Abstract
“The program generated a revenue of $969 per 60-minute class to cover the costs of the instructors’ time and overhead expenses.”
Introduction
Nutrition and lifestyle play a key role in the pathogenesis and management of obesity and type 2 diabetes, and dietary and lifestyle changes are recommended for all people with these conditions.1,2 The American Diabetes Association Standards of Care advise weight loss for persons with type 2 diabetes, as it is associated with improved glycemic control and reduced complications. 3 Randomized, controlled trials have shown that low-fat plant-based diets lead to significant weight loss and improvements in glycemic control in type 2 diabetes.4-9
However, research findings have not translated into changes in clinical practice, due in part to limited nutrition education among medical providers, and limitations on office visit duration and reimbursement for nutrition education or support.10-16 Medical nutrition therapy with a registered dietitian may be limited due to geographical location or inconsistent insurance coverage. 17 Currently, Medicare coverage for medical nutrition therapy for diabetes is limited to 4 to 5 hours annually, despite the fact that ongoing therapy results in better outcomes.18,19
Shared medical appointments (SMAs) have been effective in providing nutrition education and improving clinical parameters in type 2 diabetes. 20 However, SMAs have limitations. First, recruitment can be challenging for practices with smaller patient panels. Second, insurance coverage for frequent office visits, such as those required in SMA-based group education programs, may be inconsistent. Third, most health insurance plans—with the exception of Medicaid—may require copays ranging from $15-$50 per shared medical appointment.21,22 Lastly, SMAs require administrative time for documentation and medical coding and billing.
To overcome these barriers, we developed a novel nutrition education program designed to work efficiently in the clinical setting. The program goals were to provide accessible, live, ongoing nutrition education and support for individuals with type 2 diabetes using a cost structure that is economically viable for participants and providers. The program was structured as a self-pay program to obviate the need for insurance coverage or prior authorization. This study was conducted to assess the effects of this program.
Methods
Study Design and Eligibility
This nonrandomized clinical trial was conducted in a primary care setting from 2023 to 2024. However, the program was not limited to patients with a pre-existing provider-patient relationship with the facilitators and non-patients were allowed to enroll. Participants were recruited in 2 cohorts via Physicians Committee social media advertisements and email notifications to Physicians Committee members. Adults with a previous diagnosis of type 2 diabetes were eligible to participate. Exclusion criteria included the use of recreational drugs in the past 6 months, pregnancy, unstable medical or psychiatric illness, lack of English fluency, inability to maintain current medication regimen, and inability or unwillingness to participate in all components of the study. The study protocol was approved by the Advarra Institutional Review Board. All participants gave informed consent.
The study design did not include a control group for several reasons. First, the goal of the program was not to test the efficacy of plant-based diets in type 2 diabetes or weight management, as that has already been established in randomized clinical trials. Rather, the goal of the study was to investigate whether it is feasible to provide nutrition education efficiently in clinical practice, a question that randomization does not address. Second, it was essential to model an economically viable program the way that it would be run in clinical practice. As such, the program involved a self-pay cost for participants and randomizing them to no intervention or a delayed intervention was not practical in this context. Lastly, due to ethical considerations, a control group was not part of the study design as members of such a control group would be denied access to a nutrition-based intervention that has proven efficacy in type 2 diabetes and weight management.
Study Intervention
The program cost $399 per person and included 12 online weekly classes, laboratory testing for hemoglobin A1c (HbA1c) and lipid levels at baseline and at 12 weeks, a digital body weight scale, and informative books. Each weekly 60-minute class was led by a practicing internal medicine physician and a registered dietitian and included education about nutrition’s role in diabetes, plant-based eating, practical tips for grocery shopping, cooking, and dining out, and opportunities for peer-to-peer mentoring and support.
The weekly sessions were structured as educational meetings, and the participants were advised to continue care with their primary care providers. The study team notified the primary care providers about the participants’ enrollment in the program. The study team reviewed the laboratory results with the participants, and the participants were asked to notify the study team and their primary care providers of any adverse events. However, the study team did not manage the participants’ medications, adverse events, or laboratory test results, and participants signed a waiver acknowledging that they would follow-up with their primary care providers for management.
All participants were asked to consume a low-fat plant-based diet for the duration of the 12-week study period. The prescribed diet was comprised of whole grains, vegetables, legumes, and fruits with no restriction on energy intake, while excluding all foods containing animal products and minimizing the intake of high-fat plant-based foods (added oils, nuts, seeds, olives, coconut, solid chocolate, avocados, high-fat vegan meats, vegan cheeses). Participants were advised to limit alcoholic beverages to 1 and 2 per day for women and men, respectively. The study diet derives approximately 10% of energy from fat, approximately 10 to 15% of energy from protein, and the remainder from complex carbohydrates. Participants were advised to consume a commercially available supplement containing at least 100 micrograms of vitamin B12 daily. Food items and meals were not provided, and participants handled their own food preparation and purchases. Participants were asked to maintain their pre-enrollment exercise and medication regimen and to not make any medication changes unless advised by their healthcare provider.
Outcomes
The main outcomes included assessing the feasibility of the program, specifically whether participants would be willing to self-pay for a nutrition education program for type 2 diabetes and the number of program participants and completers. Other main outcomes included HbA1c, body weight, total cholesterol, low-density lipoprotein (LDL) cholesterol, and medication usage. Except for self-reported height measurements at baseline, all clinical variables were assessed at baseline and at week 12. All self-reported data were collected via Qualtrics and email. Participants self-reported their body weights using the digital scales provided to them. Measurements of HbA1c and total and LDL cholesterol levels were completed at LabCorp using standard methodology after overnight fasting of 12 hours. Information about the dosages of unique medications was collected by self-report.
Information regarding adverse events during the 12-week study period was collected by self-report and participants were advised to report any adverse events to the study team and their healthcare providers. The study team prepared a monthly data safety and monitoring report that was reviewed by an independent safety officer and study statistician.
Power Analysis
The prespecified sample size for the trial was based on two-sided significance testing using a one-sample t-test with an alpha level of 0.05. The sample size was calculated based on the primary outcome of change in HbA1c from baseline to 12 weeks. Based on previous studies, the assumed magnitude of change in HbA1c was a decrease of −0.5 units on the low-fat vegan diet in 12 weeks. Assuming the standard deviation of the change in HbA1c is no more than 1.0, and thus more generally assuming an effect size of one half of a standard deviation, 44 participants completing the intervention were required for 90% power to detect a significant treatment effect. Assuming an attrition of up to 20%, the required sample size was a minimum of 53 total recruited participants for 90% power.
Statistical Analysis
Outcomes are reported as means with two-sided 95% confidence limits. Correspondingly, significances in changes from baseline to 12 weeks are evaluated using a one-sample t-test that is two-sided with Type I error of 0.05. As this analysis is considered hypothesis-generating, formal adjustment was not done for the multiple comparisons being reported. Analyses were carried out using SAS Version 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Participant Characteristics
Of 83 individuals screened telephonically, 76 met the participation criteria and were enrolled, and 58 completed the study program (Figure 1). The mean age was 63.4 years for the 58 participants who completed the program, and 69% were women (Table 1). Per self-report, 53% of participants who completed the program were not following a plant-based diet prior to enrollment, and 69% and 50% of participants had been using medications for diabetes or dyslipidemia, respectively. Of note, 4 participants were using glucagon-like peptide-1 receptor agonists (GLP-1 RAs) at the time of enrollment. Flow of participants into the nutrition intervention program. Baseline Characteristics of all Participants Who Completed the Program.
Economic Variables
The program cost $399 per person and included 12 weekly classes, laboratory assessments for HbA1c and lipid panel before and after the program, a digital body weight scale, and helpful books. The total revenue for the 2 cohorts, with 76 participants altogether, was $30,324. After deducting the costs of the laboratory tests ($53), digital body weight scale ($20), and helpful books ($20), the participant cost and practice revenue for each weekly 60-minute class were $25.50 and $969, respectively.
Clinical Variables
Results for all Participants.
SI conversions: To convert HbA1c to proportion of total hemoglobin, multiply by 0.01; total, HDL, LDL cholesterol to mmol/L, multiply by 0.0259; triglycerides to mmol/L, multiply by 0.0113.
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
aP values are for change.
Results for Participants Not Using GLP-1 RAs and Without Changes in Diabetes or Lipid-Lowering Medications During Study Period.
SI conversions: To convert HbA1c to proportion of total hemoglobin, multiply by 0.01; total, HDL, LDL cholesterol to mmol/L, multiply by 0.0259; triglycerides to mmol/L, multiply by 0.0113.
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); GLP-1 RA, glucagon-like peptide-1 receptor agonist; HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
aP values are for change.
Subgroup Analysis.
SI conversions: To convert HbA1c to proportion of total hemoglobin, multiply by 0.01; total, HDL, LDL cholesterol to mmol/L, multiply by 0.0259; triglycerides to mmol/L, multiply by 0.0113.
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); GLP-1 RA, glucagon-like peptide-1 receptor agonist; HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
aP values are for change.
Adverse Events
The most common self-reported adverse events were 63 episodes of hypoglycemia (capillary blood sugar less than 70 mg/dL) experienced by 11 participants that were corrected with intake of high-sugar foods and adjustments in medications. Two participants experienced adverse events (urinary tract infection; asymptomatic coronary artery disease screening followed by cardiac catheterization and coronary artery bypass surgery) that were unrelated to the nutrition intervention.
Discussion
The current study shows that it is feasible to implement a novel plant-based nutrition education program for type 2 diabetes that overcomes the barriers commonly encountered in clinical practice—lack of provider knowledge, short duration of office visits, limited and inconsistent health insurance coverage for medical nutrition therapy, access issues related to geography, and economic sustainability for participants and providers. The program also led to significant reductions in body weight, HbA1c, diabetes medication use, and total and LDL cholesterol levels.
The 12 weekly nutrition education classes were online, thereby providing easy access to participants in different geographical areas. Participants received a total of 720 minutes of nutrition education led by an interdisciplinary team of a practicing internal medicine physician and a registered dietitian over a 12-week period, which is considerably more than the 300-minute annual limit set forth by Medicare for medical nutrition therapy for diabetes. 18 Another advantage of the program was that the live classes were structured in a group format and group programs are more effective for weight loss than individual-based ones.23-25
An essential component of any nutrition education program is that it be financially viable for the participants and clinicians, and the study finds that the program is cost-effective for both. 26 The participant cost for each weekly 60-minute class was $25.50, an amount in line with typical copays for an office visit with a primary care ($15-$25) or specialty care ($30-$50) provider.21,22 The program generated a revenue of $969 per 60-minute class to cover the costs of the instructors’ time and overhead expenses.
The study finds that the program is clinically effective and led to reductions in body weight, medication usage, HbA1c, and total and LDL cholesterol levels. Overall, the mean body weight and HbA1c levels decreased amongst all participants regardless of the baseline diet (plant-based vs non-plant-based), although the mean decreases in body weight and HbA1c were greater in the group that was not following a plant-based diet prior to the program. Out of the 58 participants who completed the program, 13 (22%) decreased the usage of diabetes medications with concurrent reductions in body weight and HbA1c levels. Improvements in HbA1c and body weight with simultaneous reductions in diabetes medications can lead to substantial cost savings for individuals and the healthcare system and minimize medication-related adverse events.27-29
Amongst all participants, there were no significant changes in lipid levels. This is most likely because lipid-lowering medications are commonly recommended for primary and secondary prevention of atherosclerotic cardiovascular disease in diabetes and 50% of participants were taking statins during the study period, which may create a floor effect for total and LDL cholesterol values. 30 However, amongst participants who were not using statins or GLP-1 RAs and who did not experience medication changes during the study period, there were significant reductions in total and LDL cholesterol levels.
Strengths and Limitations
The study has several important strengths. The weekly group classes were held online, eliminating the need for office space and allowing easy access across different locations and time zones. The group classes allowed for peer-to-peer mentoring and support and were led by an interdisciplinary team comprised of a practicing internal medicine physician and a registered dietitian well versed in diabetes care and plant-based nutrition. The 12-week program length allows sufficient time to support behavioral and dietary changes and to detect changes in HbA1c and cholesterol levels. Additionally, 60 minutes of weekly nutrition education over 12 weeks entails a time commitment that may be realistic for most patients and providers. The program modeled a self-pay fee structure that practitioners can implement in clinical practice.
The study also has some limitations. The study design did not include a control group for the reasons outlined above. Of the 76 participants enrolled in the program, 18 (6 dropped out; 12 lost to follow-up) did not complete the program. While the program’s attrition rate of 23.6% is lower than those seen in the literature (31% to 69%) for weight-loss programs, further investigation is needed to explore factors associated with attrition and program completion.31-33 Information about baseline diet and adherence to study diet was collected by self-report. While this approach has limitations, it also simulates what is practical in a clinical setting. Self-reported heights were used in body mass index (BMI) calculations and research shows that self-reported heights are prone to errors. 34 The study participants self-selected into the education program and may therefore be more motivated to make dietary changes and not representative of the general population. However, this also reflects the nature of self-selection in clinical interventions. The study duration was 12 weeks and does not provide an estimate of the program’s impact over a longer duration of time. The study does not separate whether the benefits in clinical variables are due to a low-fat plant-based diet or group support and mentoring. The study did not assess changes in medication expenses, and the self-pay cost of $399 ($25.50 per class) may not be affordable for some individuals. Also, multiple outcomes and subgroups were evaluated, raising the issue of falsely positive significant findings. We note, however, that many changes were significant at the 0.001 or 0.0001 levels; for these, false positive conclusions are unlikely after considering the multiple data looks.
Conclusion
This nonrandomized clinical trial found that it is feasible to implement a plant-based nutrition education program for type 2 diabetes in a primary care setting that is readily accessible, provides sufficient time for education and support, and is cost-effective for participants and providers. The nutrition education program led to reductions in body weight, HbA1c, diabetes medication usage, and total and LDL cholesterol levels and can be implemented in an outpatient setting to improve care in type 2 diabetes.
Footnotes
Author Contributions
Drs. Rahman, Becker and Shannon Gray were involved in the design and conduct of the study; data collection, management, analysis, and interpretation; manuscript preparation, review, and submission. Dr Holubkov was involved in the statistical and power analysis of the study and served on the data safety monitoring committee. Dr Loomis was involved in data analysis and interpretation. Dr Barnard was involved in the design, data analysis, interpretation, manuscript review, and submission.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Drs. Rahman and Loomis and Shannon Gray are practicing clinicians at the Barnard Medical Center. Drs. Rahman, Loomis, Becker and Shannon Gray are employees of the Physicians Committee for Responsible Medicine, a nonprofit organization providing educational, research, and medical services related to nutrition. Dr Rahman has authored several books on nutrition and receives royalties from these sources. Dr Holubkov has received personal fees from the Physicians Committee for Responsible Medicine during the conduct and analysis of the study. Dr Barnard is an Adjunct Professor of Medicine at the George Washington University School of Medicine. He serves without compensation as president of the Physicians Committee for Responsible Medicine and Barnard Medical Center in Washington, DC, nonprofit organizations providing educational, research, and medical services related to nutrition. He writes books and articles and gives lectures related to nutrition and health and has received royalties and honoraria from these sources.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Physicians Committee for Responsible Medicine.
Trial Registration
ClinicalTrials.gov Identifier: 
Ethical Statement
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.