhood obesity. Thus, the purpose of this study was to assess the mediating role of family functioning in the relationship between overweight children’s behavior problems and parental feeding practices. We hypothesized that higher childhood behavior problems (measured via total score on the Child Behavior Check List) would be associated with lower parental engagement in health-promoting feeding practices (measured via the Perceived Responsibility, Restriction, and Monitoring subscales of the Child Feeding Questionnaire), and that higher family dysfunction (measured via the Global Family Functioning scale on the Family Assessment Device), would drive (mediate) that relationship. In order to examine alternative pathways between these variables, we also investigated alternative mediation models with childhood behavior problems mediating the relationship between family functioning and parent feeding practices.

METHODS

Participants and procedures

Participants included 220 treatment-seeking parents of children (ages 7–12) with overweight or obesity (BMI ≥ 85th percentile for age and gender) and who were residing in a rural county in the Southeastern United States. This study utilized pre-treatment data from a larger research project, the Extension Family Lifestyle Intervention Project (E-FLIP for Kids), which examined the effectiveness of year-long family-based weight management interventions for overweight and obese children. Parent and child dyads were recruited from rural communities using diverse methods, including direct mailings, brochure distributions through public schools and pediatrician offices, and community presentation [32]. Families completed an initial phone screening followed by an in-person visit to complete eligibility screening, informed consent and assent procedures and baseline assessment measures of height and weight. Parents completed questionnaires at baseline assessing child behavioral functioning, parent feeding practices, and family functioning. The governing Institutional Review Board approved the larger study.

Measures

Demographic information

Parents completed a demographics questionnaire that included family background information, such as parent/child age, sex, race, marital status, and family income.

Child Behavior Checklist

The Child Behavior Checklist (CBCL) is a commonly-used parent rating scale that assesses both internalizing and externalizing behavioral problems in children 4–18 years of age [33]. Extensive reliability and validity data have been reported on the CBCL, including excellent internal consistency and test–retest reliability, a stable factor structure, and positive relations with other measures of childhood behavior [34]. Higher T-scores indicate greater childhood behavior problems. The total behavior problems subscale was used in this analysis.

Child Feeding Questionnaire

The Child Feeding Questionnaire (CFQ) is a parent-completed measure of parental attitudes, beliefs, and practices about child feeding. It has been shown to have good reliability and validity as well as sound psychometric structure [35]. Three subscales were used in this analysis: (a) perceived responsibility, (b) restriction, and (c) monitoring. Parental responsibility (three items) assessed parent perceptions of their responsibility for feeding (e.g., when your child is at home, how often are you responsible for feeding him/her?). Parent monitoring (three items) assessed the extent to which parents’ perceived oversight of their child’s eating (e.g., How much do you keep track of high fat foods that your child eats?). Restriction (eight items) assessed the extent to which parents restricted their child’s access to foods (e.g., I intentionally keep some foods out of my child’s reach). Higher scores indicate greater use of these different parental feeding practices (e.g., greater monitoring, responsibility taking, and restriction of unhealthy food items).

Family Assessment Device

The Family Assessment Device (FAD) is a 60-item measure that provides an assessment of family functioning along seven dimensions: problem solving, communication, affective regulation, affective involvement, roles, behavioral control, and global family functioning. Each item is rated on a four-point Likert scale. Parents were asked to respond to each item in terms of “how well it describes your own family.” The global family functioning scale (12 items) was used in the current study; lower scores indicate better family functioning. Results from previous studies have suggested that the FAD is a reliable and valid instrument it has adequate internal reliability and a strong factor structure among both clinical and nonclinical groups [36–38]. Specifically, previous psychometric studies have shown the test–retest reliability for the global family functioning subscale to be 0.71 and the internal consistency of the global family functioning scale to be 0.86 [37,38].

Statistical analyses

Analyses were conducted with SPSS, version 22. A total of 223 participants were included in the original sample, but three participants were excluded due to missing data. The means, standard deviations, and frequencies for demographic variables and variables of interest were calculated for the remaining sample (n = 220). In order to identify covariates, bivariate Pearson correlations and analysis of variances were used to test the associations between youth age, BMI z-score, sex, race, and annual household income, with the outcome variable of interest.

The SPSS macro PROCESS (model 4) was used to analyze three mediation models using bootstrapping [39]. Bootstrapping is a statistical method that involves drawing repeated samples from the data with replacement in order to gain multiple estimates of the indirect effect [39]. Using this method, we generated 5,000 bootstrapped samples. Advantages to using this statistical approach to testing mediation over Baron and Kenny’s approach is that it does not make the often erroneous assumption of normality for the direct effects, and multiple mediators and moderators can be tested simultaneously [40]. Additionally, Type II error is reduced because fewer inferential tests are required [41]. Bootstrapping bases significant results upon finding that the 95% confidence interval (CI) does not contain zero.

The first model, with identified covariates, was defined with youth behavioral problems (total score on the CBCL) as the independent variable, parental perception of responsibility for their child’s weight as the dependent variable, and family functioning as the mediator. Two more mediation models were used with the same IV and mediator as used in the first model, but the dependent variable varied between each model based on the different subscales on the CFQ (parent monitoring of child food intake and parental restriction of children’s access to food).

Three additional mediation models were conducted to determine if a relationship held with child behavior problems mediating the relationship between family functioning and various parent feeding practices. The first model included family functioning as the independent variable, youth behavioral problems (total score on the CBCL) as the mediator, and parental perception of responsibility for child’s weight as the dependent variable. The additional models included the same independent variable and mediator, but the dependent variable changed for each model based on the different subscales on the CFQ (parent monitoring of food intake, and parental restriction of children’s access to food). We report unstandardized B weights to describe the relations between variables within the mediation models.

RESULTS

The mean age of child participants was 10.32 (SD = 1.39) years; 53.64% of the sample was female. The majority (90.45%) of the children in the sample had obesity. BMI percentiles ranged from 88.76 to 99.86 (M = 97.92, SD = 2.08). The primary parents who completed the questionnaires were mothers (92.2%) and married (66.4%). The median family income was between $40,000 and $59,999, and the mean age of parents was 40 years. The majority of parents were obese with a mean BMI of 35.21 (SD = 8.39). The sample consisted of non-Hispanic white (64.09%), non-Hispanic African American (12.73%), Hispanic white (7.27%), and other (8.18%) in addition to those who did not disclose their ethnicity/race (7.73%).

The CBCL mean score (T-score = 56.25) was below the clinical cut off of 60 (T-score), and suggests this sample did not exhibit significant behavioral problems. Of our sample, 34.1% fell within the elevated range on the CBCL. The global family functioning mean score (20.99) was below the significantly elevated cutoff score of 26.04 and therefore did not indicate significant family functioning problems. Of our sample, 10.9% fell within the elevated range on the FAD. The CFQ does not have clinical cutoffs, but the CFQ subscale means obtained in the current study are similar to those in previous studies [35]. Complete descriptive statistics for parent report on the CBCL, FAD, and CFQ are recorded in Table 1.

Table 1.

Descriptive characteristics of key independent and dependent variables

Measure	Mean	SD	Range Min-Max
Total Behavior Problems (CBCL)	56.25	9.57	31–76
Global Family Functioning (FAD)	20.99	5.12	12–39
Parental Perception of Responsibility for Child Weight (CFQ subscale)	11.70	2.27	4–15
Parental Restriction of Children’s Access to Food (CFQ subscale)	31.02	4.8	8–40
Parental Monitoring of Child Food Intake (CFQ subscale)	11.00	2.63	3–15

CBCL Child Behavior Checklist; CFQ Child Feeding Questionnaire; FAD Family Assessment Device.

Mediation model predicting parental perception of responsibility for child weight

The first mediation model assessed if overall family functioning mediated the relationship between total child behavior problems and parental perception of personal responsibility for child weight (Fig. 1). Child age and family income were correlated with parental perception of responsibility (B = −0.23, p = .043; B = −0.351, p < .001) and were therefore included in the following mediation model as covariates. The model showed that total child behavior problems were positively related to overall family functioning (B = 0.1763, t = 3.51, p < .0001); therefore, greater childhood behavior problems was associated with poorer family functioning. Further, poorer family functioning was associated with less responsibility taking of children’s food choices (B = −0.087, t = −2.79, p = .0057). Finally, child behavior problems and parental perception of responsibility were not related within the mediation model. The estimated indirect effect (−0.015) of child behavior problems on parental perception of responsibility for child weight through family functioning was statistically significant as shown by the bootstrapped 95% CI that did not contain zero (−0.032, −0.005). Overall, this analysis showed that poorer family functioning significantly mediated the relationship between higher child behavior problems and lower parental perception of responsibility for child weight. The model accounted for 12% of the variance in parental perception of responsibility (p = .0001).

Fig 1.

The mediation model shows that family functioning mediates the relationship between child behavior problems and parental perception of responsibility for child weight. Note. Estimate of the indirect effect (−0.015); BC bootstrapped 95% confidence interval (−0.032, −0.005).

Mediation model predicting parental monitoring of child food intake

The second mediation model examined if overall family functioning mediated the relationship between total child behavior problems and parental monitoring of child food intake (Fig. 2). None of the demographic variables were significantly related to monitoring. Therefore, no covariates were included in mediation model two. The model showed that greater child behavior problems were associated with worse family functioning (B= 0.1882, t = 5.27, p < .001). Further, family functioning was negatively related to parental monitoring of child food intake (B = −0.13, t = −3.65, p = .0003); therefore, worse family functioning was associated with less parental monitoring of children’s food intake. Finally, child behavior problems and parental monitoring were not related within the mediation model. The estimated indirect effect (−0.0247) of child behavior problems on parental monitoring of child food intake through family functioning was statistically significant as shown by the bootstrapped 95% CI that did not contain zero (−0.0424, −0.0093). Overall, this analysis showed that poorer family functioning significantly mediated the relationship between higher child behavior problems and lower parental monitoring. The model accounted for 7% of the variance in parental monitoring of child food intake (p = .0009).

Fig 2.

The mediation model shows that family functioning mediates the relationship between child behavior problems and parental monitoring for child food intake. Note. Estimate of the indirect effect (−0.0247); BC bootstrapped 95% confidence interval (−0.0424, −0.0093).

Mediation model predicting parental restriction of children’s access to food

The third mediation model was tested to determine if overall family functioning mediated the relationship between total child behavior problems and parent use of restrictive feeding practices to limit children’s access to food. This model (F = 0.9232, p = .3989) was not significant as shown by the bootstrapped CI, which included zero (−0.0361, 0.0201).

Alternative mediation models predicting parent feeding practices

A series of alternative mediation models were tested to determine if a relationship held with child behavior problems mediating the relationship between family functioning and various parent feeding practices. Child behavior problems did not mediate the relationship between family functioning and parental perception of responsibility shown by the bootstrapped CI including zero (−0.0219, 0.0221). Child behavior problems also did not mediate the relationship between family functioning and parental monitoring (CI: −0.0239, 0.0297). Lastly, child behavior problems did not mediate the relationship between family functioning and parental restriction (CI: −0.0147, 0.0905).

DISCUSSION

It is critical to gain a better understanding of the obesogenic environmental risk factors for childhood obesity, as identifying these factors may help to refine treatment and prevention efforts to reduce the childhood obesity epidemic in the USA. While the literature contains a wealth of information on different types of parental feeding practices and related outcomes, there remains a dearth of information regarding the connections between child behavioral problems, family functioning, and parental feeding practices. Approximately 113 million children worldwide have disruptive behavioral disorders (i.e., oppositional-defiant disorder, conduct disorder, etc.) and approximately 13% have a diagnosed psychological illness (i.e., anxiety, depressive, or attention-deficit hyperactivity) [42]. Since behavioral problems are associated with an increased risk of obesity, it is imperative to hone in on obesogenic risk factors that may interact with child behavioral problems [18].

The purpose of the current study was to examine the mediating role of family functioning in the relationship between children’s behavior problems and parental feeding practices in a sample of families with children with overweight or obesity. The results confirm our hypotheses that poorer family functioning partially mediated the relationship between greater child behavior problems and lower levels of parent feeding practices shown to be protective against weight gain (parental perception of responsibility and parent monitoring of children’s unhealthy food choices). Through conducting a series of alternative mediation models, we were able to establish that the opposite relationship did not hold. That is, child behavior did not mediate the relationship between family functioning and parental feeding practices. Thus, this suggests that may be initial child behavior problems that drive decrements in family functioning to ultimately influence the way parents interact with their children around food. As lower parental monitoring and responsibility taking are related to higher risk for childhood overweight and obesity [13], this model highlights a new potential environmental obesogenic risk factor for childhood obesity. However, longitudinal research is needed to support causality of these relationships.

The pre-existing literature indicates child behavioral problems predict future development of childhood obesity, as well as a relationship between poor family functioning and child obesity [18,23]. Our findings add to this by showing a mediating effect of poor family functioning on the relationship between elevated child behavior problems and certain parental feeding practices. Additionally, it suggests a potential order to these events, with child behavior problems driving family functioning to impact parent feeding, but not vice versa. While the explained variances in our two models are relatively small, they highlight potential risk factors that add to already established obesogenic risks. Therefore, the results of this study suggest additional obesogenic environmental risk factors (e.g., the combined factors of child behavior problems and poor family functioning). While these results need to be replicated with longitudinal research before definitive conclusions can be drawn, these data suggest that clinicians should assess family functioning in children with both behavior problems and overweight/obesity, as they may be at greater risk for unhealthy parental feeding practices (e.g., less monitoring of foods and less responsibility of children’s food intake). There may be opportunity to intervene upon family communication and parental feeding behaviors for children who have behavioral problems, particularly for children who are already seeking treatment for overweight or obesity.

Family functioning was not found to be a significant mediator on the relationship between child behavior problems and parental perceived restriction of children’s unhealthy eating. Parents who perceive their children as having obesity are more likely to restrict their children’s eating [16]; thus, we would have expected the majority of parents in this treatment-seeking population to score high on this measure. Family functioning, therefore, may not have as much of an impact on parental restriction of unhealthy foods in a population of children with obesity and who have behavioral problems, as compared to monitoring and responsibility taking.

On the other hand, our null finding is not entirely surprising, given that parental restriction implies some type of child monitoring (e.g., in order to restrict what your child is eating, you must first be aware of their foods), and our results indicated that child behavioral problems and family functioning are related to less parental monitoring of children’s eating. Another possible explanation for our findings is that parents of families with more conflict may be more prone to give their children preferable foods as a way to soothe them (e.g., “my child is more likely to quiet down/behave if I let him eat chips”), as opposed to restricting preferable foods (i.e., foods high in sugars and fats) [21,22]. There currently exists a dearth in the literature describing the characteristics of this phenomenon (e.g., emotional feeding behaviors), specifically what characteristics of families often lead to the presence of these behaviors.

There exist a few limitations to this study. First, all of the data are parent self-report. Thus, all measures collected are subject to the limitations of subjective report. Further, the data are cross-sectional; therefore, we are unable to determine causality between variables. The sample also consists of only children with overweight or obesity, and the sample consists of children from one geographic region. Therefore, the results cannot generalize to all children. However, given the high prevalence of overweight and obesity in the current population [1], the results may be generalizable to over one third of U.S. families. Finally, the effect sizes are relatively small for the mediation models. Obesity is the result of a complex environmental and genetic etiology, and while child behavior problems and family functioning may not be the sole drivers of obesity, they point to a potential risk factor impacting the obesogenic environment.

As childhood obesity is a national epidemic, future research should consider further investigating the relationships between these variables and how they may impact future intervention programs. First, future research is needed to determine longitudinal and therefore possibly causal relationships between these variables. If research supports the causal association between child behavior problems and family functioning, there are a few suggestions for pediatric obesity prevention and treatment. Behavioral family lifestyle interventions are currently the gold standard treatment for pediatric obesity, and they offer an opportunity to provide parents with the education and tools to manage problematic childhood behaviors and family communication as it relates to obesity-related behavior (e.g., meals). Future interventions should consider specific psychoeducation and coaching strategies to improve family communication, parental monitoring, and responsibility taking of children’s food intake. Furthermore, future research should consider examining which characteristics of families lead to the emotional feeding practices, a modifiable environmental obesogenic factor, and potentially a reaction to certain child behaviors.

CONCLUSION

Childhood obesity is a major national epidemic and it is crucial that we better understand modifiable contributors to the obesogenic environment to better prevent and treat pediatric obesity. The results from this study may point to the combined risk factor of perceived child behavioral problems and poor family functioning as contributing to unhealthy parent feeding practices. Professionals working with children with overweight or obesity may consider screening for these additional factors as points of intervention to improve the obesogenic environment.

Compliance with Ethical Standards

Funding: This study was funded by the National Institute of Diabetes and Digestive and Kidney Diseases (DK082374).

Conflicts of Interest: All authors declare that they have no conflicts of interest.

Ethical Approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent: Informed consent was obtained from all individual participants included in the study.