Qualitative Comparative Analysis of Program and Participant Factors That Explain Success in a Micropolitan Pediatric Weight Management Intervention

Abstract

Background:

Building Healthy Families (BHF) was developed through a community–academic partnership to provide a 12-week family-based obesity treatment program. Nine cohorts of BHF have been delivered in multiple micropolitan settings between 2009 and 2016, but participant outcomes have varied. This study sought to explore the variation in BHF outcomes to identify the necessary and sufficient conditions that are associated with larger 12-week reductions in BMI z-scores.

Methods:

A qualitative comparative analysis was used to determine potential causal conditions or combination of conditions associated with larger reductions in BMI z-score. Seventy-five participants with 12-week outcome data were rank ordered as the 10 most successful participants (largest reductions in BMI z-score; m ± std = −0.64 ± 0.18) and the 10 least successful participants (smallest reductions or an increase in BMI z-score; m ± std = 0.02 ± 0.04). The conditions selected for analysis were identified based on theory and the delivery team's experience with implementing BHF.

Results:

Necessary conditions (i.e., present in all highly successful participants, but also some less successful participants) included children with high attendance and self-regulation, at least one adult with high attendance and self-regulation, a mother who lost weight during the program and achieved clinically meaningful weight loss. Sufficient conditions (i.e., present in only the highly successful participants) included mothers with self-regulation scores >45% (range 46.7%–98.2%), and children with a combination of high attendance (72%–100%) and self-regulation scores ≥45% (45%–92.7%).

Conclusion:

Program implementers should continue to focus on encouraging high attendance and emphasize the necessity of enacting self-regulation strategies at both the child and parent level.

Introduction

The prevalence of childhood obesity remains high, and disparities exist throughout geographical areas.^1,2 Children and adolescents have 26% greater odds of obesity if they reside in a rural area compared with their urban counterparts.² Furthermore, the majority of pediatric weight management interventions (PWMIs) were developed and based in larger cities and urban areas delivered through interdisciplinary teams in a hospital or medical center, and the most recent childhood obesity treatment recommendations do not address medically underserved areas.³ As a result, the degree to which these or other efficacious PWMIs have been translated to, and are feasible in, typical rural or micropolitan areas (i.e., cities with <50,000 residents) is unclear.^3,4

Fortunately, decades of efficacy trials have yielded a vast body of evidence for the elements necessary to treat childhood obesity. Interventions that focus on the family and parents as agents of change through improved dietary intake, increasing physical activity, and behavioral strategies delivered through an interdisciplinary team (e.g., primary care provider, behavioral interventionist, registered dietitian, exercise physiologist, and a care coordinator) have efficaciously reduced child weight status.^5–9 However, those who reside in rural or micropolitan communities may not experience the benefits of PWMI due to differences in the availability of interdisciplinary health care teams and community resources needed for implementation.^4,5

Given the differences in community resources and expertise available for childhood obesity treatment between metropolitan communities and micropolitan and surrounding rural communities, there is a need to determine where these communities should focus limited resources by examining the conditions that lead to intervention success for program participants. Understanding the potential causal factors that lead to program success may help to guide program strategy development that if optimized could enhance the impact of local PWMI implementation. Based on the previous literature, this could include examining (1) program factors, (2) child characteristics, and (3) parent characteristics.^10–14

Unfortunately, it is difficult to determine which of these factors are necessary and sufficient to produce pediatric weight management success. Further, given the limited information specific to program delivery in rural or underserved areas, even less is known about which factors are most important in these settings. Conventional statistical models may not capture the causality of conditions or variability in outcomes within the available small sample sizes. However, innovative methods such as qualitative comparative analyses (QCAs) can be used to create a specific set-theoretical approach to analyzing complex social phenomena.¹⁵

The approach allows for a systematic, cross-case comparison among a small number of cases to understand the necessary (i.e., present in all highly successful participants and some unsuccessful participants) and sufficient (i.e., present in only highly successful participants) factors that increase the likelihood of supporting changes in child weight status to inform adaptations and speed the translation of evidence-based PWMI in underserved areas.^16,17

Building Healthy Families (BHF), an adapted evidence- and family-based PWMI,¹⁰ was developed and implemented in a midwestern micropolitan city (population ∼33,464) through a community–academic partnership to provide a treatment option for families with children who have obesity in underserved areas. Cohort 6 was implemented in an additional community setting (population ∼5495) 30 miles from the original site at the university using the developed program materials from the education implementation team.

Cohort 9 was implemented in two additional community settings (population ∼3359–7698) within 65 to 100 miles from the original program site at the university. With the exception of cohort 6, all cohorts of BHF were delivered by the same education implementation team that codeveloped the program (e.g., program coordinator; university researcher, nutrition coordinator; community dietician, behavior modification coordinator; community clinical behavioral psychologist, and academic exercise scientist) and held at the university.

Cohort 9 was delivered face to face at the university location by the education implementation team and via synchronous video over Zoom to two additional micropolitan communities. Each additional community that connected via Zoom held the weekly sessions at a central location in the community, and utilized a local program coordinator to establish relationships with the families and the university education implementation team.

These communities also used a local physical activity coordinator to facilitate the family physical activity sessions. Children and their families were eligible to participate in BHF if the child was between 6 and 12 years of age and had a BMI percentile ranking at or above the 95th percentile. Families were recruited throughout all nine cohorts using local newspaper advertisements, physician referrals, and BMI report cards through the public schools.

BHF consists of 12 weekly face-to-face group sessions and a 6-month follow-up session, 2 hours in duration, to meet the recommended ≥26 contact hours with the enrolled families and the education implementation team. The intervention was adapted from Epstein et al.’ efficacious Traffic Light Eating Plan.^6,13,18

Weekly interactions focus on key behavior change strategies (e.g., goal setting, self-monitoring, rewards/contingency management, role modeling, and stimulus control/modifying the environment), improved dietary intake, increased physical activity, and decreased sedentary time. Families engaged in nutrition education, behavior modification, and physical activity education together and independently depending on the topic and activity. BHF has helped children to change their weight status with a similar magnitude of change to those seen in efficacy trials (BMI z-score reduction of ≥0.25).^{6,10,13,19,20}

The purpose of this study was to explore the variation in BHF program conditions, child characteristics, and parent characteristics that distinguish between participants with the largest BMI z-score reductions and those with the smallest reductions or an increase in BMI z-score. The overarching goal of this study is to identify key program, child, and parent factors that may predict successful changes in weight status, and to use these findings in communicating core intervention components that are sufficient to achieve success.

Methods

To determine necessary and sufficient factors that may increase the likelihood of participant success in BHF, we used a QCA methodology. This QCA used a retrospective study design to analyze and compare participant outcomes over nine cohorts of families participating in BHF from 2009 to 2016 with program, child, and parent factors that may be necessary and sufficient to predict child success in reducing weight status (measured by BMI z-scores). From the available secondary data, several potential causal conditions were identified based on delivery team experience with implementing BHF and evidence-based recommendations for PWMIs.^{3,6,14,21–25} The BHF program was approved by the University of Nebraska at Kearney Institutional Review Board (IRB#: 090408-1).

Participation Selection

BHF was delivered to 1 to 2 cohorts of families per year consisting of 5 to 12 families per cohort, for a total of nine cohorts. Eighty-two percent of families completed 12 weeks of BHF resulting in data for 75 children, mean age 9.9 ± 2.5 years with an initial BMI percentile of 97.0 ± 3.3. The 10 children with the largest reductions in 12-week BMI z-scores and the 10 children with the smallest reductions or an increase in BMI z-scores were selected along with their parents or guardians (n = 32) to be included in the analysis. The 10 cases selected for analysis defined as “most successful” reduced their BMI z-scores on average by −0.64 (±0.18), and the 10 cases defined as “least successful” showed a small increase in BMI z-score (0.02 ± 0.04).

Data Collection and Coding

Secondary, deidentified data collected over the nine cohorts of BHF were used in this analysis. Data were derived pragmatically from information gathered during the implementation of BHF for all nine cohorts (n = 75 children). Four abstractors from the BHF implementation team and evaluation team were used to obtain data from 20 child participant records. Each participant's records and the records for the participating adult were collected for analysis. Data available for condition selection included (demographics, program cohort attended, program attendance, self-monitoring completion scores, and 12-week weight change; Table 1).

Table 1.

Description of Conditions That May Account for Higher or Lower Success in Building Healthy Families and How Those Conditions Were Operationalized

Condition	Decision rule	Final coding
Child participant characteristics
Change in BMI z-score	Difference in BMI z-score from baseline to 3 months.	(0) If least successful, (1) if most successful
Most successful participants	Those who had the largest reductions in BMI z-score (−0.46 to −0.97).
Least successful participants	Those who had the smallest reductions in BMI z-score or an increase (0.11 to −0.02).
Self-regulation score	Self-monitoring completion scores for physical activity and red foods. Low self-regulation considered ≤44%, high self-regulation considered ≥45%.	(0) If low self-regulation, (1) if high self-regulation
Attendance	Attendance percentage of child participant. Low attendance considered ≤70%, high attendance considered ≥71%.	(0) If low attendance, (1) if high attendance
Age	Age 6–9 considered younger participants, age 10–12 considered older participants.	(0) If younger, (1) if older
Sex	Biological sex of child.	(0) If male, (1) if female
Race/ethnicity	Race/ethnicity of child.	(0) If Non-Hispanic White, (1) if Hispanic
Parent participant characteristics
Participating parent	Who was the supporting parent in the program?	(0) If mother only, (1) if both parents
Self-regulation	Self-monitoring completion scores for physical activity and red foods. Low self-regulation considered ≤44%, high self-regulation considered ≥45%.	(0) If low self-regulation, (1) if high self-regulation
Attendance	Attendance percentage of parent. Low attendance considered ≤70%, high attendance considered ≥71%.	(0) If low attendance, (1) if high attendance
Weight loss	Did the supporting parent lose weight over the 3-month program?	(0) If no, (1) if yes
Clinically meaningful weight loss	Did the supporting parent achieve clinically meaningful weight loss?	(0) If no, (1) if yes
Age	Age of supporting parent. Age ≤44 considered younger, age ≥45 considered older.	(0) If younger, (1) if older
Sex	Biological sex of supporting parent.	(0) If male, (1) if female
Race/ethnicity	Race/ethnicity of supporting parent.	(0) If Non-Hispanic White, (1) if Hispanic
Program conditions
Cohort	Cohort participant attended.	Cohorts 1–9
Cohort size	Size of cohort. Participants ≤9 considered low, ≥10 participants considered high.	(0) If low, (1) if high
Cohort mean BMI z-score	Mean BMI z-score change ≤0.24 considered low, ≥0.25 considered high.	(0) If low, (1) if high

Self-monitoring was completed by both children and parents using Epstein's Habit Books for cohorts 1–6. Participants recorded their daily food and physical activity on a weekly basis, and provided their total weekly minutes of physical activity and their red foods. Cohorts 7–9 piloted self-monitoring using technology for tracking red foods (MyFitnessPal and Kurbo) and physical activity (FitBits and Omrons) for tracking steps per day. All cohorts were provided a self-regulation score based on the proportion of days self-monitoring was completed.

A crisp set qualitative comparative analysis (csQCA) was used to allow researchers to easily interpret the results by dichotomizing the conditions to better understand the causality of the complex conditions that facilitate the implementation of BHF that could lead to successful participant outcomes. The calibration process uses decision rules to consider how a case is related to the condition, and was determined based on theory and the delivery team experience with implementing BHF.^15,17 The conditions were identified as those (1) that met the condition and (0) those that did not meet the condition.

A raw data table was generated to list all possible configurations of conditions, and a code book was created by researchers using the available data to detail the conditions, decision rules, and coding calibration (Table 1). Researchers worked independently at first, and then collaboratively to compare their work and reach a consensus on the decision rules and configuration patterns. The configurations that were possible but not represented in the sample were eliminated from further analysis as logical remainders.¹⁶

Data Analysis

After calibration, a truth table was generated to list the possible conditions present in the sample to determine the necessary and sufficient conditions or combination of conditions for larger reductions in participant BMI z-scores (Table 2). A condition was considered necessary if it was always present in the top 10 successful participants, but alone it does not guarantee the outcome.

Table 2.

Truth Table

	Change in BMI z-score	Cohort	Child attendance	Child self-regulation	Participating adult	Mom attendance	Dad attendance	Mom weight loss	Mom clinically meaningful	Dad weight loss	Dad clinically meaningful	Mom self-regulation	Dad self-regulation
Most successful participants
1	−0.97	4	1 ^C	1 ^C	1	1	1	1	1	1	1	1	1
2	−0.89	4	1 ^C	1 ^C	0	1	0	1	1	0	0	1	0
3	−0.80	1	1 ^C	1 ^C	1	1	1	1	1	1	1	1	1
4	−0.61	2	1 ^C	1 ^C	0	1	0	1	1	0	0	1	0
5	−0.60	7	1 ^C	1 ^C	1	1	0	1	1	1	1	1	1
6	−0.58	3	1 ^C	1 ^C	0	1	0	1	1	0	0	1	0
7	−0.51	2	1 ^C	1 ^C	1	1	1	1	1	1	1	1	1
8	−0.51	6	1 ^C	1 ^C	1	1	1	1	1	1	1	1	1
9	−0.51	8	1 ^C	1 ^C	1	1	1	1	1	0	0	1	1
10	−0.46	2	1 ^C	1 ^C	1	0	1	1	1	1	1	1	1
Least successful participants
1	0.11	7	1	0	1	1	1	1	1	1	1	0	0
2	0.07	1	0	0	0	0	0	1	0	0	0	0	0
3	0.04	9	1	0	1	1	0	1	0	0	0	0	0
4	0.01	1	0	1	1	0	0	0	0	1	1	0	0
5	0.01	2	0	0	0	0	0	0	0	0	0	0	0
6	0.01	7	1	0	0	1	0	1	1	0	0	0	0
7	−0.01	6	1	0	0	1	0	1	0	0	0	0	0
8	−0.01	7	1	0	0	1	0	1	1	0	0	0	0
9	−0.02	5	0	0	1	0	0	1	1	1	1	0	0
10	−0.02	5	0	1	1	1	1	1	0	1	1	0	1

Bold numbers represent the necessary conditions for the outcome. Bold and italic numbers represent the sufficient condition for the outcome. 1^C represents the sufficient combination of conditions for the outcome.

A condition or a combination of conditions was considered sufficient if it was only present in the top 10 successful participants. Participants were first examined to identify which conditions were individually necessary or sufficient for larger reductions in participant BMI z-score. When a single condition was not present, combinations of conditions were assessed. This process was done until all participants with the largest reductions in weight status were considered.

Results

The mean age of child participants was 9.7 ± 2.0 years, and the mean age for the participating adult was 42.7 ± 9.3. The child participants were 60% male (n = 12) and 40% female (n = 8), and the adult participants were 37.5% male (n = 12) and 62.5% female (n = 20). Overall, child participants and their participating adults represented 80% Non-Hispanic White (n = 43) and 20% Hispanic (n = 11) (Table 3 Participant Characteristics).

Table 3.

Participant Characteristics

Participant descriptive characteristics	Most successful participants Reductions in BMI z-score (−0.46 to −0.97)	Least successful participants Reductions in BMI z-score (−0.01 to −0.02)	Total
Percent children with reduced BMI z-score	n = 10 (100%)	n = 4 (40%)	n = 14 (70%)
Child age	9.6 ± 1.4	9.9 ± 2.5
Child sex
Female	4 (40%)	4 (40%)	n = 8 (40%)
Male	6 (60%)	6 (60%)	n = 12 (60%)
Child race/ethnicity
Non-Hispanic White	9 (90%)	7 (70%)	n = 16 (80%)
Hispanic	1 (10%)	3 (30%)	n = 4 (20%)
Self-regulation scores (% complete)
Child	65.8 ± 18.2	22.9 ± 23.3
Parent	64.4 ± 20.6	23.9 ± 17
Participating fathers	7	5	n = 12
Participating mothers	10	10	n = 20
Both parents intervention participation^a	n = 7 (70%)	n = 5 (50%)	n = 12 (60%)
Mother only intervention participation	n = 3 (30%)	n = 5 (50%)	n = 8 (40%)
Father age	40 ± 4.5	48.9 ± 13.5
Mother age	40.5 ± 5.5	43.7 ± 10.6
Father average % weight loss	8.3 ± 5.5	8.0 ± 1.6
Mother average % weight loss	8.9 ± 2.7	4.3 ± 2.4
Father average weight loss (lbs)	−19.9 ± 14.1	−14.3 ± 2.2
Mother average weight loss (lbs)	−14.3 ± 4.5	−10.6 ± 6.5

Child attendance can vary from parent participation if both parents did not attend a given session or if a child did not attend with the parents.

One set of grandparents were included in the analysis, and considered as the mother and father participants.

Although there were more Hispanic participants among the least successful participants, there was not a clear pattern present for race/ethnicity among the most successful and least successful participants. Table 4 provides more details on the race/ethnicity of participants across cohorts, which is consistent with the demographics of the micropolitan communities BHF was implemented.

Table 4.

Cohorts 1–9 Characteristics

	Cohort 1	Cohort 2	Cohort 3	Cohort 4	Cohort 5	Cohort 6	Cohort 7	Cohort 8	Cohort 9
Cohort size—child participants	11	12	11	9	5	9	5	5	8
Number of participants with clinically significant reductions in BMI z-score	6	8	8	6	1	3	2	1	2
Percentage of participants with clinically significant reductions in BMI z-score (≥0.25)	55	67	73	67	20	33	40	20	25
Cohort mean change in BMI z-score	−0.27	−0.31	−0.33	−0.41	−0.11	−0.22	−0.19	−0.23	−0.16
Cohort mean age	9.9	10.2	9.6	8.7	9.1	9.7	8.6	10.5	8.3
Sex
Female	4	6	7	7	2	3	2	2	5
Male	7	6	4	2	3	6	3	3	3
Race/ethnicity
Non-Hispanic White	10	11	9	8	4	8	3	5	8
Hispanic	1	1	2	1	1	1	2	0	0

Among the participants selected for this analysis, 70% of the child cases (n = 14) had an overall reduction in BMI z-scores (−0.01 to −0.97). The most successful cases achieved reductions in BMI z-scores ranging from −0.46 to −0.97. Parental weight loss appeared to be a strong indicator of child success in the BHF intervention. Specifically, among the most successful cases, 100% of mothers lost weight and achieved clinically meaningful weight loss (>5% of initial body weight). Whereas among the least successful cases only 40% of the mothers achieved clinically meaningful weight loss. Overall, weight loss among parents was achieved by 90% of the mothers and 83% of the fathers, and clinically meaningful weight loss was achieved by 70% of mothers and 83% of fathers (Table 3).

An overview of the factors that appeared to have some relationship with the most and least successful participants is provided in Table 1. While some cohorts (3, 4, and 8) were represented by the most successful participants and not in the group of least successful participants, there was also not a clear pattern present for cohort attended and success in BHF—suggesting that program factors were likely consistent across cohorts (see Table 4 for details). Overall, factors that demonstrated some pattern of presence and program success focused on child and parent engagement, enactment of self-regulation strategies, and successful parent weight loss.

Necessary conditions for successful child weight loss (i.e., factors that characterized all of the highly successful participants, but also some of the less successful participants) included children with high attendance and self-regulation, at least one adult with high attendance and high self-regulation, and a mother who lost weight during the program and achieved clinically meaningful weight loss (≥5% initial body weight). Fathers' self-regulation did not meet the conditions to be necessary or sufficient; however, 100% of the most successful participants who had a father engaged in the program (n = 6) achieved high self-regulation compared with 20% of the least successful participants.

Two sufficient conditions or combination of conditions (i.e., patterns found in only the highly successful participants) were identified. The first sufficient condition was having a mother with high self-regulation scores (46.7%–98.2%). The second sufficient combination of conditions included children with both high attendance (72%–100%) and high self-regulation scores (55%–87%). All of the most successful participants met these criteria, while none of the least successful participants did.

Discussion

The BHF program has demonstrated effectiveness at reducing participant BMI z-scores with nearly half (49%) of the participants achieving clinically significant reductions over nine cohorts with a similar magnitude of change to those seen in efficacy trials (BMI z-score reduction of ≥0.25).^{6,10,13,19,20} We applied a QCA to a selection of the most successful and the least successful child participants to explore which intervention factors were associated with the variation in participant weight status outcomes. Findings from this study suggest that there are program features and participant engagement factors that are likely critical for success in pediatric obesity treatment interventions.

Similar to other obesity treatment interventions, parental weight loss was also a strong indicator of child success in BHF.^7,12,20,25 Successful weight loss by parents of children in the most successful cases could be linked to changes in the home environment and shared behavioral changes made through the program.²⁵ Although attendance is important for successful outcomes in obesity treatment interventions, our findings suggest that high child attendance will have no impact on their success if parental self-regulation is low.

Of note, children among the least successful participants with high attendance had low self-regulation and a supporting parent with low self-regulation. Those children with high self-regulation but low attendance did not attend as many sessions as the supporting parent; emphasizing the importance of family participation for successful outcomes in obesity treatment interventions.

Parents and guardians play a crucial role in developing and supporting health behaviors in children. The degree of parent engagement in pediatric obesity treatment interventions has been associated with child weight status outcomes.^7,21,22 High engagement in treatment interventions provides an opportunity for parents and guardians to model positive health behaviors and implement changes within the home environment.²¹ Previous studies assessing positive outliers (i.e., children who have been successful at changing health behaviors and reducing BMI) have underscored the importance of parental behavior change and support in changing child behaviors that facilitate successful reductions in BMI.^26,27

Parents act as important agents of change to impact child weight status.^7,12 However, limited studies are available that demonstrate the specific behavioral strategies in family-based treatment interventions that lead to successful child and parent weight change. Greater adherence to intervention behavioral strategies by parents has demonstrated greater reductions in child weight status.^21,25,28

Participating in family-based interventions allows parents to model behavioral strategies, and for the child to practice new behaviors among peers and family while receiving feedback on their learned behaviors.^3,23,25 Wilfley et al.²³ found three parent variables associated with successful child weight status outcomes—monitoring and goal setting to support the child, home environment to support the child, and healthy behaviors with peers.³

Throughout the nine cohorts of BHF, program implementers used daily habit books where participants would hand write, or in some cohorts electronically log, their food intake while identifying the number of red foods consumed, their daily physical activity levels, and number of days they met their program goals.

Similar to our results, previous studies have determined that participants who regularly use self-monitoring logs and filled them out completely were among the most successful cases.²⁴ Of note, the BHF program based self-regulation on the achievement of program-defined goals and self-monitoring. Thus, our results could be interpreted to suggest that regular engagement in high self-regulation relative to a specific program goal is a key ingredient to change.

Our findings indicated that high self-regulation was a key indicator for successful weight outcomes for both the child and the participating parent. The presence of high self-regulation in both children and the participating parent among the most successful cases suggests that child adherence to behavioral strategies is also a key factor influencing the parent and child's success. The mother's self-regulation score was determined as a sufficient condition for child success in BHF.

Further, while not necessary or sufficient, 100% of the fathers who were engaged in the program achieved high self-regulation scores among the most successful cases. The majority of the parent participants were mothers which could have led to this finding or the absence of engagement from the father could also have been a factor for children being unsuccessful in BHF. This suggests that future implementation efforts should be structured to encourage self-regulation strategies of both the children and the participating parents in the intervention and to develop strategies focused on adherence to self-monitoring behavior change goals.

This analysis allowed for a systematic, cross-case comparison among a small sample size to identify the conditions that led to larger reductions in BMI z-score of participants in a micropolitan childhood obesity treatment program where other studies may have failed to capture these nuances quantitatively. A limitation of this QCA is that it is retrospective in nature, and researchers were limited to examining the data that were collected during the original trial. Further, the results from this study are limited to program completers, and do not provide insight on program features that could be enhanced to reduce attrition and reach those in need of obesity treatment.

Our findings, which are similar in magnitude to other successful childhood obesity interventions tested in metropolitan areas, suggest that future community implementers of BHF should communicate the benefits of high attendance, maintaining self-regulation skills, and the supporting parent's success to maximize effectiveness. This study provides implications for improving pediatric obesity treatment programs and understanding the potential causal factors that lead to program success in a micropolitan community to enhance implementation.

Conclusion

Program implementers should continue to focus on encouraging high attendance, and underscore the necessity of enacting self-regulation strategies at both the child and parent level. Better knowledge of treatment outcome predictors could facilitate effective adaptations to translate BHF in other micropolitan or rural settings in need of childhood obesity treatment options.

Footnotes

Funding Information

This work was supported in part by the Nebraska Research Initiative, Rural Futures Institute of Nebraska, Blue Cross Blue Shield of Nebraska, Sentinel Health, and the Centers for Disease Control and Prevention of the U.S. Department of Health and Human Services (U18DP006431). The contents are those of the authors and do not necessarily represent the official views of, nor an endorsement, by CDC/HHS, or the U.S. Government or any of the aforementioned organizations.

Author Disclosure Statement

No competing financial interests exist.

References

Ogden

, Fryar

, Hales

, et al. Differences in obesity prevalence by demographics and urbanization in US children and adolescents, 2013–2016. JAMA, 2019; 319:2410–2418.

Johnson JA

III

, Johnson

. Urban-rural differences in childhood and adolescent obesity in the United States: A systematic review and meta-analysis. Child Obes, 2015; 11:233–241.

Wilfley

, Staiano

, Altman

, et al. Improving access and systems of care for evidence-based childhood obesity treatment: Conference key findings and next steps. Obesity (Silver Spring), 2017; 25:16–29.

Findholt

, Davis

, Michael

. Perceived barriers, resources, and training needs of rural primary care providers relevant to the management of childhood obesity. J Rural Health, 2013; 29 Suppl 1:s17–s24.

Lim

, Follansbee-Junger

, Crawford

, Janicke

. Treatment outcome research in rural pediatric populations: The challenge of recruitment. J Pediatr Psychol, 2010; 36:696–707.

Epstein

, Paluch

, Roemmich

, Beecher

. Family-based obesity treatment, then and now: Twenty-five years of pediatric obesity treatment. Health Psychol, 2007; 26:381–391.

Golan

Parents as agents of change in childhood obesity—From research to practice. Int J Pediatr Obes, 2006; 1:66–76.

Golan

, Weizman

, Fainaru

. Impact of treatment for childhood obesity on parental risk factors for cardiovascular disease. Prev Med, 1999; 29:519–526.

Golan

, Weizman

, Apter

, Fainaru

. Parents as the exclusive agents of change in the treatment of childhood obesity. Am J Clin Nutr, 1998; 67:1130–1135.

10.

Ruebel

, Heelan

, Bartee

, Foster

. Outcomes of a family based pediatric obesity program—Preliminary results. Int J Exerc Sci, 2011; 4:217–228.

11.

Damschroder

, Aron

, Keith

, et al. Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implement Sci, 2009; 4:50.

12.

Golan

, Crow

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

13.

Epstein

, Valoski

, Wing

, McCurley

. Ten-year outcomes of behavioral family-based treatment for childhood obesity. Health Psychol, 1994; 13:373–383.

14.

Israel

, Silverman

, Solotar

. The relationship between adherence and weight loss in behavioral treatment program for overweight children. Behav Ther, 1988; 19:25–33.

15.

Devers

, Lallemand

, Burton

, et al. Using qualitative comparative analysis to study patient-centered medical homes. Washington

D.C.

, The Urban

Institute.

2013. Available at https://www.urban.org/research/publication/using-qualitative-comparative-analysis-qca-study-patient-centered-medical-homes/view/full_report Last accessed October 6, 2019.

16.

Kane

, Lewis

, Williams

, Kahwati

. Using qualitative comparative analysis to understand and quantify translation and implementation. Transl Behav Med, 2014; 4:201–208.

17.

Ragin

CC.

Using qualitative comparative analysis to study causal complexity. Health Serv Res, 1999; 34:1225–139.

18.

Epstein

LH.

Family-based behavioural intervention for obese children. Int J Obes Relat Metab Disord, 1996; 20 Suppl 1:S14–S21.

19.

Epstein

, Wing

, Penner

, et al. Effects of family-based behavioral treatment on obese 5-to-8 year-old children. Behav Ther, 1985; 16:205–212.

20.

Epstein

, Wing

, Koeske

, et al. Child and parent weight loss in family-based behavior modification programs. J Consult Clin Psychol, 1981; 49:674–685.

21.

Janicke

, Steele

, Gayes

, et al. Systematic review and meta-analysis of comprehensive behavioral family lifestyle interventions addressing pediatric obesity. J Pediatr Psychol, 2014; 39:809–825.

22.

Kitzmann

, Dalton WT

III

, Stanley

, et al. Lifestyle interventions for youth who are overweight: A meta-analytic review. Health Psychol, 2010; 29:91–101.

23.

Wilfley

, Saelens

, Stein

, et al. Dose, content, and mediators of family-based treatment for childhood obesity. A multisite randomized clinical trial. JAMA Pediatr, 2017; 171:1151–1159.

24.

Burke

, Wang

, Sevick

. Self-monitoring in weight loss: A systematic review of the literature. J Am Diet Assoc, 2011; 111:92–102.

25.

Wrotniak

, Epstein

, Paluch

, Roemmich

. Parent weight change as a predictor of child weight change in family-based behavioral obesity treatment. Arch Pediatr Adolesc Med, 2004; 158:342–347.

26.

Sharifi

, Marshall

, Goldman

, et al. Exploring approaches and patient-centered outcomes from positive outliers in childhood obesity. Acad Pediatr, 2014; 14:646–655.

27.

Sharifi

, Marshall

, Goldman

, et al. Engaging in the development of obesity interventions: Exploring outcomes that matter most among obesity positive outliers. Patient Educ Couns, 2015; 98:1393–1401.

28.

Faith

, Van Horn

, Appel

, et al. Evaluating parents and adult caregivers as “agents of change” for treating obese children: Evidence for parent behavior change strategies and research gaps: A scientific statement from the American Heart Association. Circulation, 2012; 125:1186–1207.