Susan M. Schembre’s research while affiliated with Georgetown University and other places

Background The Data-Driven Fasting (DDF) app implements glucose-guided eating (GGE), an innovative dietary intervention that encourages individuals to eat when their glucose level, measured via glucometer or continuous glucose monitor, falls below a personalized threshold to improve metabolic health. Clinical trials using GGE, facilitated by paper logging of glucose and hunger symptoms, have shown promising results. Objective This study aimed to describe user demographics, app engagement, adherence to glucose monitoring, and the resulting impact on weight and glucose levels. Methods Data from 6197 users who logged at least 2 days of preprandial glucose readings were analyzed over their first 30 days of app use. App engagement and changes in body weight and fasting glucose levels by baseline weight and diabetes status were examined. Users rated their preprandial hunger on a 5-point scale. Results Participants used the app for a median of 19 (IQR 9-28) days, with a median of 7 (IQR 3-13) weight entries and 52 (IQR 25-82) glucose entries. On days when the app was used, it was used a median of 1.8 (IQR 1.4-2.1) times. A significant inverse association was observed between perceived hunger and preprandial glucose concentrations, with hunger decreasing by 0.22 units for every 1 mmol/L increase in glucose (95% CI −0.23 to −0.21; P <.001). Last observation carried forward analysis resulted in weight loss of 0.7 (95% CI −0.8 to −0.6) kg in the normal weight category, 1 (95% CI −1.1 to −0.9) kg in the overweight category, and 1.2 (95% CI −1.3 to −1.1) kg in the obese category. All weight changes nearly doubled when analyzed using a per-protocol (completers) analysis. Fasting glucose levels increased by 0.11 (95% CI 0.09-0.12) mmol/L in the normal range and decreased by 0.14 (95% CI −0.16 to −0.12) mmol/L in the prediabetes range and by 0.5 (95% CI −0.58 to −0.42) mmol/L in the diabetes range. Per-protocol analysis showed fasting glucose reductions of 0.26 (SD 4.7) mg/dL in the prediabetes range and 0.94 (16.9) mg/dL in the diabetes range. Conclusions The implementation of GGE through the DDF app in a real-world setting led to consistent weight loss across all weight categories and significant improvements in fasting glucose levels for users with prediabetes and diabetes. This study underscores the potential of the GGE to facilitate improved metabolic health.

Background Continuous glucose monitoring (CGM) holds potential as a precision public health intervention, offering personalised insights into how diet and physical activity affect glucose levels. Nevertheless, the efficacy of using CGM in populations with and without diabetes to support behaviour change and behaviour-driven outcomes remains unclear. This systematic review and meta-analysis examines whether using CGM-based feedback to support behaviour change affects glycaemic, anthropometric, and behavioural outcomes in adults with and without diabetes. Methods Ovid MEDLINE, Cochrane Central Register of Controlled Trials, Elsevier Embase, EBSCOhost PsycINFO, and ProQuest Dissertations & Theses Global were searched through January 2024. Eligible studies were randomised controlled trials in adults that implemented CGM-based feedback in at least one study arm compared to a control without CGM feedback. Dual screening, data extraction, and bias assessment were conducted independently. Mean differences in outcomes between intervention and comparison groups were analysed using generic inverse variance models and random effects. Robustness of pooled estimates from random-effects models was considered with sensitivity and subgroup analyses. Results Twenty-five clinical trials with 2996 participants were included. Most studies were conducted in adults with type 2 diabetes (n = 17/25; 68%), followed by type 1 diabetes (n = 3/25, 12%), gestational diabetes (n = 3/25, 12%), and obesity (n = 3/25, 12%). Eleven (44%) studies reported CGM-affiliated conflicts of interest. Interventions incorporating CGM-based feedback reduced HbA1c by 0.28% (95% CI 0.15, 0.42, p < 0.001; I² = 88%), and increased time in range by 7.4% (95% CI 2.0, 12.8, p < 0.008; I² = 80.5%) compared to arms without CGM, with non-significant effects on time above range, BMI, and weight. Sensitivity analyses showed consistent mean differences in HbA1c across different conditions, and differences between subgroups were non-significant. Only 4/25 studies evaluated the effect of CGM on dietary changes; 5/25 evaluated physical activity. Conclusions This evidence synthesis found favourable, though modest, effects of CGM-based feedback on glycaemic control in adults with and without diabetes. Further research is needed to establish the behaviours and behavioural mechanisms driving the observed effects across diverse populations. Trial registration CRD42024514135.

Conflicting clinical trial results on omega-3 highly unsaturated fatty acids (n-3 HUFA) have prompted uncertainty about their cardioprotective effects. While the VITAL trial found no overall cardiovascular benefit from n-3 HUFA supplementation, its substantial African American (AfAm) enrollment provided a unique opportunity to explore racial differences in response to n-3 HUFA supplementation. The current observational study aimed to simulate randomized clinical trial (RCT) conditions by matching 3766 AfAm and 15,553 non-Hispanic White (NHW) individuals from the VITAL trial utilizing propensity score matching to address the limitations related to differences in confounding variables between the two groups. Within matched groups (3766 AfAm and 3766 NHW), n-3 HUFA supplementation’s impact on myocardial infarction (MI), stroke, and cardiovascular disease (CVD) mortality was assessed. A weighted decision tree analysis revealed belonging to the n-3 supplementation group as the most significant predictor of MI among AfAm but not NHW. Further logistic regression using the LASSO method and bootstrap estimation of standard errors indicated n-3 supplementation significantly lowered MI risk in AfAm (OR 0.17, 95% CI [0.048, 0.60]), with no such effect in NHW. This study underscores the critical need for future RCT to explore racial disparities in MI risk associated with n-3 HUFA supplementation and highlights potential causal differences between supplementation health outcomes in AfAm versus NHW populations.

BACKGROUND The Data-Driven Fasting (DDF) app implements glucose-guided eating (GGE), an innovative dietary intervention that encourages individuals to eat when their glucose level falls below a personalised threshold to improve metabolic health. Clinical trials using GGE, facilitated by paper logging of glucose and hunger symptoms, have shown promising results. However, the real-world effectiveness and adherence to GGE supported by a mobile app remain unexplored. OBJECTIVE To describe DDF users’ demographics, app engagement, adherence to preprandial glucose monitoring, and the resulting impact on weight and glucose levels. METHODS Data were collected over three years from users of the Data-Driven Fasting (DDF) app. The analysis covered the first 30 days of app use and included users with at least two days of preprandial glucose entries. App engagement and changes in body weight and fasting glucose levels by weight and diabetes status were examined. RESULTS 6197 people used the DDF app for at least two days. Participants used the app for a median of 19 days (25th, 75th percentiles: 9, 28 days), with a median of 7 weight entries (25th, 75th percentiles: 3, 13 entries) and 52 glucose entries (25th, 75th percentiles: 25, 82 entries). Last observation carried forward analysis revealed a weight loss of 0.7 kg (95% CI -0.8, -0.6) in the normal weight category, 1.0 kg (95% CI -1.1, -0.9) in the overweight category, and 1.2 kg (95% CI -1.3, -1.1) in the obese category. Fasting glucose levels increased by 0.11 mmol/L (95% CI 0.09, 0.12) in the normal range, decreased by 0.14 mmol/L (95% CI -0.16, -0.12) in the prediabetes range, and decreased by 0.50 mmol/L (95% CI -0.58, -0.42) in the diabetes range. CONCLUSIONS The implementation of GGE through the DDF app in a real-world setting led to consistent weight loss across all weight categories and significant improvements in fasting glucose levels for users with prediabetes and diabetes. This study underscores the potential of the GGE to facilitate improved metabolic health. CLINICALTRIAL

The 2020–2025 Dietary Guidelines for Americans recommend limiting intakes of saturated fat and added sugars (SF/AS) to <10% total energy. Data-driven approaches to identify sources of SF/AS are needed to meet these goals. We propose using a population-based approach to identify the leading food and beverage sources of SF/AS consumed by US adults. Foods and beverages reported as consumed were assessed from two, 24 h dietary recalls (24HRDR) from 36,378 adults aged 19 years and older from the 2005–2018 National Health and Nutrition Examination Survey. Intakes of SF/AS were aggregated across both 24HRDR to identify What We Eat in America food categories accounting for ≥90% of SF/AS, respectively, by the total population and within population subgroups. Data were weighted to estimate a nationally representative sample. Ninety-five discrete food categories accounted for ≥90% of the total SF/AS intakes for >88% of the representative sample of U.S. adults. The top sources of SF were cheese, pizza, ice cream, and eggs. The leading sources of AS were soft drinks, tea, fruit drinks, and cakes and pies. This analysis reflects a parsimonious approach to reliably identify foods and beverages that contribute to SF/AS intakes in U.S. adults.

Background Amidst the escalating prevalence of glucose-related chronic diseases, the advancements, potential uses, and growing accessibility of continuous glucose monitors (CGM) have piqued the interest of healthcare providers, consumers, and health behaviour researchers. Yet, there is a paucity of literature characterising the use of CGM in behavioural intervention research. This scoping review aims to describe targeted populations, health behaviours, health-related outcomes, and CGM protocols in randomised controlled trials (RCTs) that employed CGM to support health behaviour change. Methods We searched Ovid MEDLINE, Elsevier Embase, Cochrane Central Register of Controlled Trials, EBSCOhost PsycINFO, and ProQuest Dissertations & Theses Global from inception to January 2024 for RCTs of behavioural interventions conducted in adults that incorporated CGM-based biological feedback. Citation searching was also performed. The review protocol was registered (https://doi.org/10.17605/OSF.IO/SJREA). Findings Collectively, 5389 citations were obtained from databases and citation searching, 3995 articles were screened, and 31 were deemed eligible and included in the review. Most studies (n = 20/31, 65%) included adults with type 2 diabetes and reported HbA1c as an outcome (n = 29/31, 94%). CGM was most commonly used in interventions to target changes in diet (n = 27/31, 87%) and/or physical activity (n = 16/31, 52%). 42% (n = 13/31) of studies provided prospective CGM-based guidance on diet or activity, while 61% (n = 19/31) included retrospective CGM-based guidance. CGM data was typically unblinded (n = 24/31, 77%) and CGM-based biological feedback was most often provided through the CGM and two-way communication (n = 12/31, 39%). Communication typically occurred in-person (n = 13/31, 42%) once per CGM wear (n = 13/31; 42%). Conclusions This scoping review reveals a predominant focus on diabetes in CGM-based interventions, pointing out a research gap in its wider application for behaviour change. Future research should expand the evidence base to support the use of CGM as a behaviour change tool and establish best practices for its implementation. Trial registration doi.org/10.17605/OSF.IO/SJREA.

The National Diabetes Prevention Program (DPP) promotes lifestyle changes to prevent diabetes. However, only one-third of DPP participants achieve weight loss goals, and changes in diet are limited. Continuous glucose monitoring (CGM) has shown potential to raise awareness about the effects of diet and activity on glucose among people with diabetes, yet the feasibility of including CGM in behavioral interventions for people with prediabetes has not been explored. This study assessed the feasibility of adding a brief CGM intervention to the Arizona Cooperative Extension National DPP. Extension DPP participants were invited to participate in a single CGM-based education session and subsequent 10-day CGM wear period, during which participants reflected on diet and physical activity behaviors occurring prior to and after hyperglycemic events. Following the intervention, participants completed a CGM acceptability survey and participated in a focus group reflecting on facilitators and barriers to CGM use and its utility as a behavior change tool. A priori feasibility benchmarks included opt-in participation rates≥50%, education session attendance≥80%, acceptability scores≥80%, and greater advantages than disadvantages of CGM emerging from focus groups, as analyzed using the Key Point Summary (KPS) method. Thirty-five DPP members were invited to participate; 27 (77%) consented, and 24 of 27 (89%) attended the brief CGM education session. Median survey scores indicated high acceptability of CGM (median=5, range=1–5), with nearly all (n=23/24, 96%) participants believing that CGM should be offered as part of the DPP. In focus groups, participants described how CGM helped them make behavior changes to improve their glucose (e.g., reduced portion sizes, increased activity around eating events, and meditation). In conclusion, adding a single CGM-based education session and 10-day CGM wear to the DPP was feasible and acceptable. Future research will establish the efficacy of adding CGM to the DPP on participant health outcomes and behaviors.

The 2020-2025 Dietary Guidelines for Americans recommends limiting intakes of saturated fat and added sugars (SF/AS) to 88% of the representative sample of U.S. adults. The top sources of SF were cheese, pizza, ice cream, and eggs. The leading sources of AS were soft drinks, tea, fruit drinks, and cakes and pies. This analysis reflects a parsimonious approach to reliably identify foods and beverages that contribute to SF/AS intakes in U.S. adults.

Background High glycemic variability (GV) is a biomarker of cancer risk, even in the absence of diabetes. The emerging concept of chrononutrition suggests that modifying meal timing can favorably impact metabolic risk factors linked to diet-related chronic disease, including breast cancer. Here, we examined the potential of eating when glucose levels are near personalized fasting thresholds (low-glucose eating, LGE), a novel form of timed-eating, to reduce GV in women without diabetes, who are at risk for postmenopausal breast cancer. Methods In this exploratory analysis of our 16-week weight loss randomized controlled trial, we included 17 non-Hispanic, white, postmenopausal women (average age = 60.7 ± 5.8 years, BMI = 34.5 ± 6.1 kg/m², HbA1c = 5.7 ± 0.3%). Participants were those who, as part of the parent study, provided 3–7 days of blinded, continuous glucose monitoring data and image-assisted, timestamped food records at weeks 0 and 16. Pearson’s correlation and multivariate regression were used to assess associations between LGE and GV, controlling for concurrent weight changes. Results Increases in LGE were associated with multiple unfavorable measures of GV including reductions in CGM glucose mean, CONGA, LI, J-Index, HBGI, ADDR, and time spent in a severe GV pattern (r = −0.81 to −0.49; ps < 0.044) and with increases in favorable measures of GV including M-value and LBGI (r = 0.59, 0.62; ps < 0.013). These associations remained significant after adjusting for weight changes. Conclusion Low-glucose eating is associated with improvements in glycemic variability, independent of concurrent weight reductions, suggesting it may be beneficial for GV-related disease prevention. Further research in a larger, more diverse sample with poor metabolic health is warranted. Clinical trial registration: ClinicalTrials.gov, NCT03546972.