A Mixed-Methods Examination of Referral Processes to Clinic–Community Partnership Programs for the Treatment of Childhood Obesity

Abstract

Background:

Partnerships with community programs have been used to improve access to obesity care and address clinical barriers to childhood obesity management; however, little is known about the program referral process. The objective of this study was to identify factors that affect the referral from clinics to community-based programs.

Methods:

Active Recreation through Community-Healthcare Engagement Study (ARCHES) is a mixed-method, implementation study designed to test the feasibility of establishing clinic–community partnerships to treat childhood obesity. We collected clinical referral and program attendance data from the six ARCHES clinic–community partnerships and conducted semistructured interviews (n = 19) with key stakeholders. Logistic regression models were used to identify referral characteristics associated with ever attending a community program. We used deductive thematic analysis to examine contextual factors affecting the clinical referral and subsequent attendance at the community programs.

Results:

Patients referred from individual providers [odds ratio (OR): 3.20, 95% confidence interval (CI): 1.08–9.48], specialty clinics (OR: 2.73, 95% CI: 1.48–5.05), and community wellness clinics (OR: 3.42, 95% CI: 1.05–11.13), had greater odds of ever attending the programs compared with patients from primary care clinics. Patients referred to cohort-based programs compared with open enrollment programs had greater odds of ever attending the programs. Stakeholders emphasized the value of communication within the partnership and with patients in clinical settings. Effective provider communication with patients involved engaging and program endorsing conversations to explain the value of the program.

Conclusions:

We identified factors that may improve the referral process in clinic–community partnerships to provide resources to primary care providers looking to address childhood obesity.

Clinical Trial Registration number: NCT03246763.

Introduction

The prevalence of childhood obesity in the United States continues to rise and remains a public health concern for our nation's youth and their care providers.¹ Within the United States, obesity disproportionately affects racial minority and low-income populations, predisposing them to the negative health consequences associated with obesity.^1,2 United States Preventative Services Task Force (USPSTF) guidelines for clinical treatment recommend that pediatric patients with obesity receive 26 hours of counseling, treatment, and education over a 6-month period.^3,4 These guidelines are intended to reduce disparities, improve treatment outcomes, and ultimately reduce the prevalence of child obesity.

Primary care providers should screen for overweight and obesity using CDC age- and gender-specific BMI growth curves. However, beyond diagnosis, they lack the necessary time to deliver comprehensive behavioral counseling, do not receive adequate reimbursement for obesity-related counseling, and lack actionable and evidence-based treatment programs for their patients.^5–7 Although the 26 hours treatment recommendation is often difficult to achieve in clinical settings, partnering with community programs provide an opportunity to enhance patient care, utilize existing resources, and optimize their clinical time with patients.^8,9

Clinic–community partnerships have proven successful at extending clinical reach and improving patient care. For instance, successful clinic–community partnerships have provided patients with mental health services, access to vaccinations, smoking cessation programs, and resources to address food insecurity.^10–13 There are also innovative clinic–community partnerships such as MEND Foundation (Mind, Exercise, Nutrition, Do it!) and Bull City Fit (a partnership between Durham Parks and Recreation and Duke Healthy Lifestyles Clinic) that have demonstrated efficacy in addressing childhood obesity and meeting obesity treatment guidelines both nationally and internationally.^14–17 However, despite the demonstrated effectiveness of community-based programs at addressing childhood obesity, previous studies have identified the challenge of recruiting and retaining eligible patients referred from partner clinics.¹⁸ As such, key questions remain about the best way to connect patients with community-based resources and the factors that affect clinical practices' referral to, and patients' use of these programs and partnerships.^10,19–24

Active Recreation through Community-Healthcare Engagement Study (ARCHES) is a mixed-method, large scale implementation study intended to assess the feasibility of implementing clinic–community partnerships to address child obesity in a wide range of geographic locations throughout North Carolina. Analyzing existing clinic–community partnerships, such as those implemented through ARCHES, provides an opportunity to identify clinic-specific barriers to engaging patients in these evidence-based community programs. Therefore, the aim of this article is to identify the key factors that affect the referral process between a clinic and community program to inform future implementations using a similar partnership model.

Methods

We conducted a mixed-methods study, including a secondary analysis of referral-specific data from the ARCHES study, supplemented with in-person semistructured interviews of key implementation stakeholders (n = 19) to provide context and insights into the observed referral patterns and outcomes. The Duke University Health System Institutional Review Board reviewed and approved all study procedures.

Active Recreation through Community-Healthcare Engagement Study (ARCHES)

ARCHES is a hybrid implementation–effectiveness trial (Clinical Trials: https://clinicaltrials.gov/ct2/show/NCT03246763?term=active+recreation&draw=2&rank=1) that aimed to test the feasibility of implementing an evidence-based intervention that delivers the USPSTF-recommended treatment in geographically diverse settings.²⁵ The intervention paired local pediatric practices with a community partner, typically the local Parks & Recreation organization, to deliver comprehensive screening and counseling in the medical setting, and family-inclusive, culturally tailored, accessible physical activity and nutrition sessions at the recreation center. Providers screened patients for obesity, provided obesity-related counseling in the clinic, and referred patients to a community program. The ARCHES implementation required the community program to host at least two sessions per week. All sessions include structured physical activity and programs were encouraged to also provide opportunities for nutrition education. As part of the ARCHES implementation strategy, clinic and community partners along with a connector (i.e., program coordinator responsible for facilitating communication and program activities) came together to discuss criteria for eligibility that were appropriate for their local context and agreed on a shared strategy for providers to refer patients to the community program and appropriate program activities. The ARCHES team provided each partner with a “Trail Guide” that served to guide the partnership and included all materials needed to implement the program, including a sample referral form and standardized program curriculum that could be adapted to meet the needs of the community. Site stakeholders also participated in obesity-related training provided by the ARCHES team.

Participants

Between January 2016 and March 2020, primary care providers referred 1193 children ages 5–17 years with overweight or obesity to six community programs across eight NC counties. The NC Department of Commerce ranked three counties among the most distressed counties in NC and four were classified as rural.^26,27

Measures

Referrals

Providers screened patients for obesity during their standard clinic appointments and completed a form to refer eligible patients to the community program. The site connector received completed referral forms, including patient contact information. Connectors entered the patient data from the referral form into a secure REDCap database, which was also used to record subsequent program attendance and other research data. We classified patients based on whether they ever attended the community program (yes/no).

Site referral characteristics

The referral processes differed slightly by site because each site was responsible for designing and implementing a referral process that best worked for them. Figure 1 illustrates the referral process for each site. We tracked referral process characteristics, including types of clinics making the referral, connector origin, presence of a clinical referral coordinator, presence of a program orientation, and program model. Each referring clinic was classified as primary care, specialty clinic, school-based, individual provider, or community wellness clinic, and connector origin was based on the organization the connector worked for parks and recreation, clinic, or a tertiary community group. The presence of a clinical referral coordinator (yes or no) refers to the clinic having an additional person to assist the clinic and the connector in managing referrals. The presence of a program orientation (yes or no) indicates whether the site had an orientation for patients before attending the community program. For program model, we classified programs as either continuous (i.e., ongoing) or cohort (i.e., specific beginning and ending dates).

Figure 1.

Detailed referral processes for each program site.

Stakeholder interviews

Trained research staff conducted and audiorecorded 19 semistructured interviews with stakeholders, including referring providers (n = 5), connectors (n = 7), and recreation staff (n = 7) from each site. ARCHES investigators developed an interview guide to investigate the implementation of the clinic–community model. For this study, we focused on discussion specific to the referral process. We asked stakeholders about their experience with the referral process, how their method worked for their site, and what, if any, were barriers or facilitators.

Analysis

To describe the referrals and subsequent attendance for each referring clinic and site, an attendance to referral ratios was calculated by dividing the number of patients who ever attended the community program by the total number of patients referred. Frequencies were used to describe categorical site referral characteristics. A logistic regression model (PROC GLIMMIX), with a random effect to account for site-level clustering, was used to identify site referral characteristics associated with patients ever attending the community program. Site orientation was excluded from the model because only one site did not have an orientation. p-Values <0.05 were considered statistically significant. All analyses were performed in SAS 9.4 (Cary, NC).

A transcription service transcribed verbatim stakeholder interviews. Transcripts were uploaded into QSR NVivo12 for thematic analysis using both inductive and deductive approaches.²⁸ Two team members (E.A. and C.N.) developed a codebook using the interview guides and independently coded all transcripts with frequent meetings to resolve coding discrepancies. For this study, codes relevant to the referral process were reviewed to identify specific themes and factors that affected the referral processes. To provide further context for the identified themes, themes were examined by site-level attendance to referral ratios.

Results

Table 1 describes the characteristics of the referral process at each of the ARCHES sites. Two of the six sites only had one referring clinic with a range of 1 to 10 referring groups across the six sites. All but one site (Wake ENERGIZE) had primary care clinics for referring partners. Three sites had specialty clinics (e.g., weight management, diabetes), whereas two sites had school-based referral partners. Three of the six program sites had at least one clinic with a clinical referral coordinator. All three of the clinics with a referral coordinator were primary care clinics. All sites had a connector, either from a clinic, Parks and Recreation, or another clinical group (e.g., a hospital-based community outreach program and two community health coalitions), to facilitate communication between the clinic and the community organization. Finally, two sites offered cohort-based enrollment with the remainder offering continuous enrollment for participants.

Table 1.

Outline and Description of Site-Based Referral Process Characteristics

Characteristic	Wake ENERGIZE	Adventure Rowan!	Fayetteville Fit	MARCHES	Goldsboro PRIDE	BCF at WD Hill
Orientation	YES	YES	YES	YES	YES	NO
Clinical referral coordinator^a	NO	YES	NO	NO	YES	YES
Connector origin²	Clinic	Community coalition	Community coalition	Hospital-based community outreach	Parks and recreation	Clinic
Total number of referring clinics	1	1	6	10	2	5
Number of referring clinics by clinic type
Primary care	—	1	5	6	1	3
Specialty clinic	1	—	1	—	—	1
School-based clinic	—	—	—	1	1	—
Community wellness clinic	—	—	—	1	—	1
Individual provider	—	—	—	2	—	—
Program enrollment type³	Cohort	Cohort	Continuous	Continuous	Continuous	Continuous

If at least one clinic at a site had a clinical referral coordinator, this was noted as YES for the whole site.

Hospital-Based Community Outreach groups and Community Coalitions are classified as Other Group for the purpose of further process analysis.

Cohort-based enrollment programs enroll participants in groups with a set start date, whereas Continuous enrollment programs allow new patients to enroll or start the program at any time.

Table 2 shows the attendance to referral ratios for each referring clinic and site. The average attendance to referral ratio was 0.24 (range 0.09–0.36) indicating that ∼24% of those referred attended the community program at least once. The Wake ENERGIZE and Adventure Rowan! programs had low numbers of referrals, but the highest attendance to referral ratios (R = 0.36, N_referral = 45; R = 0.36, N_referral = 81, respectively), whereas the BCF at WD Hill program had the most referrals, yet the lowest attendance to referral ratio (R = 0.09, N_referral = 367).

Table 2.

Site-Based and Clinic Type-Based Attendance to Referral Ratios^d

Site/Clinic	Referral (n)^a	Attendance (n)^b	Attendance to referral ratio^c
Site totals
Wake ENERGIZE	45	16	0.36
Adventure Rowan!	81	29	0.36
Fayetteville Fit	292	85	0.29
MARCHES^e	292	63	0.22
Goldsboro PRIDE^e	113	13	0.12
BCF at WD Hill	367	33	0.09
Clinic type totals
School-based clinic	182	7	0.04
Specialty clinic	480	105	0.22
Primary care	359	91	0.25
Community wellness clinic	33	11	0.33
Individual provider	46	18	0.39

Referral is defined as the number of participants who were referred from the providing clinic.

Attendance is defined as the number of participants who attended the program at least once.

Number of participants who ever attended divided by the number of referrals.

Given that this was a Duke orchestrated research study, some Duke pediatric providers were able to refer to other ARCHES community programs. These values have been included for the purpose of the study but the clinics were not considered part of the official referral process for those sites.

Missing values: MARCHES—Six patients were referred to the program but did not have a referring provider listed. Four of those six attended a MARCHES program session. Goldsboro PRIDE—Three patients were referred to the program without a documented referring provider. All three of these participants attended the program.

ARCHES, Active Recreation through Community-Healthcare Engagement Study.

Table 3 displays odds ratios that predict the likelihood of attending a program based on referral process characteristics. Patients referred from a specialty clinic [odds ratio (OR): 2.73, 95% confidence interval (CI): 1.48–5.05], community wellness clinic (OR: 3.42, 95% CI: 1.05–11.13), or an individual provider (OR: 3.20, 95% CI: 1.08–9.48) had greater odds of attending a program at least once compared with those referred from a pediatric primary care clinic. In contrast, individuals referred from school-based clinics had significantly lower odds of ever attending a program (OR: 0.15, 95% CI: 0.04–0.52). Referrals from clinics at sites with Connectors sourced from clinics were less likely to ever attend an ARCHES program (OR: 0.24, 95% CI: 0.09–0.62) compared with sites with Connectors from other organizations. Patients referred to cohort programs had significantly greater odds of ever attending the program compared with patients referred to a continuous enrollment program (OR: 3.05, 95% CI: 1.16–8.04).

Table 3.

Odds Ratios Predicting Attendance Based on Clinical/Referral Process Characteristics

Variable	OR	95% CI
Clinic type
Primary care	[Reference]
Specialty clinic	2.73	1.48–5.05
School-based clinic	0.15	0.04–0.52
Community wellness clinic	3.42	1.05–11.13
Individual provider	3.20	1.08–9.48
Connector origin
Clinic	0.24	0.09–0.62
Parks and recreation	1.09	0.28–4.28
Other	[Reference]
Clinic referral coordinator
No coordinator	0.76	0.32–1.85
Yes coordinator	[Reference]
Program type
Cohort-based program	3.05	1.16–8.04
Continuous enrollment	[Reference]

CI, confidence interval; OR, odds ratio.

Referral Factors

The thematic analysis of the interview data revealed important themes that affect the referral process and the stakeholders' perceptions of their process. The themes we identified were communication between stakeholders; clinician engagement with patients; connector engagement; the value of a simple, clear referral process; and the presence of a clinical stakeholder champion and/or a clinical referral coordinator. Table 4 outlines and summarizes these themes and provides representative illustrative quotes. The reported thematic results were referenced across multiple sites, indicating their importance to the referral process.

Table 4.

Themes Identified from Interview Analysis, Brief Description of the Theme, and Illustrative Quotes to Support the Thematic Findings

Theme	Description of theme	Quotes
Communication between stakeholders	Clinical stakeholders and connectors emphasized the value of effective communication with each other in the referral process. Connectors often believed that clinic personnel were hard to reach when they had questions about referrals or patients. Clinical stakeholders felt that communication was successful and frequent.	“They don't have time to do that. […] Like, it's delegated down to the connectors. So, I never met them. […] But the connections gotta be good because they keep referring people, so it's a good thing.”—Connector“So, it's just—it's a lot of trust that goes in between the connector—oh, really all three phases of it—the connector, the community partner, and the clinical partnership—is trusting that each person is doing their part of it. And if the kids aren't coming, they're just not coming.”—Community partner
Clinician engagement with patients	The clinical provider's description of the program and engagement with the patients was perceived as important for program participation at all six sites.These conversations were intended to get patients interested in the clinician-supported program and provider explanations differed between clinical practices and between providers within a practice.	“She would refer people all the time. The only problem was it's really important for parents to be there when the kid is referred in terms of referral rates because if the—she would refer kids, but the parent wouldn't be present because they would just be at school, and I would call the parent and they might not have gotten the paper in the mail or the paper from their kid. So, they'd be like, ‘What? I don't know what this program is.’”—Connector“… the parents weren't there so, the lack of that parent contact makes it increasingly difficult because when you're basically cold calling the parent telling them that their child is eligible for their program, their question is, ‘How do you know my child is eligible? And who gave permission for my child's information to be shared with you?’”—Connector
Connector engagement	The program connector often contacted patients after they received the clinical referral to remind patients of the program. Some sites referenced the importance of timely communication after referral as a way to optimize program engagement.	Interviewer: “Were there any challenges with the referral process, especially when you guys first started?”Interviewee: “No, no. Pretty easy. Referred, and they usually call them almost immediately. Yeah. Yeah, and I had great results.”—Clinical stakeholder“We just kinda talk about what our program is. The biggest draw to families is that a lot of these kids, they're kind of scared or put away from team sports or school activities because of the competitive nature of it and the not-great support that they get there. So, kind of just explaining that our program is family-friendly…”—Connector
Clinical referral coordinator	Four sites referenced the value of an additional in-clinic referral coordinator to help smooth communication and facilitate the referral process. One connector referenced the value of a clinical referral coordinator without actually having one already.	“So, just being easy to communicate with and having a point person at each practice. It's doesn't have to be the provider. It could be the referral person. It could be a nurse that is supportive to call and be like, ‘This person doesn't have a phone number’ or ‘This person hasn't showed up in a while.’ Having a point person is important because sometimes it's hard to get them.”—Connector“It was a little more time for our front office staff who were responsible for implementing the {ARCHES} referral.”—Clinical stakeholder
Simple, clear process	Two sites discussed their referral processes and described them as simple and clear. They specifically referenced the benefit of a single form or having a choice of method for providers as important.	“The simplicity of the one-page referral and sending it to a referral person because, that way, we're not trying to take it places and do stuff because it's—that's gonna make people less likely to do it, if they have to do—more with it other than just fill it out and stick it in the referral box.”—Clinical stakeholder“We did an electronic form and I think we did one on our own, then we borrowed the one that was already created, and then we also had a paper form. So having multiple options. Some providers would rather just grab it off the wall and do it there. Some providers would rather do the smart phrase and print it out. So those were positives, too.”—Clinical stakeholder
Clinical stakeholder champion	Some clinical stakeholders mentioned the difficulty of remembering to refer to the program. At these sites, a clinical stakeholder champion to remind providers of the program as a resource helped increase referrals.	“Again, just constantly reminding providers that—to refer patients, to educate the patients at first.”—Clinical stakeholder“And, again, the biggest thing is getting all my providers to recognize and take the time to do it.”—Clinical stakeholder“Why wouldn't you participate? All you have to do is fill out a referral form and do your patients a favor. So, it wasn't probably any more time for our providers”—Clinical stakeholder

Discussion

This study provides key information to help better inform successful implementation of existing recommendations for child obesity treatment. While intensive community-based programs are effective in reducing child BMI, underutilization and poor retention are known barriers.¹⁸ Our study describes several key implementation strategies that have the potential to increase utilization of evidence-based interventions and allow better access to care. First, we found that cohort-based programs improve the patient odds of attending the program compared with continuous enrollment programs. Additionally, patients referred from individual providers, specialty clinics, and community wellness clinics had greater odds of ever attending the programs when compared with primary care clinics. This reinforces the qualitative finding that effective communication is essential within this clinic–community partnership model, specifically, communication between referring clinicians and their patients. Finally, we found that consistent communication between stakeholders in the clinic–community partnership facilitates effective patient–provider conversations. These findings can help expand the care-providing capacity of clinicians to meet patient needs within a community setting.

First, our findings suggest that provider enthusiasm and engagement during referral conversations with patients and their parents can encourage families to attend the program. From clinic-specific analysis and stakeholder feedback, we found that when providers do not explain the referral, refer solely based on BMI, or do not demonstrate a strong endorsement for the program, conversion from referral to program attendance is less likely. Previous research suggests that lack of parent engagement in and parent perceptions of community-based programs are significant barriers to first-time attendance.²⁹ We found that patients referred from school-based clinics had significantly lower odds of ever attending the community program. Qualitative findings suggest this may be attributed to insufficient communication with parents by school-based providers. One connector described uncomfortable conversations with parents of children referred from school-based clinics. These parents were reportedly caught “off-guard” and “offended” and had not been informed of their child's referral to the program. In contrast, specialty clinics and individual providers were more successful at referring patients to programs, likely because these providers were more invested in the program and were more likely to have discussions about the program with patients and their families in the clinic. This suggests that it is important that providers directly communicate with parents to describe the program, why their child is eligible, and highlight the value of such a program.

Although our findings suggest the quality of communication with patient families is integral to a successful referral to a community program, there are other unique aspects of programs that may also influence attendance. For example, despite the large number of referrals at BCF at WD Hill, even from a specialty clinic, there was low attendance. This site specifically targeted adolescents (11–17 years), so the limited engagement could be linked to the complexities of adolescent attendance and motivation.^30–32 Engaging patients and their families in clinic is vital for a successful referral, even if other factors may affect the likelihood of program attendance. Patient engagement should vary based on the target audience; for adolescent patients, the conversation could be directed at the patient and for younger pediatric patients, the conversation could be focused more on the parents to encourage attendance. Future interventions could provide training to providers to promote consistent and engaging descriptions of the community program that clearly explain next steps and the program's value for social, emotional, and physical health, and that successfully promote the program to patients and their families. Provider language could include “I trust this program and it would be good for your and your family's health,” or similar program-endorsing language.

Second, our findings suggest that increased provider engagement can be facilitated by improved communication within the partnership. Sharing stories, seeing photos or videos, and getting feedback from program operators can “close the loop” on communications and allow providers to accurately portray the program to potential participants. Our findings support previous conclusions that communication between stakeholders is an essential component of establishing and maintaining partnerships.^{18,20,22–24} In the ARCHES model, the connector position was designed to facilitate this interstakeholder contact and was perceived as an essential component of the model by all involved site stakeholders. Nonclinical stakeholders felt that communication between clinics and the connector was frequent and efficient, whereas connectors often cited a perceived lack of engagement from clinic partners. Clinical referral coordinators in some clinics assisted in reinforcing communication between clinics and the connector. Additionally, clinical stakeholder champions, providers who lobbied for the program within clinics, were also important for encouraging referrals and communicating progress with other stakeholders. The connector, clinical referral coordinator, and clinical stakeholder champions all worked to improve communication from the clinic.

Third, we found that programs with a finite beginning and ending date for participants, creating a “cohort,” had greater odds of attendance compared with programs with open enrollment. Cohort programs may motivate participation through group accountability and relationships with peers. For instance, starting the program as a group may motivate program attendance because individuals may form social relationships and feel a responsibility to participate as part of the group. Additionally, the concept of scarcity, the expectation that the opportunity to start participation will disappear, may explain the improved engagement with cohort programs.³³ Open enrollment programs allow referred participants to potentially postpone attendance since they can start at any time and it may be that finding the motivation to attend that first session could be a large hurdle for many families. While our findings suggest that a cohort-style program may promote better program attendance, we recognize that this may not be possible in some contexts. In this case, programs could host monthly orientations so that participants start the program with their orientation group or could use team-based language when describing cohorts to increase the sense of group responsibility.

Although we did not assess the value of including electronic health records (EHR) or online care networks into the ARCHES communication chain, interviewed stakeholders mentioned the value of a simple, clear referral process. Research assessing referral process from clinics to community programs is often concerned with incorporating EHR or online care networks into their method for frequent referrals.^18,34 An EHR-based referral method has the potential to streamline communication between clinical and community partners and increase the efficiency of transfer of the patient referral form to the connector or community group. An EHR-based referral system could also prompt or remind providers to communicate effectively with patients during the process. In future implementations of clinic–community referral-based programs, incorporation of EHR could be explored as a potential way to communicate about referred patients and facilitate communication.

Our study has several strengths, including referral data from a diverse set of clinic–community partnerships designed to treat childhood obesity and the use of a mixed methods approach, yet there were several limitations. First, it is possible that unmeasured community-level variables may have contributed to referral and attendance at each site (e.g., program location, community resources, or prior affiliation with community partners). Future studies should assess community-level factors such as access to transportation, language differences between program leaders and potential attendees and rurality or distance between participant residence and program location, and should focus on incorporating cultural awareness when engaging participants. Second, we were unable to assess patient perception of the referral process and barriers to program attendance. Factors such as time, family resources, competing activities, or barriers associated with racial minority or socioeconomic status, could have influenced program attendance.³⁵ Future studies could assess whether this implementation model could alleviate barriers in obesity care based on racial minority or socioeconomic status. Finally, although we had stakeholder representatives from each site, we were unable to interview all stakeholders, specifically clinical providers from each of the referring clinics, which could have provided more details about the referral process at each clinic.

Conclusion

This research demonstrates the potential value of clinic–community partnerships and provides direction for future referral-based partnerships to improve patient care and access to childhood obesity treatment outside of clinic. Overall, strategies that prioritize effective conversation with parents, stakeholder communication, a simple and clear referral process, and potentially cohort-based enrollment programs, will be the most productive methods of converting referral to program attendance. Clinic–community partnerships should prioritize communication with patients in clinic and between partners using strategies such as including a connector, clinical champion, and a clinical referral coordinator. These models of clinic–community partnerships have the capacity to alleviate strain on clinical practices while providing evidence-based care to patients with obesity.

Footnotes

Authors' Contributions

E.A. made substantial contributions to analysis and interpretation of the work as well as the conception and design of the project. She drafted the article, revised it critically, and gave final approval. A.S. made substantial contributions to the design and conception of work as well as the acquisition, analysis, and interpretation of data. She revised the article critically and gave final approval. K.G. made substantial contributions to the conception and design of work, revised the article critically, and gave final approval. J.J. made substantial contributions to the conception or design of work, revised the article critically, and gave final approval. C.W. made substantial contributions to the conception and design of work as well as the analysis and interpretation of data. She revised the article critically, contributed to drafting the work, and gave final approval. C.L. made substantial contributions to interpretation and analysis of data for the work, revised the article critically, and gave final approval. R.F. made substantial contributions to the analysis and interpretation of the work, revised the article critically, and gave final approval. J.H. made substantial contributions to the conception and design of work and the analysis of data. She also revised critically and gave final approval. S.A. made substantial contributions to the conception and design of work and the acquisition, analysis, and interpretation of the data. She drafted and revised the article critically and gave final approval. C.N. made substantial contributions to the conception and design of work as well as the acquisition, analysis, and interpretation of the data. He drafted and revised the article critically and gave final approval.

Acknowledgments

The authors would like to acknowledge the site coordinators, partners, and participants for their role in making this study possible.

Funding Information

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by The Duke Endowment.

Author Disclosure Statement

No competing financial interests exist.

References

Skinner

, Ravanbakht

, Skelton

, et al. Prevalence of obesity and severe obesity in US children, 1999–2016. Pediatrics, 2018; 141:e20173459.

Caprio

, Daniels

, Drewnowski

, et al. Influence of race, ethnicity, and culture on childhood obesity: Implications for prevention and treatment: A consensus statement of Shaping America's Health and the Obesity Society. Diabetes Care, 2008; 31:2211–2221.

US Preventive Services Task

Force

, Grossman

, Bibbins-Domingo

, et al. Screening for obesity in children and adolescents: US Preventive Services Task Force Recommendation Statement. JAMA, 2017; 317:2417–2426.

Barlow

, Expert

Committee

. Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: Summary report. Pediatrics, 2007; 120 Suppl 4:S164–S192.

Spivack

, Swietlik

, Alessandrini

, Faith

. Primary care providers' knowledge, practices, and perceived barriers to the treatment and prevention of childhood obesity. Obesity (Silver Spring), 2010; 18:1341–1347.

Perrin

, Flower

, Garrett

, Ammerman

. Preventing and treating obesity: Pediatricians' self-efficacy, barriers, resources, and advocacy. Ambul Pediatr, 2005; 5:150–156.

Morais

, Kelly

, Bost

, Vaidya

. Characteristics of correctly identified pediatric obesity and overweight status and management in an Academic General Pediatric Clinic. Clin Pediatr (Phila), 2018; 57:1168–1175.

Sequist

, Taveras

. Clinic–community linkages for high-value care. N Engl J Med, 2014; 371:2148–2150.

Brandling

, House

. Social prescribing in general practice: Adding meaning to medicine. Br J Gen Pract, 2009; 59:454–456.

10.

Berge

, Everts

. Family-based interventions targeting childhood obesity: A meta-analysis. Child Obes, 2011; 7:110–121.

11.

Sacher

, Kolotourou

, Poupakis

, et al. Addressing childhood obesity in low-income, ethnically diverse families: Outcomes and peer effects of MEND 7–13 when delivered at scale in US communities. Int J Obes, 2019; 43:91–102.

12.

Law

, Cole

, Cummins

, et al. A pragmatic evaluation of a family-based intervention for childhood overweight and obesity. Public Health Res, 2014; 2:1–184.

13.

Hoffman

, Frerichs

, Story

, et al. An integrated clinic-community partnership for child obesity treatment: A randomized pilot trial. Pediatrics, 2018; 141:e20171444.

14.

Porterfield

, Hinnant

, Kane

, et al. Linkages between clinical practices and community organizations for prevention: A literature review and environmental scan. Am J Public Health, 2012; 102(S3):S375–S382.

15.

Doucette

, Kent

, Seegmiller

, et al. Feasibility of a coordinated human papillomavirus (HPV) vaccination program between a Medical Clinic and a Community Pharmacy. Pharmacy (Basel), 2019; 7:91.

16.

Marpadga

, Fernandez

, Leung

, et al. Challenges and successes with food resource referrals for food-insecure patients with diabetes. Perm J, 2019; 23:18-097.

17.

Reid

, Fallon

, Sonne

, et al. Smoking cessation treatment in community-based substance abuse rehabilitation programs. J Subst Abuse Treat, 2008; 35:68–77.

18.

Etz

, Cohen

, Woolf

, et al. Bridging primary care practices and communities to promote healthy behaviors. Am J Prev Med, 2008; 35(5 Suppl):S390–S397.

19.

Fishleder

, Petrescu-Prahova

, Harris

, et al. Bridging the gap after physical therapy: Clinical–community linkages with older adult physical activity programs. Innov Aging, 2018; 2:igy006.

20.

Brownson

, O'Toole

, Shetty

, et al. Clinic-community partnerships: A foundation for providing community supports for diabetes care and self-management. Diabetes Spectr, 2007; 20:209–214.

21.

Ackermann

. Description of an integrated framework for building linkages among primary care clinics and community organizations for the prevention of type 2 diabetes: Emerging themes from the CC-Link study. Chronic Illn, 2010; 6:89–100.

22.

Pescheny

, Pappas

, Randhawa

. Facilitators and barriers of implementing and delivering social prescribing services: A systematic review. BMC Health Serv Res, 2018; 18:86.

23.

Kemner

, Donaldson

, Swank

, Brennan

. Partnership and Community Capacity Characteristics in 49 Sites Implementing Healthy Eating and Active Living Interventions. Available at https://oce-ovid-com.proxy.lib.duke.edu/article/00124784-201505001-00005/HTML Last accessed April 15, 2020.

24.

Schwartz

, Vitolins

, Case

, et al. The YMCA healthy, fit, and strong program: A community-based, family-centered, low-cost obesity prevention/treatment pilot study. Child Obes, 2012; 8:577–582.

25.

Curran

, Bauer

, Mittman

, et al. Effectiveness-implementation hybrid designs. Med Care, 2012; 50:217–226.

26.

NC Commerce: County Distress Rankings (Tiers). Available at https://www.nccommerce.com/grants-incentives/county-distress-rankings-tiers Last accessed June 19, 2020.

27.

About Us. NC Rural Center. Available at https://www.ncruralcenter.org/about-us/ Last accessed June 19, 2020.

28.

Hsieh

H-F

, Shannon

. Three approaches to qualitative content analysis. Qual Health Res, 2005; 15:1277–1288.

29.

Kelleher

, Davoren

, Harrington

, et al. Barriers and facilitators to initial and continued attendance at community-based lifestyle programmes among families of overweight and obese children: A systematic review. Obes Rev, 2017; 18:183–194.

30.

Ryan

. The peer group as a context for the development of young adolescent motivation and achievement. Child Dev, 2001; 72:1135–1150.

31.

Patrick

, Ryan

, Alfeld-Liro

, et al. Adolescents' Commitment to developing talent: The role of peers in continuing motivation for sports and the arts. J Youth Adolesc, 1999; 28:741–763.

32.

Brown

, Dovico

, Garner-Edwards

, et al. Predictors of engagement in a pediatric weight management clinic after referral. Child Obes, 2020; 16:238–243.

33.

Asch

, Rosin

. Engineering social incentives for health. N Engl J Med, 2016; 375:2511–2513.

34.

Chambers

, Wylie-Rosett

, Blank

, et al. Increasing referrals to a YMCA-based diabetes prevention program: Effects of electronic referral system modification and provider education in federally qualified health centers. Prev Chronic Dis, 2015; 12:E189.

35.

Cui

, Seburg

, Sherwood

, et al. Recruitment and retention in obesity prevention and treatment trials targeting minority or low-income children: A review of the clinical trials registration database. Trials, 2015; 16:564.