Scholarly Commons @ Baystate Health
Not a member yet
8474 research outputs found
Sort by
Bias and Oversight in Clinical AI: A Review of Decision Support Tools and Equity Frameworks
No full textArtificial intelligence (AI) decision support tools (DSTs) are increasingly used across clinical settings to improve efficiency and support decision-making. However, these tools risk perpetuating existing healthcare disparities if not designed and implemented with transparency, equity, and cultural sensitivity. This review explores how racial and ethnic biases manifest within AI-driven DSTs and evaluates the role of governance frameworks in mitigating such harms. It examines the implications of biased algorithms, presents case examples highlighting disparities in tool performance, and critically assesses the adequacy of current national and international regulatory guidance. The review reports that bias can stem from unrepresentative training datasets, exclusion of equity auditing in design, and the absence of mandated transparency in reporting. Although several frameworks exist to guide development and reporting, few are mandatory, and most do not include equity as a core criterion. The current UK and US regulatory models are decentralized and lack mechanisms to systematically detect or prevent bias. To prevent biased tools from entering practice, equity must be structurally embedded across the AI lifecycle. Embedding equity into AI tools requires standardized subgroup performance reporting, mandating fairness assessments, and establishing national and global governance standards to ensure AI tools serve all populations equitably.
Keywords: algorithmic bias; artificial intelligence; clinical decision support tools; digital health; ethnic disparities; healthcare inequity; racial bias
Recurrent Gastrointestinal Bleeding in Patients With Atrial Fibrillation Treated With Left Atrial Appendage Occlusion
No full textBackground: Concomitant atrial fibrillation (AF) and gastrointestinal (GI) bleeding present a clinical challenge due to recurrent bleeding risk associated with anticoagulation for AF-associated stroke prevention. Left atrial appendage occlusion (LAAO) offers an alternative stroke prevention strategy, but its impact on recurrent GI bleeding remains unknown.
Methods: This retrospective, multicenter cohort study used the TriNetX database to identify adults with AF on oral anticoagulation and a history of GI bleeding. Patients were stratified by treatment with or without LAAO. 1:1 propensity score matching (PSM) was employed. The primary outcome was recurrent GI bleeding. Cox regression analysis was used to generate hazard ratios (HRs) with 95% confidence intervals (CIs). Odds ratios (ORs) were used to evaluate effect sizes between groups. Kaplan-Meier curves were used for time-to-event analyses.
Results: After PSM, 9259 patients were compared in each group. Odds of recurrent GI bleeding were consistently lower in patients undergoing LAAO than without LAAO across all follow-up intervals: at 3 months (OR 0.84; 95% CI 0.78-0.91), 6 months, 1 year, 3 years, and 5 years (OR 0.87; 95% CI 0.82-0.92). Kaplan-Meier analysis demonstrated significantly lower risk of recurrent GI bleeding with LAAO (HR 0.80; 95% CI 0.76-0.84; p \u3c 0.01).
Conclusion: In patients with AF and prior GI bleeding, LAAO was associated with a significantly lower risk of recurrent GI bleeding at short-term and long-term time intervals.
Keywords: LAAO; anticoagulation; atrial fibrillation; gastrointestinal bleeding
Editorial: Seeing clearly now: Imaging guidance modalities for left atrial appendage occlusion
No full textGlandular Crowding of the Endometrium: Clinical Outcomes of a Diagnostic Gray Zone
No full textEndometrioid intraepithelial neoplasia (EIN) is a well-defined premalignant lesion of the endometrium, characterized by crowded glands exceeding the amount of stroma, cytologic atypia distinct from background endometrium, and a sufficiently sized lesion (often designated as 1 mm). However, some endometrial samples contain foci of crowded glands that fall short of these criteria. These diagnostically indeterminate areas, often described as atypical glandular crowding, present challenges in both interpretation and clinical management. To assess their significance, we conducted a retrospective review of endometrial specimens diagnosed with atypical glandular crowding over an 8-yr period (2016-2024) at Baystate Medical Center. Cases were identified through a free-text database search and included when at least one follow-up specimen was available within 1 yr of the index biopsy. Demographic, clinical, and histopathologic data were recorded, and outcomes were classified as benign, EIN, or carcinoma. Among 79,998 gynecologic specimens, 167 cases (0.2%) met the inclusion criteria, with 127 unique patients analyzed. Within 1 yr, 36 patients (28.3%) were diagnosed with EIN (n=31) or endometrioid carcinoma (n=5), most within 4 mo of the initial biopsy. Notably, roughly one-quarter of these cases arose within endometrial polyps, and their risk of EIN or carcinoma (~25%) was similar to that of nonpolypoid lesions. Thus, atypical glandular crowding carries a substantial short-term risk of a premalignant or malignant diagnosis on follow-up, with ~1 in 4 patients being diagnosed with EIN or carcinoma within a year. Recognition of this subdiagnostic but clinically significant pattern should prompt timely re-sampling, as early follow-up often reveals occult neoplastic lesions not captured in the initial biopsy.
Keywords: Atypical glandular crowding; Endometrial carcinoma; Endometrioid intraepithelial neoplasia (EIN)
Improving Pediatric Concussion Management in the Primary Care Setting
No full textBackground and objectives: Pediatric concussion is a common condition, yet limited standardization in its evaluation and management within primary care-where most patients initially present-can result in delayed recovery, prolonged symptoms, inconsistent guidance on returning to school and play/sport, and increased disparities in care. We aimed to increase to 70% the proportion of patients with concussion presenting to primary care who (1) completed a standardized concussion symptom scale (PCSS); (2) received educational and instructional materials after the initial visit; and (3) were seen for concussion follow-up within 2 weeks if indicated.
Methods: At 2 hospital-affiliated primary care clinics, we implemented a multidisciplinary process improvement initiative using Plan-Do-Study-Act cycles within the Model for Improvement framework. Interventions included creating a clinical pathway, integrating the PCSS into the electronic medical record, creating a standardized instruction auto-text, and conducting staff education. We monitored monthly screening, education/instruction, and follow-up rates using statistical process control charts. Measures were stratified by patient language, race/ethnicity, and insurance status.
Results: PCSS use increased from 37% to 85% (1348 total visits), distribution of education/instructions increased from 42% to 83%, and follow-up visit completion improved from 47% to 71%. Notably, disparities based on language and insurance status observed at baseline in PCSS use and distribution of written instructions were attenuated. Emergency department use was unchanged.
Conclusions: This multifaceted interdisciplinary improvement effort standardized concussion evaluation and management in primary care, improving adherence to best practices and reducing disparities. Future work should investigate the impact of care standardization on clinical outcomes and patient recovery
Isolated Lateral Calf Compartment Syndrome from a Sports Injury: A Case Report and Review of the Literature
No full textCase: A 17-year-old lacrosse player presented 2 days after ankle injury with increasing lateral calf pain and lateral ankle numbness. Testing demonstrated a lateral compartment pressure of 134 mm Hg. The patient underwent emergent anterior and lateral compartment release. The peroneus longus and brevis were dusky, but regained perfusion. The patient\u27s serum creatine kinase (CK) spiked initially because of reperfusion injury but ultimately normalized.
Conclusion: Isolated nontraumatic lateral compartment syndrome is extremely rare. Compartment release in a timely fashion restores perfusion to the compartment; however, CK is released and needs to be diluted to prevent renal injury.
Keywords: adolescent; isolated lateral compartment syndrome; male; peroneus longus tear
Outcomes of heart failure hospitalizations at urban teaching vs. Non-Teaching Hospitals: A Nationwide Propensity Score Matched Analysis in the United States
No full textAims: Heart failure (HF) continues to be a major cause of morbidity and mortality worldwide, placing a significant burden on healthcare systems. Differences in the outcomes of HF hospitalizations for adults in teaching vs. non-teaching hospitals in urban settings are uncertain. We evaluated outcomes of HF hospitalizations in patients hospitalized in urban teaching vs. non-teaching hospitals in the United States.
Methods and results: HF hospitalizations were abstracted from the 2016 through 2022 Nationwide Readmissions Database and stratified into urban teaching vs. urban non-teaching hospital settings. Propensity-score matching was used to control for baseline differences between teaching and non-teaching hospital cohorts. Logistic regression and lognormal models were estimated to assess differences in inpatient mortality, length of stay (LOS), total costs, complications, and 30-day and 90-day all-cause readmissions. A total of 7 558 299 weighted HF hospitalizations were included in the analysis, of which 76.3% involved urban teaching hospitalizations. Compared to urban non-teaching, HF hospitalizations in urban teaching hospitals were associated with significantly higher odds of inpatient mortality (OR 1.19, 95% confidence interval [CI]: 1.16-1.22), complications including cardiogenic shock, cardiac arrest, Intra-aortic balloon pump, cardiopulmonary resuscitation, extracorporeal membrane oxygenation and mechanical ventilation use as well as more extended hospital LOS, higher total costs, palliative care consultation and readmissions rates (all P \u3c 0.001). 30-day and 90-day all-cause readmission rates were also statistically higher in teaching hospitals, though the difference was clinically non-significant.
Conclusion: HF hospitalizations in urban teaching hospitals were associated with higher inpatient mortality, complication rates, resource utilization, and palliative care consult rates than urban non-teaching hospitals.
Keywords: Healthcare outcomes; Heart Failure; Hospitalizations; Teaching; United States; Urban
Editorial: Seeing clearly now: Imaging guidance modalities for left atrial appendage occlusion
No full textConscious Sedation Versus General Anesthesia for Percutaneous Left Atrial Appendage Occlusion: A Systematic Review and Meta-Analysis
No full textBackground: Percutaneous left atrial appendage occlusion (LAAO) has traditionally been performed under general anesthesia (GA) to maintain patient comfort and immobility, especially in the setting of transesophageal echocardiography.
Aims: We aimed to compare the safety and efficacy of conscious sedation (CS) compared to GA for LAAO, concurrent with the expansion of intracardiac echocardiography guidance.
Methods: A systematic search of the PubMed, Embase, Cochrane Central, and Scopus databases was conducted through December 2024 for studies comparing CS versus GA for LAAO. For outcomes of procedural complications and characteristics, random-effects models were used to calculate the mean difference (MD), odds ratio (OR), and risk ratio (RR) effect estimates with 95% confidence intervals (CIs).
Results: Four studies with 1540 patients undergoing LAAO were included (CS = 678, GA = 862). CS was associated with significantly shorter total procedural time [MD -11.95 min; 95% CI -19.52 to -2.78; p = 0.009] and a lower volume of contrast media [MD -31.90 mL; 95% CI -56.72 to -7.08; p = 0.01]. No significant differences were noted for total fluoroscopy time, total length of hospital stay, device success, all-cause mortality, cardiovascular mortality, stroke, device-related thrombus, and peri-device leak \u3e 5 mm.
Conclusion: Percutaneous LAAO using CS allows for faster procedures and lower contrast media volumes compared to GA without compromising device success and clinical outcomes. Randomized studies with longer follow-ups are necessary to understand the long-term outcomes of percutaneous LAAO using CS.
Keywords: atrial fibrillation; conscious sedation; general anesthesia; left atrial appendage occlusion; meta‐analysis
