Background: Radiological imaging has improved our insight into how obesity and sarcopenia impacts acute pancreatitis via several measured variables. However, we lack understanding of the association between social determinants of health and these variables within the acute pancreatitis population. Methods: This study included patients at a single tertiary care center between 1 January 2008 and 31 December 2021. Measurements of visceral adiposity (VA), subcutaneous adiposity (SA), the ratio of visceral to total adiposity (VA/TA), and degree of sarcopenia via psoas muscle Hounsfield unit average calculation (HUAC) were obtained on CT scans performed at presentation. Using geocoded patient data, we calculated the social vulnerability index (SVI) from CDC metrics. Descriptive and regression analyses were performed utilizing clinical and radiological data. Results: In 484 patients with 592 acute pancreatitis-related hospitalization, median (IQR) VA was 176 (100–251), SA was 209.5 (138.5–307), VA/TA ratio was 43.5 (32.3–55.3), and HUAC was 51.3 (44.4–58.9). For our primary outcome, geospatial analyses showed a reverse association between VA and SVI with a coefficient of −9.0 (p = 0.04) after adjustment for age, health care behaviors (i.e., active smoking and drinking), and CCI, suggesting residence in areas with higher SVI is linked to lower VA. However, VA/TA, SA, and HUAC showed no significant association with SVI. The SVI subdomain of socioeconomic status had significant association with VA (−39.78 (95% CI: −75.88–−3.70), p = 0.03) after adjustments. For our secondary outcome, acute pancreatitis severity had significant association with higher VA (p ≤ 0.001), VA/TA (p ≤ 0.001), and lower HUAC (p ≤ 0.001). When comparing single vs. recurrent hospitalization patients, there was significantly higher median VA with recurrences (VA-single acute pancreatitis: 149 (77.4–233) vs. VA-recurrent acute pancreatitis: 177 (108–256); p = 0.04). Conclusions: In this study we found that patients residing in more socially vulnerable areas had lower visceral adiposity. This paradoxical result potentially conferred a protective effect against severe and recurrent acute pancreatitis; however, this was not found to be statistically significant.

The BIDMC Optimization & Standardization for Treatment Outcomes in New England for Helicobacter Pylori (BOSTON-HP) study included 1002 patients with confirmed Helicobacter pylori (HP) via gastric histopathology from 2016-2022 to evaluate the effects of a Multidisciplinary Division Wide Program. Our previous data showed that demographic factors, such as race/ethnicity and preferred language, are associated with treatment, test-of-cure, and eradication outcomes. We hypothesize that given an unlikely even geographic distribution of demographic factors, HP outcomes may cluster geographically. The BOSTON-HP cohort patient addresses were transformed to georeferenced coordinates which were mapped using QGIS (3.30 's-Hertogenbosch) onto OpenStreetMap. We calculated the maximum Euclidean distance matrix to assign heatmap weights for each outcome, and calculated global spatial autocorrelation to evaluate the extent of spatial clustering. Figure 1. BOSTON-HP Treatment Cohort. Figure 2. Untreated in the BOSTON-HP Cohort. A moderate and significant spatial correlation existed between the likelihood of HP treatment and the geographic location of the patient’s residence. Identification of groups for targeted quality improvement interventions (RLRD) and models of success within the system (SCMC) can help improve treatment outcomes and reduce geographic disparities. Javier Villafuerte-Gálvez, MD, Milky Way Biosciences: Partial Research Support.

Background: Clostridioides difficile infection (CDI) imposes a burden that extends well beyond the gastrointestinal tract, yet existing outcome measures only partially capture the patient experience. We used frontier large language models (LLMs) on patient and caregiver narratives at scale to describe how burden shifts with disease course. Methods: We analyzed 189 testimonials from the Peggy Lillis Foundation corpus, sorted into four cohorts with recurrence ( r ) and fulminant ( f ) severity as axes (rfCDI, fCDI, rCDI, non-rfCDI). Two independent LLMs coded eight thematic domains, four fulminant flags, thirteen emerging semantic fields, the dominant dimension, and narrative arcs. Two clinicians independently coded a subset for inter-rater reliability (PABAK, Gwet’s AC1). Results: Treatment trajectory was the dominant theme in recurrent disease, whereas death and near-death dominated non-recurrent fulminant narratives. Psychological burden was near-universal in fulminant disease (98.0% in rfCDI, 97.2% in fCDI). Caregiver and bereavement content concentrated in fCDI (66.7%). Diagnostic failure was frequent across recurrent cohorts (47.6 - 56.1%). Bacteriotherapy tracked recurrence (60.2% rfCDI versus 5.6% fCDI). Financial, mental-health, and caregiver burdens were prominent and are currently unaddressed by guidelines. Human-human reliability was substantial (PABAK 0.79 for semantic fields, 0.76 for domains); arc coding was least reliable. Conclusions: Patient narratives reveal a course-dependent, multidimensional burden in CDI. Concrete gaps exist between what patients prioritize, what guidelines recommend, and what therapy access provides. Frontier-LLM coding, validated against clinicians, offers a reproducible route to translate these priorities into research, care, and policy. Key Points: Patient narratives reveal a course-dependent CDI burden: recurrence organizes around treatment, fulminant disease around survival, and bereavement. Psychological, financial, and caregiver domains important to patients remain largely unaddressed by current guidelines.

Background: Clostridioides difficile infection (CDI) imposes a burden that extends well beyond the gastrointestinal tract, yet existing outcome measures only partially capture the patient experience. We used frontier large language models (LLMs) on patient and caregiver narratives at scale to describe how burden shifts with disease course. Methods: We analyzed 189 testimonials from the Peggy Lillis Foundation corpus, sorted into 4 cohorts with recurrence (r) and fulminant (f) severity as axes (rfCDI, fCDI, rCDI, non-rfCDI). Two independent LLMs coded 8 thematic domains, 4 fulminant flags, 13 emerging semantic fields, the dominant dimension, and narrative arcs. Two physicians independently coded a subset for inter-rater reliability (PABAK, Gwet's AC1). Results: Treatment trajectory was the dominant theme in recurrent disease, whereas death and near-death dominated nonrecurrent fulminant narratives. Psychological burden was near-universal in fulminant disease (98.0% in rfCDI, 97.2% in fCDI). Caregiver and bereavement content concentrated in fCDI (66.7%). Diagnostic failure was frequent across recurrent cohorts (47.6–56.1%). Mentions of bacteriotherapy (fecal microbiota transplantation) concentrated in recurrent cohorts (60.2% rfCDI versus 5.6% fCDI). Financial, mental-health, and caregiver burdens were prominent and remain largely unaddressed by guidelines. Human-human reliability was substantial (PABAK 0.79 for semantic fields, 0.76 for domains); arc coding was least reliable. Conclusions: Patient narratives reveal a course-dependent, multidimensional burden in CDI. Concrete gaps exist between what patients prioritize, what guidelines recommend, and what therapy access provides. Frontier-LLM coding, validated against clinicians, offers a reproducible route to translate these priorities into research, care, and policy.