Examining the effect of parent participation in an adult weight management program on changes in children's weight

Children of parents participating in weight management programs (WMPs) are more likely to adopt their parents' weight control practices. Little is known about the weight outcomes of children who have a parent participating in a WMP. This study aimed to assess this relationship. Children 2–17 years of age with a parent who participated in a WMP were included in the study. Multilevel linear mixed‐effects regression models were used, stratified by child weight status at the time of parental WMP participation (healthy weight, overweight/obesity) to determine change in children's BMIz from before to after parents WMP participation, including covariates of parent BMI and parental feeding practices. Parents (N = 77) were mostly white (76%) and female (84%). Children (N = 114) had a mean age of 10.5 ± 4.6; 47% had overweight or obesity. Children with overweight or obesity prior to their parent's WMP had a decrease in BMIz (−0.68) after the WMP while children with a healthy weight had no significant change. Children with overweight or obesity had a decrease in BMIz from before to after parent's participation in a WMP. Further research is needed to understand changes in family eating practices that occur during and after parent WMP participation.


Summary
Children of parents participating in weight management programs (WMPs) are more likely to adopt their parents' weight control practices. Little is known about the weight outcomes of children who have a parent participating in a WMP. This study aimed to assess this relationship. Children 2-17 years of age with a parent who participated in a WMP were included in the study. Multilevel linear mixed-effects regression models were used, stratified by child weight status at the time of parental WMP participation (healthy weight, overweight/obesity) to determine change in children's BMIz from before to after parents WMP participation, including covariates of parent BMI and parental feeding practices. Parents (N = 77) were mostly white (76%) and female (84%). Children (N = 114) had a mean age of 10.5 ± 4.6; 47% had overweight or obesity. Children with overweight or obesity prior to their parent's WMP had a decrease in BMIz (À0.68) after the WMP while children with a healthy weight had no significant change. Children with overweight or obesity had a decrease in BMIz from before to after parent's participation in a WMP. Further research is needed to understand changes in family eating practices that occur during and after parent WMP participation. What is already known about this subject • Parents provide the foundation of eating and exercise behaviours for their children.
• Children of parents participating in weight management programs are more likely to adopt their parents' weight control practices.

What this study adds
• Children with overweight or obesity prior to their parent's weight management program had a decrease in BMIz (À0.68) after the parent's participation.
• Children with a healthy weight had no significant change in BMIz after parent's participation in a weight management program.
• Further research is needed to understand changes in family eating practices that occur during and after parental weight management program participation.

| INTRODUCTION
One-third of children and adolescents in the United States (US) have overweight or obesity. 1 These children are five times more likely to have obesity in adulthood than those who do not have obesity during childhood. 2 Risk factors for obesity in childhood include physical inactivity, low socioeconomic status, and controlling parental feeding practices (e.g., restriction, monitoring). 3 Additionally, parental obesity remains one of the strongest risk factors for overweight and obesity in children. 4 Healthcare practitioners and researchers struggle to find effective interventions that promote short and long-term behaviour change leading to healthy weight management for children and families.
More than half of US adults are actively trying to lose weight, 5 and 65% have been recommended for weight loss by their physician. 6 Children of parents who are participating in weight management programs (WMPs) are more likely to use the same weight control practices as their parents. 7,8 In one study, parents reported that their child utilized the following weight control practices in the past year: 12% dieted, 12% ate very little food, 7% used food substitutes, 10% skipped meals, 29% increased fruit and vegetables, 25% reduced calories, 22% reduced snacking, 21% reduced fat intake, and 45% exercised. Children were more likely to utilize weight control practices if their parent utilized the same weight control practice and if children had obesity and were older. 8 Additionally, parents who participate in family-based, paediatric WMPs often lose weight with their child; increased parental weight loss in these programs is associated with increased child weight loss. 9,10 One recently published pilot study examining changes in child weight and behaviours after parents' participation in a medical WMP found that of the 13 dyads who completed an assessment 1 year later, BMI zscore decreased from 0.97 to 0.75. 11 Despite cross-sectional evidence about children of parents in WMPs, 8 little is known about the effect of parental participation in adult-based WMPs on the child's subsequent weight changes. This study aimed to (1) explore weight change in children from before to after their parents' participation in a WMP, and (2) examine the association of parental BMI and feeding practices during the WMP on changes in children's weight. The hypothesis of this study was that parental participation in a WMP would result in a decrease in children's BMI z-score after parental participation, and that increased parental BMI and parental restriction and pressure to eat would be associated with less favourable weight outcomes in the child.

| Study design and participants
This was a retrospective cohort study. Adult participants at a single institution from a previous study (N = 150) who attended Wake Forest Baptist Health's 6-month WMP between May 2017 and July 2017 were included if they had at least one child between the ages of 2 and less than 18 years old living at home. Custody of the child was not assessed as part of the initial study, but children were required to live in the home at least 4 days a week. Participants of the index study were contacted via phone and provided verbal consent for their children's participation in this study. Child assent was not required as this was a retrospective study of existing EHR data. For all children (ages 2-17) who were living in the home at the time of the parents WMP participation, parents provided children's name(s), date(s) of birth, and primary care provider. This information was used to locate children in the EHR, from which their data were extracted. Children were excluded if they had any chronic medical condition that affects growth, significant developmental delay, or intellectual disability. Children were also excluded if they did not have at least one height/weight in the electronic health record (EHR) within 1 year both before and after parental participation.
Information was received from 77 of the 150 parents; 38 did not answer or return our phone calls and emails, 25 declined to participate, 9 did not have accurate contact information, and 1 was deceased. Information was obtained regarding 120 children, 114 of whom met eligibility criteria. The Institutional Review Board of Wake Forest School of Medicine approved the study protocol (IRB00058514).

| Study instruments and measurements
As part of the cross-sectional index study in 2017 during their participation in the WMP, parents completed the Child Feeding Questionnaire (CFQ). Parents responded on a Likert scale (range 1-5) and a mean score was calculated for each subscale, with higher scores indicating higher perceived responsibility, restriction, pressure, and monitoring. 12 Parental demographics obtained from the index study included age, sex, height, weight, race, household income, and health insurance. Children's demographics and anthropometrics were collected from the EHR. Data from 3 years before the index study (May 2014) through 3 years after (August 2020) were obtained, including sex, visit date, height, and weight. Age-and sex-specific percentiles and BMI z-scores (BMIz) were calculated, according to standardized CDC growth charts.
For this analysis, based on sample size and data distribution, child weight status at the time closest to the index study and parental WMP participation was categorized into underweight/healthy weight (BMI <85th percentile) and overweight /obesity (BMI ≥85th percentile).   T A B L E 2 Multilevel linear mixed-effects regression models assessing change in BMIz for all child participants (model 1) and stratified by child weight status (models 2 and 3). deviation (SD) = 5.8] years. Three-quarters had an annual household income of at least $60 000 US dollars (Table 1). Children (N = 114)

| Study sample and characteristics
had an average age of 10.5 years (SD = 4.6). A majority (53%) had underweight or a healthy weight status at the initiation of parent participation in a WMP while 47% had overweight or obesity (Table 1). Parent's report of their feeding practices is presented in Table 1.  Figure 1a). Children were more likely to have an increase in BMIz from before to after the WMP if their parent

| Multivariate analyses
reported increased concern about their child's weight at the time of

| DISCUSSION
Children of parents in WMPs were more likely to have a decrease in BMIz if they had overweight or obesity compared to children with underweight or healthy weight. In prior cross-sectional observational studies, children were more likely to utilize weight control practices, including healthy (e.g., exercising and increasing fruit and vegetable intake) and unhealthy behaviours (e.g., dieting, skipping meals, and using a food substitute) if children had overweight or obesity. 7,8 Given the results of the present study, it is possible that parental feeding attitudes, beliefs, and practices may have an effect on child weight outcomes. Specifically, this study found that parental restriction of access to palatable foods (all children) and monitoring of their child's dietary intake (children with overweight/obesity) was associated with a significant decrease in BMIz. However, the CFQ was not repeated, and it is possible that the feeding practices and beliefs of parents changed over time with participation in their WMP. Additional studies are needed to assess the stability of feeding practices. In other work with repeated measures of the CFQ with parents in a WMP, significant decreases in parental restrictive feeding were observed from parent initiation in the WMP over 12 months, with no differences observed based on child weight status; however, increased restrictive feeding was observed specifically for parents of female children. 11 Regarding child feeding practices, previous studies have consistently shown that parental restriction is associated with child weight gain, with restriction defined as parents exerting control over the child's eating by restricting access to certain desired types and amounts of foods. 3 This relationship between restriction of food leading to increased caloric intake and weight gain may be due to the child's defiance of parental rules when restricted foods are available and self-control is not actively exercised. However, the present study found contradictory results. Children of parents who exhibited restriction had a decrease in BMIz. It is possible that this may be due to selflearned food restriction and caloric control, though prospective studies are needed to measure child behaviours in response to parent's behaviours and their correlation to both parent and child BMI trends .
This study was not without limitations. Feeding practices and beliefs were reported by the parents, which could result in respondent bias, specifically under-reporting of unhealthy feeding practices. Only half of eligible parents from the index study participated in this study, resulting in potential selection bias. The CFQ was also completed by the parent at the start of their WMP, and it is possible that feeding practices and beliefs evolved over time. Additionally, the sample size was relatively small, from a single area, and the index study was crosssectional so no data are available on parent weight or behaviour changes as a results of the program. A high percentage of parents included in this study were healthcare employees. Workers in healthcare, specifically nursing, psychiatric, and home health aides, have a higher prevalence of obesity than the general public. 13 This is related to the level of physical activity at work, work-related psychosocial stressors, long work hours, and time constraints outside of work. 14,15 Additionally, parents included in this study were primarily mothers, which may not reflect other parent-child dyads and family dynamics and could further decrease the generalizability of results.
Given the high rates of adult participation in WMPs, it is crucial for parents to model healthy weight management behaviours and utilize healthy feeding practices. Adult WMPs that include messaging and education around family healthy feeding practices and physical activity behaviours are encouraged.

AUTHOR CONTRIBUTIONS
Dr. Pham conceptualized and designed the study, contributed to data collection, and drafted the initial manuscript. Drs. Skelton, Pratt, and Lewis conceptualized and designed the study, contributed to acquisition, analysis and interpretation of the data, and reviewed and revised the manuscript critically for important intellectual content. Dr. Brown conceptualized and designed the study, completed data analysis, contributed to the acquisition and interpretation of data, and reviewed and revised the manuscript critically for important intellectual content.