Integration of a lifestyle modification intervention for women with overweight and obesity in a gynecologic oncology practice.

Participant flow is presented in Figure 1 . A total of 164 gynecologic oncology patients were referred to the LMI during the study timeframe. Of those referred, 82 did not enroll in the intervention, making the reach of the intervention to be 50% of its intended target. Two patients deferred enrollment until 2023, which is outside the timeframe of this study, 40 could not be reached by phone, 36 declined, and 5 did not show for their individual pre-intervention session. Reasons for decline were distance from hospital when the LMI was offered in-person, work schedule conflict, too busy, and not interested. A zip code map is shown in Figure 2 depicting the distance from the clinic for both the patients who enrolled (blue) and those who were referred but did not enroll (orange). Of the 163 patients with zip code data, 52 (31.90%) resided more than 25 miles from the cancer center; 17 (10.43%) resided 25–49 miles from the cancer center, 25 (15.33%) resided 50–99 miles from the center, and 10 (6.13%) resided more than 100 miles from the center. There were no statistical differences between the number of patients who enrolled and those who did not enroll in the LMI based on distance from the center ( p >0.54), suggesting locality did not impact the reach of the LMI. A total of 31 patients (19.02%) resided in rural-designated zip codes, of which 13 enrolled in the LMI. There was no significant difference in enrollment rate based on rural classification ( p > .32). Demographic data, baseline clinical, and oncologic histories of the 82 participants who enrolled in the intervention are documented in Table 1 . The median age was 57 years (range: 30 to 73 years), with most patients (68.3%) being white. Diet programs (e.g., Weight Watchers, Jenny Craig) were the most commonly (74.1% of sample) reported method of past weight loss attempts, followed by weight loss medications (27.1%) and weight loss surgery (3.7%). Review of medical records indicated that about 20% of the sample were prescribed an FDA-approved weight loss medication (e.g., phentermine, semaglutide) while enrolled in the LMI. Endometrial malignancy and hyperplasia were the most common cancer-related diagnoses (53.6% and 13.4%, respectively). Ovarian/fallopian tube cancer (13.4%) was the next most common, followed by cervical cancer (4.9%) and vulvar/vaginal cancer or VAIN/VIN (6.1%). Seven participants had no oncologic history but were referred to the gynecologic oncology clinic for a family history of ovarian cancer (n = 1), for risk reducing surgery in the setting of BRCA mutation carrier status (n = 3), a pelvic angiomyxoma (n =1), a benign adnexal mass (n = 1), and endometriosis (n = 1). Most patients with cancer had stage I disease, with six patients having active disease while enrolled in the intervention. Surgery was the most common treatment (81.7%) followed by chemotherapy (23.2%) and radiation (17.1%). Approximately 36 months ( SD = 39.59 months) elapsed, on average, between the end of all active gynecologic oncology treatment and starting the LMI (range: 1–180 months). Participants attended a median of 15 sessions ( M = 13.11, SD = 5.79), with a minimum of 1 session attended and a maximum of 22 sessions attended. 8 of the 82 participants enrolled in the LMI dropped out during the group portion of the intervention, with another 12 patients discontinuing after completion of the group. A total of 62 patients completed the entire 6-month intervention, resulting in a retention rate of 75.6%. Baseline weights varied widely ( M = 111.72 kgs, Mdn = 108.30 kgs, range: 65.63–225.44 kgs) with a mean BMI of 41.51 kg/m 2 ( SD = 9.36 kg/m 2 ; Mdn = 40.35 kg/m 2 ) . The repeated-measures ANOVA revealed a significant main effect of time on total body weight, F (1, 61) = 28.48, p < .001, η p 2 = .32, and BMI, , F (1, 61) = 29.61, p < .001, η p 2 = .33 (see Table 2 ). Post hoc tests indicated that the decrease in weight was significant from baseline to post-group ( p < .001) and post-intervention ( p < .001) and from post-group to post-intervention ( p < .05). More specifically, participants lost an average of 3.04 kgs during the initial 3-month group intervention (baseline to post-group), with almost one quarter losing ≥ 5% of their initial body weight. Participants lost an additional 1.14 kgs during the final 3-months of the intervention involving individual sessions (post-group to post-intervention). Overall, the LMI resulted in a mean weight loss of 4.19 kgs ( p < .001), with 40.30% of participants losing at least 5% of their initial body weight by the end of the 6-month intervention.

We performed a retrospective review of a LMI for gynecologic oncology patients at a large university-based cancer center in an urban setting. Retrospective data were collected from nine cohorts of patients who enrolled in the intervention from September 2019 through February 2023. The study was approved by the Institutional Review Board at Washington University in St Louis. The primary outcome was the percentage of patients who lost at least 5% of their initial body weight during the intervention. Secondary outcomes included feasibility of the intervention based on the enrollment and retention rates, and contextual analysis of the geographic factors implementing enrollment. Gynecologic oncology patients with a body mass index (BMI) > 25 kg/m 2 , a history of any stage of disease, including pre-cancerous conditions (e.g., endometrial hyperplasia) or non-malignant conditions (e.g., fibroids and ovarian cyst) were eligible for the intervention. Patients were either passively recruited for the intervention via flyers in waiting and exam rooms or were actively referred by their oncologists. Ultimately, those recruited via flyers were required to request referral from their oncologists to be assessed for eligibility for the LMI. These referrals were placed via electronic medical record, typically at the time of a scheduled gynecologic oncology follow up visit. Cohorts were limited to no more than 15 patients to encourage discussion and ensure that patients had adequate opportunity to develop a practitioner-patient alliance with the group facilitators. The LMI implemented was an adaptation of the evidence-based Intensive Behavioral Therapy protocol. 12 This protocol involves multiple individual counseling visits completed over the course of a year in a primary care setting. The protocol was modified to be offered in a mixed format (i.e., group and individual sessions) and to be delivered remotely. These modifications were made to increase the number of patients who could enroll in the program while minimizing cost and to increase reach to underserved populations (e.g., rural, low income). The duration of the intervention was also shortened to 6 months by increasing the frequency and length of visits. An initial, 60-minute, individual session was conducted by the psychologist facilitating the intervention to provide overview of the program, establish program goals, and enhance motivation and commitment to behavior change. Participants then participated in 12 consecutive weekly 60-minute group sessions followed by three months of 30-minute, individual sessions that were gradually spaced farther apart. The intervention consisted of dietary counseling, physical activity, and behavioral modification. All aspects of the implemented LMI adhered with recommended components of lifestyle interventions put forth by the 2013 Guidelines for Managing Overweight and Obesity in Adults. 14 Previous RCTs have demonstrated the feasibility of similar interventions offered in a group format and delivered remotely. 24 , 25 Sessions were staffed by a licensed clinical psychologist and psychology trainee, with a dietitian consulting throughout the intervention. The initial cohort completed the entire intervention in-person, with the second cohort completing a hybrid in-person and virtual model due to COVID-19. All subsequent cohorts completed the intervention entirely remotely. The in-person sessions took place in a classroom setting at the Siteman Cancer Center. The virtual sessions were Zoom-based, and all patients and group facilitator(s) shared the Zoom meeting. This intervention was offered at no cost to patients, which is standard for most support services at Siteman Cancer Center. The nutritional component of the intervention focused on reducing caloric intake while working towards a more balanced, quality diet. The Harvard Healthy Eating Plate model was used as a guide for patients to create a healthy diet ( https://www.hsph.harvard.edu/nutritionsource/healthy-eating-plate/ ). Additional topics addressed were keeping a daily food diary, understanding nutritional labels, meal planning, reducing portion size, eating mindfully, and coping with social eating. Behavioral therapy focused on coping strategies, including stimulus control, problem solving, cognitive restructuring, and stress management. Patients were also provided information on energy expenditure in the forms of daily lifestyle activity and moderate intensity physical activity. Patients were encouraged to live a more active daily lifestyle and were provided with an exercise and strength program that gradually increased the amount of moderate physical activity completed every other week. Patients were educated about strategies to improve exercise consistency, including exercise tracking programs and exercise buddies. Feasibility of the LMI in a real-world clinical setting was assessed based on the RE-AIM framework. 22 Reach of the intervention was evaluated by the percentage of patients referred to the intervention who completed at least one session of the intervention (i.e., enrollment rate). Additionally, the geographic reach of the intervention was determined using de-identified zip code data from the referral base. These were mapped by distance from the clinic to provide context to the geographic factors impacting reach such as rural-urban inequality. The Federal Rural Health Policy list of zip codes eligible for rural health grants was used to determine which zip codes were located in rural areas. 26 The primary criterion for evaluating the effectiveness of the LMI was the percentage of the sample that achieved at least a 5% reduction of initial body weight. A 5% reduction in body weight is widely accepted as a clinically meaningful criterion due to its significant health benefits across multiple metabolic and cardiovascular risk factors. 27 , 28 Self-reported weights were used, given that the in-person setting for the initial visits was not equipped with scales and subsequent visits were primarily virtual. While discrepancies between self-reported and measured weight exist, prior studies suggest that self-reported data remain valid for identifying overall weight change trends. 29 Patients self-reported baseline weights at the start of the intervention, from which initial BMIs were computed using the height documented in patients’ medical records. Weight loss was assessed at the middle of the LMI (i.e., post group) and at the end of the LMI 18 . Based on the primary criterion for evaluating effectiveness, a dichotomous variable was generated from patients’ weight loss data at two time-points—the end of the initial 3 months of group sessions and at the end of the entire 6-month intervention. 30 Implementation of the intervention was evaluated by the number of sessions attended and the rate of retention in the intervention. Retention rate was defined as the percentage of patients enrolled in the intervention who completed the entire 6-month intervention. Geographic Information System (GIS) software used zip code data to display geographical information for all patients referred to the program. Not all patients had 9-digit ZIP code data, so 5-digit codes were used. One patient who did not enroll in the LMI was excluded from this set of analysis due to missing zip code data. SPSS version 25 was used to compute all analyses. Chi-square tests were conducted to examine rural-urban inequalities by testing differences in enrollment rate by geographical location. Percentage of the sample who met criterion for effectiveness of the intervention (>5% total body weight lost) was computed as the primary effectiveness outcome. A repeated measures ANOVA was also performed to evaluate reduction in total body weight as a function of time (baseline, post-group, post-intervention). Planned comparisons were conducted for significant main effect of time on weight loss. Only participants who completed the entire intervention (n = 62) were included in this set of analyses.

It is well established that there is a strong link between the occurrence of gynecologic cancers, particularly endometrial cancer, and excessive body fat. 31 , 32 Although randomized controlled trials have shown intensive LMI to be an efficacious approach for weight loss among women with gynecologic cancers, efforts to translate and implement LMI into clinical settings have been scarce. The results of our study demonstrate successful implementation and effectiveness of a remotely-delivered, hybrid based (i.e., group and individual sessions) LMI integrated into a real-world gynecologic oncology clinic, and to assess the geographic context of its reach. We found that the LMI was feasible with an enrollment rate around 50% and retention greater than 75%. The remote delivery of the intervention was able to successfully cut across inequality between urban and rural areas, increasing the reach of the intervention. Moreover, the LMI was found to be effective in helping patients lose a mean of 4.19 kgs during the intervention, with 40.3% of patients losing at least 5% of their initial body weight. The present study highlights that practical management of weight among women treated in gynecologic oncology practices is possible. Our findings suggest that LMI for overweight and obesity can be effective in a clinical practice with more heterogenous populations and fewer in-clinic resources. The amount of weight loss and the percentage of patients to achieve clinically meaningful weight loss (≥5% initial body weight) was similar to that observed in well-funded trials with more rigorous research designs. For instance, Von Gruenigen et al. conducted a 6-month randomized control trial of an LMI among endometrial cancer survivors with obesity, known as the SUCCEED trial, that resulted in an average weight loss of 3.0 kg (3% change from baseline) in the intervention arm. 19 It is encouraging to find that the LMI could be modified to require fewer clinical resources without sacrificing outcomes. While a comprehensive weight loss approach involving a multidisciplinary team is the gold standard of care, not all cancer centers are able to offer such services. 33 Our study demonstrates that a single professional can effectively deliver a LMI for weight loss where resources are scant. Additional implementation studies are required to examine other ways (e.g. weekly vs bimonthly sessions) to offer weight management interventions in an effective and cost-effective manner. Retention can be a major challenge for behavioral weight management interventions and directly impact their effectiveness. 34 , 35 The present study observed good session attendance and a relatively high (75.6%) retention rate similar to that reported in previous control trials, further supporting the implementation of such weight loss interventions. 16 , 19 While no conclusions can be drawn from the present study on contributors to the successful implementation, one important aspect of the intervention design that may have enabled patients to adhere to the intervention and engage in it longer was its remote delivery. Time constraint is a common barrier to behavioral weight loss interventions, particularly among patients with endometrial cancer. 7 , 36 The remote delivery of the intervention increased the accessibility and flexibility of the intervention by eliminating extra travel time to the cancer center and the need to coordinate transportation or childcare. In turn, participants likely had an easier time attending sessions and were more capable of dedicating necessary time to making healthier habits. Another potential contributor to the high level of retention is the group format of the intervention. In general, LMI for weight loss offered in groups are more effective than those offered on an individual basis and tend to be associated with lower attrition. 37 , 38 39 Group delivery may improve retention by promoting participant interaction and cohesion and offering support to those who are socially isolated. 40 Qualitative studies exploring factors influencing adherence and retention may provide important insights into ways to improve the implementation of behavioral weight loss interventions in gynecologic oncology clinics. The low enrollment rate (50%) observed in the present study illustrates the difficulties common to engaging patients in health behavior change programs. 41 , 42 The present findings show that the gap between rural and urban areas can be overcome using remote delivery. Greater geographic distance makes weight loss intervention less accessible and introduces barriers for patients to maintain participation. 43 Our geographic information data revealed that over 30% of the referral base for the present study resided over 25 miles from the cancer center, and nearly 1 in 5 resided in zip codes that have been classified as rural. However, enrollment in the intervention was independent of distance from the cancer center, suggesting that patients did not view locality as a hindrance to participation. Other important barriers to enrolling in the LMI may include persistent treatment side effects, particularly fatigue, and insufficient patient education are key obstacles to engagement after referral. 44 , 45 In a review of the literature, Yannitsos and colleagues (2019) found that receiving advice and guidance from healthcare professionals plays a crucial role in enrolling prostate cancer patients in a lifestyle modification. While time constraints and lack of formal obesity medicine training are common for busy gynecologic oncologists, having a practical and free-of-charge strategy to offer, with a straightforward EMR-based referral system, has improved our ease of communication about the importance of weight management. While pharmacologic weight loss approaches are showing increasing effectiveness, notably only 20% of our cohort received a prescription, and not typically from a gynecologic oncologist. Bariatric surgery continues to be an underutilized strategy in our gynecologic oncology population, considered by only 3.7% of our cohort here. 7 Accessibility and sustainability of the GLP1-receptor agonist class of drugs in this population should be studied further, but the importance of baseline lifestyle modification to pharmacologic management should not be forgotten. Implementation of weight management strategies into a gynecologic oncology care—whether in the prevention, treatment or survivorship phase of care, remains challenging. The heterogeneity of the sample from which the data was drawn is both a strength and a limitation. The ORBIT model for developing behavioral treatments for chronic conditions specifically emphasizes the need to prioritize testing interventions in diverse populations during the effectiveness and implementation phase (Phase 4) of developing behavioral treatments. 21 The sample reflects the breadth of a modern-day gynecologic oncology clinic, including women with a broad range of ages, as young as 30 and as old as 73, and about a third identified with a race other than white. While endometrial hyperplasia or early-stage endometrial cancer comprised 67% of our cohort, a third of the sample had some other type of cancer, and almost 20% of the sample had more advanced disease. While the variability in age, diagnosis, and treatment timeline may introduce heterogeneity in our outcomes, it also enhances the external validity and real-world applicability of the LMI for weight loss among women seen in gynecologic oncology clinics. Regardless of diagnosis, gynecologic oncologists play a key role in motivating healthy behaviors in patients after an experience of cancer. We acknowledge that our results could be skewed by selection bias, in that the participants who did enroll were more likely to succeed, given that 50% of the referral base did not enroll in the intervention. The assessment of weight change was reliant on self-report due to the remote delivery of the intervention, making it susceptible to recall and response biases. Multi-method approaches to assessing weight change will strengthen future studies by reducing potential sources of bias. It also is unclear how well the observed weight loss was maintained given the lack of post-intervention follow up. Future studies will benefit from a more comprehensive assessment of lifestyle changes and anthropometric measures with particular attention on long-term follow up. Finally, the single-site nature of our study limits the generalizability of the findings. In particular, the ability to offer the LMI at no expense may be unrealistic at other sites, making patient cost and reimbursement additional barriers not encountered in the present study. We look forward to more rigorous testing of this intervention in a randomized, multi-site trial for premenopausal women with atypical endometrial hyperplasia or early-stage endometrial cancer and obesity ( NCT05903131 ), where we aim to address these weaknesses. We conclude that lifestyle modification approaches to weight loss among patients in a gynecologic oncology practice with overweight or obesity provide practical and accessible strategies for addressing obesity without loss in effectiveness. While the amount of weight loss was less than that typically observed among healthy individuals with obesity 14 , they are comparable to randomized controlled trials testing LMI among gynecologic patients 18 . More research is needed on its perceived benefits for participants and how it can best be combined with other weight loss strategies to address the epidemic of obesity in this population.

Obesity impacts care in gynecologic oncology in a myriad of ways. The obesity epidemic has led to increased incidences of certain gynecologic cancers, increased risks of surgical complications, and poor long-term outcomes of cancer survivors. 1 – 3 Obesity results in a 4-fold to 10-fold greater risk of women developing endometrial cancer compared to those without obesity. 4 , 5 Cancer survivors with obesity are also six times more likely to die of non-cancer related causes. 2 Guidelines exist that stipulate the incorporation of weight management into oncologic care. 6 However, behavioral, pharmacologic, and surgical weight loss interventions have had varying degrees of success within gynecologic oncology. While bariatric surgery may be the most effective tool in achieving sustained weight loss and decreasing the risk of developing endometrial cancer, it is used in <5% of women with endometrial cancer. 7 – 9 The accessibility of pharmacologic options is limited due to cost from lack of insurance coverage, and there remains insufficient data on the safety and longevity of pharmacotherapy for weight loss. 10 Lifestyle modification interventions (LMI) remain perhaps the most accessible option for those who are not candidates for pharmacologic or surgical approaches or who do not wish to pursue these approaches. 11 Comprehensive LMI for obesity are high intensity (14 or more sessions in a six-month period), multi-component programs that include diet, physical exercise, and behavioral therapy. 10 , 12 , 13 These interventions are designed to help individuals achieve a 5% or greater weight loss by consuming a reduced-calorie diet, increasing physical activity, and acquiring behavioral coping strategies. Individuals lose an average of 5–8% of their initial body weight after a six-month LMI, which in turn leads to clinically significant improvement in health and quality of life. 14 Recent studies have provided preliminary evidence for the feasibility and efficacy of LMI in helping survivors of different types of cancer, including gynecological cancers, eat healthier, be more physically active, lose weight, and maintain weight loss. 15 – 18 Von Gruenigen and colleagues demonstrated that their LMI helped survivors of endometrial cancer lose more weight sustainably over the 12-month intervention compared to the control group. 19 Not only have these trials demonstrated successful weight loss from LMI but also improvements in fatigue, physical functioning, and overall quality of life. 19 , 20 A crucial next step to evaluate the utility of these LMI is to examine how well they disseminate to real-world clinical settings. Feasibility and implementation studies play a critical role in the translation of behavioral interventions for chronic diseases, ensuring they are both practical and effective in real-world settings. 21 The Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework is a commonly used model for feasibility testing. This framework highlights the importance of evaluating an interventions’ reach, effectiveness, and implementation across diverse populations and settings to enhance its external validity. 22 Without feasibility testing, LMI for weight loss may demonstrate efficacy in controlled trials but fail in clinical practice due to low adherence and retention, limited scalability, and resource constraints. 23 As such, the primary aim of this study was to evaluate the feasibility of a remotely delivered, group-based LMI offered to patients in a real-life clinical setting. Secondary aims involved assessing whether the remote nature of the intervention could overcome rural-urban inequalities, thereby, increasing the LMI’s reach.