Predicting serious postoperative complications and evaluating racial fairness in machine learning algorithms for metabolic and bariatric surgery.

BACKGROUND: Predicting the risk of complications is critical in metabolic and bariatric surgery (MBS). OBJECTIVES: To develop machine learning (ML) models to predict serious postoperative complications of MBS and evaluate racial fairness of the models. SETTING: Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) national database, United States. METHODS: We developed logistic regression, random forest (RF), gradient-boosted tree (GBT), and XGBoost model using the MBSAQIP Participant Use Data File from 2016 to 2020. To address the class imbalance, we randomly undersampled the complication-negative class to match the complication-positive class. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC), precision, recall, and F1 score. Fairness across White and non-White patient groups was assessed using equal opportunity difference and disparate impact metrics. RESULTS: A total of 40,858 patients were included after undersampling the complication-negative class. The XGBoost model was the best-performing model in terms of AUROC; however, the difference was not statistically significant. While the F1 score and precision did not vary significantly across models, the RF exhibited better recall compared to the logistic regression. Surgery type was the most important feature to predict complications, followed by operative time. The logistic regression model had the best fairness metrics for race. CONCLUSIONS: The XGBoost model achieved the highest AUROC, albeit without a statistically significant difference. The RF may be useful when recall is the primary concern. Undersampling of the privileged group may improve the fairness of boosted tree models.

Investigate-Design-Practice-Reflect: An Iterative Community-Engaged Action Process to Improve Population Health.

BACKGROUND: Community-based coalitions are a common strategy for community engagement efforts targeting the improvement of a variety of population health outcomes. The typical processes that coalitions follow to organize efforts include steps that are sequential, slow, and time intensive. These processes also limit local decision-making to the selection of evidence-based policies or programs. METHODS: We present a process control theory-based Community Action Process, Investigate-Design-Practice-Reflect (IDPR), where community hubs (i.e., coalitions) organize agile efforts in a non-sequential, rapid, and efficient manner to harness local assets and data to make decisions regarding the provision and production of population health services. Using qualitative methods, we illustrate and analyze the use of IDPR in a one community case study as part of Wellscapes, a Type 3-hybrid implementation-effectiveness community randomized controlled trial to improve children's population health physical activity. RESULTS: We found community members followed the IDPR Community Action Process to rapidly design, organize, deliver, and receive feedback on a community-based, children's population physical activity prototype, an afterschool Play-in-the-Park opportunity for all children. DISCUSSION: Following IDPR afforded the community coalition timely learning through feedback within a process that coordinated decisions regarding what community services met community needs (provision decisions) and how to organize the production of the population health services (production decisions).

Inequality in Physical Activity in Organized Group Settings for Children: A Cross-Sectional Study.

BACKGROUND: Adult-led organized settings for children (eg, classrooms) provide opportunities for physical activity (PA). The structure of setting time may influence inequalities (ie, unequalness) in the distribution of PA. This study examined differences in PA inequality by setting and time-segment purpose in time-segmented organized group settings for children. METHODS: PA and setting meetings were assessed using accelerometer and video observation data from school, before-/after-school, and youth club groups (n = 30) for third- through sixth-grade children (n = 699) in 2 rural US communities. Meetings (n = 130) were time-segmented into smaller units (sessions; n = 835). Each session was assigned a purpose code (eg, PA). Accelerometer data were paired with the meetings and sessions, and the Gini coefficient quantified inequality in activity counts and moderate to vigorous PA minutes for each segment. Beta generalized estimating equations examined differences in PA inequality by setting and session purpose. RESULTS: Activity count inequality was lowest (P < .05) during youth club meetings (Gini = 0.17, 95% CI, 0.14-0.20), and inequality in moderate to vigorous PA minutes was greatest (P < .01) during school (Gini = 0.34, 95% CI, 0.30-0.38). Organized PA sessions (Gini = 0.20, 95% CI, 0.17-0.23) had lower activity count inequality (P < .0001) than academic (Gini = 0.30, 95% CI, 0.27-0.34), enrichment (Gini = 0.31, 95% CI, 0.27-0.36), and nonactive recreation (Gini = 0.30, 95% CI, 0.25-0.34) sessions. Inequality in moderate to vigorous PA minutes was lower (P < .05) in organized PA (Gini = 0.26, 95% CI, 0.20-0.32) and free play (Gini = 0.28, 95% CI, 0.19-0.39) than other sessions. CONCLUSIONS: PA inequality differed by setting time structure, with lower inequality during organized PA sessions. The Gini coefficient can illuminate PA inequalities in organized settings and may inform population PA improvement efforts in rural communities.

Inequality in Children's Physical Activity Across Time-Segmented Youth Sport Practice.

Purpose: Youth sport (YS) is a community system for promoting children's physical activity (PA). Studies have examined mean PA during YS practices, but few have examined inequalities in the distribution of PA among children during practice time. This study examined PA inequality in time-segmented YS practices and differences in inequality by time segment characteristics. Methods: Children's PA and YS practices were examined using accelerometer and video observation data from a sample of YS teams (n = 36 teams, n = 101 practices) for third- through sixth-grade children (n = 392), approximately eight to 12 years old, in two rural U.S. communities. Practices were time-segmented into smaller units (episodes; n = 991). Episodes were assigned codes for purpose (e.g. warm-up), member arrangement (e.g. whole group), and setting demand (i.e. fosters participation, creates exclusion). Group accelerometer data were paired with episodes, and the Gini coefficient quantified inequality in activity counts and minutes of moderate-to-vigorous PA (MVPA). Beta generalized estimating equations examined the influence of episode structure on PA inequality. Results: Warm-up (Gini = 0.22), fitness (Gini = 0.24), and sport skill (Gini = 0.24) episodes had significantly lower inequality (p < .05) in activity counts than other purpose types. Management (Gini = 0.32) and strategy (Gini = 0.40) episodes had significantly greater inequality (p < .05) in MVPA minutes than other purpose types. Episodes fostering participation (Gini = 0.32) had significantly lower activity count inequality (p < .05) than episodes creating exclusion (Gini = 0.35). Conclusion: PA inequality among children during YS varied by practice structure. Metrics such as the Gini coefficient can illuminate inequalities in PA and may be useful for guiding efforts to improve population PA in children. Trial Registration: This study is registered at www.clinicaltrials.gov (Identifier: NCT03380143).

First modified Delphi consensus statement on robotic-assisted da Vinci sleeve gastrectomy.

BACKGROUND: Sleeve gastrectomy (SG) is the most commonly performed metabolic and bariatric surgery (MBS) procedure. Technical considerations related to the performance of SG are well established and reported in the literature but not in relation to robotic-assisted (RA) SG. We report the results of the first modified Delphi consensus-building exercise addressing technical considerations of RA da Vinci (dV) SG. OBJECTIVES: Develop best practices for the performance of robotic-assisted da Vinci sleeve gastrectomy. SETTING: Survey based consensus statement. METHODS: A consensus building committee (CBC) was created comprising 10 experts in the field of RA surgery and MBS based on strict selection criteria. The CBC developed 49 consensus statements which were then shared with 240 experts in RA surgery. Our stopping criterion was stability in responses (≤15%). The consensus cut point was 70%. RESULTS: The overall response rate was 49%. In the first round of voting, there was consensus agreement on 25 statements (51%), consensus disagreement on 14 (28%), and no consensus on the remaining statements (21%). In the second round of voting, we reached agreement on 3 additional statements. Experts recommended the use of the number of pauses generated by the stapler to guide choice of staple height (91.2%) and to upsize the staple height when using buttressing (92%). There was also consensus (81.4%) that the use of the closed staple height of 1.00 mm (white) is acceptable and that stapling of the antrum using a 1.5-mm staple (blue load) is also acceptable (73%). CONCLUSIONS: Collective expert opinion structured through a modified Delphi consensus statement presents a practical guide for surgeons interested in performing dV-SG.

Assessment of arterial supply to the stomach after bariatric surgery using multidetector CT arteriography.

PURPOSE: To describe residual arterial supply to the stomach after bariatric surgery via a systematic arterial-phase CT assessment approach that can aid in diagnosis and treatment of postoperative complications and facilitate planning for future procedures. METHODS: Arterial-phase CT of 46 patients who underwent Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG) at 3 academic institutions were retrospectively reviewed to assess patency of left gastric artery (LGA), right gastric artery (RGA), gastroepiploic artery (GEA), and left inferior phrenic artery (LIPA) and presence of gastric perforators. RESULTS: In 25 RYGB and 21 SG patients, mean diameters were LGA 2.2 ± 0.4 mm, RGA 1.6 ± 0.5 mm, and GEA 1.7 ± 0.4 mm. On RYGB scans, all LGAs, RGAs, and 24/25 (96%) of GEAs were identified. Excellent to good patency was seen in 20/25 (80%) LGAs, 21/25 (84%) RGAs, and 23/24 (96%) GEAs. On SG scans, all LGAs, 18/21 (86%) of RGAs, and 20/21 (95%) GEAs were identified. Excellent to good patency was seen in 17/21 (81%) LGAs, 15/18 (83%) RGAs, and 20/20 (100%) GEAs. In terms of gastric perforators, LGA supply was seen on 23/25 (92%) of RYGB and 17/17 (100%) of SG scans. RGA supply was seen on 13/21 (62%) RYGB and 9/18 (50%) SG scans. GEA supply was seen on 19/23 (83%) RYGB scans. No gastric supply via GEA was seen on SG scans. CONCLUSION: In this study, arterial supply to the stomach through the LGA was consistently identified in all RYGB and SG cases, indicating an uncomplicated surgical approach with regard to preserving the LGA. Dedicated CT angiography protocol or catheter-directed angiography is recommended for accurate and comprehensive assessment of the gastric blood supply, particularly before surgical re-intervention.

Community-engaged Systems for Population Health Improvement: A Novel Approach to Improve Diabetes Outcomes in Rural Communities.

BACKGROUND: Approaches to prevent and manage diabetes at a community population level are hindered because current strategies are not aligned with the structure and function of a community system. We describe a community-driven process based on local data and rapid prototyping as an alternative approach to create diabetes prevention and care management solutions appropriate for each community. We report on the process and provide baseline data for a 3-year case study initiative to improve diabetes outcomes in two rural Nebraska communities. METHODS: We developed an iterative design process based on the assumption that decentralized decision-making using local data feedback and monitoring will lead to the innovation of local sustainable solutions. Coalitions act as community innovation hubs and meet monthly to work through a facilitated design process. Six core diabetes measures will be tracked over the course of the project using the electronic health record from community clinics as a proxy for the entire community. RESULTS: Baseline data indicate two-thirds of the population in both communities are at risk for prediabetes based on age and body mass index. However, only a fraction (35% and 12%) of those at risk have been screened. This information led both coalitions to focus on improving screening rates in their communities. DISCUSSION: In order to move a complex system towards an optimal state (e.g., improved diabetes outcomes), stakeholders must have access to continuous feedback of accurate, pertinent information in order to make informed decisions. Conventional approaches of implementing evidence-based interventions do not facilitate this process.

Development of a Mindfulness-Based Intervention for Narcolepsy: A Feasibility Study.

STUDY OBJECTIVES: Mindfulness-based interventions (MBI) have been shown to improve psychosocial functioning in medical populations but have not been studied in narcolepsy. This study examined the feasibility and acceptability of an MBI that was adapted for narcolepsy, including three variations in program length. METHODS: Adults with narcolepsy (N = 60) were randomized to MBI groups of varying durations: brief (4 weeks), standard (8 weeks), or extended (12 weeks). Participants completed assessments at baseline, 4 weeks, 8 weeks, and 12 weeks. To assess feasibility and acceptability, primary outcomes included attendance, meditation practice, and data completeness. Additionally, participants completed measures of mindfulness, self-compassion, mood, sleep, psychosocial functioning, and cognition. An effect size of Cohen's d ≥ 0.5 was used as the pre-specified benchmark for a minimal clinically important difference (MCID). RESULTS: The attendance, meditation, and data completeness benchmarks were met by 71.7%, 61.7%, and 78.3% of participants, respectively. Higher proportions of the brief and extended groups met these benchmarks compared to the standard group. All groups met the MCID for mindfulness, self-compassion, self-efficacy for managing emotions, positive psychosocial impact, global mental health, and fatigue. Standard and extended groups met the MCID for anxiety and depression, and extended group met the MCID for additional measures including social and cognitive functioning, daytime sleepiness, hypersomnia symptoms, and hypersomnia-related functioning. CONCLUSION: Results suggest that the remote delivery and data collection methods are feasible to employ in future clinical trials, and it appears that the extended MBI provides the most favorable clinical impact while maintaining attendance and engagement in meditation practice.

Indicators of community physical activity resources and opportunities and variation by community sociodemographic characteristics: A scoping review.

Objective: This scoping review synthesizes studies examining community-level variability in physical activity resource (assets) and opportunity (organized group physical activity services) availability by community sociodemographic characteristics to describe methodologies for measuring resources/opportunities, indicators characterizing availability, and associations between community-level sociodemographic characteristics and availability. Methods: A systematic search was conducted in MEDLINE, CINAHL, PsycINFO, and Scopus for literature through 2022. Eligible studies quantitatively examined measures of physical activity resource/opportunity availability by community-level racial, ethnic, and/or socioeconomic characteristics within geospatially defined communities. Extracted data included: community geospatial definitions, sociodemographic characteristics assessed, methodologies for measuring and indicators of community physical activity resource/opportunity availability, and study findings. Results: Among the 46 included studies, community geospatial units were defined by 28 different community boundaries (e.g., town), and 13% of studies were conducted in rural areas. Nearly all (98%) studies measured community-level socioeconomic status, and 45% of studies measured race/ethnicity. A total of 41 indicators of physical activity resource/opportunity availability were identified. Most studies (91%) assessed built environment resources (e.g., parks), while 8.7% of studies assessed opportunities (e.g., programs). Of 141 associations/differences between community sociodemographic characteristics and resource/opportunity availability, 29.8% indicated greater availability in communities of higher socioeconomic status or lower prevalence of minority populations. The remaining findings were in the opposite direction (9.2%), non-significant (36.9%), or mixed (24.1%). Conclusions: Variability in physical activity resources/opportunities by community sociodemographic characteristics was not consistently evident. However, the indicators synthesized may be useful for informing population health improvement efforts by illuminating the physical and social conditions impacting population physical activity outcomes.

American Society for Metabolic and Bariatric Surgery 2022 estimate of metabolic and bariatric procedures performed in the United States.

BACKGROUND: Metabolic and bariatric surgery (MBS), despite being the most effective durable treatment for obesity, remains underused as approximately 1% of all qualified patients undergo surgery. The American Society for Metabolic and Bariatric Surgery established a Numbers Taskforce to specify the annual rate of obesity treatment interventions utilization and to determine if patients in need are receiving appropriate treatment. OBJECTIVE: To provide the best estimated number of metabolic and bariatric procedures being performed in the United States in 2022. SETTING: United States. METHODS: We reviewed data from the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program and National Surgical Quality Improvement Program. In addition, data from industry and state databases were used to estimate activity at non-accredited centers. Data from 2022 were compared mainly with data from the previous 2 years. RESULTS: Compared with 2021, the total number of MBS performed in 2022 increased from approximately 262,893 to 280,000. The sleeve gastrectomy (SG) continues to be the most commonly performed procedure. The gastric bypass procedure trend remained relatively stable. The percentage of revision procedures and biliopancreatic diversion with duodenal switch procedures increased slightly. Intragastric balloon placement increased from the previous year. Endoscopic sleeve gastroplasty increased in numbers. CONCLUSIONS: There was a 6.5% increase in MBS volume from 2021 to 2022 and a 41% increase from 2020, which demonstrates a recovery from the COVID-19 pandemic. SG continues to be the most dominant MBS procedure.

The SAGES MASTERS program bariatric surgery pathway selects 10 seminal publications on revisional bariatrics.

BACKGROUND: The Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) Masters Program designated bariatric surgery as a clinical pathway. Among the tiers of the Masters Program, revisional bariatric surgery is the highest tier of "mastery" within the pathway. This article presents the top 10 seminal studies representing the current landscape of revisional bariatrics. METHODS: The literature was systematically searched and seminal articles designated by consensus agreement of the SAGES Metabolic and Bariatric Surgery committee using multiple criteria, including impact on the field, citation frequency, and expert opinion. Articles were reviewed by committee members and presented in summarized fashion. RESULTS: The top 10 papers are presented in grouped thematic categories covering the early evolution of revisional bariatrics, changing criteria for reoperative bariatric surgery, divergence of revision versus conversion bariatric surgery, and recent technologic innovations in revisional bariatric surgery. Each summary is presented with expert appraisal and commentary. CONCLUSION: These seminal papers represent a snapshot of the dynamic field of revisional bariatric surgery and emphasize the need to not only remain current with contemporary trends but also keep a patient-oriented perspective on patient and intervention selection for optimal success.

Sociodemographic influences on youth sport participation and physical activity among children living within concentrated Hispanic/Latino rural communities.

Introduction: Lack of physical activity (PA) among children living in rural communities is a documented public health problem. Although studies have examined community conditions defined by a rural-urban dichotomy, few have investigated rural community conditions with a concentration of Hispanic/Latino people. This cross-sectional study examined sociodemographic characteristics associated with youth sport (YS) participation and daily PA among children living within concentrated Hispanic/Latino rural U.S. Midwest communities. Methods: During spring 2022, 97% of 3rd-6th grade children (n = 281, aged approximately 8-12 years) attending school in rural Midwestern communities (n = 2) with >50% concentration of Hispanic students participated in the Wellscapes Project, a community randomized trial. Participants completed the Youth Activity Profile and supplemental National Survey of Children's Health questions assessing PA behaviors and YS participation. Caregivers of a subsample of children (n = 215; males, n = 93; females, n = 122) consented to pair their child's survey results with school enrollment records (e.g., free/reduced lunch status and race and ethnicity). Mixed models with community as a random effect examined main and interaction effects of grade, sex, ethnoracial status, and family income on YS participation and these sociodemographic characteristics and YS participation on daily moderate-to-vigorous PA (MVPA). Results: Approximately half of children participated in YS. Non-Hispanic White children (n = 82) were over five times more likely to participate in YS than Hispanic peers (n = 133) (OR = 5.54, 95% CI = 2.64-11.61, p < 0.001). YS participants accumulated 8.3 ± 2.3 more minutes of daily MVPA than non-participants (p < 0.001). Sixth graders, females, and Hispanic children reported lower daily MVPA than comparison groups (p < 0.05). Significant interaction effects on daily MVPA between grade and ethnoracial status (F(3, 204) = 3.04, p = 0.030) were also found. Discussion: Disparities in sport participation and PA outcomes based on sociodemographic characteristics exist among children living in ethnoracially diverse rural communities. Strategies to promote YS participation, including community structural changes, may help reduce PA disparities. The research provides valuable insights for policymakers, public health professionals, and community members to address YS participation barriers, not limited to cost, while considering other PA-promotion efforts to improve child population health.

Predicting operative time for metabolic and bariatric surgery using machine learning models: a retrospective observational study.

BACKGROUND: Predicting operative time is essential for scheduling surgery and managing the operating room. This study aimed to develop machine learning (ML) models to predict the operative time for metabolic and bariatric surgery (MBS) and to compare each model. METHODS: The authors used the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database between 2016 and 2020 to develop ML models, including linear regression, random forest, support vector machine, gradient-boosted tree, and XGBoost model. Patient characteristics and surgical features were included as variables in the model. The authors used the mean absolute error, root mean square error, and R 2 score to evaluate model performance. The authors identified the 10 most important variables in the best-performing model using the Shapley Additive exPlanations algorithm. RESULTS: In total, 668 723 patients were included in the study. The XGBoost model outperformed the other ML models, with the lowest root mean square error and highest R 2 score. Random forest performed better than linear regression. The relative performance of the ML algorithms remained consistent across the models, regardless of the surgery type. The surgery type and surgical approach were the most important features to predict the operative time; specifically, sleeve gastrectomy (vs. Roux-en-Y gastric bypass) and the laparoscopic approach (vs. robotic-assisted approach) were associated with a shorter operative time. CONCLUSIONS: The XGBoost model best predicted the operative time for MBS among the ML models examined. Our findings can be useful in managing the operating room scheduling and in developing software tools to predict the operative times of MBS in clinical settings.

Psychometric Analysis of the Rapid Eating Assessment for Participants-Short Form to Evaluate Dietary Quality in a Pre-Surgical Bariatric Population.

Dietary quality and eating behaviors are essential to evaluating bariatric surgery candidates. The Rapid Eating Assessment for Participants-Short Form (REAP-S) is a previously validated measure of dietary quality suited for use in primary care. This study aimed to evaluate the psychometric properties of the REAP-S in a pre-surgical bariatric population. This study included data from one academic medical center from August 2020 to August 2022. Variables included socio-demographics, the REAP-S, mental health, and assessments of appetitive traits. Statistical methods included Cronbach's alpha, confirmatory factor analysis (CFA), and multivariable analyses. A total of 587 adult patients were included in this analysis. The mean score for the REAP-S was 28.32 (SD: 4.02), indicative of relatively moderate dietary quality. The internal consistency of the REAP-S was moderate, with a Cronbach's alpha of 0.65. The three-factor CFA model resulted in a comparative fit index of 0.91. Race (p = 0.01), body mass index (p = 0.01), food fussiness (p < 0.0001), food responsiveness (p = 0.005), and socially desirable responses (p = 0.003) were significantly associated with the total REAP-S score. Although the REAP-S's original purpose was to assess dietary quality within a primary care population, it shows promise for application within a bariatric surgery-seeking population.

Associations of Food Addiction Symptomatology and Disordered Eating Behaviors in a Pre-Surgical Bariatric Population.

The construct of food addiction (FA) has been highly debated in recent years particularly in the fields of disordered eating, medical weight management, and bariatric surgery. Some researchers have argued that FA symptoms are distinct, highly prevalent, and present a barrier for patients seeking medical treatment for obesity. The purpose of this study is to evaluate the cross-sectional associations between FA symptomatology, binge eating disorder (BED) and other appetitive traits, as well as dietary quality in a sample of adults with obesity seeking bariatric surgery. This post hoc analysis was conducted on a prospectively collected dataset from August 2020 to August 2022 at a single academic medical center. Descriptive statistics were used to characterize the sample. Additional analyses included: correlation coefficients, multivariable linear regression, and analysis of variance. A total of 587 patients were included in this analysis with low average scores for FA symptoms (mean: 1.48; standard deviation (SD): 2.15). Those with no BED symptoms had the lowest average FA symptoms scores (mean: 0.87; SD: 1.52) and those with both bingeing and LOCE had the highest average scores (mean: 3.35; SD: 2.81). This finding supports the hypothesis that, while related, FA and BED may represent different cognitions and behaviors.

Performance of PROMIS Measures to Assess Fatigue and Sleep Symptom Severity Among African American Patients Newly Diagnosed With Obstructive Sleep Apnea.

This pilot study examined the concurrent validity of Patient-Reported Outcomes Measurement Information System (PROMIS), Short Form, measures with the longer Multidimensional Fatigue Inventory among patients living with obstructive sleep apnea (OSA). A total of 26 African American patients living with prediabetes and newly diagnosed with OSA completed the six-item short form versions of PROMIS Fatigue and PROMIS Sleep Disturbance, and the longer 20-item Multidimensional Fatigue Inventory. Both PROMIS Fatigue and Sleep Disturbance scales demonstrated high reliability with Cronbach's α of .91 and .92, respectively. PROMIS Fatigue scores were significantly correlated with Multidimensional Fatigue Inventory scores (rs = .53; p = .006) and demonstrated concurrent validity. However, PROMIS Sleep Disturbance scores and Multidimensional Fatigue Inventory scores were not associated with one another. The brief PROMIS Fatigue scale is a useful, succinct approach to assess fatigue severity among diverse patient populations living with OSA. This study is among the first to evaluate the performance of PROMIS Fatigue in a sample living with OSA.