Oral Abstracts

Background: Technology holds promise for delivery of accessible, individualized obesity prevention and treatment to youth. This review examined the efficacy of recent technology-based prevention and treatment interventions on weight outcomes in youth. Methods: Seven databases were searched from 2014 to 2020 following PRISMA guidelines. Inclusion criteria were: participants aged 1-18y, use of technology in an intervention for prevention/treatment of overweight/obesity; weight outcome; and randomized controlled trial (RCT). Random effects models with inverse variance weighting estimated pooled mean effects separately for treatment and prevention interventions. Subgroup analyses examined the effect of technology type (telehealth or technology-based) and delivery (to parent, child, or both). Results: A total of 3406 records were screened for inclusion; 55 studies representing 54 unique RCTs met inclusion criteria. Most (89%) of the included articles were of high or moderate quality. Meta-analyses of 20 prevention RCTs did not show a significant effect of prevention interventions on pediatric weight outcomes (d=0.05, p=0.52). The pooled mean effect of 31 treatment RCTs showed a small but significant effect on weight outcomes (d= ‒0.13, p=0.001), although 79% of treatment RCTs did not find significant differences between treatment and comparator. There were significantly greater effects for treatment interventions delivered to child or both child and parent compared to parent-delivered treatment interventions, but no differences for technology type. Conclusions: Recent technology-based interventions for the treatment of pediatric obesity show small effects on weight; however, evidence is inconclusive on the efficacy of prevention interventions. Researchers should examine cost-effectiveness of interventions for pediatric weight management to provide important insights regarding the cost advantages of technology-based interventions. Recommendations for augmenting the effects of digital solutions will be discussed.

Background: Dopamine (DA) neurons in the ventral tegmental area (VTA) modulate motivation to engage in physical activity and food intake, dual behaviors that are disrupted in obesity. However, the functional heterogeneity of VTA DA neurons, and lack of molecular markers to distinguish them, has prevented determination of which DA neurons might be leveraged to support weight loss behaviors. We identified a distinct subset of VTA DA neurons defined by their expression of neurotensin receptor-1 (NtsR1) and hypothesized that they modulate feeding restraint and physical activity. Methods: To test this hypothesis, we expressed Designer Receptors Exclusively Activated by Designer Drugs (DREADDs) in VTA NtsR1 neurons of normal weight and diet-induced obese mice, permitting in vivo activation of VTA NtsR1 neurons by treatment with the DREADD-ligand clozapine-N-oxide (CNO). Results: Acute activation of VTA NtsR1 neurons (24hr) in normal weight and diet-induced obese mice significantly decreased body weight by reducing food intake and increasing physical activity. Moreover, the weight loss of diet-induced obese mice was maintained with daily activation of VTA NtsR1 over 7d. Intriguingly, activation of VTA NtsR1 neurons also suppressed how much mice would work to obtain sucrose rewards even when there was high motivation to consume (e.g. in fasted mice), and did not induce vasodepressor or hypothermic responses. Conclusions: Thus, activation of VTA NtsR1 neurons during states of high-appetitive drive promotes dual behaviors that support weight loss without causing adverse cardiorespiratory effects. In the future, strategies to selectively modulating VTA NtsR1 neurons could provide a safe, effective avenue for weight loss.
Background: Triple Negative Breast Cancer (TNBC) is an aggressive and highly metastatic obesity-related cancer with low responsiveness to currently available treatment regimens. As such, therapies that address the metabolic consequences of obesity and slow tumor growth may have value in the treatment of TNBC. We evaluated the antitumor effects of BAM15, a mitochondrial uncoupler with therapeutic potential in the treatment of obesity, in an immunocompetent mouse model of TNBC. Methods: Cell proliferation, migration (scratch-wound), apoptosis (Caspase-3/7), and respiration (Seahorse and O2k) were evaluated in vitro after treatment with BAM15. Orthotopically injected (E0771 cells) female C57BL/6J mice with obesity, were randomized by tumor volume and body weight to 2.5 weeks of high fat diet (HFD; 60% kcal from fat), BAM15 (BAM15; 0.1% w/w BAM15 in HFD), or calorie restriction to match the body weight of BAM15 animals (CR; ~40% restriction of HFD). Food intake, body weight, temperature, and composition, and tumor measurements were obtained. Tissue and circulating concentrations of BAM15 were measured via LC/MS. Mitochondrial capacity was determined in fresh tumor homogenates. Ki67+ cells were visualized by immunohistochemically staining fixed tumor sections. Results: BAM15 reduced cell proliferation, migration, and ATP production while increasing apoptosis in vitro. In mice, BAM15 reduced body weight without altering food intake relative to HFD. Tumor growth was reduced by CR, but the effect was greater with BAM15 treatment. Tumor respiratory capacity and Ki67+ staining was reduced in BAM15 treated animals relative to CR. Conclusions: BAM15 altered hallmarks of TNBC including reducing proliferation, migration, respiration and increasing apoptosis and was efficacious in reducing tumor growth and proliferation in obese mice with breast cancer. BAM15-mediated mitochondrial uncoupling may be an effective strategy for the prevention and treatment of obesity-related TNBC.

Background:
We examined to what extent the association between neighborhood socioeconomic status (NSES) and obesity among the Dutch adult general population is moderated by exposure to payfor-use physical activity (PA) facilities and fast-food (FF) outlets. Methods: Data from the baseline adult Lifelines cohort, Statistics Netherlands and LISA were used. Individuals' residential addresses were linked to neighborhood data and pay-for-use PA facility (0, 1, or >1 within 1 km) and FF outlet locations (0, 1, or >1 within 1 km) using geocoding. Multivariable logistic regression was used to investigate the association between NSES (categorised as low, middle and high) and obesity (Body Mass Index ≥ 30.0), adjusting for neighborhood address density, age, sex, education, partner status, occupational prestige, and income. If the interaction term with NSES was significant, analysis was stratified for having 0, 1, >1 pay-for-use PA facility and 0, 1, or >1 FF outlet within 1 km around the residential address. Results: Obesity was present among 14.7% of the 89,160 participants. People living in low NSES or middle NSES more often had obesity than people living in high NSES (OR (95%CI): 1.56 (1.48 to 1.64), OR (95%CI):1.22 (1.16 to 1.28), respectively). The association between low NSES and obesity was weaker in neighborhoods without any FF outlet within 1 km (OR (95%CI)=1.32 (1.10 to 1.59)) compared to neighborhoods with one FF outlet (OR (95%CI)=1.58 (1.37 to 1.82)) or more than one FF outlet (OR (95%CI): 1.52 (1.43 to 1.62)). Exposure to pay-for-use PA facilities did not moderate the association between NSES and obesity. Conclusions: People living in low NSES more often had obesity than people living in high NSES. This association became weaker if no FF outlets are available within 1km around the residential address. These results indicate that health disparities may be reduced if fewer FF outlets are available.
Background: The dramatic increase in the postprandial levels of gut peptides after vertical sleeve gastrectomy (VSG) has been hypothesized to be a critical mechanism of surgical success because of their roles in satiety and glucose homeostasis. VSG increases the number of enteroendocrine cells (EEC) in the intestinal epithelium, and we speculated that the fate of intestinal stem cell (ISCs) was changed toward EEC differentiation after VSG. In addition, VSG increases the level of specific bile acid subspecies, and the beneficial effects of VSG are blocked in a mouse model lacking bile acid receptor. Therefore, we hypothesized that ISCs express bile acid receptors and the increase in specific bile acid subspecies level after VSG drives ISC fate toward enteroendocrine differentiation. Methods: We performed VSG on a genetically-modified mouse model expressing eGFP in its Lgr5 gene, a specific ISC/progenitor marker gene. We then performed RNAseq on the GFPpositive cells sorted from the jejunal epithelia, where we had seen an increased number of glucagon-like peptide-1 (GLP-1)-expressing cells. We then treated specific bile acids subspecies that increased after VSG to a mouse intestinal organoid model to determine the level of enteroendocrine differentiation. Results: VSG increased the number of GLP-1 and peptide YY (PYY)-expressing cells in the mouse intestinal segments. RNAseq data revealed that VSG altered explicitly ISC expression of several genes that regulate intestinal secretory cell development (e.g., Notch gene family and their targets), including EEC differentiation. We observed multiple bile acids receptor (e.g., Fxr) expression in the ISCs, and treatment of specific bile acid subspecies to mouse intestinal organoid increased the number of GLP-1-expressing enteroendocrine cells. Conclusions: We suggest that the increased levels of specific bile acid subspecies after VSG impact the ISC fate towards EECs, and this is critical to bariatric surgery outcomes.
Background: Sleeve Gastrectomy (SG) increases GLP-1 and bile acid (BA) levels. GLP-1 is released from intestinal L cells with luminal activation of the G protein-coupled receptor, TGR5 in the intestine. Levels of individual BAs in the GI tract following SG and their TGR5 affinity are unknown. We hypothesized that SG induces a specific, gut-based BA that affects systemic glucose metabolism. Methods: UPLC-MS was used to identify and quantify BAs in cecal contents of diet induced obese (DIO) Bl/6J mice after SG or sham surgery and in stool from humans pre and post SG. Metabolites' ability to agonize TGR5 and be a GLP-1 secretagogue were tested in HEK293T and NCI-H716 cells. Acute in vivo effects were tested via direct luminal gut perfusion and by oral gavage in DIO mice. Results: An endogenous cholic acid (CA) metabolite, cholic acid-7-sulfate (CA7S), was increased in the gut of mice and humans after SG. CA7S was absent in systemic or portal circulation and thus, was gut restricted. CA7S is a potent TGR5 agonist (EC50 0.17 µM vs 12.22 µM for CA7S and CA, respectively), induces TGR5 expression (3-fold vs DMSO p = 0.01), and induces a TGR5-dependent increase in GLP-1 expression in vitro. Trans-duodenal and colonic infusion of CA7S in DIO mice increased systemic GLP-1 (2-fold p=0.1), decreased fasting glucose (1.5-fold p=0.02), and increased colonic TGR5 expression (15-fold p=0.03). A single dose of oral CA7S increased systemic GLP-1 (1.5-fold p=0.02) and increased the rate of glucose clearance during oral glucose challenge -27% (p=0.01) and 43% (p=0.004) reduction in AUC and iAUC, respectively. These changes were partially abrogated by lentiviral shR-NA-mediated knockdown of the GLP-1 receptor, and completely abrogated by TGR5 knockdown. Conclusions: CA7S is a novel, naturally occurring, and gut restricted TGR5 agonist upregulated after SG in both mice and humans. It exerts systemic glucoregulatory effects in a TGR5dependent manner. Targeting the CA7S-TGR5 axis may recapitulate the metabolic benefits of SG.

Background:
The effectiveness of family-based weight loss treatment (FBT) for children with severe obesity is unclear. This study compares changes in relative weight and adiposity, following FBT, for children with severe obesity versus those with overweight (OW) or Class I obesity (C1). Given variability in the literature and critiques of some metrics being less sensitive to change in children with severe obesity (particularly BMI z-score (zBMI) which attenuates at high weights), multiple outcome metrics were examined. Methods: Children ages 7-11 (N=160) and a parent completed 12 months of FBT and maintenance intervention as well as anthropometric and adiposity (DEXA) assessments. Severe obesity was defined as a BMI>120% of the 95th percentile for age and sex. Four change indices were used: 2 for relative weight (zBMI & percent overweight [PctOw]) and 2 for adiposity (body fat percentage [BF%; kgfat/kgtotal] & fat mass index [FMI; kgfat/m2]). Weight group by time interactions using repeated-measures ANOVAs were examined. Results: At post-treatment, 40% of children who had severe obesity at baseline no longer did (31% C1; 8% OW; 1% healthy weight). PctOw (Severe: -15.6±15.7 vs. C1/OW: -14.6±11.5, p=0.63) and FMI (Severe: -1.6±2.4 vs. C1/OW: -1. 9±1.8, p=0.55) showed similar magnitude of changes among both weight groups, whereas zBMI (Severe:-0.3±0.4 vs. C1/OW: -0.5±0.4, p=0.001) and BF% (Severe: -3.9±5.3 vs. C1/OW: -5.7±5.6, p=0.037) showed smaller changes among the children with severe obesity. Conclusions: FBT can benefit children with severe obesity and, depending on the measures used, produce similar weight/adiposity reductions in these children as they do for those with lower levels of excess weight. Future studies should determine the best weight/adiposity metric of improved health among children with severe obesity and the potential benefit of a more intensive intervention, as many children with severe obesity still had high relative weights post-treatment.

Binge Watching During COVID: Associations With Stress and Body Weight
Anahys Aghababian Baltimore MD, Kimberly Smith, PhD Baltimore MD, Elena Jansen, PhD Baltimore MD, Gita Thapaliya, PhD Baltimore MD, Jennifer Sadler, PhD Baltimore, Liuyi Chen Baltimore MD, Susan Carnell, PhD Baltimore MD Background: Binge watching (BW) is a widespread phenomenon with potential to impact energy balance and thereby body weight. Previous studies show stress is associated with TV viewing and increased weight, but the influence of stress on BW and weight is understudied. The COVID-19 pandemic and lockdown measures created conditions conducive to BW and increased stress. Here we investigated relationships between BW, stress, and weight by examining reports of BW during (concurrent report) and before (retrospective report) the pandemic. Methods: 439 adults (age: 37.0±9.1y; BMI: 28.0±7.3 kg/m 2 ) completed an online survey assessing BW behaviors, COVID-related stress, and BMI. Of the total sample, 167 (38.0%) were normal weight, 154 (35.1%) overweight, and 121 (27.6%) obese. We ran repeated measures ANOVAs with time (pre-COVID vs now), and stress tertiles and weight groups as between-subjects factors. Results: The duration of a typical BW increased from 3.3±1.9h (range 0-10.3h) before the COVID crisis to 3.9±2.1h (range 0-10.5h) during the crisis (p<.001). BW frequency also increased (p<.001) with the number of participants reporting BW 3-7x a week increasing from n=57 (9.8%) to n=136 (29.2%). BW frequency was greater with higher stress (p=.023), driven by higher values in the high stress group compared to the low stress groups (p=.027). A time x stress x weight group interaction was identified such that reported increases in BW were greatest in individuals with obesity and high stress levels (p=0.045). Higher stress was associated with eating while BW (p=.001), where the number of participants reporting often/always eating while BW was greatest in the high stress group (high stress: 45.7%, n=53; medium stress: 24.8%, n=32; low stress 20.9%, n=18). Conclusions: Binge watching increased during the pandemic, with larger increases for individuals with obesity and high stress, and concomitant effects on eating, highlighting a potential target for interventions to minimize obesogenic impact of the ongoing crisis.
Background: Recent advancements in three-dimensional optical (3DO) imaging have made this technology readily accessible and cost-effective in fitness centers and clinical settings across the world. Advanced statistical methods have also been used to convert digital anthropometry and body shape avatars into accurate and precise body composition (BC) measurements. In this study, we validate BC estimates from 3DO scans against the criterion dual energy X-ray absorptiometry (DXA) in a multiethnic adult cohort with a broad age and BMI range. Methods: The Shape Up! Adults Study spans three sites: Pennington Biomedical Research Center (PBRC), University of California San Francisco (UCSF), and University of Hawaii Cancer Center (UHCC). Each participant underwent whole body 3DO (Styku S100) and DXA (Hologic Horizon A/Discovery A) scans. The PBRC data was provided to the manufacturer to create BC prediction equations from automated anthropometry measures for lean mass (LM), fat mass (FM), percent body fat (PBF), and visceral adipose tissue (VAT). The UCSF and UHCC data were used for independent validation. All 3DO BC measurements were compared against DXA as the criterion method. Results: 296 participants (155 female) from PBRC, and 204 participants (116 female) from UCSF and UHCC received both scans with a BMI range of 14.2 to 52.5 and five different ethnicities. In the training data, 3DO estimates were highly associated to their DXA equivalents. Background: Caloric intake increases during sleep restriction (SR). The current study examined the effects of two separate bouts of SR on objective caloric intake to determine if these SR effects cumulatively increase with repeated exposure. The number of recovery nights between bouts was systematically manipulated to determine if intervening recovery sleep affected the intake response. Methods: Healthy adults (N=45, 33.9±8.9y; 28 male) participated in a laboratory-controlled protocol. Participants underwent 2 baseline nights (10h time-in-bed (TIB)/night; 22:00-08:00) followed by 10 SR nights (4h TIB/night; 04:00-08:00). The 10 SR nights were divided into two bouts of 5 consecutive nights each and were separated by 1 (n=13), 3 (n=12) or 5 (n=10) recovery nights (12h TIB/night; 22:00-10:00). Control condition participants (n=10) were permitted 10h TIB (22:00-08:00) on all nights. Food and drink consumption were ad libitum. Caloric intake was recorded daily and analyzed using The Food Processor SQL program. Mixed-model ANOVAs (Day x Condition) and Intraclass Correlation Coefficients (ICCs) were used for analyses. Results: Compared to baseline, sleep-restricted participants increased caloric intake during both SR bouts (SR1-3: p<0.001; SR6-8: p<0.001). Caloric intake did not differ between SR bouts (p=0.18); the ICC for caloric intake between bouts was almost perfect (0.95). The Day*Condition interaction was not significant (p=0.27). Participants in the control condition did not exhibit changes in caloric intake across corresponding days (p=0.40). Latenight caloric intake (22:00-03:59) did not differ between SR bouts and there was no Day*Condition interaction (ps>0.34). The ICCs comparing late-night intake between bouts were substantial (% daily intake: 0.65, kcal: 0.87). Conclusions: Daily and late-night intake responses were comparable between SR bouts and did not differ by the number of intervening recovery nights, suggesting that increased time spent awake is the primary driver of increased caloric intake.

Background:
The 2010 Affordable Care Act included a provision requiring retail food establishments, including supermarkets, to post calorie information for prepared foods. Implementation of calorie labeling may spur product reformulation, but this has not been well-studied.
We aimed to examine changes in calorie content of prepared foods sold at two large supermarket chains after the chains implemented calorie labels on prepared foods. Methods: The supermarket chains provided us with information on the calorie content and labeling status of all items sold by the chains between July 2015 and January 2019. We used difference-in-differences (DD) models, allowing for differential trends by labeling status, to examine changes in calorie content (kcal/item) from pre-to post-labeling among prepared foods (products designated to receive calorie labels) compared to changes among comparison foods (packaged products similar to prepared foods but not subject to the new labeling requirement). We examined items that were continuously available throughout the study period and ran analyses separately for bakery items (e.g., muffins, n=1,890 unique items) and entrées (e.g., rotisserie chicken, n=2,664 unique items). Results: In the pre-labeling period, trends in calorie content were relatively stable for both prepared (bakery: +3.4 kcal/year; entrées: -0.8 kcal/year) and comparison (bakery: +0.5 kcal/year; entrées: -1.4 kcal/year) items. After labeling implementation, calorie content of prepared bakery items decreased modestly relative to comparison bakery items (DD, -13.2 kcal/item; 95% CI: -27.0, 0.7; p=0.06; ~1% change). We did not observe significant changes in the calorie content of prepared entrées (DD,9.0 kcal/item;22.6;p=0.20;~1% change). Conclusions: In this longitudinal study of two supermarket chains, calorie labeling was associated with a very small reduction in calorie content of prepared bakery items but not associated with changes in calorie content of prepared entrées.

Oral 021
Cardiovascular Effects of Self-Selected Exercise Intensity in youth with Obesity: A RCT Wagner Prado, PhD San Bernardino CA, Tercio Barros, MsC, Thiago Tenorio, PhD, Raphael Ritti-Dias, PhD, Antonio Germano-Soares, MsC, James Hill, PhD Birmingham, Babu Balagopal, PhD Jacksonville FL, Ricardo Freitas-Dias, PhD Background: Self-selected intensity of exercise (SEI), enabling the participants choosing their own exercise intensity, has been proposed as an alternative approach to treat people with obesity. There is a scarcity of data regarding its benefits cardiovascular outcomes, particularly in children and adolescents with obesity. The aim of this study was to compare the effects of SEI vs. pre-determined intensity (PEI) on blood pressure and arterial stiffness in adolescents with obesity. Methods: Thirty-seven adolescents (25 boys), 13 to 18 years old, with obesity (body mass index (BMI) ≥ 95 th ) were randomly allocated into SEI (n = 18; 12 boys) or PEI (n = 19; 13 boys) groups. Adolescents from both groups exercised for 35 minutes on a treadmill, 3x/week, for 12 weeks. SEI permitted the participants to set the treadmill speed at the beginning of the sessions and make changes every 5 minutes. PEI adolescents were trained at an intensity set at 60-70% of heart rate (HR) reserve. Brachial and central BP, pulse pressure, and reflected wave measures (augmentation index) and arterial stiffness (applanation tonometry) were determined at baseline and after the 12 weeks. Results: Both groups reduced brachial systolic BP (SEI = -4.12 ± 10.64 mmHg and PEI = -4.01 ± 12.22 mmHg; P = 0.04), central systolic BP (SEI = -4.92 ± 6.21mmHg and PEI = -4.61 ± 3.73 mmHg; P = 0.01) and central pulse pressure (SEI -4.47 ± 4.76 mmHg and PEI = -3.51 ± 4.29 mmHg; P = 0.02), without differences between groups. No changes in augmentation index and carotid-femoral pulse wave were observed in both PEI and SEI (P > 0.05). Conclusions: Aerobic training performed at self-selected intensity induced similar changes on various cardiovascular outcomes compared to pre-determined exercise intensity in adolescents with obesity. The data from the current study are important and suggest potential for adjustments to designing more effective and viable exercise protocols for adolescents with obesity.

Oral 022
Central Obesity and Puberty Associated with Caudate Response During Anticipation but not Reward Grace Shearrer, PhD Chapel Hill NC, Afroditi Papantoni, BS Chapel Hill NC, Jennifer Sadler, PhD Baltimore, Kyle Burger, PhD, RD, MPH Chapel Hill Background: Heightened striatal response to anticipation of reward, relative to reward receipt is well established as a characteristic of adult obesity and has been implicated in adolescent risk taking. How the striatal response pattern interacts with body composition and pubertal development is unknown. We hypothesized that children with high body mass index (BMI) and waist circumference (WC) would show higher anticipatory response, versus reward receipt, in the caudate and putamen compared to children with average BMI and WC. The positive difference between anticipation and receipt of reward would be exaggerated during mid-puberty. Methods: Data is from the Adolescent Brain Cognition and Development (ABCD) study. Participants were included if they were not taking any medications, had no serious illnesses, and had complete monetary incentive delay scanning data, BMI, waist circumference, and self-reported Pubertal Development Scale data. 844 ABCD participants met inclusion criteria. Within the bilateral caudate and putamen, anticipation of a reward was compared to receipt of a reward (anticipation vs. reward). General linear models assessed a three-way interaction between Pubertal Development Score by standardized BMI (BMIsds) by standardized waist circumference (WCsds) on anticipation vs. reward controlling for sex, race, socio-economic status, age, scanner type, and head motion. Significance was set at p<0.05. Results: The interaction between pubertal development, BMIsds, and WCsds was positively related to anticipation versus receipt of a reward in the left caudate (beta= 0.41, SE= 0.004, p= 0.03). Specifically, with increased pubertal development, children with central obesity (high BMI and high WC) showed a greater response in the left caudate to anticipation of a reward versus the receipt of a reward compared to children with average BMIsds and WCsds (beta= 0.36, SE= 0.02, p= 0.01). Conclusions: This analysis refines previous literature showing heightened striatal response to anticipatory reward cues, compared to receipt, is associated with increasing BMI; implicating that body composition and stage of pubertal development are critical to consider. Preventing trunk fat accumulation early in childhood may impart a more favorable reward learning response throughout puberty.

Oral 023
Central α-Klotho Increases POMC Neuron Activity and Resting Energy Expenditure in Mice Taylor Landry, MS Greenville NC, Peixin Li, Brenton Laing, Wyatt Bunner, Zhijian Rao, Anna Bass, Quincy Gilliard, Hu Huang, PhD Greenville NC Background: We recently identified the centrally circulating α-klotho protein as a novel hypothalamic regulator of food intake and glucose metabolism. The current study aimed to determine central α-klotho's effects on resting energy expenditure (EE) and identify novel neuronal and molecular effectors involved in central α-klotho function. Methods: To investigate the effects of central α-klotho treatment on EE, indirect calorimetry was performed using the TSE Phenomaster metabolic cages in diet-induced obesity (DIO) mice receiving 2 days intracerebroventricular (ICV) injections of α-klotho (2.0ug) or vehicle (2.0uL). Immunohistochemical staining for cFOS, proopiomelanocortin (POMC) neurons, and Ki67 was performed in the arcuate nucleus (ARC) to determine the effects of ICV α-klotho on POMC neuron activity and progenitor cells. Lastly, ICV pretreatment with fibroblast growth factor receptor (FGFR) or PI3kinase inhibitors was performed to determine the cell signaling involved in α-klotho-mediated regulation of hypothalamic POMC neurons. Results: In DIO mice, 2 days ICV α-klotho treatment significantly increased mean oxygen consumption (5288±400 vs. 5862±610 ml/kg lean mass/h) and carbon dioxide production (3489±310 vs. 3881±357 ml/kg lean mass/h), with no effects on locomotor activity. Acute ICV α-klotho increased cFOS expression in POMC neurons 2.5-fold in the ARC of fasted, but not fed, mice. Inhibition of FGFR or PI3kinase signaling abolished these effects. Lastly, 7 days ICV α-klotho treatment increased the number of Ki-67-expressing cells 3-fold in the ARC, suggesting increased activated progenitor cells. Conclusions: Our results demonstrate that central α-klotho improves energy balance by increasing energy expenditure in DIO mice. Furthermore, these data identify POMC neurons as novel neuronal targets of central α-klotho and provide evidence of a prominent central α-klotho-FGFR-PI3kinase signaling axis in homeostatic regulation of metabolism.

Background:
In an effort to reduce growing burdens of obesity and diabetes, South Africa implemented a sugar-content based tax called the Health Promotion Levy in April 2018, one of the first SSB taxes to be based on each gram of sugar (beyond 4g/100mL). This before-and-after study estimates changes in taxed and untaxed beverage intake one year after the tax, examining separately for the first time the role of reformulation distinct from behavioral changes in SSB intake. Methods: We collected single day 24h dietary recalls from repeat cross-sectional surveys of adults aged 18-39 years in Langa, South Africa. Participants were recruited in February-March 2018 (pre-tax, N=2,467) and Feb-Mar 2019 (post-tax, N=2,540) using door-to-door sampling. We developed time specific food composition tables (FCTs) for South African beverages before and after the tax, linked with diet recalls. We estimated beverage intake using a two-part model, with a probit model for the first part and a generalized linear model with log-link for the second part. Results: Among taxed beverages, sugar intake decreased significantly (p<0.0001) from 28.8 (95% CI 27.3-30.4) pre-tax to 19.8 g/capita/day (95% CI 18.5-21.1) post-tax. Energy intake decreased (p<0.0001) from 121 (95% CI 114-127) pre-tax to 82 kcal/capita/day (95% CI 76-87) post-tax. Volume intake decreased (p<0.0001) from 315 (95% CI 297-332) pre-tax to 198 mL/capita/day (95% CI 185-211) post-tax. Behavioral change accounted for reductions of 24% energy, 22% sugar, and 23% volume, while reformulation accounted for additional reductions of 8% energy, 9% sugar, and 14% volume from taxed Obesity www.obesityjournal.org Obesity | VOLUME 28 | SUPPLEMENT S2 | NOVEMBER 2020 13 Oral Abstracts beverages. Conclusions: Using a large sample of a high-consuming, low-income population, we found large reductions in taxed beverage intake, separating the effects of behavioral change from reformulation. Because policies such as taxes can incentivize reformulation, our use of an up-to-date FCT that reflects a rapidly changing food supply is novel and important for evaluating policy effects on intake.

Oral 025
Changes In January 2017, Philadelphia implemented a 1.5 cent/ oz tax on sugar-and artificially-sweetened beverages, which led to a large decline in sales of taxed beverages. Evidence for the effect of this tax on the price and sales of specific beverage types is needed to better understand the health implications of the tax. Methods: We obtained sales data from large, chain food retailers in Philadelphia, Baltimore (control city), and counties that bordered Philadelphia in 4-week periods from January 2016 to December 2017. We used a difference-in-differences approach to examine changes in prices and sales of specific beverage types in Philadelphia one year after tax implementation compared to Baltimore. We also estimated how much of the decline in specific beverage sales in Philadelphia was offset by increases in sales across the Philadelphia border. Results: The analytic sample included 291 stores in Philadelphia, Baltimore, and border counties; 5.9 billion oz of beverages were sold across the Philadelphia stores in 2016 (pre-tax). In Philadelphia, the tax passthrough to prices varied by beverage type and ranged from 0.06 cents/ oz (95% CI: -0.12, 0.25) for coffee-based drinks to 1.02 cents/oz (95% CI: 0.98, 1.05; 68% pass-through) for regular soda compared to price changes in Baltimore. Volume sales of taxed beverages decreased by >45% for all beverage types except for coffee-based drinks and energy drinks (15% and 17% decrease, respectively), and decreased most for flavored water, diet soda, and regular soda (62%, 60%, and 53% decreases, respectively). Increases in border county shopping offset Philadelphia's volume sales decrease by <25% across all beverage types except for regular and diet soda, which were offset by 36% and 35%, respectively. Conclusions: Philadelphia's beverage tax was passed through to prices to varying degrees based on beverage type, but was generally associated with large reductions in volume sales of a range of both sugar-and artificially-sweetened beverages. Oral 027

Cumulative Social Stress in Kindergarten and Obesity Risk, ECLS-K:2011
Zerleen Quader Atlanta GA, Shakira Suglia, ScD, MS Atlanta GA Background: Social stress has been shown to be associated with childhood obesity. However, few studies have used longitudinal data to assess the risk of developing obesity in a nationally representative sample of non-obese children. This study aims to determine if exposure to stressors in kindergarten is associated with an increased risk of obesity prior to adolescence. Methods: Data were from the Early Childhood Longitudinal Study, Kindergarten Class of 2010-2011, a nationally representative sample of kindergartners followed through 5th grade. Parents completed questionnaires at baseline, and child height and weight were measured annually and used to calculate body mass index. A cumulative social stress score was created using nine factors: food insecurity, at or below poverty level, mother not married at child's birth, high residential mobility, high aggravation in parenting, parental depression, harsh discipline, seeking professional help for drugs or emotional problems, and serious financial difficulty since birth. The final analytic sample was 7,568 children who were not obese at baseline. Log-binomial regression was used to determine risk of obesity in fifth grade by social stress score, stratified by sex. Results: Sixty-eight percent of children were exposed to 2 or more social stressors at baseline and 13% percent were obese by 5th grade. Among those with no stressors, 8% became obese, and among those with 5 or more stressors, 17% became obese. After adjustment for race/ethnicity and parent education level, the greatest risk of obesity was for those with 5 or more stressors compared to no stressors. Boys had a 26% greater risk of obesity (95%CI: 0.88, 1.81) and girls had a 75% greater risk of obesity Background: Well-established sociodemographic differences in weight loss success are often attributed to barriers in accessing intensive, in-person lifestyle treatment. We posited that addressing access barriers by delivering intensive lifestyle treatment remotely, facilitated by mhealth tools, might overcome usual racial, sex, and age differences in early weight loss success. Methods: We performed preliminary analyses of initial weight loss success in the SMART Weight Loss Management Study, a stepped-care trial that initially randomized participants equally to either smartphone self-monitoring app alone (App) or app plus coaching (App+C), followed by re-randomization to either moderate or vigorous treatment intensification if weight loss was less than 0.5 lb/week. Results: Participants were 346 adults, (77% female), between ages 19 and 60 (m=40.8, sd=11.1) with overweight or obesity (84% obese). Logistic regression comparing the effects of initial treatment, sex, age, and race on the probability of needing treatment intensification showed no effects of beginning treatment with App vs. App+C, but treatment intensification was more likely to be needed by Black participants (χ2(1) = 8.37, p < .01), females (χ2 (1) = 4.79, p < .05), and younger participants (χ2 (1) = 8.53, p < .01). Conclusions: Usual racial, sex, and age differences in early weight loss success persist in the face of mobile obesity treatment tools, and do not appear to be ameliorated by including coaching in the initial treatment assignment. Background: Global obesity rates are on the rise, resulting in a growing threat to public health. Pharmacotherapies that safely reduce body weight in obesity remain elusive, partially due to our incomplete knowledge of the complex neuronal mechanisms that control food choice. Similarly, we know little about the mechanisms by which consumption of palatable food can transition from controlled to compulsive, thereby driving the development of obesity. The lateral hypothalamus (LH) is a brain region considered a critical node in the maintenance of energy homeostasis. We hypothesized that obesity is accompanied by changes to LH function and this in turn impacts a range of downstream brain structures leading to obesity associated behavioral abnormalities. Methods: Single cell sequencing, electrophysiology, chemogenetics, whole brain clearing, optogenetics, photometric recording of neuronal activity in vivo, rodent behavioral feeding assays. Results: Single cell sequencing of LH RNA transcripts from mice revealed excitatory neurons show marked obesity-associated alterations in gene expression. Specifically, transcripts from genes associated with glutamatergic neurotransmission were significantly lower in tissue harvested from obese mice. Using electrophysiology and whole brain clearing with automated quantification of synaptic contacts, we found this transcriptional plasticity accompanied the emergence of LH glutamatergic hypofunction in the LH and a profound restructuring of hypothalamic output to a range of brain loci including the lateral habenula, periaqueductal gray and dorsal raphe nucleus. Finally, using transgenic mice and viral manipulation of neuronal function in vivo we linked this LH glutamatergic hypofunction in individual circuits to the emergence of discrete obesity associated behaviors. Conclusions: Our research posits rescuing excitatory hypothalamic output as a novel therapeutic strategy to combat the food related motivational deficits that emerge in obesity. Obesity | VOLUME 28 | SUPPLEMENT S2 | NOVEMBER 2020 www.obesityjournal.org

Oral Abstracts
Oral 034

Effect of Macronutrient Content on Meal and Macronutrient Pattern Variation in Ad Libitum Feeding
Tomás Cabeza de Baca, PhD Phoenix AZ, Paolo Piaggi, PhD, FTOS Phoenix AZ, Marci Gluck, PhD Phoenix AZ, Jonathan Krakoff, MD Phoenix AZ, Susanne Votruba, PhD, RD Phoenix AZ Background: Macronutrient intake of protein and carbohydrates is purported to be tightly regulated across individual diets. In contrast, it is hypothesized that fat intake has no such regulatory mechanisms. The current study aims to examine whether there is evidence of macronutrient regulation across meals. Methods: Healthy volunteers (n=287; 177m; Age=35±11; BMI=32±8) were admitted as inpatients to our CRU. Macronutrient meal content (grams) and energy intake (Kcal) were quantified by specialized food processing software and collected on an hourly basis over a three-day period using a validated ad libitum vend paradigm. Lagged multi-level models accounting for age, sex, race/ethnicity, fat and fat free mass were fitted to examine the impact of prior macronutrient content on subsequent meals. Results: Greater carbohydrate (B=11.86, p=0.0006) and decreased protein (B=-16.51, p=0.0042) intake predicted increased energy intake in subsequent meals. Additionally, prior increased carbohydrate (B=0.32, p=0.02) and decreased protein (B=-0.54, p=0.02) predicted increased protein intake in the next meal. The same pattern was found for carbohydrates, such that more carbohydrates (B=2.17, p<0.0001) and decreased protein (B=-2.14, p=0.0006) intake predicted a greater consumption of carbohydrate at the following meal. There was no significant effect of fat content in the models. Conclusions: These results indicate that carbohydrate and protein composition of prior meals may influence energy intake and macronutrient content of future meals. Clinical interventions may consider regulating macronutrient content of meals as a target for weight loss. Background: Time-restricted eating (TRE) is a novel type of intermittent fasting that involves eating within a period of 10 hours or less. Pilot studies suggest that TRE reduces body weight, improves glycemic control, and may enhance fat loss in humans. Here, we performed the first largescale randomized controlled trial to test whether TRE improves weight loss (WL), fat loss, and cardiometabolic risk factors, compared to continuous energy restriction. Methods: Ninety patients of the UAB Weight Loss Medicine Clinic with obesity were randomized to either practice early time-restricted eating (TRE: eat between 7 am -3 pm; 8-hr eating period) or to eat throughout the day (≥12 hours; CONT), while following the clinic's standard-of-care WL treatment for 14 weeks. Outcome measures including DXA were assessed at weeks 0 and 14. All p-values are adjusted for race, sex and age except an unadjusted Wilcoxon p-value for WL (reported as median [IQR]) due to a non-parametric distribution. Results: Seventy participants (mean±SD: age=43 ± 11 years, BMI=39.6 ± 6.7 kg/m 2 ) completed the intervention and were included in the analysis. The TRE and CONT groups were adherent 5.9 ± 0.7 and 6.3 ± 0.8 days/week, respectively (p=0.06). TRE did not increase absolute (median[IQR]: 5.2[3.1-8.3] vs 4.1[1.5-6.9] kg; p=0.17) or relative weight loss (12 ± 9 vs. 10 ± 9% of initial body weight; p=0.27). However, TRE increased trunk fat loss (-3.2 ± 1.8 vs. -2.0 ± 2.2 kg; p=0.03) and tended to increase total fat loss (-5.2 ± 3.4 vs. -3.4 ± 3.5 kg; p=0.09), although it did not affect the percent of weight lost as fat (75 ± 17 vs. 78 ± 14%; p=0.61), relative to CONT. TRE also decreased diastolic blood pressure (-6 ± 9 vs. -1 ± 6 mm Hg; p=0.01) and tended to decrease heart rate (-7 ± 12 vs.

Effect of Obesity and Metabolic Disease on Severity of SARS-CoV-2 Infection
-2 ± 8 bpm, p=0.10), fasting insulin (-6.9 ± 11.2 vs. -0.7 ± 14.4 mU/mL; p=0.08) and HOMA-IR (-4.0 ± 2.7 vs. -4.4 ± 4.3, p=0.10), but changes in waist circumference, fasting glucose, HbA1c, and lipids did not differ between groups (all p≥0.15). Conclusions: Eating within an eight-hour window early in the day did not increase weight loss relative to conventional energy restriction, but it did increase abdominal fat loss and tended to increase total fat loss. TRE may also improve diastolic blood pressure, fasting insulin, and insulin resistance, but it was not superior to conventional energy restriction for reducing cholesterol and glucose levels.
Oral 037 Oral Abstracts promoting healthy diet with a specific focus on reducing fat calories was successful in achieving healthy gestational weight gain (GWG) within recommended guidelines. Whether these measurable effects are significantly associated with ultra-processed food (UPF) intake was the focus of this investigation. Methods: Maternal weights were measured at weeks 14 and 35 of gestational age (GA) (n=157). UPF is presented as a percentage of energy intake (PEI-UPF) and was derived from the Automated Self-Administered-24 (ASA-24) collected at 35 GA. Food codes provided by the Food and Nutrient Dietary Database through the ASA-24 were coded by number and type of ingredients, and grouped as unprocessed, culinary ingredient, processed and ultra-processed. Differences in PEI-UPF at 14 and 35 weeks GA were analyzed using t-tests. Logistic regression analyzed the relationships between PEI-UPF and excessive GWG, defined by sample median (7.4kg) due to the intervention effects; covariates and interaction terms were also explored. Results: There were no differences for PEI-UPF between 14 and 35 wks GA. Women with obesity and higher PEI-UPF intake were more likely to gain 7.4kg between 14 and 35 GA (OR:91.6, p=0.047, CI:1.064-7884.1) and non-white women were more likely to have excessive GWG between 14 and 35 GA (OR:2.5, p=0.023, CI:1.13-5.51). The model (p<0.001) explained 31.4% of the variance. Conclusions: Women with obesity and consuming higher ultra-processed food intake and non-white women were more likely to have excessive GWG in this cohort.

Effect of Weight Promoting Medications in a Real-Life Outpatient Setting
Monica Gobrial, LM Morselli, MD PhD Milwaukee WI, Zhuping Garacci, Srividya Kidambi, MD Milwaukee WI Background: Obesity prevalence has increased significantly in the past 50 years. We examined the effect of certain classes of medications on weight (WT) in the general population followed in out-patient clinics over a multi-year period. Methods: Electronic medical records were used to extract data (2012-2019) to identify individuals aged 18-70 years, with WT recorded within 3 months prior to starting an obesogenic medication, and with at least one WT recorded per consecutive follow-up year in an outpatient clinic. Among these, individuals prescribed and continued on a medication associated with WT gain during the follow-up period were identified (referred to as cases going forward). Patients with no obesogenic medication prescription throughout the follow-up years served as controls. Results: 8169 individuals (34% obese) were identified, of which 4515 (40% obese) were started on an obesogenic medication. The remaining (n=3654, 26% obese) were controls. Baseline (BL) and subsequent WTs of cases were 86±22 kg (BL), 86±22 (year 1, n=4515), 88±23 (year 2, n=687), 89±23 (year 3, n=279), and 90±19 (year 4, n=133). BL and subsequent WTs of controls were 81±22 kg (BL), 80±20 (year 1, n=3654), 80±20 (year 2, n=1206), 79±19 (year 3, n=574), and 79±20 (year 4, n=327). WT increases during each follow-up year among cases were significantly higher compared to controls after adjusting for baseline WT (p<0.0001). Evaluation of subgroups of medications after adjusting for BL WT and follow-up year, there was statistically significant weight increase with psychotropic medications (p<.0001), reproductive hormones (p<.0001), as well as cardiovascular medications (p=0.0001) compared to controls. Conclusions: In this multiyear follow-up study, several medication classes resulted in sustained WT gain compared to individuals not taking these medications after adjusting for BL weight.
Temporal associations of certain classes of medications with WT gain should be monitored and alternatives that are WT-neutral should be considered.

Effectiveness of Pharmacotherapy for the Treatment of Obesity in an Urban, Safety Net Population
Eric Kim, BA San Francisco CA, Zoe Cheng Orinda CA, Maria Gutierrez, RD, Diana Alba, MD, Sarah Kim, MD San Francisco CA Background: Published obesity interventions for racial minorities and low-income populations are limited and predominantly involve diet and exercise. Additionally, minorities are largely underrepresented in studies of pharmacotherapy for weight loss. Thus, data supporting weight loss medication treatment in racially diverse, low income populations is lacking. In 2015, in the San Francisco safety net, we established a Weight Management Clinic to treat class 2-3 obesity. Medical therapy, primarily phentermine + topiramate (PT) and/or a glucagon-like peptide-1 agonist (GL), was offered as an option to patients. Methods: We conducted a retrospective chart review of patients seen between 11/2015 and 6/2019 who took medical therapy (PT, GL, or GL plus PT) for at least 90 days and had at least 1 year of follow up. Paired t-tests were used to assess weight, blood pressure (BP) and A1C change. ANOVA was used to compare baseline differences between groups. Results: 72 received medications: GL (n=30), PT (n=32), or GL+PT (n=10) for average of 13.9 +/-8.2 months. At baseline, 76% were female, average age was 49 +/-11yr, and average BMI 45 +/-9 kg/m2. Primary languages were: 33% Spanish, 63% English, 1% Arabic. Self-reported race/ethnicity of patients were: 58% Hispanic, 28% White, 8% Black, 4% Asian, and 1% Pacific Islander. GL recipients were significantly older, but BMI and BP were similar across treatment groups. Percent weight loss at one year was significant with all treatments: GL 5.5% (p=0.0003), PT 9.5% (p<0.0001) and GL+PT 7.6% (p=0.002). HbA1c decreased by 0.4% (p=0.03) in the GL and GL+PT groups. There was no significant BP change in any group. Conclusions: Social determinants of health such as low socioeconomic status and race are associated with higher rates of obesity. Despite this, our findings suggest that GLP-1 agonists and phentermine + topiramate combination are still effective weight loss medications in a diverse, multilingual safety net population.
weight Oral 041 Methods: Twenty adults without diabetes spent 4 continuous weeks as inpatients at the NIH Clinical Center where they were fed ad libitum either an animal-based, LC diet (75% fat, 10% carbohydrates, 15% protein) or a plant-based, LF diet (75% carbohydrates, 10% fat, 15% protein) for 14 days each, in random order. Participants were served three meals daily and additional snacks amounting to 200% of their calculated energy requirements. Subjects were instructed to eat as much or as little as they desired. Body weight was measured using a calibrated scale and body fat and fat-free masses were measured using dual-energy X-ray absorptiometry, respectively. Results:

Effects of Ad Libitum Low Carbohydrate Versus Low Fat Diets on Body Weight and Composition
We enrolled 9 women and 11 men, aged (mean±SE)  Background: Among U.S. adults, many view obesity as matter of personal choice and responsibility. In 2013 the AMA ruled obesity is a complex chronic disease, but skepticism remains. Bias that denies the physiology of obesity and blames persons with obesity (PwO) harms patients. The present study aimed to detect possible shifts in public opinion about obesity and blame directed at PwO. Methods: A total of U.S. 7,076 adults completed online surveys -3,530 in Nov 2017 and 3,546 in May 2020. Each respondent answered only one question -either a multiple choice question to describe the type of problem they believe obesity to be (personal problem of bad choices; community problem of bad food and inactivity; a medical problem; or none of the above) or a 5-point Likert scale item to indicate their agreement regarding whether or not a) obesity is the fault of a PwO; b) obesity is a problem because PwO are blamed for the disease instead of receiving needed medical help; or c) they would interview a PwO for a job. Results: Between 2017 and 2020, there was little change in participants' views that obesity is a matter of personal choice or responsibility; 35% of respondents endorsed this view in 2017, versus 34% in 2020. However, more respondents in 2020 agreed with the narrative that PwO need less blame and more medical help, which increased from 30% to 42%. Respondents were nominally less likely to assign blame to PwO (a drop from 31% in 2017 to 25% in 2020) and more likely to agree that obesity is not their fault (up from 11% in 2017 to 16% in 2020). Finally, 66% of respondents in 2020 indicated they would interview PwO for a job, up from 58% in 2017. Conclusions: Personal responsibility remains a dominant view of obesity among U.S. adults, despite its classification as a disease by the AMA. However, fewer respondents in 2020 explicitly blame PwO, deny the need for medical care, or say they would discriminate against PwO in employment. Oral 047

Food Insecurity and SNAP Affect Food Reinforcement for Cookies and Fruit in Low-income Preschoolers
Sally Eagleton, MS University Park PA, Jennifer Temple, PhD Buffalo NY, Jennifer Savage, PhD University Park PA Background: Food insecurity (FI) and the relative reinforcing value (RRV) of food, an index of motivation to eat, have been associated with child obesity risk. Evidence suggests that FI may increase RRV in adults. This study examined whether the RRV of high-energy-dense (cookies) and low-energy-dense (fruit) foods is higher in FI children and children in households receiving Supplemental Nutrition Assistance Program (SNAP) benefits. Methods: 130 Head Start preschoolers completed the RRV task; height and weight were measured to calculate BMI%ile. Caregivers reported on household and child-level FI and SNAP participation. For the RRV task, children pressed one button to earn cookies and one button to earn fruit on concurrent independent progressive-ratio schedules (work to obtain a reward doubles after each reward is earned). RRV was operationalized as the maximum schedule reached (pmax) for each food; a higher pmax score reflects greater motivation to earn that food. Analysis of covariance examined RRV by weight status and the effect of FI and SNAP on RRV before and after adjusting for age, sex, BMI, and classroom. Results: 41% of households were food insecure, 22% reported child-level FI, and 77% of families received SNAP benefits. 25% of children had overweight and 17% had obesity. Children with obesity had higher cookie pmax compared to children without obesity (LSMEAN±SE; obese: 5.79±0.44; nonobese: 4.82±0.20, p=0.05). In unadjusted models, child-level FI was associated with lower fruit pmax (p=0.048), and remained marginally significant in adjusted models (p=0.07). Household FI was not associated with cookie or fruit pmax. However, among food secure households, when exploring SNAP participation as a moderator, cookie pmax was higher for children in households receiving SNAP benefits (p=0.01). Conclusions: Food insecure children were less motivated to earn fruit whereas children receiving SNAP benefits in food secure households were more motivated to earn cookies. Oral Abstracts viewed food and non-food pictures, in a fed and fasted state. DNA was extracted from saliva and FTO rs9939609 (A=risk allele) was genotyped (AA n=14, AT n=43, TT n=24). Whole brain analyses (posterior probability>98.75%, k 3 10) with control for age, gender and SES composite score. Results: In the fasted condition, the AA vs. TT group showed greater activation to food vs non-food cues in middle occipital gyrus, primary visual cortex, middle temporal gyrus and orbitofrontal cortex, while the AT vs. TT group showed greater activation in putamen, primary visual cortex and cerebellum. In the fed condition, the AA vs. TT group showed decreased activation in middle frontal and inferior parietal cortex, while the AT vs. TT group showed increased activation in angular gyrus, lingual gyrus, and insula. Within-group comparisons of fed and fasted conditions revealed greater frontocingulate activation in the fed condition for the TT group. Conclusions: Our results support influence of FTO variants on satiety-related modulation of neural responses to food cues among diverse adolescents.

Oral 050
Functional Architecture of Central Amygdala Neuropeptide Networks During Feeding Andrew Hardaway, PhD Birmingham AL Background: Obesity affects more than a third of US adults and is an important risk factor for cardiovascular disease, cancer, and type II diabetes. Brain systems dedicated to promoting and reducing food intake mediate the imbalance between energy intake and expenditure. Many brain circuits in the hypothalamus and hindbrain are critical for setting hunger or satiety states, shifting homeostatic need to drive caloric deficit or surplus, whereas limbic circuits in the amygdala are critical for suffusing food with motivational information. The central amygdala, a conserved output amygdala subdivision, expresses multiple opioid neuropeptides including prepronociceptin. Methods: Using laboratory mice, we used in vivo fiber photometry with the calcium indicator GCaMP7 to study how the central amygdala is activated during operant palatable food consumption. In addition, we used a combination of anatomical, electrophysiological, chemogenetics and, optogenetic techniques to determine how central amygdala neurons that express prepronociceptin(PnocCeA), a precursor to the opioid-like neuropeptide nociceptin, are activated by and functionally contribute to palatable food consumption and reward. Results: Our data demonstrate that the central amygdala is activated during both appetitive and consummatory behavior and that PnocCeA neurons are activated during palatable food consumption. Cell ablation and chemogenetic experiments demonstrated that PnocCeA are required for the consumption of palatable food specifically. PnocCeA neurons send axonal projections to the ventral BNST, PBN, and NTS and activation of these outputs is sufficient to produce reward-like behavior and positive valence. Conclusions: The CeA is a critical brain nucleus for feeding that is activated by discrete epochs of feeding behavior. PnocCeA neurons are a novel target substrate for the development of treatment for obesity and binge eating.

Oral 051
Health Background: High levels of red meat consumption are associated with increased risk of cardiovascular disease and obesity, are also damaging to the environment. Health warning messages are effective for reducing consumption of products like sugary drinks but have not been applied to red meat. This study explored potential health and environmental warning messages to reduce red meat consumption. Methods: A national convenience sample of US red meat consumers (n = 1,199; mean age 45 years) completed an online survey in 2020. Participants were randomized to view a series of either health or environmental warning messages (between-subjects factor) about the risks associated with eating red meat. Messages were presented in random order (within-subjects factor; 8 health messages or 10 environmental messages). Participants rated each warning message on a validated 3-item scale for perceived message effectiveness (PME), ranging from 1 (low) to 5 (high). Participants then rated their intention to reduce their red meat consumption in the next 7 days. Results: Health warning messages elicited higher PME ratings than environmental messages ( Health warnings may be more effective than environmental messages at discouraging red meat consumption. Additional research will clarify consumer reactions to these warnings in retail environments.

High Fat Diet Attenuates Stress-Induced Hypophagia and Activation of Glp1r+ Lateral Septum Neurons
Michelle Bales, PhD Nashville TN, Samuel Centanni, PhD, Julio Ayala, PhD Nashville TN Background: Stress in humans can promote overconsumption of calorically dense substances which can lead to obesity. Similarly, stress typically promotes hypophagia in chow-fed rodents, but this can be attenuated when rodents are given access to a high fat diet (HFD) or sucrose. The aim of this study is to identify whether HFD interferes with mechanisms that promote stress hypophagia. Activation of the glucagon-like peptide-1 receptor (Glp1r) in the lateral septum (LS) is required for restraint stress hypophagia in lean mice. Restraint stress stimulates neuronal activity in unidentified neurons in the LS, and access to sucrose attenuates LS neuronal activity and hypophagia in response to restraint stress. Therefore, we hypothesize that LS Glp1r neurons are activated following restraint stress, but this activation is impaired in mice fed a HFD, thus contributing to attenuated stress hypophagia. Methods: Food intake and LS Glp1r neuronal activity were measured in response to 1h restraint stress in lean and HFD-fed mice. Food intake was measured in C57Bl/6J mice using a Sable Systems Promethion. LS Glp1r neuronal activity was measured via fiber photometry in Glp1r-Cre mice targeted with a Cre-dependent GCaMP sensor to the LS. Results: Lean mice showed the expected hypophagic response following restraint stress (RS) compared to non-restrained controls (NRS; N=12/group; p<0.001). There was no significant difference in food intake between RS and NRS in HFD-fed mice (N=12/group; p>0.9999). Restrained lean mice had significantly higher neuronal activity than NRS lean mice (N=4/group; frequency of spikes with an amplitude >2.91 SD from baseline [z-score]; p=0.04; average amplitude z-score; p=0.015) and there was no significant difference between neuronal activity in RS vs. NRS HFD-fed mice (N=4/group; spike frequency; p=0.111; amplitude; p=0.121). Conclusions: In sum, reduced LS Glp1r neuronal activity during stress could contribute to impaired stress hypophagia in HFD-fed mice.

Oral 053
High Background: Adopted preference and intake of high fat over low fat foods is an evolutionarily conserved mechanism for promoting growth and development. However, the brain's natural responses to foods of varying caloric load represent a major hurdle to dieting and maintaining a healthy body mass index in our current obesogenic environment. Agouti-related peptide (AgRP)-expressing neurons of the arcuate nucleus of the hypothalamus (ARC) and dopaminergic (DA) neurons of the ventral tegmental area (VTA) are critical regulators of the homeostatic and hedonic motivational drives of food intake, respectively. Methods: We used access to a 60% high fat diet (HFD) to induce devaluation of home cage standard diet (SD) chow followed by a forced abstinence from HFD. Using a longitudinal design and in vivo fiber photometry, we recorded the activity of ARC AgRP neurons and VTA DA neurons in response to SD and HFD exposure. ARC AgRP responses to orexigenic and anorectic agents, as well as gastric infusion of calories, were also measured using fiber photometry. We assessed HFD-induced effects on functional ARC AgRP and VTA DA neuron feeding drives through optogenetic stimulation. Results: HFD exposure blunts homeostatic and hedonic neural responses to SD, even after food-deprivation, and these effects persisted beyond palatable food withdrawal. Similarly, endogenous responses of ARC AgRP neurons to the hunger hormone ghrelin were diminished following HFD, highlighting combinatory dysregulation of feeding drives through altered responses to the sensory detection of food as well as to the body's peripheral cues of hunger. Furthermore, this devaluation could not be rescued through optogenetic activation of either orexigenic ARC AgRP or reward-encoding VTA DA neurons. Conclusions: These data reveal a neural basis behind the challenges of making healthy eating decisions during dieting after high fat diet exposure.

Oral 054
High Reciprocally, this inflammation is known to induce further hyperphagia, exacerbating weight gain. However, it remains unknown whether and how HFD-induced inflammation affects hypothalamic appetite-regulating neurons, such as melanin-concentrating hormone (MCH) neurons. Thus, we sought to determine whether inflammatory factors could influence the electrical activity of MCH neurons during HFD feeding. Methods: Rats were fed a standard chow or an HFD for up to 14 weeks and then in-vitro patch clamp electrophysiology or immunohistochemistry was performed. Results: We found that HFD induces a robust activation of MCH neurons compared to chow, characterized by increased cell excitability. This activation occurred before the onset of excess weight gain and worsened with time. Furthermore, the degree of excitation was not correlated with caloric intake, body weight or fat mass. These results suggest that MCH neuron activation is not secondary to obesity. We also determined that the HFD effect is mimicked by prostaglandin E2 (PGE2), an inflammatory mediator, while reversed by inhibition of cyclooxygenase-2 (COX2), an enzyme responsible for PGE2 production. Further, antagonizing the PGE2 EP2 receptor (EP2R), but not other PGE2 receptors, blocks both the PGE2-and HFD-induced activation of MCH neurons. Immunohistochemistry found that EP2R is localized on MCH neurons whereas COX2 is expressed primarily by microglia. Together, this suggests that HFD induces microglial PGE2 production in the hypothalamus, which activates MCH neurons via EP2R. Conclusions: Our study shows that HFD increases an inflammatory mediator in the hypothalamus that can directly activate appetite-promoting neurons. As MCH neurons promote food intake and weight gain, this mechanism may contribute to obesity.
Background: Objective: How the impact of the COVID-19 stayat-home orders have impacted physical, mental and financial health among those with obesity is unknown. The aim of this study was to explore the health implications of COVID-19 among adults with obesity. Methods: Methods: A retrospective medical chart review identified patients with obesity from an obesity medicine clinic and a community-based metabolic and bariatric surgery (MBS) practice. Patients completed an online survey that assessed COVID-19 status, and physical and mental health behaviors during stay-at-home orders. Logistic regression models examined the impact of these orders on anxiety and depression by ethnic group. Background: Current recommendations suggest effective obesity treatment requires ≥ 26 hours of lifestyle modification over 6-12 months. However, implementing treatment at this intensity is difficult, particularly in rural and low-income communities. Partnerships between primary care pediatricians and parks and recreation departments provide both clinical care and community-based family activity opportunities. We examined a strategy for implementation of such partnerships for delivery of child obesity treatment. Methods: We used a mixed-methods, hybrid type 3 implementation-effectiveness study in eight North Carolina counties varying in geography and demographics. We measured program referrals, attendance, engagement, fidelity, and adaptation, and individual-level behavior and health changes, using program data and individual survey responses. Results: All eight sites successfully implemented their partnership, and 1,193 patients referred by primary care to their community-based program (mean age 12 years). The partnerships identified high-needs patients: 63% had severe obesity (30% class II, 33% class III); 47% were Black and 9% other race; 32% identified as Hispanic; 43% reported food insecurity. Of referred patients, 20% attended a community program session at least once (n=241 children), with 85% continuing attendance. Among engaged participants (those attending 2+ sessions), 32% met the 26-hour treatment recommendation. The mean attendance was 16.5 participants (5.7 patients, 6.1 parents, and 5.3 siblings) each session, and sessions included indoor and outdoor activities, as well as Obesity | VOLUME 28 | SUPPLEMENT S2 | NOVEMBER 2020 www.obesityjournal.org Oral Abstracts nutrition programming. Conclusions: A clinic-community partnership can effectively deliver recommended obesity treatment and engages a wide range of populations. Partnerships are consistently able to meet the USPSTF recommendations for treatment contact time. A comprehensive implementation strategy can support the development of clinic-community partnerships in rural and underserved communities lacking specialty care. . Diabetes was not associated with risk of ICU transfer or mortality and did not significantly interact with obesity to impact either clinical outcome. Conclusions: Patients with obesity had poorer outcomes even after controlling for prognostic factors. The increased mortality among patients with obesity was not explained by increased level of inflammation nor metabolic derangements as evidenced by Type 2 diabetes. Studies are needed to define the mechanism of the effect. Hypotheses include preponderance of adipose tissue, infiltration of ACE2 receptors in adipose tissue, and obesity hypoventiliation.

Influence of Adiposity and Fat Distribution on Insulin Sensitivity by Race/Ethnicity in US Children
Lauren Fowler, PhD Birmingham AL, Jose Fernandez, PhD Birmingham AL, Barbara Gower, PhD Birmingham AL Background: African-American (AA) and Hispanic-American (HA) children are at a disproportionately high risk for obesity and Type 2 Diabetes relative to their European-American (EA) counterparts. While the adverse effects of increased overall adiposity and unfavorable body fat distribution on insulin sensitivity (SI) have been well-studied in adults, racial/ethnic differences in their contributions to SI have not been clearly defined in pediatric populations. The objective of this study was to determine whether obesity-related predictors of SI differed by race/ethnicity in a sample of 84 EA, 79 AA, and 74 HA non-diabetic children. Methods: SI was assessed with a frequently-sampled intravenous glucose tolerance test and minimal modeling, while body composition was measured using dual-energy X-ray absorptiometry. Multiple linear regression was used to compare associations of SI with anthropometric and body composition measures, adjusted for age, sex, and pubertal stage. Results: Results indicated that body mass, BMI percentile, percent body fat, total fat, and leg fat were significantly associated with SI in EA, AA, and HA (P<0.05). In contrast, waist hip ratio and ratio of leg fat to total fat independently predicted SI in HA (P<0.001 for both associations), but not EA or AA. Relative to EA, all associations with SI were stronger in HA (Pinteraction<0.05), while associations did not differ between EA and AA (Pinteraction>0.10 for all). Conclusions: These results suggest that in pediatric populations, measures of overall adiposity and fat mass may be stronger determinants of SI in HA, while deleterious effects of upper body adiposity and relatively low leg fat on SI may be unique to HA. These findings can aid in early identification of HA children who are at increased risk of Type 2 Diabetes due to obesity.

Influence of Familial Obesity Risk and Current Overweight on Brain Structure in Adolescents
Gita Thapaliya, PhD Baltimore MD, Liuyi Chen Baltimore MD, Afroditi Papantoni, BS Chapel Hill NC, Susan Carnell, PhD Baltimore MD Background: Obesity is familially-transmitted, via genetic and environmental pathways. Obesity in adolescents, and familial obesity risk in adults, have been associated with brain structure alterations. To identify neural correlates of familial obesity risk without confounding by current weight, we compared currently lean adolescents varying by maternal weight, as well as currently obese/ overweight adolescents.  19F). Participants underwent MRI scans to acquire T1-weighted anatomical images. Regional grey and white matter volume (GMV, WMV) were assessed using voxel-based morphometry (VBM), and cortical thickness (CT) using the computational anatomy toolbox (CAT12), in Statistical Parametric Mapping (SPM) software. All 3 measures were corrected for age and sex, with additional total intracranial volume correction for GMV+WMV. Results: The lean-HR (vs. lean-LR) group demonstrated structural alterations in postcentral gyrus (somatosensory cortex; GMV+WMV loss, cortical thinning), opercular cortex (taste cortex; GMV+WMV loss), ACC (GMV loss, cortical thinning), and precuneus (cortical thinning). Comparisons of the lean and obese/overweight groups showed further alterations in postcentral gyrus, posterior cingulate gyrus and right middle temporal gyrus. Conclusions: Structural brain alterations associated with Oral Abstracts familial obesity risk could predict future obesity and are potential targets for neurobehaviorally-informed obesity prevention.

Insulin Enhances Presynaptic Glutamate Release Via Opioid Receptor-Mediated Disinhibition
Carrie Ferrario, PhD Ann Arbor MI, Tracy Fetterly, PhD Ann Arbor, MI MI, Max Oginsky, PhD, Allison Nieto, Yanaira Alonso-Caraballo, PhD Background: Insulin influences activity in brain centers that mediate food-seeking, reward and motivation in humans. However, nothing is known about how insulin influences excitatory transmission in regions like the nucleus accumbens (NAc) which govern food-seeking behavior and motivational processes in the adult brain. Further, insulin dysregulation that accompanies obesity is linked to cognitive decline, depression, anxiety, and aberrant motivation that also rely on NAc excitatory transmission. Methods: Using a combination of whole-cell patch clamp and biochemical approaches we determined how insulin affects NAc glutamatergic transmission in adult male rats without and with obesity. Obesity was induced by 8 weeks of free access to 60% high fat diet (Open Source Diets D12492). Results: In non-obese males, we show that insulin receptor activation increases presynaptic glutamate release via a previously unidentified form of opioid receptor-mediated disinhibition. In contrast, activation of IGF receptors by insulin decreases presynaptic glutamate release in adult male rats. Furthermore, obesity results in a loss of insulin receptor-mediated increases and a reduction in NAc insulin receptor surface expression, while preserving reductions in transmission mediated by IGRFs. Conclusions: These results provide the first insights into how insulin influences excitatory transmission in the adult brain, they provide foundational information about opioid-mediated regulation of NAc glutamatergic transmission, and have broad implications for the regulation of motivation and reward related processes by peripheral hormones that influence weight gain and obesity. Background: Current theories indicate that food reward largely impacts weight regulation over the lifespan. Cross-sectional studies support that both high food preference and/or weak taste sensitivity relate to increased acute meal intake. However, the impact of preference and taste sensitivity on dietary intake patterns and BMI change over time is largely understudied. We hypothesize that weak taste sensitivity will be associated with increased taste preference, increased dietary intake and higher weight over time. Methods: We tested longitudinal relationships between sugar and fat preference, sweet taste and fat sensitivity, and self-reported habitual dietary intake, on BMI percentile in a sample of 105 healthy-weight adolescents (baseline: BMI%tile 57.7±24.1; age 14-16 y; 55.2% female) using repeated measures mixed effects models. For four years, annual assessments included: taste sensitivity via triangle fat and sugar discrimination tests; taste preferences via visual analog scale for four milkshakes that varied in sugar and fat content, and dietary intake, as assessed by the reduced Block Food Frequency Questionnaire (BFFQ). Significance was set at p < 0.05. Results: Weaker taste sensitivity to sugar was associated with higher preference for a high-sugar milkshake (p<0.001) throughout the 4-year period. Moreover, weaker sugar sensitivity at baseline predicted increases in BMI%tile over time (p=0.03). Percent fat consumed per day was positively associated with high-fat/low-sugar milkshake preference (p=0.021), and was negatively associated with low-fat/ low-sugar milkshake preference over 4 years (p=0.047). Conclusions: Results support theories that the combination of weaker sugar taste sensitivity and high preference for sweet may both contribute to overeating and that preference for fat parallels consumption patterns over time. In sum, the interrelations between taste sensitivity, preference, and food intake patterns provide a deeper understanding of food-reward related predictors of obesity, relative to food reward alone. Oral 067

Long-term Impact of the Philadelphia Beverage Tax on Prices and Purchases at Independent Stores Sara Bleich, PhD Boston MA
Background: This study examined whether Philadelphia's 1.5 cent-per-ounce excise tax on sugar-and artificially-sweetened beverages had a sustained impact on beverage prices, beverage purchases, and high sugar food purchases two years after tax implementation, at small, independent stores. Methods: We used a difference-indifferences approach to compare changes in beverage prices, beverage purchases, and high sugar food purchases (candy, sweet snacks) at independent stores in Philadelphia and Baltimore (a non-taxed control) before tax implementation and two years after tax implementation. The sample included 130 stores with price data and 6,495 customer purchases. We measured the change in: 1) mean price (¢/oz) of taxed and non-taxed beverages, 2) purchases (ounces) of taxed and non-taxed beverages, and 3) total calories from beverages and high sugar foods. Results: Compared to Baltimore, taxed beverages in Philadelphia had a price increase of 1.91¢/oz (127.2% tax pass-through; 95%CI:1.65 to 2.16, p<.001) two years after tax implementation, while non-taxed beverages had no statistically significant price change. Purchases of taxed beverages declined by 7.48 ounces or 50.5% decline (95% CI:-11.4 to -3.6, p<.001) in Philadelphia compared to Baltimore, and there was no significant change in the purchase of non-taxed beverage ounces. Absolute declines in taxed beverage purchases were larger among customers shopping in low-income neighborhoods and individuals with lower education levels (-9.20oz and -5.79oz). Total calories from sugary beverages and food declined significantly by 88 calories or -33.4% overall (95%CI:-154 to -21, p=.020) and by 118 calories or -41.0% (95%CI: -209 to -27, p=.022) among customers shopping in low-income neighborhoods in Philadelphia compared to Baltimore. Conclusions: Beverage excise taxes may be an effective policy tool to sustainably decrease purchases of sweetened drinks and calories from sugar in independent stores, particularly among populations at higher risk for SSB consumption.
Background: There is currently little research documenting the content of marketing inside of quick service restaurants (QSRs) in the United States. The purpose of this study was to examine marketing techniques currently used in QSRs that may influence children, and whether these techniques differ in lower-income and/or minority neighborhoods compared with higher income and/or predominantly white neighborhoods. Methods: Using standard mystery shopper methodology, trained RAs visited n=165 QSRs (33 from each of the top five national QSR chains) located throughout New England in communities that varied by income (% below the poverty line) and race/ethnicity. Marketing techniques inside of the QSRs were examined using a validated environmental assessment tool. Mixed methods ANOVA was used to examine differences in the number and types of techniques promoting healthy or less healthy options by income and race/ethnicity, with QSR chain as a random effect and adjusting for size of the QSR (i.e. number of seats) and community demographics. Results: Overall, only 6.5% of QSRs had healthy ads (range 0-1 ads) while 94.7% of QSRs had unhealthy ads (range 0-14 ads) that were visible while waiting in line to order. Over half (55%) of QSRs had ads that promoted overeating (e.g. supersizing [range 0-3 ads]). Overall, there were no significant differences in the number ads by community income. However, there were more ads with cartoon and movie characters in communities with a greater percent minority population; each 20% increase in the percent of the population that was Black and/or Hispanic was associated with a 0.4 (p=0.03) increase in the number of ads with cartoon characters and Obesity www.obesityjournal.org Obesity | VOLUME 28 | SUPPLEMENT S2 | NOVEMBER 2020 27 Oral Abstracts a 0.6 (p=0.01) increase in the number of ads with movie characters. Conclusions: There is currently a substantial number of ads promoting less healthy food items in QSRs. Additionally, QSRs may have a tendency to market more unhealthy foods using child-directed ads in minority communities. , and 20 controls with overweight (OW) (n=20, age:36.0±14.2y, BMI:29.5±2.6 kg/ m 2 ), at baseline and 3 months post-surgery (RYGB, VSG) or post-baseline (OW). Whole brain group-level analyses, using Bayesian posterior inference, were applied to contrast images generated from first-level analyses of food vs. non-food activation from baseline to 3 months post-surgery/baseline for RYGB patients vs. OW, and VSG vs. OW (posterior probability > 98.75%, extent threshold k >= 10). Results:

Metabolic Adaptation Delays Time to Reach Weight Loss Goals
The RYGB (vs. OW) group showed markedly decreased post-surgical activation to food cues in lingual gyrus, medial prefrontal cortex, precentral gyrus and middle temporal gyrus. The VSG (vs. OW) group showed a pattern of decreased post-surgical activation to food cues in precentral gyrus and precuneus together with increased activation in inferior parietal lobule and postcentral gyrus. High-calorie food cue wanting ratings decreased in RYGB and VSG groups. Conclusions: Our findings support differential impacts of RYGB and VSG on food cue responses 3 months post-surgery. These differences may be attributable to differing mechanisms of action and could explain divergent outcomes of RYGB and VSG.

Oral 072
No Effect of Ultra-processed Food Intake on Neonatal Lean or Fat Free Mass Measured by QMR or PEAPOD Kathryn Whyte, PhD, RD New York NY, Isobel Contento, PhD, CDN, Xavier Pi-Sunyer, MD MPH New York NY, Dympna Gallagher, EdD New York NY Background: LIFT (Lifestyle Intervention for Two) trial found that intervening in women with overweight and obesity through behaviors promoting healthy diet and physical activity to control gestational weight gain (GWG) resulted in neonates with similar fat and greater lean mass at birth. Whether these measurable effects are significantly associated with ultra-processed food (UPF) intake defined by the NOVA guidelines, was the focus of this investigation. Methods: Neonatal body composition was assessed by Infant QMR and PEAPOD at birth (n=129). UPF is presented as a percentage of energy intake (PEI-UPF) and was derived from the Automated Self-Administered-24 (ASA-24) collected at week 35. Food codes provided by the Food and Nutrient Dietary Database via ASA-24 were assigned a number 1 through 4 to one of the NOVA guidelines categories according to number and type of ingredients as: (1) unprocessed, (2) culinary ingredient, (3) processed, and (4) ultra-processed. Linear regression was used to determine relationships between PEI-UPF and neonatal lean mass (LM; QMR) and fat-free mass (FFM; PEAPOD Background: Mounting evidence suggests beverage taxes reduce sweetened beverage purchases, but data from real-world taxes on whether consumers consequently increase purchases of snacks, beverage concentrates, or alcoholic beverages are lacking. We examined whether Philadelphia's 1.5 cent-per-ounce beverage tax was associated with substitution to three kinds of hypothesized substitutes: 1) snacks (candy, sweet snacks, salty snacks), 2) non-taxed beverage concentrates, and 3) alcohol (wine, spirits). Methods: Using commercial retail sales data and a difference-in-differences approach, we tested changes in logged sales of snacks and beverage concentrates between 2016 (pre-tax) and 2017 (post-tax) at chain retail stores in Philadelphia (n=180) and Baltimore (non-taxed control city; n=60), and in logged volume of alcohol at liquor stores in Philadelphia (n=44) and nearby Pennsylvania counties (alternate control; n=66). Analyses were stratified by store type because tax pass-through varies 43-104% by store type. Additional analyses of the snacks and beverage concentrates tested logged volume changes and triple differences compared to control products. Results: Across store type, there were no statistically significant increases in logged sales of snacks or volume of wine and spirits in Philadelphia compared to control stores. Analyses showed some substitution to non-taxed beverage concentrates in supermarkets (24% increase in volume and 8% increase in sales relative to other food), despite supermarkets having the lowest tax pass-through rate. Conclusions: This is the first real-world study addressing whether sweetened beverage taxes lead to substitution with snacks, beverage concentrates and alcohol. At the population level, we see no evidence that Philadelphia's decline in taxed beverage purchases is offset by increases in snacks, or wine and spirits purchasing, but there is some evidence of substitution to beverage concentrates in supermarkets. Future studies should explore individual-level purchasing changes. Background: Ghrelin, when administered, potently increases food intake. Yet, high plasma ghrelin is not routinely associated with obesity. Rather, low plasma ghrelin occurs most often in settings of obesity. Indeed, obesity-associated reduction of plasma ghrelin, together with ghrelin resistance, have been proposed as potential processes employed to limit even greater obesity. In the current study, we investigated a potential mechanism underlying obesity-associated reduction of plasma ghrelin. In particular, we tested whether hyperinsulinemia during obesity acts via ghrelin cell-expressed insulin receptors to suppress ghrelin release. Methods: We generated ghrelin cell-selective insulin receptor-knock out (GIRKO) mice by crossing ghrelin-Cre mice with floxed-insulin receptor mice. Four wk-old male GIRKO and wild-type littermates were fed 42% high fat diet (HFD) for 16 wks. We assessed body weights and food intake weekly and blood glucose and plasma ghrelin, insulin, and LEAP2 (the ghrelin receptor antagonist) monthly. Results: HFD raised plasma insulin in both GIRKO and WT mice (63% increase by 8 wks-of-age in both groups). While plasma ghrelin fell in WT mice, it was higher in GIRKO mice (by 188 ± 29% at 4 wks-of-age; by 221 ± 55% at 8 wks-of-age) and did not fall over time. Despite the markedly elevated plasma ghrelin in GIRKO mice, their body weights, food intake, and blood glucose were similar to those of WT mice. Plasma LEAP2, which is a proposed contributor to ghrelin resistance, was similar in GIRKO and WT mice. Conclusions: These data suggest that insulin, acting via ghrelin cell-expressed insulin receptors, is a key determinant of plasma ghrelin levels, especially in the setting of obesity in which it acts to restrict ghrelin secretion. Despite the elevated plasma ghrelin resulting from ghrelin cellselective insulin receptor deletion, body weight, food intake, and blood glucose remain unaffected in this experimental setting, suggesting persistence of ghrelin resistance.

Obesity-Associated Plasma Ghrelin Reduction Is Mediated by Ghrelin Cell-Expressed Insulin Receptors
Background: Controlling child feeding practices, restriction and pressure to eat, influence the development of adolescent obesity and eating-related behaviors. Adolescents from households at lowincome are at heightened risk for obesity and may be particularly susceptible to obesogenic eating patterns due to greater availability of high energy dense foods.
Research has yet to investigate how the parent and adolescent perspective of feeding practices may be differentially related to weight and eating behaviors. The present study sought to examine the association between controlling child feeding practices (restriction and pressure to eat) and adolescent weight and obesogenic eating patterns among adolescents from a lowincome background. Methods: Parent and adolescent (13-17 yrs) dyads from low-income households (N=73) participated in a larger study (R03HD079509). Objective height and weight measured. Adolescents and their parents completed questionnaires assessing eating patterns and perception of parental use of controlling feeding practices, restriction and pressure to eat. Results: In bootstrapped linear regression models, parent-report restriction [95% CI(3.84, 18.08)] and pressure to eat (-14.42, -.04) were significantly associated with adolescent weight. Adolescent report of restriction was significantly associated with obesogenic eating patterns, emotional (.09, 2.56) and external eating (.15, 3.19). Adolescent report of pressure to eat was significantly associated with both emotional (.32, 2.79) and external eating (.04, 2.56). Parent and adolescent report of restriction and pressure to eat were not significantly associated. Conclusions: Results suggest parent perception is more salient in association with adolescent weight status, while adolescent perception is more salient in association with adolescent obesogenic eating patterns. Feeding practices and adolescents' perception of these feeding practices could serve as a future target for intervention with adolescents from low-income households.
Oral 077 Background: Obesity is a chronic, multifactorial, heterogeneous disease associated with variable response to treatments. Our multidisciplinary team tested the hypothesis that pathophysiological characterization elucidates obesity subgroups and enhances response to obesity therapy. Methods: Obesity phenotypes measured were by body composition (DEXA), resting energy expenditure (Indirect calorimetry), satiety (appetite score by visual analog scale and gastric emptying by scintigraphy), satiation (kcal intake at ad libitum buffet meal), hedonic eating behavior (questionnaires Eating Inventory, WEL, HADS), and physical activity questionnaires. These variables were used to classify obesity phenotypes and to guide anti-obesity medications in a pragmatic 12-month trial comparing phenotype-guided treatment to standard approach. Results: A total of 570 patients with obesity have been phenotyped to date [(mean ± SEM): age 39 ± 0.5 years old, BMI 37 ± 0.3 kg/m 2 , 72% females]. Using a priori determined cutoff of the 75th percentile for the measured variables, we identified the prevalence of each obesity phenotype: hungry brain -abnormal satiation (32%), emotional hungerhedonic eating (21%), hungry gut -abnormal satiety (32%), slow burn -abnormal metabolism (21%). Two or more phenotypes were documented in 27% of patients (mixed) and 15% in had no specific phenotype. In a 268-participant pragmatic trial, there was no statistical difference in the demographics, comorbidities, percentage of type of medications-used or adverse events in the two treatment cohorts. Phenotype-guided obesity pharmacotherapy resulted in 79% of patients achieving clinically significant weight loss at one year (>10%) compared to 35% of patients in the non-phenotype approach (p<0.001); and mean weight loss at 12 months was 16.1% in the phenotype-guided group compared to 9.2% in the non-phenotype-guided group (p<0.001). Conclusions: Biological and behavioral phenotypes of obesity identify mechanisms that reduces obesity heterogeneity and can be targeted to enhance weight loss.

Phenotype-Guided Compared To Standard Use of Pharmacotherapy Enhances 1 Year Weight Loss In Obesity
Background: Behavioral lifestyle interventions are the foundation of adolescent obesity treatment. Tailoring an intervention using adolescent stakeholder engagement during the development process could improve intervention effectiveness. Methods: Adolescents ages 14-19 with BMI ≥85th percentile-for-sex-and-age participated in focus groups and were asked for their preferences regarding components of a weight loss program. A codebook was generated based on transcript emergent themes and concepts. Coded transcripts were analyzed in MAXQDA2020. Results: Ten sex-specific focus groups (F=6, M=4) were conducted with 41 adolescents (F=28, M=13). Adolescents wanted an instructor with prior weight loss experience who was relatable. Females had more concerns about participating in a weight loss program and preferred group sessions to be sex-stratified, which some even felt should extend to the instructor. Both sexes preferred parental involvement to be optional, as some parents were described as helpful, while others were perceived as a hindrance to participant success. Males and females identified incentives, engaging activities, and electronic communication (e.g., reminders and group chats) as core components for program engagement, with females placing emphasis on socializing and relationship building. Although adolescents acknowledged nutrition is an important factor in weight loss, discussions about desired program components focused primarily on exercise. Adolescents advised that programs avoid a focus on "weight loss," and instead emphasize "healthy lifestyle," which represents a more comprehensive goal of targeting both physical and mental well-being for participants. Conclusions: Females expressed more concerns about attending a weight loss program, preferred to be in same-sex intervention groups, and emphasized a communal program focus centered around friendship and support. Interventions for adolescents with overweight/obesity may be more engaging if desired program characteristics are incorporated.
Oral 079 Background: High obesity rates in children coincide with increased incidence of non-alcoholic fatty liver disease (NAFLD). Understanding the genetics underlying susceptibility to NAFLD could guide intervention strategies. The Protein Kinase A (PKA) signaling pathway is central in regulating energy balance, glucose homeostasis and lipid metabolism, and its dysfunction has been seen in related abnormalities. We hypothesized that variants in PKA subunit genes may be associated with NAFLD susceptibility. Methods: 457 youths from the Yale Obesity Clinic were incorporated into the Pathogenesis of Youth Onset Diabetes (PYOD) cohort (NCT01967849), where MRI determination of hepatic fat fraction (HFF%), oral glucose tolerance testing (OGTT), anthropometric and biochemical tests were conducted. An HFF% of >5.5% was diagnosed as NAFLD. Exome sequencing of germline DNA for PKA subunit genes was performed at the NICHD Molecular Genomics Core; variants were confirmed by Sanger sequencing. FRET and PKA enzymatic assays were performed in transfected HEK293 cells. In silico structural analysis was done. Clinical data were analyzed accounting for NAFLD and BMI z-score. Results: We identified the PRKAR1B p.R115K variant in 4 nonrelated, obese or overweight African American subjects, all without NAFLD. Variant frequency was about 70 times that of the African American population and was associated with favorable circulating lipoprotein levels, e.g. increased large LDL (p=0.003), compared to other African Americans without NAFLD. FRET analysis showed stronger interaction between the R1β mutant and catalytic subunit Cα compared to wildtype (p<.001) and basal PKA enzymatic activity was decreased (p<.001). Structural analysis potentially shows tighter binding of cAMP at one cAMP binding site. Conclusions: The data, with the limitation of small numbers, suggest that the PRKAR1B p.R115K variant may favorably modulate lipoprotein profile in African American youth and protect them from developing NAFLD whether they are obese or not.

Oral Abstracts
Clearer ways to identify specific barriers for individual patients, and provide support to overcome those, will likely be needed to improve retention in weight management. The relationship between poverty and obesity poses a significant risk across the lifespan as both obesity and wealth inequality in the US continue to climb. This study investigated potential mechanisms for this relationship by examining the effects of experimentally manipulated financial gains and losses among parents and children from diverse socioeconomic backgrounds. Methods: A stratified sample of 106 families (53 children aged 7-10, 53 adolescents aged 15-17), from both high-and low-income households, visited our laboratory for three separate appointments. Each appointment included the experimental manipulation of financial gains and losses and computer-based DD and RRV tasks. Linear mixed models were used to examine responses to the manipulation among those with and without food insecurity and those receiving/not receiving public assistance in terms of DDT and RRV, controlling for covariates. Results: Among parents, those reporting food insecurity were significantly more sensitive to the manipulation and showed and increase in their RRV of food in response to a financial loss (p=.028). Among adolescents, those reporting food insecurity had an overall higher RRV of food (p=.045), but did not respond to the manipulation. In terms of DD, parents receiving any type of government assistance showed an increase in DD in response to manipulated financial gains (p=.020). Adolescents receiving government assistance had and trend for higher DD (p=.077), but this did not change in the response to the manipulation. Children, regardless of background, responded to the manipulation by decreasing DD in response to financial gains (p=.023). Conclusions: These results highlight the consistent relationship between food insecurity and RRV of food among adults and adolescents. These results also suggest that receipt of government assistance may affect DD over time and differentially across the lifespan.

Oral 085
The Impact of Obesity The adverse association of obesity with productivity needs to be considered by employers in designing employee health benefits.

Oral 087
The OPTIFAST Very Low Calorie Diet (VLCD) Achieves Significant Type 2 Diabetes Remission Evan Keller, BS Pittsburgh PA, David Rometo, MD Pittsburgh PA Background: A Very Low Calorie Diet using meal replacements has been shown to achieve T2DM remission. We present here our retrospective analysis of T2DM remission, euglycemia on medication, and glycemic control on medication, and their association with % weight loss (WL) in an OPTIFAST 6 month program. Methods: The program consists of 13 weeks of meal replacement (~800 kcal/day), followed by 3 months of food re-introduction. Background: Pulmonary function is compromised in 40-80% of obese individuals. Respiratory muscle weakness clinically contributes to this phenomenon; but its pathophysiology is not well defined. We previously used non-invasive ultrasound (US) to show that long-term diet-induced obesity (DIO) progressively impairs diaphragm motion in mice.
Ex vivo isometric force testing demonstrated that contractile dysfunction intrinsic to the muscle corresponded with changes seen on US. Intra-diaphragmatic adipocyte number and polymerized collagen content increased with DIO; and both parameters quantitatively correlated with contractile force deficits. All adipocytes and many collagen-depositing cells in the obese diaphragm arose from mesenchymal fibroadipogenic progenitors (FAPs). Thrombospondin-1 (THBS-1) is a circulating matricellular protein that increases with obesity, functions as a mesenchymal cell mitogen and has been linked to muscle injury in inflammatory myopathies. We hypothesized that THBS-1 drives FAPmediated diaphragm remodeling and contractile dysfunction in obesity.

Methods:
To test the hypothesis, we compared effects of a 6-month DIO time course on diaphragm physiology (US, isometric force measurement), anatomy, and gene expression in wild-type (WT) and Thbs1 knockout (Thbs1-/-) mice. Results: In WT mice, diaphragm excursion amplitude, inspiratory velocity and expiratory velocity (measured on US) progressively declined, reaching 50% of baseline after 6 months DIO. These parameters were unchanged in Thbs1-/-mice despite similar weight gain. Isometric specific force was also higher in diaphragm isolates from Thbs1-/-versus WT animals. Moreover, Thbs1-/-mice had fewer intra-diaphragmatic adipocytes and reduced expression of genes encoding adipocyte markers and collagen species. Finally, in primary culture, THBS-1 treatment induced FAP proliferation and collagen deposition. Conclusions: These data highlight THBS-1 as a novel FAP mitogen and mediator of obesity-associated respiratory muscle dysfunction. Background: Effective obesity prevention and treatment interventions are lacking in the United States, especially for impoverished youth at risk for health disparities. Schools are recognized as a good place to intervene, yet schools rarely implement evidence based programs. Methods: Using a pragmatic quasi-experimental design, we report on the effectiveness of adding structured nutrition education and a minimum requirement for physical activity to standard middle school on-site afterschool programming. Programming was evaluated over three school years (2016,2017,2018) to determine: 1) the impact on healthy weight and target healthy behavior attainment; and 2) whether target behavior improvement was associated with academic achievement. Results: Of 76 students enrolled, 62% were Hispanic, 46% were girls, 80% had overweight or obesity (54% had obesity or severe obesity). There were positive, non-significant trends in the proportion of students with improvements in four elements: BMI Z-score, fruit and vegetable consumption, sugar free beverages consumption, and unhealthy snack food consumption. Significant improvements in academic achievement were associated with improved vegetable consumption (p=0.013) and decreased fast food consumption (p=0.054), and improvements in academic achievement were not significantly associated with school lunch consumption, sugar free beverage consumption, and physical activity. Sleep behavior improvement was significantly associated with BMI Z-score decrease (Beta=-0.05, 95% CI (-0.1, -0.003), p=0.038).

Conclusions:
Enhancing existing afterschool programming with health oriented elements demonstrates trends toward improvement in several health behaviors, weight outcomes, and academic achievement. Research with greater sample size will allow these promising strategies to bring us closer to solving issue of obesity in impoverished communities. Of the 74 abstracts screened, 6 were excluded; of the 68 left, 60 were excluded, because these imaged parts of the body other than the brain (27/60); were not about obesity/diabetes (24/60); did not involve imaging (5/60) or were not an meta-analysis of imaging (5/60). This yielded 8 imaging meta-analyses of obesity (6/8) or Type 2 Diabetes (2/8). Results: In the 6 neuroimaging meta-analyses of obesity, we found the majority of studies conducted in the US reported the sex and the age of their sample (41/43, 95%). However, only 17/43 studies (40%) reported the racial diversity of their sample. The proportion of African-Americans represented in these studies (16%) greatly under-represented the rate of obesity in African-Americans in the USA (46.8%). In the two meta-analyses examining Type 2 Diabetes, only 1 of 14 US studies (7%) reported the racial diversity of their sample, with 1/14 studies reporting a predominately (82.2%) Caucasian sample. This sampling greatly underrepresented the rate of Type 2 Diabetes in American Indian/Alaska Natives (14.7%), Hispanics (12.5%), African-Americans (11.7%) and Asian Americans (9.2%), overrepresenting non-Hispanic whites (7.5%). Conclusions: The lack of reporting of the racial diversity of samples in neuroimaging studies included in meta-analyses of obesity and Diabetes Mellitus, Type 2, continue to fail to reflect the reality of these rates across diverse groups in the USA. This pilot and feasibility study has 2 important findings: 1) MRI can be used to obtain reliable and reproducible measures of BAT in non-sedated infants and 2) MRI measurement of BAT provides physiologically relevant data. Infants with a lower FF (indicating greater metabolic activity of BAT) had more FFM which reflects the increased need for heat production due to less insulating FM. MRI provides an accurate methodology to quantify BAT in infants and allows for further understanding of BAT and EE in humans.

Verification of Keratinocyte-Associated Protein 3 as a Novel Gene for Adiposity and Body Weight
Alexandria Szalanczy Winston Salem NC, Osborne Seshie, Emily Goff Denton TX, Michael Grzybowski, MS Milwaukee WI, Aron Geurts, PhD, Leah Solberg Woods, PhD NC Background: Obesity poses a significant health burden in the United States, and investigating obesity-causing genes remains an important area of research. A previous study from our lab used outbred heterogeneous stock rats to identify Keratinocyte-associated protein 3, Krtcap3, as a candidate gene within a quantitative trait locus for adiposity. Increased expression of Krtcap3 correlated with decreased retroperitoneal fat pad weight (RetroFat). Here we seek to confirm that Krtcap3 expression affects adiposity traits. Methods: We used CRISPR/Cas9 to develop an in vivo whole-body knock-out rat model of Krtcap3. Heterozygotes were bred to obtain wild-type (WT) and knock-out (KO) rats. Experimental rats were placed onto a high-fat (HFD; 60% kcal fat) or low-fat diet (LFD; 10% kcal fat) at six weeks of age and were maintained on the diet for 13 weeks, with weekly measures of body weight and food intake. From 10-12 weeks on diet we measured body composition, energy expenditure, and glucose tolerance, with one test a week. We hypothesized that Krtcap3 KO rats will have increased adiposity and a worse metabolic phenotype. Results: We found that male and female KO rats have significantly increased body weight at six weeks of age. Body weight differences are also noted over time, with the strongest effect in females on HFD. KO females on HFD show significantly increased fat mass and lean mass relative to WT females. Surprisingly, there are no differences in fat mass between WT and KO male rats, although there is a trend toward increased lean mass in KO on HFD. Both KO males and females on HFD, but not LFD, exhibit trends toward increased food consumption. There are no differences in energy expenditure and glucose tolerance between WT and KO rats independent of sex or diet. Conclusions: These studies support a role of Krtcap3 in regulating body weight and fat mass, particularly in females under HFD conditions. Future studies will investigate the mechanisms by which Krtcap3 is acting.

VTA Amylin-Induced Attenuation of Motivated Feeding Involves Prefrontal Cortex Projections
Caroline Geisler, PhD Philadelphia PA, Jane Gaisinsky, BS PA, Arianna White, Heath Schmidt, PhD, Matthew Hayes, PhD Philadelphia PA Background: A better understanding of the neural mechanisms regulating satiety in response to highly palatable foods is essential to improve dysregulated feeding behavior in obesity. The β-cell derived hormone amylin induces postprandial satiation and improves energy homeostasis through central actions at multiple nuclei including the ventral tegmental area (VTA). The VTA projects to the prefrontal cortex (PFC) where information about food reward contributes to high-level decision making. Accordingly, we hypothesize that activation of VTA-to-PFC neurons underlies amylin-mediated decreases in motivated feeding behaviors. Methods: To test this hypothesis we used a dual viral strategy to knockdown (KD) expression of the core component of the amylin receptor, calcitonin receptor (CTR), specifically in VTA-to-PFC projecting neurons. 12-week old male rats received prelimbic PFC injections of a retrograde Cre-expressing AAV, and VTA injections of either a control AAV or a Cre-dependent shRNA-CTR that resulted in CTR KD in VTA-to-PFC projecting neurons. Rats were maintained on a choice diet with access to both chow and 60% high fat diet (Research Diets). To directly test motivated feeding behavior we assessed progressive ratio (PR) lever presses for sucrose pellets in classic operant conditioning boxes. Results: In non-food deprived rats, CTR KD increased the number of sucrose pellets earned during a PR test. Biweekly recordings of 48h ad libitum choice food intake revealed that 9 weeks post-virus, CTR KD increased HFD and decreased chow intake. Lastly, CTR KD attenuated the acute (1h) hypophagic response to intraperitoneal amylin (2µg/kg) and the CTR agonist, salmon calcitonin (2µg/kg). Conclusions: These result support that endogenous CTR signaling in VTA-to-PFC neurons decreases motivation and consumption of rewarding foods. Importantly, amylin signaling is preserved in obesity, supporting the hypothesis that enhancing central amylin signaling is an attractive therapeutic strategy to treat hyperphagia in obesity. Background: To describe weight outcomes of the NU-HOME program, a family-meal focused randomized controlled trial (RCT) designed to prevent child obesity in rural communities. The NU-HOME study is an innovative and collaborative effort of academic researchers, health systems researchers and interventionists and community leaders. Methods: The rural RCT included 114 child (M=8.9 yrs [SD=1.0], 59% girls, 93% white; 46% overweight/obese) and parent (M=37.8 yrs [SD=5.3], 97% women, 97% white, 76% overweight/obese) dyads. In 2017-2018 and 2018-2019, research staff measured height, weight and body fat at baseline (BL) and post-intervention (PI). Families were randomized to intervention (n=58) or control (n=56) groups. Designed with Social Cognitive Theory and community engagement, the NU-HOME program included seven monthly sessions and four goal-setting calls focused on family nutrition education and meal planning and preparation skills and was delivered in community settings. Multiple linear and logistic regression models tested PI (n=98, complete data) child BMIz-score, percent body fat, and overweight/obesity status by treatment group, adjusted for BL values and demographics. Similar analyses were run with four additional participants with interim weight data that could be carried forward for missing PI values (n=102). Results: Although in the right direction, no statistically significant treatment effects were seen for child BMIz or overweight/obesity status. However, a promising reduction in boys' percent body fat (-2.1%, 95% CI (-4.84, 0.63)) was associated with the intervention. Conclusions: Making significant impacts on child BMIz and overweight/obese status remains challenging in community settings. BL levels of weightrelated outcomes in children are strong drivers of future weight status. Intensive family-focused programming may require extensive societal, policy and environmental interventions to mitigate complex factors that lead to excess child weight gain.

Oral 98
Weight Research among adults suggests that weight stigma is associated with poor cardiometabolic health and inflammatory markers. However, little is known about these relationships among youth. Therefore, we examined associations between weight-based teasing (WBT) and metabolic and inflammatory markers among two samples of youth: 1) a non-treatment-seeking sample across the weight spectrum and 2) a weight loss treatment-seeking sample with obesity. Methods: Fasting weight, height, adiposity, waist circumference, and blood pressure were measured. Fasting blood samples were collected for metabolic (triglycerides, HDL-C, and glucose) and inflammatory analytes (erythrocyte sedimentation rate in both studies and high-sensitivity C-reactive protein in study 1 only). Youths completed a questionnaire assessing history of perceived WBT. Metabolic and inflammatory indices were compared between those with and without reported WBT using a series of ANCOVAs and MANCOVAs, adjusting for age, sex, race, and fat mass (kg). Given the potential overlap between WBT, depression, and cardiometabolic health, all models were repeated including depressive symptoms as an additional covariate. Findings remained consistent after adjusting for depressive symptoms. Conclusions: Reported WBT was not associated with cardiometabolic health in non-treatment seeking or treatment-seeking youth. It is possible that the physical effects of WBT victimization in youth emerge in adulthood. Longitudinal research is needed to elucidate associations between WBT and health across the lifespan.

Weight Enumeration in State Anti-Bullying Laws and Risks of Weight-Based Bullying Among LGBTQ Youth
Leah Lessard, PhD Hartford CT, Ryan Watson, PhD, Hannah Schacter, PhD, Christopher Wheldon, PhD, Rebecca Puhl, PhD Hartford CT Background: Weight-based bullying is the most prevalent form of peer harassment and carries significant health consequences -particularly among adolescents who identify as a sexual and/or gender minority (SGM). Despite the increasing use of legislation to address bullying in schools, there has been limited empirical examination on the efficacy of anti-bullying laws in reducing the prevalence of weight-based bullying and its adverse health sequelae. We tested whether anti-bullying legislation that includes weight as a protected class (i.e., enumeration) was associated with lower rates and health risks of weight-based bullying for SGM adolescents. Methods: Data on weight-based bullying and health risk were collected from the 2017 LGBTQ National Teen Survey (N=17,112; Mage=15.57, SDage=1.27) and linked to state anti-bullying legislation (coded based on whether weight was, or was not, enumerated) obtained from the U.S. Department of Health and Human Services. Multilevel analyses were conducted. Results: There was substantial variation across states in the rates of weight-based bullying (range=49%-86%). Three states had anti-bullying laws that provided special protections for individuals based on weight (i.e., weight enumeration). Weight-based bullying occurred less frequently, but was associated with greater health risk (i.e., more unhealthy eating behaviors, stress, and lower perceived health) for SGM adolescents in states five years to three years, and WTP over $12,000 to reduce the wait from five years to a year or less. Simulation modelling showed that one quarter (26%) of the adult population living with severe obesity are willing to pay to some degree for surgery with one in five (18%) WTP up to $20,000 to undergo bariatric surgery in Canada within 12 months.
Conclusions: This study demonstrates that while the majority of Canadians are willing to wait longer to have bariatric surgery provided through the publicaly funded health care system, one in five Canadians affected by obesity are willing to pay a significant amount rather than wait in the public system. Given the reduced access to bariatric surgery as a treatment for severe obestiy in Canada, it may be viable to expand the provision of this service offered through private clinics for those willing to pay in order to reduce wait times and positively impact the health of Canadians.

Oral 103
Effect of Obesogenic Medication on Weight-and Fitness-Change Outcomes: Evidence from the Look AHEAD Study Rena C. Moon Orlando FL, and Zakaria Almuwaqqat Atlanta GA Objective: This study evaluates whether obesogenic medications may decrease the effectiveness of lifestyle interventions. The authors of this study hypothesized that participants who took obesogenic medications would be less responsive to the intervention in the Look AHEAD trial. Methods: In the trial, 5,145 participants with overweight or obesity, aged 45 to 76 years with type 2 diabetes, were randomly assigned to an intervention (vs. support and education). In this analysis, the association of exposure to obesogenic medications and successful weight loss (≥5% and ≥10% of total weight) and fitness gain (≥1 and ≥2 metabolic equivalents) was examined. For each outcome, multiple logistic regression models were fitted. Results: Analytic sample sizes were 4,496 for weight-change analyses and 4,051 for fitness-change analyses. After adjusting for covariates, exposure to one or more obesogenic medications significantly decreased the odds of achieving ≥5% weight loss by 32% (odds ratio [OR] 0.68) and achieving ≥10% weight loss by 19% (OR 0.81). The association was dose-dependent-participants using two or more medications were less likely to achieve weight loss than those using one medication. Obesogenic medication exposure was not associated with decreased odds of achieving fitness gain overall.

Conclusions:
The results suggest that exposure to obesogenic medications could hinder successful weight loss in a lifestyle intervention for people with diabetes. Obesity www.obesityjournal.org Obesity | VOLUME 28 | SUPPLEMENT S2 | NOVEMBER 2020 39

Oral Abstracts
Preoperative screening is recommended, but some screening tools lack validation, and their relative performance is unclear in this population. The study objective was to compare the ability of four existing tools (STOPBANG, NO-OSAS, No-Apnea, and