Impact of a Community-Based Weight Management Program in a North Carolina Health Care System

Obesity affects 42% of US adults, with persons of low socioeconomic status disproportionately being burdened.^{1, 2} Unfortunately, most treatment still occurs in centralized multidisciplinary centers or commercial clinics with high out-of-pocket costs. ³

As most health care interactions occur in primary care practices, providing weight management services in the same locations enhances patient access to obesity care and is expected to improve health outcomes.^{4, 5} However, the time constraints in primary care practices and lack of obesity-specific training are major barriers to obesity treatment in communities.^6-10 Moreover, clinical guidelines underscore that intense behavioral therapy alone is insufficient to achieve sustained weight loss for most patients, necessitating the consideration of antiobesity medication (AOM) when bariatric surgery is declined. ^11-15

To increase access, we piloted a weight management program (WMP) utilizing obesity medicine-trained primary care providers (PCPs) nested in primary care practices. We sought to evaluate the effectiveness of such a nested WMP for clinically meaningful weight loss in community-based practices as a scalable model for delivering comprehensive obesity care.

The WMP was piloted in three primary care clinics. Three PCPs certified in weight management by the American Board of Obesity Medicine were assigned visits dedicated to obesity care, charging primary care copays. Staffing included registered dieticians (RD) and licensed clinical social workers (LCSW) as part of a primary care medical home. All clinical staff were existing employees. We informed primary care physicians at the three pilot sites of our study. Physicians then identified patients and referred them to our weight management pilot program.

The physician and RD evaluated all new patients. Evaluations included weight history, eating behaviors, physical activity, sleep, stress, physical exam, and metabolic data. Patients with uncontrolled mental health diagnoses were referred to LCSW. When AOMs were initiated, preference was given to FDA-approved agents. Generic medication combinations were prescribed to reduce cost when possible.

The program offered targeted lifestyle counseling (TLC; 6-8 sessions per year) plus AOMs. Physicians individualized interventions by applying the American Association of Clinical Endocrinology’s guidelines for obesity management.¹⁶ Providers offered bariatric surgery to patients meeting criteria.¹⁷ Patients received nutrition instructions emphasizing reducing ultraprocessed foods and following a Mediterranean-style diet rather than restricting calories due to extensive literature supporting the diet’s positive impact on cardiometabolic risk factors and overall health.^18-20 This approach has been shown to reduce total caloric intake, reduced the complexity of dietary counseling, and reflected general patient and professional preference.¹⁸ Providers used motivational interviewing techniques to encourage patient-centered SMART goals for dietary choices, physical activity, stress management, and sleep improvement.^{21, 22}

The obesity-certified PCPs were involved in every follow-up visit. Visits were 20 minutes in duration. The visits were primarily structured to address lifestyle changes. Goals were reviewed, positive behaviors were reinforced, and strategies were discussed to work on unmet goals. We also titrated antiobesity medications. Since weight loss has a profound impact on so many comorbidities, we regularly addressed significant conditions, such as hypertension, nonalcoholic fatty liver disease, obstructive sleep apnea, and type 2 diabetes, and titrated down or removed cardiometabolic medications. We offered in-person and virtual visits.

The program maintained financial stability for the clinics. All three clinicians were able to meet their goals for standard work relative value units (wRVU). The physicians used time-based, outpatient Current Procedural Terminology (CPT) codes for evaluation and management (E/M) services, the RDs used medical nutrition therapy codes, and the LCSW used behavioral health codes.

Patients enrolled in the WMP were tracked in an obesity registry. Health and demographic data were extracted and analyzed for patients 18 or older with body mass indexes (BMIs) above 27 kg/m² and adequate program exposure, defined as at least four visits.²³ BMI inclusion was based on current practice guidelines.¹⁶ The comparison group was patients with similar age and BMI criteria (Table 1) but only one assessment visit in the WMP and no follow-up visits or interventions. The study excluded patients who had had bariatric surgery within 6 months of the study period (Figure 1).

The primary outcomes were the percent weight loss postintervention and the proportion of patients who achieved a clinically significant TBWL of 5% or greater. We analyzed the data using Stata/SE software version 16.1.²⁴ We performed paired t tests to assess differences in continuous variables. We established statistical significance at ɑ=0.05. We compared the latest weight and BMI to the baseline values at the program's start to determine changes in outcomes. The University of North Carolina at Chapel Hill’s Institutional Review Board approved the study.

From March 2019 through October 2020, the WMP’s three pilot clinics served 550 new patients over 1,925 visits. The mean patient age was 49.0 years (SD 13.3), with a starting weight and BMI of 110.2 kg (SD 23.5) and 40.0 kg/m² (SD 7.7), respectively (Table 1).

For inclusion in the analysis, 209 patients had adequate program exposure (≥4 visits) and 66 patients were in the comparison group (assessment without follow-up). The mean weight change was -5.7% (SD 5.8%), average attendance was 6.3 visits (SD 2.6), and mean follow-up was 241 days (SD 127). Fifty three percent (n=111) achieved at least 5% TBWL with 6.8 mean visits. Twenty percent of patients achieved 10% TBWL or more. The comparison group had a mean weight gain of 1.5% over an average of 407 days (SD 158; Table 2). The weight loss appeared to co-relate with the number of visits completed, with patients achieving TBWL of at least 5% between the fourth and fifth visit (Figure 2). AOM utilization is shown in Figure 3.

The purpose of our WMP was to explore the impact of obesity medicine-trained PCPs on durable weight loss in our communities. During the WMP pilot, our patients achieved a clinically meaningful weight loss of approximately 5% after attending at least four visits. Half of the patients (53%) achieved 5% weight loss compared to the marginal weight gain in individuals who only received the initial assessment. These results make our community-based WMP on par with programs in more controlled, resource-intense settings,^{25, 26} and did not require high-intensity counseling (>2 sessions/month) or referrals to specialty-based WMPs, which the current literature suggests are necessary for achieving such weight loss in primary care.⁶ We believe that our model’s leverage of obesity-trained PCPs in primary care practices can increase access to weight management services and meaningful weight loss among socioeconomically diverse patient populations.

There is growing evidence that moderate weight loss can significantly improve outcomes of many obesity-related comorbidities.^{27, 28} Accordingly, we believe that obesity treatment strategies can and should include community-based solutions to improve access to care and improve obesity-related outcomes. Increased access through primary care-based services has the potential to address specialty-based WMPs’ insufficient capacity within our health care system to address the rapidly increasing prevalence of obesity.

This study's strength is its achievement of significant weight loss in real-world community settings serving patients from diverse socioeconomic and underrepresented racial groups. Limitations of this study include its initiation as a quality improvement project with a small number of patients; the lack of a control group; its underrepresentation of males and Latino ethnicity; limited AOM insurance coverage; and the COVID-19 pandemic beginning midway through the study.

We do not think the weight management certification of the physicians is a critical component of the intervention. We have found that having easy access to obesity medicine-certified mentors is the key to providing effective, evidence-based obesity treatment in low-resource community settings. For example, at our institution, we used an information dissemination technique similar to the ECHO (Extension of Community Healthcare Outcomes) Model.²⁹ We have monthly, virtual obesity medicine meetings with community clinicians discussing the latest obesity literature and difficult clinical cases. These meetings also build relationships such that PCPs can have access to experienced obesity clinicians. Also, our institution is sponsoring a local obesity conference for primary care professionals to teach effective weight management tools to be used by them in the community setting.

In conclusion, this approach represents a potential scalable model for expanding access to comprehensive, effective obesity treatment for the general population. The next steps include comparing community-based results to matched controls and to specialty-based obesity clinic outcomes at our institution.

We thank Darren DeWalt, MD, MPH; John Batsis, MD; and Young E. Whang, MD, PhD, for reviewing and editing the manuscript. We also want to thank Sachin Gupta, MD; Shawn Armstrong; Ellen Lawrence, RD; and Candice Hunt for help in implementing this program.

Program evaluation support was provided by a training grant from the UNC Institute for Healthcare Quality Improvement.

This project was presented orally and as a poster at the Obesity Medicine Association’s Overcoming Obesity Fall 2020 Conference (Virtual) on October 7, 2020.

Dr Machineni reports personal fees and others from Novo Nordisk and personal fees from Rhythm Pharmaceuticals outside the submitted work. Dr Davis reports personal fees (speaking fees) from Novo Nordisk outside of the submitted work.