Development of a Measure to Assess Parent Perceptions of Barriers to Child Weight Management

Abstract

Background:

Families face many barriers related to creating or maintaining a healthy lifestyle, which likely contributes to the prevalence of childhood obesity in the United States; however, no measure currently exists to examine these barriers. This study developed a quantitative measure of parents' perceptions of barriers to family healthy lifestyle.

Methods:

Parents of children between the ages of 7–17 were recruited using Amazon Mturk (n = 812). Exploratory and confirmatory factor analyses and preliminary convergent validity analyses were conducted.

Results:

Four factors emerged in the final measure: Parent Disengagement, Cost and Built Environment, Lack of Family Support, and Family Time Constraints, with the final 19-item measure having good initial psychometric properties, including reliability and validity.

Conclusions:

Future research is needed to examine whether this measure may be used in clinical practice to identify barriers to a healthy family lifestyle, to tailor interventions for families.

Approximately one-third of children and adolescents in the United States are either overweight or obese.¹ Obesity in childhood puts children at risk for negative psychosocial² and physical health outcomes.¹ Given these negative outcomes, it is crucial to understand existing barriers to maintaining a healthy lifestyle in children and adolescents. A small, but growing body of evidence suggests that many children and families do not feel well equipped to address barriers (e.g., limited access to resources and lack of time availability) contributing to childhood obesity.^3,4 Within the family context, examining barriers to establishing or maintaining a healthy lifestyle might aid in understanding factors that are related to childhood weight gain.⁵ This study sought to develop a quantitative measure of parents' perceived barriers to a healthy lifestyle within the family context.

Potential Barriers to Pediatric Weight Management

Parental support and engagement, such as role modeling, encouragement, and monitoring of child health behaviors are important for child weight management. Parental support is associated with greater engagement in healthy behaviors⁶ and weight control intervention involving parents has consistently been found to be effective for children's weight loss.⁷ However, a lack of this parental support and involvement may decrease a child's ability to maintain healthy weight-related behaviors.⁷

Beyond family behaviors, the larger context of the environment may also contain facets considered to be barriers to pediatric weight management. One such barrier that may exist in the larger context is the cost of healthy food as one of the most commonly cited barriers to healthy eating.⁸ Uncertain access to food and costs of healthy foods and physical activity facilities may be salient barriers to healthy behaviors in low-income families.⁹ Related to physical activity, low-income areas are more likely to have fewer safe outdoor areas for play. Parents' safety and anxiety related to child safety prevent physical activity,¹⁰ and children living in these areas may be at greater obesity risk as a result.¹¹

A small, but growing body of previous research has included the development of measures sampling constructs that may be conceptualized as barriers to weight management (such as time demands).¹² However, to our knowledge, no measure has been developed to quantitatively examine multiple facets of barriers to healthy weight management within the family context. This study therefore sought to extend the existing body of research by specifically identifying domains that may act as barriers to pediatric weight management within the family context. Previous research has examined barriers qualitatively^3,13 or barriers specific to weight loss (i.e., barriers to physical activity in obese children).¹⁴ By examining barriers to healthy family lifestyle within a nonclinical sample, the factors hindering families' abilities to support a healthy lifestyle for their child can be identified and addressed before the child becomes overweight or obese.

This Study

As shown through previous research, many families experience barriers affecting healthy lifestyles in children, yet no quantitative measure exists to assess these perceived barriers. The goal of this study was to develop a quantitative measure of parents' perception of barriers to healthy weight management in children to elicit these specific barriers to aide in prevention and intervention for a healthy family lifestyle.

Method

Participants

Parents of children aged 7–17 (n = 812; M parent age = 37.2) were recruited online using Amazon Mechanical Turk (MTurk). The majority of participants were mothers (64.7%) and Caucasian (79.0%; children were identified as 74.1% Caucasian). The median household income of the sample was $54,000, which is consistent with the median US income in 2015 of $56,516.¹⁵ Inclusion criteria comprised the following: (1) US residency, (2) English fluency, and (3) at least one child between the ages of 7 and 17. Please see Table 1 for descriptive characteristics of demographic variables.

Table 1.

Descriptive Characteristics of Demographic Variables (n = 812)

Characteristic	n (or Mean)	% (or SD)
Child characteristics
Biological sex
Female	388	47.8%
Race
White	602	74.1%
Black or African American	92	11.3%
Asian or Pacific Islander	33	4.0%
American Indian or Eskimo	10	1.2%
Other	75	9.1%
Ethnicity
Hispanic or Latino	84	10.3%
Not Hispanic or Latino	728	89.6%
Age	M = 11.07	SD = 3.48
BMI percentile for age and gender	M = 66.36	SD = 32.29
Under/healthy weight (≤85%ile)	463	57.0%
Overweight (85%ile to 95%ile)	121	14.9%
Obese (>95%ile)	221	27.2%
Parent characteristics
Biological sex
Female	525	64.7%
Ethnicity
White	641	78.9%
Black or African American	92	11.3%
Asian or Pacific Islander	37	4.6%
American Indian or Eskimo	13	1.6%
Other	29	3.5%
Age	M = 37.21	SD = 8.30
BMI	M = 28.09	SD = 8.09
Under/healthy weight (below 24.9)	481	59.2%
Overweight (25.0–29.9)	80	9.9%
Obese (30 or higher)	251	30.9%
Household Income	M = 62, 646	SD = 42,553

All percentages may not equate to 100 due to rounding.

M, sample mean; SD, standard deviation.

Procedure

Parents/guardians were recruited through MTurk (an online forum to recruit individuals to complete surveys) and compensated $1.00 for their participation in a 30-minute survey. Participants completed the Barriers measure being developed within this study, as well as measures used for convergent validity within this study. Previous research has found MTurk participants are more diverse, recruited more rapidly, and the data are equally reliable compared to traditional methods of data collection.^16–18 Parents/guardians in this study were asked to answer all questions thinking of their oldest child within the 7–17 age range for consistency. Parents/guardians who did not choose the correct answers for all validity questions (e.g., “Is the earth round?”) included in the survey were excluded from the sample (20 participants). Less than 1% of participants (n = 4) had missing data; therefore, listwise deletion was used for these participants. The Kent State IRB approved this study.

Measures

Demographics

The demographic measure developed for this study included 41 items concerning basic demographic characteristics of both the parent and child (e.g., age, gender, education, and income). Parents reported their own and their child's height and weight. Child height and weight were used to calculate BMI percentile for age and gender using CDC guidelines.¹⁹

Convergent validity

Time demands questionnaire

The Time Demands Questionnaire (TDQ) is a 9-item measure of family stressors related to time demands¹² (e.g., “I feel too busy with work or other demands”). Rated on a Likert scale from 1 (Strongly Disagree) to 4 (Strongly Agree), higher total scores indicate greater time demands related to family mealtime. Consistent with previous research,¹² this study found good reliability for the TDQ (α = 0.89).

Core food security module

The USDA Core Food Security Module is a validated scale measuring the severity of household food insecurity and hunger within the past 12 months.²⁰ Example item: “In the last 12 months, the food that we bought just didn't last and we didn't have money to get more.” The responses to this 18-item food security scale were used to calculate the 12-month food security scale. Higher scores indicate higher levels of food insecurity.

Weight control strategies scale

The Weight Control Strategies Scale (WCSS) is a 30-item validated self-report measure of weight control behaviors²¹ (e.g., “I scheduled exercise into my day”). Each item is rated on a Likert scale from 0 (never) to 4 (always). This measure exhibited good reliability in this sample (α = 0.95), consistent with previous research.²¹

Healthy habits assessment

The Healthy Habits Assessment is a 6-item measure that asks about a variety of children's healthy habits reflective of the current American Academy of Pediatrics Guidelines.²² Each item is scored on a scale from 1 (lower levels of healthy habit) to 4 (high levels of healthy habit). A total score of the 6 items was created by summing the scores from each question, with higher scores indicating higher levels of healthy habits. Example item: “My child has sweet drinks (cola, sweet tea, juice, sports drinks, other juice drinks): (1) more than 2 a day, (2) 2 a day, (3) 1 a day, (4) not very often.”

Barriers to pediatric weight management questionnaire (BPWM)

For the first phase of measure development, experts in pediatric obesity and weight management (e.g., three hospital-based doctoral level psychologists [i.e., PsyD and PhD]) in multidisciplinary pediatric weight management clinics across the Northeast and Midwestern United States were asked to identify domains they perceive as barriers to engaging in weight management behaviors, as well as domains commonly cited by parents as barriers to maintaining a healthy family weight-related lifestyle (e.g., eating fast food and limited resources for physical activity). Each of these pediatric psychologists has had multiple years of experience working with youth within a behavioral weight management setting. Example items were written to tap these identified constructs and were sent back to pediatric obesity experts for review, with additional items added by experts, where deemed necessary. A review of relevant literature to identify barriers in maintaining healthy weight status in clinical and community samples was also conducted. Correlates of BMI (e.g., fast food consumption and parental support), which may serve as barriers in families, were also identified through an extensive literature review. Further barriers to adherence in other pediatric populations (i.e., asthma and diabetes) were examined through literature search, as adherence to medication in these pediatric population parallels maintenance of healthy behaviors within pediatric weight management. Studies published in English were identified through literature searches on PubMed, Google Scholar, and PsycINFO using search terms related to healthy weight management in the family (pediatric, obesity, BMI, adherence, barriers, and weight). Reference sections of all relevant articles were examined for factors related to adherence and barriers in pediatric healthy weight and chronic illness.

Phase two of measure construction consisted of writing sample items to measure identified constructs. Experts were again contacted with a list of relevant constructs, as well as initial items written to tap these constructs. Experts provided feedback on the items written and were asked to provide additional items related to the relevant constructs. Based on Phases I and II, the initial measure consisted of 41 items, agreed upon by experts in pediatric obesity, tapping constructs of motivation,²³ financial access,²⁴ family support with healthy eating and exercise,²⁵ built environment access,¹¹ and parental time demands.²⁶ Please see Table 2 for a list of all items included. Items were rated on a scale from 1 (Strongly Disagree) to 5 (Strongly Agree).

Table 2.

Exploratory Factor Analysis Factor Loadings of Items onto Five Factors

Item	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5
Items dropped due to multiple loading (EFA #2)
Our family has enough time to exercise together.	—	—	—	—	—
Items not loading onto a scale (EFA #3)
My child has places to exercise when the weather is poor.	—	—	—	—	—
There is a park within walking distance of our home.	—	—	—	—	—
Our neighborhood is a safe place to play.	—	—	—	—	—
I feel motivated to help my family be healthy.	—	—	—	—	—
I feel motivated to keep healthy foods at home.	—	—	—	—	—
Usual social activities in our family do not include exercise.	—	—	—	—	—
I eat fruits and vegetables in front of my child.	—	—	—	—	—
There is a grocery store close to our house.	—	—	—	—	—
I am not interested in exercising with my family.	—	—	—	—	—
Our family time is spent watching T.V.	—	—	—	—	—
Two-item scale dropped (EFA #3)
I do not like eating healthy foods with my family.	—	—	—	—	—
I am interested in making healthy meals with my family.	—	—	—	—	—
Items kept in final EFA (EFA #4)
I compliment my child when he/she exercises.	0.99	—	—	—	—
I praise my child for physical activity.	0.83	—	—	—	—
I compliment my child when he/she makes healthy eating choices.	0.72	—	—	—	—
I talk to my child about different ways to get more physical activity.	0.71	—	—	—	—
I encourage my child to exercise regularly.	0.68	—	—	—	—
I talk to my child about how to make healthy eating choices.	0.48	—	—	—	—
If my family had more money my child could eat healthier foods.	—	0.74	—	—	—
Extracurricular sports are too expensive for my child to participate.	—	0.70	—	—	—
In our house, decisions about what to eat are based on cost.	—	0.69	—	—	—
Having a gym membership is too expensive for our family.	—	0.66	—	—	—
Fruits and vegetables are too expensive for our family to buy.	—	0.53	—	—	—
Our family always has enough food to eat.	—	−0.53	—	—	—
There are recreational facilities available to my child (e.g., basketball court, gym, and park).	—	−0.47	—	—	—
Our whole family participates in physical activity together.	—	—	0.71	—	—
Our family makes an effort to eat healthy foods.	—	—	0.69	—	—
When our family goes to restaurants we try to make healthy choices.	—	—	0.56	—	—
Our family does not focus on being healthy.	—	—	−0.53	—	—
Being healthy is important in our family.	—	—	0.45	—	—
Our family is too busy to eat meals together.	—	—	—	0.78	—
It is hard for me to find the time to prepare healthy food at home.	—	—	—	0.74	—
It is hard for my family to find the time for physical activity.	—	—	—	0.57	—
Our family eats quick foods because we don't have time to prepare a meal.	—	—	—	0.51	—
If we were less busy we would eat less prepackaged meals (e.g., mac and cheese).	—	—	—	0.50	—
Our family eats dinner at a similar time and in the same place most nights.	—	—	—	−0.42	—
Our family often eats fast food (e.g., McDonalds) to save time.	—	—	—	—	0.67
Our family keeps junk food in the house.	—	—	—	—	0.60
There are many fast food restaurants near our house.	—	—	—	—	0.51
Our family eats fast food because it is cheap.	—	—	—	—	0.49

Factor loadings below 0.4 are suppressed. Italicized items were retained in the final measure.

EFA, exploratory factor analysis.

Data Analysis

An approximate random split-half of the sample (using SPSS software approximate split-half) was used to conduct an exploratory factor analysis (EFA; n = 410) and a confirmatory factor analysis (CFA; n = 402) to allow for cross-validation of the factor structure. No significant differences were found between participants in the EFA and CFA subsamples on demographic variables.

Exploratory and confirmatory factor analyses

EFA was conducted using principle axis factoring with Promax rotation within SPSS (version 20) software to allow the factors to correlate, and a parallel analysis was conducted to determine the number of factors to extract.²⁷ Items that did not load onto any scales with at least a 0.4 standardized loading, or loaded onto more than one scale with a standardized loading were dropped from the measure. Scales with two or fewer items were also dropped due to lack of utility and reliability of two-item scales.²⁸ CFA was conducted using MPlus (version 7) software.²⁹ As the χ² statistic is highly sensitive to sample size, alternative fit statistics, were used to examine model fit within CFA. Consistent with Hu and Bentler,³⁰ the standardized root mean square residual (SRMR; <0.08 acceptable),³¹ comparative fit index (CFI; >0.90 acceptable),³² and root mean square error of approximation (RMSEA; <0.08 acceptable)³³ were used to test model fit.

Preliminary validity

To assess the convergent validity of the BPWM total score, as well as each subscale, the scales that emerged were correlated with theoretically related, previously validated measures. The BPWM total score was examined in relation to the Healthy Habits Assessment, with an expected negative relationship. It was expected that the Cost and Built Environment subscale of the BPWM would be positively related to the USDA Core Foods Security Module.²⁰ We also predicted a statistically significant difference in scores on the Cost and Built Environment subscale when comparing families above (vs. below) the median household income in the United States. In addition, we predicted a positive relationship between the Family Time Constraints subscale and the TDQ.¹² The Parent Disengagement and Lack of Family Support subscales were examined in relation to the WCSS,²¹ with an expected negative correlation, such that higher levels of parental disengagement and barriers related to lack of family support were related to fewer parental weight control strategies. It was also expected that there would be significant differences between healthy weight and overweight/obese parents on level of endorsed barriers within the family.

Results

Exploratory Factor Analysis

Results of parallel analyses for an initial EFA indicated that six factors best fit the data. A second EFA constrained the data to six factors. One item loaded onto multiple scales with a loading of 0.4 or greater and was dropped from the measure. A third EFA was conducted without this item (Table 2 displays the factor loadings after rotation). Items demonstrating factor loadings below 0.4 were dropped from the measure (10 items),³⁴ resulting in a 30-item scale. One factor consisted of only two items, decreasing the utility of the subscale; so the 2-item subscale was dropped. We reran the EFA with the remaining 28 items and a five-factor solution (accounting for 48.6% of the variance among the items. The first factor (parent disengagement) consisted of 6 items accounting for 26.0% of the variance. The second factor (cost concerns) consisted of 7 items accounting for 8.7% of the variance. The third factor (lack of family support) consisted of 5 items accounting for 6.2% of the variance. The fourth factor (time concerns) consisted of 6 items accounting for 4.3% of the variance. The fifth factor (unhealthy food choices) consisted of 4 items accounting for 3.4% of the variance.

Confirmatory Factor Analysis

An initial CFA was conducted on the remaining 28 items to cross-validate the five-factor solution obtained by the previous EFA within the second subsample. Items that did not have a standardized loading of at least 0.4 on their respective factor were removed from subsequent analyses (6 items). After these items were dropped, one subscale retained only two items. This subscale was deleted due to lack of clinical utility. A second CFA was conducted to examine the remaining four-factor structure. This model did not fit the data well, χ² (146) = 447.71, p < 0.001, CFI = 0.87, RMSEA = 0.07, and SRMR = 0.06, suggesting a lack of fit for the model due to low CFI index.

Modification indices suggested that correlating the error terms among some items would increase model fit. Thus, three model respecifications were made after examining the modification indices. Specifically, the error terms of three item pairs (26 and 15; 23 and 1; and 14 and 12) were allowed to correlate based on both conceptual and statistical information as deemed acceptable within previous research.^35–37

A final CFA model was tested following model respecification, and this model was found to have an acceptable fit, χ² (143) = 358.12, p < 0.001, CFI = 0.91, RMSEA = 0.06, and SRMR = 0.06. The final model consisted of 4 factors with a total of 19 items: Parent Disengagement (5 items), Cost and Built Environment (5 items), Lack of Family Support (4 items), and Family Time Constraints (5 items). See Table 3 for items included in the final measure (items with an “R” are reverse scored) and their final factor loadings.

Table 3.

Final Confirmatory Factor Analysis Factor Loadings for Barriers to Healthy Weight in the Family

Item	Parental disengagement	Cost and built environment	Lack of family support	Family time constraints
I praise my child for physical activity.^R	0.59	—	—	—
I compliment my child when he/she makes healthy eating choices.^R	0.58	—	—	—
I talk to my child about different ways to get more physical activity.^R	0.68	—	—	—
I encourage my child to exercise regularly.^R	0.74	—	—	—
I talk to my child about how to make healthy eating choices.^R	0.74	—	—	—
Extracurricular sports are too expensive for my child to participate.	—	0.66	—	—
In our house, decisions about what to eat are based on cost.	—	0.69	—	—
Having a gym membership is too expensive for our family.	—	0.58	—	—
Fruits and vegetables are too expensive for our family to buy.	—	0.70	—	—
Our family always has enough food to eat.^R	—	−0.55	—	—
Our whole family participates in physical activity together.^R	—	—	0.44	—
Our family makes an effort to eat healthy foods.^R	—	—	0.72	—
When our family goes to restaurants, we try to make healthy choices.^R	—	—	0.58	—
Being healthy is important in our family.^R	—	—	0.71	—
Our family is too busy to eat meals together.	—	—	—	0.54
It is hard for me to find the time to prepare healthy food at home.	—	—	—	0.75
It is hard for my family to find the time for physical activity.	—	—	—	0.70
If we were less busy, we would eat less prepackaged meals (e.g., mac and cheese).	—	—	—	0.66
Our family eats dinner at a similar time and in the same place most nights.^R	—	—	—	−0.40

Items with an “R” are reverse scored.

Preliminary Reliability and Validity

Preliminary reliability

Alpha reliability for each of the subscales fell in the acceptable to good range: Parent Disengagement α = 0.84, Cost and Built Environment α = 0.77, Lack of Family Support α = 0.69, and Family Time Constraints α = 0.78. Reliabilities around 0.7 are acceptable within newly developed research.³⁸

Preliminary validity

Correlations related to validity were conducted within the CFA split-half of the sample. Please see Table 4 for correlations between all measures. All hypothesized correlations between the Barriers total score and each of the four measures used for validation were significant in the hypothesized directions. Unexpectedly, child BMI percentile was not significantly correlated with the Barriers total score or subscales; however, parental BMI was significantly related to Cost and Built Environment and the Lack of Family Support subscales. Families reporting household income below the median household income in the United States ($52,250) demonstrated significantly higher Cost and Built Environment subscale scores (M = 2.99, SD = 0.82) than those above the median household income (M = 2.47, SD = 0.69), t(390) = 6.74, p < 0.001. There was a significant difference between healthy weight parents (M = 2.4, SD = 0.52) compared to overweight/obese parents (M = 2.51, SD = 0.50) on the Barriers Total Scale, t(389) = −2.03, p < 0.05.

Table 4.

Correlations between Barriers to Healthy Weight in the Family Subscales and Total Score with Food Security, Time Demands, Healthy Habits, and Child BMI Percentile (n = 402)

Variables	1	2	3	4	5	6	7	8	9	10	11	12
1. Parent disengagement^a	—
2. Cost and built environment^a	0.14^**	—
3. Lack of family support^a	0.49^**	0.34^**	—
4. Family time constraints^a	0.25^***	0.48^**	0.43^**	—
5. Barriers to healthy weight in the family total score	0.61^**	0.73^**	0.74^**	0.79^**	—
6. Weight control strategies scale	−0.33^**	−0.25^**	−0.48^**	−0.22^**	−0.42^**	—
7. Core food security module	0.06	0.63^**	0.18^**	0.27^**	0.43^**	−0.12^*	—
8. Time demands questionnaire	0.12^**	0.39^**	0.28^**	0.68^**	0.54^**	−0.09	0.23^**	—
9. Healthy habits assessment	−0.36^**	−0.25^**	−0.50^**	−0.33^**	−0.48^**	0.41^**	−0.21^**	−0.26^**	—
10. Child BMI percentile	0.01	0.08	0.06	0.05	0.07	0.00	0.10^*	0.08	0.09
11. Child age	0.02	0.11^*	0.22^**	0.04	0.12^*	−0.09	0.14^**	0.00	−0.26^**	−0.03
12. Parent age	0.04	0.01	0.17^**	0.07	0.09	−0.04	0.00	0.00	−0.18^**	0.05	0.47^*
13. Parent BMI	0.01	0.27^**	0.14^**	0.09	0.19^**	−0.14^**	0.20^**	0.08	−0.17^**	0.15^**	0.08	0.05

Subscales of barriers to healthy weight in the family.

p < 0.05; ^**p < 0.01; ^***p < 0.001.

Discussion

This study builds upon prior research by developing a quantitative measure of parent-report barriers to healthy weight within the family to guide interventions for maintenance of a healthy weight-related family lifestyle. Currently, no measure of barriers to pediatric weight management exists. This gap in the literature limits the rapid identification of concerning areas (i.e., potential barriers) for families presenting to weight management interventions. Furthermore, given the large proportion of US youth who are overweight or obese,¹ the absence of a measure of barriers to healthy pediatric weight management also hinders the identification of barriers common to a normative population that may be addressed through public health interventions, targeting pediatric obesity prevention and intervention. The final 19-item BPWM measure consisted of four subscales. Findings from CFA suggested an overall acceptable fit of this four-factor structure. The subscales were labeled as follows: Parent Disengagement, Cost and the Built Environment, Lack of Family Support, and Family Time Constraints. The Parent Disengagement subscale reflects the extent to which a parent encourages healthy eating and physical activity behaviors in their child. The second subscale, Lack of Family Support, is similar in reflecting family member behavior, but distinctive as it measures the extent to which the entire family makes healthy choices together (e.g., family exercises together). Cost and the Built Environment emerged as the third subscale as cost and environmental factors often overlap.^39,40 Family Time Constraints was the final subscale, which aligns with previous research suggesting that overweight parents with perceived time constraints may be less likely to prepare healthy meals,^12,41 encourage exercise, or monitor children's behaviors, possibly serving as a barrier to healthy lifestyles in children.

Measure Validation

Preliminary findings from this initial measure development study suggest that the BPWM has acceptable psychometric properties, including good internal reliability and convergent validity. In this study, the BPWM Total Score showed convergent validity in that higher levels of barriers in the family were associated with lower levels of healthy habits (e.g., screen time and fruit and vegetable consumption). Notably, this is the first study to show that parents' perceptions of barriers to child weight management are directly related to implementation of weight management habits. Subscales of the BPWM measure also demonstrated initial convergent validity, at levels consistent with previously developed measures of barriers within other chronic illnesses.^42,43 Specifically, the current measure exhibited similar levels of alpha reliability and convergent validity compared to previous measures of barriers within the pediatric diabetes literature.^42,43 Overall, this measure has appropriate validity to be used within both clinical and research populations, to identify correlates and factors affecting barriers, as well as identifying specific topics that should be addressed with families in a clinical setting.

Limitations

While the development of the BPWM shows encouraging findings, several methodological issues should be taken into account. First, data for this study were collected using a nonclinical sample, and in the future should be examined across various populations with further examinations of the factor structure. While healthy weight maintenance is important in all families, many families, including those families who are not seeking weight management intervention, face significant barriers to maintaining healthy weight-related behaviors that are not currently being addressed. Future studies should examine the utility of this Barriers measure in identifying more widespread barriers that may be addressed through public health intervention. This is especially important given that only a small percentage of overweight and obese youth in the United States receive clinical intervention for weight management. Furthermore, future research should explore the potential clinical utility of the BPWM measure within a treatment-seeking population, especially in consultation with experts across multiple disciplines addressing various facets of pediatric weight management (e.g., exercise physiology, dietitian, and medical providers).

Another limitation of the study was the split-half methodology used for validating the factor structure of the BPWM measure. While it is important to cross-validate factor structures using EFA and CFA procedures, we had to rely on samples derived from the same population. Thus, the BPWM factor structure should be examined in additional samples. Finally, this study was a single-informant study. Children may have differing perceptions of barriers they face in healthy weight management. Multi-method and multi-informant studies are needed to better understand differences in perceptions of barriers from each perspective. Specifically, a child version of this measure would be helpful in understanding barriers from multiple perspectives.

Furthermore, while subject pool diversity has been noted as a strength of recruitment through MTurk,¹⁸ recruitment through MTurk faces difficulties in recruitment similar to in-person recruitment for behavioral research (e.g., self-selection biases; biases from payment—less interpersonally oriented and less attention to detail).⁴⁴ Future studies should aim to recruit a more diverse sample, as this study included a lower percentage of individuals who identify as Hispanic or Latino than the US population. In addition, the use of an online sample limits the understanding of the role of BMI within this study, as both parental and child BMI were self-reported. Previous research has found that there is a strong correlation between self-reported and objective BMI; however, calculated BMI does tend to differ from self-report data.⁴⁵ Specific to this study, child weight status tends to be higher than the national averages, with 15% of children being classified as overweight and 27% being classified as obese, compared with 31% and 17%, respectively.¹

Future Directions

Overall, this study developed a quantitative measure of barriers to pediatric weight management that can be used in clinical and research settings. This study is an important step in development of this measure to examine the barriers that may prevent families from maintaining healthy weight-related behaviors. Addressing barriers to weight management within the family may be helpful in preventing child weight gain and can also be used as a tool for weight management intervention in children. While this study assesses four important factors affecting management of pediatric obesity, other important factors that are not encompassed within the final measure should be examined in future research (e.g., parental eating and activity behaviors). Validation within future samples should include objective outcome variables (e.g., measured BMI percentile) and validation of the measure within a clinical sample. Future research should aim to further develop this measure in other areas; specifically this measure may be valuable for rapid identification of concerns within clinical settings or may be used within population-based studies examining familial traits that may be most appropriate for public health interventions.

Footnotes

Author Disclosure Statement

No completing financial interests exist.

References

Ogden

, Carroll

, Kit

, Flegal

. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA, 2014; 311:806–814.

Kraak

, Liverman

, Koplan

. Preventing Childhood Obesity: Health in the Balance. National Academies Press, Washington, DC, 2005.

Murtagh

, Dixey

, Rudolf

. A qualitative investigation into the levers and barriers to weight loss in children: Opinions of obese children. Arch Dis Child, 2006; 91:920–923.

Uzark

, Becker

, Dielman

, et al. Perceptions held by obese children and their parents: Implications for weight control intervention. Health Educ Q, 1988; 15:185–198.

Barlow

, Dietz

. Management of child and adolescent obesity: Summary and recommendations based on reports from pediatricians, pediatric nurse practitioners, and registered dietitians. Pediatrics, 2002; 110(Supplement 1):236–238.

Epstein

, Koeske

, Wing

, Valoski

. The effect of family variables on child weight change. Health Psychol, 1986; 5:1.

Epstein

, Paluch

, Roemmich

, Beecher

. Family-based obesity treatment, then and now: Twenty-five years of pediatric obesity treatment. Health Psychol, 2007; 26:381.

Eikenberry

, Smith

Healthful eating: Perceptions, motivations, barriers, and promoters in low-income Minnesota communities. J Am Diet Assoc, 2004; 104:1158–1161.

Lee

, Harris

, Gordon-Larsen

Life course perspectives on the links between poverty and obesity during the transition to young adulthood. Popul Res Policy Rev, 2009; 28:505.

10.

Foster

, Giles-Corti

. The built environment, neighborhood crime and constrained physical activity: An exploration of inconsistent findings. Prev Med, 2008; 47:241–251.

11.

Gordon-Larsen

, Nelson

, Page

, Popkin

. Inequality in the built environment underlies key health disparities in physical activity and obesity. Pediatrics, 2006; 117:417–424.

12.

Lytle

, Hearst

, Fulkerson

, et al. Examining the relationships between family meal practices, family stressors, and the weight of youth in the family. Ann Behav Med, 2011; 41:353–362.

13.

Lucan

, Barg

, Long

. Promoters and barriers to fruit, vegetable, and fast-food consumption among urban, low-income African Americans—A qualitative approach. Am J Public Health, 2010; 100:631–635.

14.

Zabinski

, Saelens

, Stein

, et al. Overweight children's barriers to and support for physical activity. Obes Res, 2003; 11:238–246.

15.

Posey

KG.

Household Income: 2015. U.S. Department of Commerce: Economics and Statistics Administration, Washington, DC, 2016; American Community Survey Briefs.

16.

Berinsky

, Huber

, Lenz

. Evaluating online labor markets for experimental research: Amazon.com's Mechanical Turk. Pol Anal, 2012; 20:351–368.

17.

Buhrmester

, Kwang

, Gosling

. Amazon's Mechanical Turk a new source of inexpensive, yet high-quality, data?. Perspect Psychol Sci, 2011; 6:3–5.

18.

Mason

, Suri

. Conducting behavioral research on Amazon's Mechanical Turk. Behav Res Methods, 2012; 44:1–23.

19.

Kuczmarski

, Ogden

, Grummer-Strawn

, et al. CDC growth charts: United States. Adv Data, 2000:1–27.

20.

Bickel

, Nord

, Price

, et al. Guide to measuring household food security. US Department of Agriculture, Food and Nutrition Service, Office of Analysis, Nutrition, and Evaluation. 2000. Available at www fns usda gov/fsec/FILES/Guide% 20to% 20Measuring% 20Household% 20Food% 20Security (3-23-00) pdf (last accessed October 1, 2017).

21.

Pinto

, Fava

, Raynor

, et al. Development and validation of the weight control strategies scale. Obesity, 2013; 21:2429–2436.

22.

Rogers

, Motyka

. 5-2-1-0 goes to school: A pilot project testing the feasibility of schools adopting and delivering healthy messages during the school day. Pediatrics, 2009; 123(Supplement 5):S272–S276.

23.

Sothern

, Hunter

, Suskind

, et al. Motivating the obese child to move: The role of structured exercise in pediatric weight management. South Med J, 1999; 92:577–584.

24.

Ogden

, Lamb

, Carroll

, Flegal

. Obesity and socioeconomic status in children and adolescents: United States, 2005–2008. NCHS Data Brief, 2010; 51:1–8.

25.

Golan

, Crow

. Targeting parents exclusively in the treatment of childhood obesity: Long‐term results. Obesity, 2004; 12:357–361.

26.

Jabs

, Devine

. Time scarcity and food choices: An overview. Appetite, 2006; 47:196–204.

27.

Reise

, Waller

, Comrey

. Factor analysis and scale revision. Psychol Assess, 2000; 12:287.

28.

Eisinga

, Te Grotenhuis

, Pelzer

. The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown?. Int J Public Health, 2013; 58:637–642.

29.

Muthén

, Muthén

. Mplus users guide, version 7. Statistical Analysis with Latent Variables. Los Angeles, CA, 2007, p. 3.

30.

, Bentler

. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equ Modeling, 1999; 6:1–55.

31.

Kline

RB.

Principles and Practice of Structural Equation Modeling. Guilford Publications, New York, NY, 2015.

32.

Bentler

PM.

Comparative fit indexes in structural models. Psychol Bull. 1990; 107:238.

33.

Browne

, Cudeck

, Bollen

, Long

. Alternative ways of assessing model fit. Sage Focus Editions, 1993; 154:136.

34.

Osborne

, Costello

. Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Pan-Pac Manage Rev, 2009; 12:131–146.

35.

Brito

, Pearl

. A new identification condition for recursive models with correlated errors. Struct Equ Modeling, 2002; 9:459–474.

36.

Byrne

BM.

Structural Equation Modeling with Mplus: Basic Concepts, Applications, and Programming: Routledge, New York, NY, 2013.

37.

Little

TD.

Longitudinal Structural Equation Modeling. Guilford Press, New York, NY, 2013.

38.

Lance

, Butts

, Michels

. The sources of four commonly reported cutoff criteria what did they really say?. Organ Res Methods, 2006; 9:202–220.

39.

Casey

, Szeto

, Lensing

, et al. Children in food-insufficient, low-income families: Prevalence, health, and nutrition status. Arch Pediatr Adolesc Med, 2001; 155:508–514.

40.

Sallis

, Glanz

. Physical activity and food environments: Solutions to the obesity epidemic. Milbank Q, 2009; 87:123–154.

41.

Gellar

, Schrader

, Nansel

. Healthy eating practices perceptions, facilitators, and barriers among youth with diabetes. Diabetes Educ, 2007; 33:671–679.

42.

Mulvaney

, Hood

, Schlundt

, et al. Development and initial validation of the barriers to diabetes adherence measure for adolescents. Diabetes Res Clin Pract, 2011; 94:77–83.

43.

Glasgow

, Toobert

, Gillette

. Psychosocial barriers to diabetes self-management and quality of life. Diabetes Spectr, 2001; 14:33–41.

44.

Rush

, Phillips

, Panek

. Subject recruitment bias: The paid volunteer subject. Percept Mot Skills, 1978; 47:443–449.

45.

Kuczmarski

, Kuczmarski

, Najjar

. Effects of age on validity of self-reported height, weight, and body mass index: Findings from the Third National Health and Nutrition Examination Survey, 1988–1994. J Am Diet Assoc Jan 2001; 101:28–34; quiz 35–36.