Applying a Medical Wellness Group Visit Model to a Community Setting Yielding Weight Loss and Improved Laboratory Results

Abstract

Purpose:

To determine the feasibility of applying a medical wellness group (WG) model to a community setting to improve cardiometabolic health.

Design:

This quasi-experiment was designed to compare individuals participating in the WG to participants in the control group who received general lectures on nutrition, physical activity, and sleep.

Setting:

A suburb north of Boston, Massachusetts.

Participants:

Forty-five adults were in the WG and 10 in the control group.

Intervention:

Fourteen weekly 90-minute sessions, led by a physician and dietitian, focusing on nutrition, physical activity, and sleep, compared to controls receiving two 30-minute general wellness lectures provided within 3 months.

Measures:

Pre- and postweight, waist circumference, hemoglobin A_1C (HbA_1c), and serum lipids; a survey measuring beliefs, attitudes, and intentions related to behavioral change.

Analysis:

T tests examined the mean change in biometric measurements. The Wilcoxon test was used to compare the ordinal questions in baseline and final survey results. The Mann-Whitney test was used to compare final survey results between groups.

Results:

The WG demonstrated desirable difference-in-difference between groups in weight (P < .001), waist circumference (P < .001), and total cholesterol (P = .03) compared to the control group. Mean change of HbA_1c and triglycerides was not different between groups. Survey results showed that attitudes, perceived behavioral control, and feeling supported about wellness behaviors significantly improved from baseline to final visit in the WG (P = .002; P = .019, P = .006, respectively), but not among controls.

Conclusion:

Wellness groups are feasible and provide high levels of support and accountability that empower people to make behavioral changes to improve health.

Keywords

interventions physical activity weight management nutrition interventions community

Purpose

Despite widely accessible resources and billions of dollars spent on addressing obesity, the high rates of obesity continue to persist.¹ The shortage of primary care providers amplifies the need for effective, low-cost solutions to this health crisis. Medical group models have been identified as one solution.² These models have demonstrated success with providing more efficient care and work satisfaction among physicians³ and a supportive network of peers that allows for improved self-efficacy and accountability for patients.⁴ In a prior study,⁵ we found that medical group participants at 1 year had lost more weight than those in traditional care (−13.21 vs +1.94 pounds, P < .05) and sustained healthy behaviors.

Extending group visits beyond the medical setting to the community may offer convenient locations to engage more diverse people.⁶ Community-based wellness programs have been effective with diabetes prevention⁷ and in reducing smoking and increasing exercise.⁸ This study aimed to determine the feasibility of applying a medical group model to a community setting to improve cardiometabolic health. Health outcomes identified for this study were improvements in proxies for body composition (weight or waist circumference) and lipids and hemoglobin A_1c (HbA_1c). We hypothesized that leveraging support and accountability in a community would help meet a growing need for a more effective and efficient method of combating chronic diseases.

Methods

Design

This quasi-experiment was designed to implement the medical wellness group (WG) program in a community setting. We compared individuals participating in the community WG model to a control group that received general lectures on nutrition, physical activity, and sleep. We also included psychosocial measures as outcomes to better understand drivers of change and chose the theory of planned behavior (TPB) because it accounts for constructs that are associated with cues to action, which would be applicable to our community wellness intervention.

Sample

The main source of recruitment was city employees. Recruitment also included residents of this city, a suburb north of Boston, Massachusetts. Participants volunteered to participate in a 14-week WG coled by a family physician and dietitian. The control group consisted of people who worked or resided in the same city as the intervention group and had an interest in participating in the study but were not able to participate in the current intervention due to scheduling conflicts. People in the intervention or control group were eligible to participate if they were at least 18 years old and interested in wellness. All participants provided written informed consent.

Intervention

The WG model was developed by a dietitian and family physician who have led medical office-based group visits since September 2009. The success of this program has been previously described.⁵ For this study, the same physician and dietitian led the groups. For the intervention, 90-minute sessions were conducted weekly over 14 weeks from January to May 2018 and focused on the following wellness concepts: (1) nutrition (healthy carbohydrates, fats, and proteins, glycemic index, reading food labels, and food quality), (2) physical activity (goals for step counting), and (3) encouraging sufficient amount of sleep. During the intervention sessions, multiple aspects of these wellness dimensions were discussed and participants were given a goal for each week. Participants also communicated with each other about their progress with achieving weekly goals via an e-mail list. The control group received two 30-minute wellness lectures provided in February and May 2018. Control group lectures focused on the same general wellness concepts as the intervention (nutrition, physical activity, and sleep) but served as a superficial introduction to these concepts and lacked the depth of content, examples, and goal setting compared to the content received by the intervention group. All study visits took place at a local middle school in Massachusetts because this school was central in location to participants.

Outcome Measures

Anthropometric and laboratory measurements were conducted during the first and final visit for both groups. Body weight was measured using a Detecto digital scale to the nearest 0.1 pound. Height and waist circumference were measured using a Singer Vinyl tape measure to the nearest 0.1 in; 3.5 mL of blood was drawn from each participant during the baseline and final visit to measure HbA_1c, cholesterol levels, and triglycerides.

We adapted a validated survey to assess key constructs of the TPB⁹ to measure beliefs, attitudes, subjective norms, perceived behavioral control, and intentions related to behavioral change associated with dietary intake, physical activity, and sleep. According to the TPB, attitude, subjective norms, and perceived behavioral control impact one’s intention to enact a particular behavior, and behavioral intention predicts whether or not that person eventually enacts the behavior.¹⁰ Respondents answered 15 questions on a 6-point Likert scale about eating and exercise habits that fell into 5 categories: attitude (2 questions), subjective norms (6 questions, such as, “Most of my friends engage in healthy eating…”), perceived behavioral control (4 questions, such as, “I am confident that I can eat healthily…”), behavioral intention (1 question), and social support (2 questions). The response options ranged from 1: good to 6: bad, 1: pleasant to 6: unpleasant, or 1: agree to 6: disagree, in which lower scores reflected positivity toward eating healthily and higher scores were more negative. The Cronbach α for the TPB survey for our study was .86. The Cronbach α for the TPB subscales are as follows: attitude (.38), subjective norms (.77), perceived behavioral control (.88), and social support (.64). The study was approved by the institutional review board at Tufts University (approval #1711003).

Analysis

Differences in the mean change in biometric measurements between the groups were determined using t tests. The sum of the responses was scored for each subscale of the TPB. The Wilcoxon test was used to compare the baseline and final TPB survey results. The Mann-Whitney test was used to compare the final TPB survey results between the groups. Results with P < .05 were considered statistically significant.

Results

The majority of participants in the study were employees of the city (96%), female (89%), English speaking (100%), and middle aged (mean: 52.7 years). Approximately two-thirds (62%) of participants were obese and most had health insurance through the city. Forty-five adults were enrolled in the WG and 10 in the control group. Thirty-nine participants (87%) in the WG and 7 of the control group participants (70%) completed the final visit.

There were no appreciable differences between the intervention and control group with regard to age, gender, or body composition measurements at baseline. Total cholesterol was the only statistically significant difference between the groups at baseline, with intervention participants having a higher total cholesterol (mean 207.1 ± 41.6 standard deviation [SD]) compared to control participants (mean 176.0 ± 35.9 SD; P = .03).

Participants had positive statistically significant outcomes compared to controls with regard to mean change in weight (P < 0.001), waist circumference (P < .001), and total cholesterol (P = .03; see Table 1). No other changes in biometric measurements were statistically significant.

Table 1.

Mean Change in Anthropometric and Laboratory Data by Wellness and Control Groups Mean (Standard Deviation).

	All (N = 46)	Wellness Group (n = 39)	Control (n = 7)	P Value
Weight (pounds)	−10.2 (10.2)	−12.7 (7.3)	4.1 (12.7)	<.001
BMI (kg/m²)	−1.7 (1.7)	−2.1 (1.3)	0.7 (2.0)	<.001
Waist circumference (inches)	−3.1 (2.2)	−3.6 (1.9)	−0.3 (1.4)	<.001
HbA_1c (%)	−0.04 (0.2)	−0.04 (0.2)	−0.07 (0.2)	.74
Total cholesterol (mg/dL)	−9.0 (31.6)	−13.4 (25.32)	14.3 (50.7)	.03
HDL cholesterol (mg/dL)	−3.2 (10.3)	−4.8 (10.0)	5.0 (8.4)	.02
LDL cholesterol (mg/dL)	−1.8 (17.6)	−3.1 (16.6)	4.6 (22.2)	.30
Triglycerides (mg/dL)	−40.5 (69.0)	−45.1 (68.9)	−16.0 (69.5)	.31

Abbreviations: BMI, body mass index; HbA_1c, hemoglobin A_1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

The survey results for constructs related to the TPB are shown in Table 2. Attitude (P = .002), perceived behavioral control (P = .019), and social support (P = .006) significantly improved in the intervention group from baseline to final visit but did not significantly differ for the control group. Subjective norms were statistically different for controls between baseline and final visit (P = .021); however, the difference showed a decline in approval of healthier eating patterns and exercise. There were no significant differences between the WG and control group at final visit, except for subjective norms and intention.

Table 2.

Constructs of the TPB at Baseline and Final Visit in the Wellness and Control Groups (Mean ± Standard Deviation).

TPB Construct/Group	Baseline	Final	P Value	P Value Between Groups at Final Visit
Attitude
Wellness	3.6 (1.5)	2.6 (1.0)	.002	.098
Controls	3.1 (1.4)	3.6 (1.5)	.140	.098
Subjective norms
Wellness	12.5 (4.6)	11.3 (3.4)	.110	.018
Controls	11.8 (3.3)	14.9 (3.5)	.021
Perceived behavioral control
Wellness	6.4 (2.8)	5.4 (2.1)	.019	.062
Controls	5.6 (3.2)	7.7 (3.7)	.121	.062
Intention
Wellness	1.3 (0.7)	1.2 (0.5)	.262	.022
Controls	1.2 (0.4)	1.6 (0.5)	.083
Social support
Wellness	4.2 (2.7)	3.1 (2.0)	.006	.057
Controls	4.2 (2.7)	4.4 (2.1)	.549	.057

Abbreviation: TBP, theory of planned behavior.

Discussion

This study aimed to implement a community WG intervention led by a physician and dietitian to improve lifestyle factors that impact weight, HbA_1c, lipids, and perceptions and behaviors about these lifestyle recommendations. The results of this study demonstrated that participants in the WG had significant weight loss and decreased waist circumference and total cholesterol levels compared to controls. The results of the TPB surveys showed that WG participants had improved attitudes, subjective norms, perceived behavioral control, and social support related to eating healthily at the end of the study compared to the beginning. Together, these results suggest that community wellness programs can provide supportive environments that help to promote and sustain healthy behaviors that lead to weight loss.

The anthropometric changes found in the WG were clinically significant. This demonstrates the feasibility of using a community setting to deliver a medical intervention for wellness programming. Bringing the study to the community allowed for more participants to be seen in nearby settings to make it easier for participation. The county where this study took place has one primary care physician per 790 patients,¹¹ further demonstrating the need to improve the ability to reach more people.

The impact of the wellness intervention was further demonstrated by the significant changes in subjective norms and intention for the intervention group compared to controls. Social support was close to statistical significance and the differences may have been attenuated because the control group may have felt some social support from the general wellness sessions. Also, many participants (both intervention and control groups) worked in the local school system, so there could have been some contamination with the intervention participants.

Prior research suggests that delivering in-person support can help patients in a clinical setting lose weight.¹² Additionally, research has shown that community wellness programs can improve dietary behaviors¹³ and physical activity.⁶ However, rarely are community interventions led by experts, like a physician and dietitian, and few studies that take place in the community have been able to use the same rigorous biometric measurements, like lipid and HbA_1c levels, that are done in clinical settings. By recruiting patients from the community and using resources available from the medical center, this study was able to integrate the accessibility of community wellness programs with the clinical relevance of primary care office interventions. Future studies are needed to further elucidate in a larger, more racially diverse group whether the high level of expertise and support provided by professionals could be maintained.

While this feasibility study provided necessary evidence for effect sizes, caution should be taken that our study included predominantly white, non-Hispanic females. The results of this study are promising and will require additional resources to enroll the numbers needed to fully measure the impact of this community WG design. In addition, we acknowledge that attrition took place in both groups (13% in the intervention group and 30% in the control group); however, based upon the characteristics that we measured, we did not see any differences between the groups of who did not complete the study. Participants in both groups mentioned scheduling difficulties as reasons for attrition. Most of the subscales for the TPB had acceptable to good internal consistency, except for the attitude scale having an insufficient Cronbach α (.38). This low internal consistency for the attitude scale could be due to only having 2 items for this scale, so future studies may need to assess including additional items.

So What?

What is already known on this topic?

Programs that focus on behavior change to promote wellness have had varying success. There are many programs that address obesity, but the prevalence of obesity persists.

What does this article add?

Our study provides evidence that taking a medical group model into the community setting can provide support that empowers people to make behavioral changes to improve cardiometabolic health.

What are the implications for health promotion practice or research?

Our findings suggest that bringing a medical wellness group model into the community is feasible and may be a way to serve more people closer to where they work or reside. Bringing together people in a community may help to motivate people to make healthy lifestyle changes. Grassroots efforts to shift the culture of communities are needed to make progress. The medical wellness group model should also be evaluated to determine how this could be scaled up in larger cities and communities with fewer resources, recognizing that relationships between medical center staff and communities maintain mutually beneficial connections.

Footnotes

Acknowledgments

The authors would like to acknowledge the efforts of Theresa Fukuda and Anna Grossman for reviewing drafts of portions of the manuscript.

Authors’ Note

K.R.D. made a substantial contribution to the design of the work, analysis, and interpretation of the data and drafted and revised the manuscript. L.F. made a substantial contribution to the acquisition and interpretation of the data and drafted and revised the manuscript. K.H. made a substantial contribution to the concept of the work and acquisition of the data. W.A. made a substantial contribution to the concept and design of the work, acquisition and interpretation of the data, and revised the manuscript. All authors approved the version to be published.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding was provided by the Cummings Foundation and the Division of Research in the Department of Family Medicine at Tufts University School of Medicine.

ORCID iD

Kimberly R. Dong, DrPH, MS, RDN

References

Flegal

Carroll

Kit

Ogden

. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. JAMA. 2012;307(5):491–497.

Parikh

Rajendran

D’Amico

Luo

Gardiner

. Characteristics and components of medical group visits for chronic health conditions: a systematic scoping review. J Altern Complement Med. 2019;25(7):683–698.

Page

Reid

Andrews

Steiner

. Evaluation of prenatal and pediatric group visits in a residency training program. Fam Med. 2013;45(5):349–353.

Gallagher

Kirkness

Armari

Davidson

. Participants’ perspectives of a multi-component, group-based weight loss programme supplement for cardiac rehabilitation: a qualitative study. Int J Nurs Pract. 2012;18(1):28–35.

Axten

Hawkins

Tybor

Bernoff

Altman

. Impact of a novel wellness group visit model on obesity and behavior change. J Am Board Fam Med. 2017;30(6):715–723.

King

Whitt-Glover

Marquez

, et al. Physical activity promotion: highlights from the 2018 physical activity guidelines advisory committee systematic review. Med Sci Sports Exerc. 2019;51(6):1340–1353.

Mayer

Vangeepuram

Fei

, et al. Outcomes of a weight loss intervention to prevent diabetes among low-income residents of East Harlem, New York. Health Educ Behav. 2019;46(6):1073–1082.

Mattke

Liu

Caloyeras

, et al. Workplace wellness programs study: final report. Rand Health Q. 2013;3(2):7.

Ajzen

. The theory of planned behavior. Organ Behav Hum Decis. 1991;50(2):179–211.

10.

Fishbein

Ajzen

. Predicting and Changing Behavior: The Reasoned Action Approach. Milton Park, Didcot: Taylor & Francis; 2011.

11.

University of Wisconsin Health Institute and the Robert Wood Johnson Foundation and Burness. County Health Rankings & Roadmaps. 2018. https://www.countyhealthrankings.org/credits.

12.

Appel

Clark

Yeh

H-C

, et al. Comparative effectiveness of weight-loss interventions in clinical practice. N Eng J Med. 2011;365(21):1959–1968.

13.

Zhu

Shive

Zhang

Senter

Topete

, et al. The evaluation of IDEAL-REACH program to improve nutrition among Asian American community members in the Philadelphia metropolitan area. Int J Environ Res Public Health. 2019;16(17):3054.