Lifestyle Change Interventions: Effectiveness of REWIND, a Virtual, Community‐Based Weight Management Program

ABSTRACT

Importance

Despite the current enthusiasm for anti‐obesity medications, there is a critical need for effective lifestyle interventions that can be broadly implemented.

Objective

To assess the effectiveness of REWIND, an obesity and diabetes treatment program.

Design

Prospective cohort study.

Setting

Virtual, community‐based weight management program.

Participants

Two‐hundred and nineteen participants, BMI ≥ 30 kg/m², with and without type 2 diabetes (T2D).

Intervention

Three phases: “induction” with meal replacement diet to promote ≥ 15% weight loss; “transition” to everyday foods; and “maintenance” to prevent weight regain. REWIND leveraged teams including community leaders, media personalities, nationally recognized athletes and participants, to disseminate information, and foster belonging, commitment, and resilience.

Main Outcomes

The co‐primary outcomes were the change in weight at 6‐, 12‐, and 18 months. Secondary outcomes were the proportion of patients achieving ≥ 5%, 10%, 15%, and 20% weight loss at 6, 12, and 18 months, and remission rates in T2D at 6‐, 12‐, and 18‐month, respectively.

Results

Two hundred and nineteen participants (mean age 49.1 [SD 10.5] years), 37% men, 12% Black, 26% with T2D. Weight decreased from: 114.4 (20.8) kg to 96.1 (18.4) kg at 6 months, to 98.5 (18.1) kg at 12 months, and to 101.1 (20.5) kg at 18 months. Weight loss percentages at 6, 12, and 18 months were ≥ 5%: 95%, 91%, 84%; ≥ 10%: 79%, 72%, 67%; ≥ 15%: 58%, 46%, 47%; and ≥ 20%: 24%, 25%, 22%. Diabetes remission was achieved by 52% and 43% of participants with T2D at 12‐ and 18‐month, respectively.

Limitations

Single geographic region. Weight was self‐reported.

Conclusion

A virtual community‐based program for individuals with obesity and T2D facilitated substantial weight loss and T2D remission. This model has the potential for wide‐scale implementation in diverse settings.

1. Introduction

Obesity is a complex health condition and is associated with myriad adverse health outcomes including type 2 diabetes, cardiovascular disease, metabolic dysfunction‐associated steatohepatitis, cancer, and decreased health‐related quality‐of‐life [1, 2, 3]. It affects over 40% of the adult US population and costs the health system $260 billion per year [4].

Bariatric surgery results in substantial and relatively durable weight loss and improvements in many of the metabolic and vascular complications of obesity [5, 6, 7, 8, 9]. However, while generally safe, bariatric surgery is invasive, expensive, and associated with short‐ and long‐term complications [10, 11]. Although anti‐obesity medications, especially the recently approved GLP‐1 receptor agonists (GLP‐1 RAs) and dual agonists, can result in marked weight loss and improvements in many of the complications of obesity [12, 13, 14, 15, 16, 17], they are expensive and not cost‐effective as first‐line therapies for obesity [18]. As a result, coverage is highly variable among health plans [19, 20] and the medications are not covered by Medicare for obesity. When used in general practice, these medications have also proven difficult for many patients to tolerate, and medication persistence at 1 year is low [21, 22, 23]. In addition, the long‐term efficacy and safety of these drugs remain uncertain [24].

For these and other reasons, virtually all clinical practice guidelines recommend a trial of lifestyle intervention before prescribing anti‐obesity medications or referring for bariatric surgery [25]. Unfortunately, many lifestyle interventions are perfunctory and have resulted in only modest and transient weight loss [26, 27, 28]. This has led to the perception that pharmacologic or surgical interventions are inevitably required to treat obesity.

This perception persists despite evidence that highly‐structured, intensive, lifestyle modification programs such as the Diabetes Remission Clinical Trial (DiRECT) [29, 30], the Doctor Referral of Overweight People to Low Energy total diet replacement Treatment (DROPLET) Study [31], the Diabetes Intervention Accentuating Diet and Enhancing Metabolism DIADEM‐I study [32], the Optiwin trial [33] and the University of Michigan Weight Management Program (MWMP) [34, 35, 36] have achieved sustained weight loss comparable to that observed with the use of FDA‐approved GLP‐1RA‐based therapies [12, 13, 14, 15, 16, 17].

Unfortunately, when translated into usual clinical practice, the weight loss and the associated benefits of lifestyle intervention wane over time. To identify strategies to address this and other treatment challenges, the National Academy of Medicine (NAM) held a roundtable on obesity solutions that focused on the need to connect the sectors that affect treatment outcomes [37]. The sectors highlighted in the NAM report included clinicians and community entities such as schools and insurers. Another non‐traditional sector that can be harnessed to promote weight loss is the sports community.

REWIND, an integrated, virtual, community‐based lifestyle intervention program that can be widely implemented in a variety of settings, was developed based on the lessons learned from successful lifestyle interventions and the recommendations from the NAM. This report describes the feasibility and effectiveness of this program in facilitating and sustaining weight loss for up to 18 months. The present study reports the outcomes among the first 219 patients who elected to participate in REWIND. The co‐primary outcomes were self‐reported changes in weight at 6, 12, and 18 months. Secondary outcomes are the proportion of participants achieving greater than or equal to 5%, 10%, 15% and 20% change in weight at those time points. Among participants with type 2 diabetes, the change in HbA1c from baseline to 6, 12, and 18 months was also assessed. Exploratory outcomes included the proportion of participants with T2D who achieved HbA1c ≤ 7%, < 6.5%, and < 5.7% and the proportion of participants with diabetes who achieved remission of diabetes defined as an HbA1c < 6.5% off all glucose‐lowering medications for at least 3 months [38].

2. Methods

2.1. Participants

REWIND is an intensive lifestyle and behavioral weight management program modeled after the MWMP as described in detail elsewhere [34]. Since the inception of REWIND, 257 consecutive participants referred to the MWMP were offered the opportunity to participate in REWIND, and 219 (85%) chose to participate. Baseline characteristics of the participants who did and did not choose to participate in REWIND are described in Table 1 and Table S1. In general, the groups differed only in that the percentage of people with type 2 diabetes (T2D) which was higher in the REWIND group than in the clinic‐based group. All participants provided written informed consent and the study was approved by the University of Michigan Institutional Review Board and registered with ClinicalTrials.gov: NCT02043457.

TABLE 1.

Baseline characteristics of the REWIND study population.

	T2D (N = 57)	Normal (N = 76)	Pre‐DM (N = 86)	Overall (N = 219)	p‐value
Age a	53.6 (9.5)	47.0 (10.5)	49.1 (10.5)	49.3 (10.4)	0.009
Sex
Female	32 (56.1%)	51 (67.1%)	54 (62.8%)	137 (62.6%)	0.43
Male	25 (43.9%)	25 (32.9%)	32 (37.2%)	82 (37.4%)
Race
Asian	2 (3.5%)	1 (1.3%)	2 (2.3%)	5 (2.3%)	0.28
Black or African American	11 (19.3%)	3 (3.9%)	13 (15.1%)	27 (12.3%)
White	44 (77.2%)	68 (89.5%)	66 (76.7%)	178 (81.3%)
Asian Indian	0 (0%)	1 (1.3%)	1 (1.2%)	2 (0.9%)
Other	0 (0%)	2 (2.6%)	3 (3.5%)	5 (2.3%)
Unknown	0 (0%)	1 (1.3%)	1 (1.2%)	2 (0.9%)
Education
4‐year college or more	49 (86.0%)	63 (82.9%)	68 (79.1%)	180 (82.2%)	0.385
Less than college	8 (14.0%)	9 (11.8%)	12 (14.0%)	29 (13.2%)
Missing	0 (0%)	4 (5.3%)	6 (7.0%)	10 (4.6%)
Weight (kg) a	114.5 (19.1)	112.4 (19.6)	116.0 (22.8)	114.4 (20.8)	0.54
BMI (kg/m2) a	38.8 (5.4)	39.3 (4.9)	39.3 (5.5)	39.2 (5.2)	0.87
HbA1c (%) a	7.6 (1.3)	5.3 (0.2)	5.9 (0.3)	6.7 (1.4)	< 0.001

Note: Data represent mean ± standard deviation. Baseline characteristics of REWIND study population: those with T2D, normal glycemia, pre‐diabetes, and overall.

^a Mean (SD).

Participants were required to live in the state of Michigan, to have BMI ≥ 30 kg/m², to have one or more risk factors for type 2 diabetes (T2D), or be diagnosed with T2D. Exclusions were prior bariatric surgery (except for a lap band that had been removed) or planned bariatric surgery, cancer other than minor skin cancers, end‐stage kidney or liver disease, active psychiatric disease and/or substance use disorder, and current tobacco dependence. Potential participants were screened for psychiatric and substance use disorders by referring physicians, schedulers who used a decision tree algorithm to screen for these disorders, and by physicians at the first REWIND encounter. Individuals identified with these disorders at the time of referral, scheduling, or the first visit with the physician were excluded from the study. All participants had hemoglobin A1c measured at baseline and those with a prior history of T2D or newly diagnosed T2D based on A1c were asked to have their HbA1c measured by their regular care physician at 6, 12, and 18 months.

2.2. Intervention

REWIND is a multicomponent, multidisciplinary weight loss program. The program is divided into three phases: a weight loss induction (“reset”) phase during which participants consume a low‐calorie diet comprised of meal replacements to promote 15% or greater weight loss; a “transition” phase when every day food is reintroduced; and a weight maintenance (“sustain”) phase focused on promoting lasting behavioral change to prevent weight regain. The intervention is further described in Supporting Information S1: eProtocol section.

In addition to medical monitoring and counseling, the physicians discuss the science related to weight loss and long‐term weight loss maintenance, diabetes physiology, diabetes care and self‐management strategies, cardiovascular disease risk reduction, sleep, and stress management with the participants. Dietitians address daily challenges, barriers and enablers to change, strategies to optimize a balanced low‐calorie diet, and approaches to becoming more physically active. In addition to providing one‐on‐one virtual visits, the dietitians host three virtual group meetings per month. These focus on (1) the approaches and skills needed for long‐term weight control; and the facts and misconceptions about weight loss and maintenance, (2) cooking demonstrations centered around a theme, which are subsequently posted on YouTube, and (3) a “hang out” session during which participants talk about their daily lives.

REWIND technology enables continuous access to personal trackers, recipes, and modules with knowledge assessments related to nutrition, biology, and environmental factors that impact weight. Furthermore, REWIND runs SMS campaigns messaging participants with reflection questions and tips/tricks. For example, REWIND providers ask patients to tell the team about their weekly “win” or share with them a breathing video to manage stress during the holidays. They earn points in the process toward a small gift (e.g., water bottle), which further incentivizes continued participation.

Beyond the patient‐provider experience, REWIND emphasizes the importance of health and well‐being within a community. It does so by building a “team.” The team includes participants, regionally and nationally recognized community leaders, including leaders of alumnus organizations, media personalities, and celebrity athletes. These individuals are denoted as “ambassadors” who, by virtue of who they are, raise awareness about the program and inspire, motivate and challenge participants to achieve their weight and health goals. Although men and Black Americans are traditionally underrepresented in weight management and diabetes prevention and treatment programs, REWIND's partnership with individuals in the sports community including the legendary University of Michigan football player, Heisman trophy winner, former National Football League star, and ESPN football analyst, Desmond Howard, generated enthusiasm for enrolling in REWIND. Furthermore, by being part of a participants' team, ambassadors help cultivate a sense of purpose, commitment, and resilience not only to support the participant's weight and health goals but also to keep them engaged. Ambassadors periodically conduct real‐time video meetings to discuss topics around diet, discipline, and dealing with obstacles. When patients reach milestones (e.g., 10% weight loss, or 1 year in the program), they receive physical medals in the mail to symbolize their achievements. Although a fully virtual program, REWIND hosts in‐person events during which ambassadors celebrate the participants and award these medals, an experience valued by all participants. This design serves to bring an intervention (the program and its components) into a community and the community (population affected) into an intervention.

The REWIND intervention was hypothesized to be effective because it combined the essential elements of an intensive lifestyle intervention with group support. The REWIND intervention was built upon the principles espoused by the U.S. Preventive Services Task Force for effective weight loss interventions [39]. It addresses important aspects of lifestyle change including diet, physical activity, stress management, sleep, and social support. It offers structured education and guidance on healthy eating, physical activity, and behavior modification techniques; employs behavioral therapy techniques including goal setting, self‐monitoring, addressing barriers to change, and building sustainable habits; and involves frequent intensive interactions with program staff over an extended period, promoting sustained engagement and greater opportunities for habit formation.

The group provides social support and shared experiences. Individuals facing similar challenges can connect, share experiences, and offer encouragement and validation to one another. The shared experiences expose participants to different perspectives and coping strategies leading to greater insight and self‐awareness. The group environment fosters motivation and a sense of accountability supporting longer‐term adherence to healthy behaviors. The group also provides a platform for problem solving, stress management, and health communication.

Because enrollment occurred over time, data were available for analysis in all 219 participants at baseline (100%) and at 6 months. 165 participants reached the 12‐month time point and 138 of them (84%) completed the evaluation. One hundred and twenty‐two participants reached the 18‐month time point and 87 of them (71%) completed the evaluation. Participation rates were thus 100%, 84%, and 71%, at 6‐, 12, and 18‐month, respectively. Among participants with diagnosed T2D, at baseline, 6, 12, and 18 months HbA1c levels were available for 100%, 100%, 75%, and 75%, respectively.

Type 2 diabetes and pre‐diabetes were defined according to the American Diabetes Association criteria [40]. Height was taken from the electronic health record if the individual had been seen in the health system within the preceding year or was self‐reported. Weight was self‐reported. Study staff ensured that participants had home scales and participants were asked to weigh themselves at the same time each day in the same clothing and on the same scale. BMI was calculated as the weight in kilograms divided by the height in meters squared.

We assessed Health‐Related Quality‐of‐Life using the EuroQoL (EQ‐5D‐3L) and its visual analog scale (VAS) at baseline and again at 6 months. The EQ‐5D is a simple and widely used multi‐attribute utility model that assesses five dimensions: mobility, self‐care, usual activities, pain/discomfort, and anxiety/depression according to 3 levels: no problems, some problems, and extreme problems. Patient‐reported responses are weighted according to the preferences of an independent sample of judges to place the health status of the individual on a continuum between perfect health (1.0) and death (0). The accompanying VAS records the patient's self‐reported health on a vertical scale where the endpoints are labeled “best imaginable health state (1.0)” and “worst imaginable health state (0).” The point selected on the scale provides a quantitative measure of the health outcome as judged by the individual [41].

2.3. Analysis

The demographic and clinical characteristics of the study population at baseline were presented as mean ± standard deviation or count (proportion) and the outcome data relating to weight loss and HbA1c change were presented as mean and the 95% confidence interval (CI). Linear mixed effects (LME) models were used to assess changes in weight in the overall population and in A1c among participants with T2D over time.

3. Results

3.1. Patient Characteristics

Between January 2021 and the data lock in November 2023, 219 participants were enrolled. Mean age was 49.1 (10.5) years, 37% were men, 12% were Black, 82% had attended college, and 76% were married (Table 1). Mean weight was 114.4 (20.8) kg and mean BMI was 39.2 (5.2) kg/m². Most had multiple co‐morbidities with an average of four weight‐related risk factors or conditions per person, including T2D, hypertension, dyslipidemia, elevated ALT, and obstructive sleep apnea. Of the participants, 57 (26%) had T2D, 86 (39%) had pre‐diabetes, and 76 (35%) had normoglycemia. Eighty‐nine percent of the participants were taking anti‐hypertensive medications, of which 48% were taking two or more medications.

Seventy‐five percent of participants with T2D (43/57) were taking anti‐hyperglycemic medications at baseline, including 53% (30/57) who were taking two or more glucose‐lowering medications. There were 19% (11/57) who were taking insulin, and 26% (15/57) who were taking a GLP‐1 RA for glycemic control.

3.2. Effect of REWIND on Weight

All participants lost weight (Figure 1). Mean weight decreased from 114.4 (20.8) kg to 96.1 (18.4) kg at 6 months, a mean weight loss of 18.2 kg (95% CI, 17.0–19.4). Among those who had weight data at 12 months (N = 138), mean weight decreased from 115.7 (19.9) kg to 98.5 (18.1) kg. Among those (N = 87) who had weight data at 18 months, mean weight decreased from 116.9 (21.0) kg to 101.1 (20.5) kg. Based on the LME model, the mean weight loss was 15.0 kg (95% CI 13.6–16.4) and 13 kg (95% CI, 10.9–14.3) at 12 and 18 months, respectively. After adjustment for baseline weight, mean weight loss was 2.5 kg less in women than men (p = 0.042). Mean weight loss did not differ by race (p = 0.51) or diabetes status (p = 0.35).

FIGURE 1.

The bold black line represents the mean weights and standard errors of the means and the red lines represent the individual weight trajectories observed in the participants during the 18 months of study.

When assessed categorically, 95% of participants lost ≥ 5%, 79% lost ≥ 10%, 58% lost ≥ 15%, and 24% lost ≥ 20% of initial body weight at 6 months (Figure 2, left panel). Among those who had weight data at 12 months (N = 138), 91% lost ≥ 5%, 72% lost ≥ 10%, 46% lost ≥ 15%, and 25% lost ≥ 20% of initial body weight at 12 months (Figure 2, middle panel). Among those who had weight data at 18 months (N = 87), 84% lost ≥ 5%, 67% lost ≥ 10%, 47% lost ≥ 15%, and 22% lost ≥ 20% of initial body weight at 18 months (Figure 2, right panel).

FIGURE 2.

The bars represent the percentage of people who lost at least 5%, 10%, 15% or 20% of their body weight at 6 months (left panel), at 12 months (middle panel), and at 18 months (right panel).

3.3. Effects of REWIND on HbA1c in Participants With Type 2 Diabetes

Of the 57 participants with T2D, all but two had HbA1c measured at 6 months; 29 participants had HbA1c at 12 months, and 21 had HbA1c at 18 months. Despite reductions or discontinuation of glucose‐lowering medications, HbA1c decreased in participants with diabetes from a mean of 7.6 (1.3) % to 6.3 (1.1) % at 6 months, to 6.2 (1.0) % at 12 months, and to 6.3 (0.7) % at 18 months (Figure 3, left panel). The estimated mean absolute reduction in HbA1c based on the LME model was 1.3% (95% CI, 0.9–1.7) at 6 months, 1.2% (95% CI 0.78–1.58) at 12 months, and 1.0% (95% CI, 0.6–1.4) at 18 months.

FIGURE 3.

(A) The bold black line in the left panel represents the mean HbA1c and the corresponding standard error. The red lines represent the individual HbA1c trajectories observed in the participants with type 2 diabetes during the 18 months of study. The bars represent the percentage of subjects whose HbA1c was less than or equal to 7%, 6.5%, and 5.7% at 6 months (B), 12 months (C), and 18 months (D).

Based on the results of a linear mixed effects model that tested the association between percentage change in weight from baseline and change in HbA1c from baseline, 10% weight loss was associated with 10.4% decrease in HbA1c (p = 0.005).

When analyzed categorically, among the 55 participants with T2D who had HbA1c measured at 6 months, 80%, 71%, and 27% achieved HbA1c < 7.0%, < 6.5%, and < 5.7%, respectively (Figure 3, right upper panel). Among the 29 with T2D who had HbA1c measured at 12 months, 86%, 76%, and 34% achieved a HbA1c < 7.0%, < 6.5%, and < 5.7%, respectively (Figure 3, right middle panel). Among the 21 participants with T2D who had HbA1c measured at 18 months, 76%, 57%, and 19% achieved a HbA1c < 7.0%, < 6.5%, and < 5.7%, respectively (Figure 3, right lower panel).

In addition, 40% of 55 participants achieved diabetes remission at 6 months, 52% of 29 participants achieved remission at 12 months and 43% of 21 participants achieved diabetes remission at 18 months.

3.4. Assessment of Health‐Related Quality‐of‐Life

HRQoL as assessed using EQ‐5D‐derived health utility scores increased by 0.049 (95% CI, 0.029–0.069), respectively, among the 120 participants who were administered and completed the questionnaire at baseline and 6 months. A change of ≥ 0.03 in the EQ‐5D‐derived health utility score is considered to be clinically significant.

4. Discussion

REWIND, a virtual, community‐based program, resulted in an average loss of 16% of initial body weight at 6 months and 14% at 18 months. Consistent with previous studies [42, 43], weight loss was slightly less in women than men (3.0 kg) and slightly, although not significantly, less in Black compared to White participants (2.5 kg). Interestingly, weight loss did not differ among participants with type 2 diabetes, prediabetes, or normoglycemia.

The magnitude of the weight loss observed was comparable to that observed in the active intervention arms of clinical trials of GLP‐1 RA medications [12, 13, 14, 15, 16, 17] for obesity but not in the lifestyle/placebo control arms where weight loss of 2%–3% has been reported at 18 months [12, 13, 16, 17]. Indeed, most lifestyle change programs for obesity and type 2 diabetes management result in only modest weight loss [12, 13, 14, 16, 17, 26, 27, 44, 45, 46]. This is likely due to both a pervasive obesogenic environment and a failure to use comprehensive, multidisciplinary, multifaceted lifestyle change programs that offer frequent follow‐up as required for any chronic disease.

Fifteen REWIND participants with T2D were taking GLP‐1 RAs for the treatment of their diabetes at study entry, but four were no longer taking them at 18 months. Of those who continued, one reduced the dose, and the others did not change the dose they were taking throughout the period of observation. The amount of weight lost by participants taking GLP‐1 RAs did not differ from that observed in the remainder of the cohort with or without T2D who were not taking GLP‐1 RAs. Despite some additional weight loss observed when GLP‐1 RA therapy was added to conventional lifestyle therapy delivered for 8 weeks in the STEP 3 trial of semaglutide [15], taking GLP‐1 RA therapy did not appear to contribute to greater weight loss in our cohort. However, the number of subjects in the overall population taking GLP‐1 RAs was small.

The weight loss achieved with REWIND was associated with a substantial improvement in glycemic control in participants with type 2 diabetes. Despite reductions in the number and doses of glucose‐lowering medications, HbA1c decreased from 7.6% at baseline to 6.3% at 6 months, 6.2% at 12 months, and 6.3% at 18 months, resulting in an average decrease of 1.3%. Of note, the magnitude of this decrease is comparable to that observed with GLP‐1 RAs that are known to both increase insulin secretion and reduce insulin resistance [47, 48]. The similarity in the decrease in HbA1c suggests that in addition to improving insulin action, the sustained weight loss achieved with REWIND may also have been associated with an improvement in insulin secretion [35].

The present study suffers from several limitations. First, it was not a randomized controlled trial but rather a “real life” evaluation of the effectiveness of a virtual community‐based lifestyle program. There was no placebo or comparison group. Nevertheless, weight loss achieved and sustained for at least 18 months with REWIND exceeds that observed in many placebo‐controlled trials where weight loss in the placebo groups was minimal [12, 13, 14, 15, 16, 17, 44, 45, 46]. Indeed, the weight loss achieved with REWIND was comparable to that which was observed with some GLP‐1 RAs [12, 13, 14, 15, 16, 17]. Second, although efforts were made to exclude participants with psychiatric and substance use disorders, some participants with these disorders may have been included. Third, weights were self‐reported. Studies have shown that while weight loss is often underestimated compared to observed weight, the difference is small and efforts were made to ensure reliable measurements over time [49, 50]. Fourth, the program was virtual. Rothberg et al., evaluated over 1300 program participants who transitioned from a fully in‐person to a virtual program. Transition was associated with improved retention and indirectly greater weight loss at 2 years [51]. Richards et al., examined the 2‐year impact of a 12‐month digital weight management intervention implemented through the United Kingdom National Health Service to increase access to treatment for people living with obesity with or without T2D [52]. Among 1130 participants at baseline (mean age 49.9 years, 65% women, 18.1% non‐White, 78.2% with diabetes), 24.9% recorded weights at 2 year follow‐up. The mean weight loss was 13.8 kg or 11.8%. These findings further support the feasibility and effectiveness of virtual weight loss interventions.

Fifth, the population comprised 37% men and 13% individuals of color. Therefore, participants were mostly White women from the Midwest. This distribution reflects what is commonly seen in weight loss and diabetes trials. Nevertheless, our findings may not be generalizable to more diverse populations. It will be important to determine if comparable degrees of weight loss can be achieved in individuals from different geographic areas and settings. Finally, all participants were motivated to lose weight to improve their health since they either sought a weight loss program or agreed to be referred to a weight loss program. Most participants had college education. Nevertheless, since people learn in diverse ways and have varying degrees of health literacy, considerable effort was made to adapt teaching methods and support systems to each person's individual needs. It remains to be determined if, and to what extent, educational attainment and health literacy influence the effectiveness of REWIND.

REWIND employed a multifaceted approach that included one‐on‐one virtual encounters, virtual forums, and team building activities to help participants make lasting behavioral changes. Despite the many demands made by REWIND, including frequent virtual interactions with physicians, dietitians, peers, and ambassadors, ongoing participation rates were 100%, 84%, and 71% at 6, 12, and 18 months, respectively, exceeding those observed for patients using GLP‐1 RAs in community settings [19, 20, 21]. Furthermore, health‐related quality‐of‐life assessed at 6 months in the 120 subjects who agreed to complete the questionnaires showed statistically (p < 0.001) and clinically significant improvements in EQ‐5D‐3L scores (Figure S1). It should be noted, however, that participants who provided HRQoL data lost 20.1 (8.9) kg on average, slightly more than those who did not provide HRQoL data, (15.0 [7.7] kg), suggesting a possible response bias. Future studies will be needed to further address this question.

In summary, an appropriately structured and implemented, virtual, community‐based weight management program helped individuals with obesity lose a substantial amount of body weight at 6 months (16%) that was sustained for at least 18 months (14%). This degree of weight loss resulted in a marked improvement in glycemic control in people with type 2 diabetes, including diabetes remission in 43% of the diabetes participants at 18 months.

Author Contributions

All authors had full access to the data. Dr. Amy E. Rothberg wrote the first draft. All authors contributed to data analysis and editing of the manuscript and take responsibility for the integrity of the data and accuracy of the analysis.

Conflicts of Interest

Dr. Rothberg is currently the Medical Director for REWIND and she and Dr. Rizza has equity interests in REWIND. Dr. Rizza is a member of REWIND's Board of Directors.

Supporting information

Supporting Information S1

Rothberg, Amy E. , Ye Wen, Miller Nicole, and Rizza Robert. 2025. “Lifestyle Change Interventions: Effectiveness of REWIND, a Virtual, Community‐Based Weight Management Program,” Obesity Science & Practice: e70092. 10.1002/osp4.70092.