Citations

Clear

9383 citation(s) found for "Thomas, PS".

Supplementation of Tamoxifen with Low-Dose Endoxifen in Patients with Breast Cancer with Impaired Tamoxifen Metabolism (TAMENDOX): A Randomized Controlled Phase I/II Trial
Mürdter, Thomas E. , Schroth, Werner , Goetz, Matthew P. , Igel, Svitlana , Schaeffeler, Elke , Jäger, Simon , Loibl, Sibylle , Gerteis, Andreas , Pfaff, Lena , Bechtner, Christina , Wrobel, Denise , Bernhoft, Inger Marie , Bernhöft, Ilka , Bjelic-Radisic, Vesna, Burkhardt, Michael , Darsow, Maren , Deutsch, Gerhard, Ts, Dewitz, Dietz, Wolfgang, Eichbaum, Michael, Faust, Elke, Göhler, Thomas, Graßhoff, Sven-Thomas, Hackmann, John , Hänle, Claudia, Hartkopf, A, Heinrich, Bernhard, Kahl, Christoph, Kaltenecker, Gabriele, Kleine-Tebbe, Anke, Hc, Kolberg, Krauss, Thomas, Kuhn, Thomas , Kunz, Georg , Ligl-Löhner, Anette, Negwer, Martina, Rhein, Uwe, Rogmans, Gunther, Rom, Joachim , Stalzer, Gabriele, Stefek, Andrea , Strittmatter, Hans-Joachim, Terhaag, Jürgen, Uleer, Christoph , Wimmer-Freys, Karen, Zahm, Dirk-Michael, Fischer, Imma , Meisner, Christoph , Block, Michael , Tremmel, Roman , Igel, Svitlana , Schaeffeler, Elke , Jäger, Simon , Loibl, Sibylle , Gerteis, Andreas , Pfaff, Lena , Bechtner, Christina , Wrobel, Denise , Fischer, Imma , Meisner, Christoph , Block, Michael , Brauch, Hiltrud , Schwab, Matthias , Goetz, Matthew P , Tremmel, Roman , Bjelic-Radisic, Vesna , Burkhardt, Michael, Hartkopf, Andreas D , Heinrich, Bernhard , Kleine-Tebbe, Anke, Krauß, Thomas, Kühn, Thomas, Schmatloch, Sabine , Schwab, Matthias , Strittmatter, Hans-Joachim, Zahm, Dirk-Michael, Brauch, Hiltrud

Mayo Clinic

Clinical Cancer Research | October 2, 2025 | Volume: 31 | Issue: 23 | Pages: 4903-4911

10.1158/1078-0432.ccr-25-2103 | PMID 41036998 Get It

Abstract Purpose :Tamoxifen undergoes bioactivation to its active metabolite (Z)-endoxifen, which blocks estrogen-dependent breast tumor growth at high potency. We tested the feasibility and safety of supplementing standard tamoxifen therapy with low-dose (Z)-endoxifen in patients with breast cancer with compromised tamoxifen bioactivation. Patients and Methods :We conducted a prospective, interventional, three group randomized trial including 235 patients with hormone receptor–positive breast cancer who received standard tamoxifen therapy (20 mg/day). Patients were stratified by CYP2D6 genotype (n = 78), defining poor, intermediate, and normal metabolizers, or by baseline (Z)-endoxifen plasma concentration (n = 78), defining ≤15, 15 to 25, and ≥25 nmol/L. Co-treatment with (Z)-endoxifen 3 and 1.5 mg/day or placebo was performed, respectively. A control group (n = 79) received placebo regardless of metabolizer phenotype. The primary endpoint was the number of patients with (Z)-endoxifen levels >32 nmol/L after 6 weeks of treatment. Adverse events were continuously monitored. Results: A higher proportion of patients in both intervention groups achieved target concentrations >32 nmol/L compared with control (P < 0.0001). At 3 mg (Z)-endoxifen supplementation, 92.3% of CYP2D6 poor metabolizer patients and all patients with baseline (Z)-endoxifen ≤15 nmol/L achieved the target concentration. At 1.5 mg (Z)-endoxifen supplementation, 88% of CYP2D6 intermediate metabolizer patients and 95% of patients with 15 to 25 nmol/L baseline (Z)-endoxifen levels achieved the target concentration. Similar proportions of patients receiving (Z)-endoxifen (6/80, 7.5%) or placebo (8/155, 5.2%) experienced grade 3 adverse events. Conclusions: Adding low-dose (Z)-endoxifen to standard tamoxifen is safe and provides a new approach to personalized antiestrogen treatment for patients with low endoxifen plasma levels.
EACTS Expert Consensus Document on protected cardiac surgery: pre-emptive use of temporary mechanical circulatory support in adult cardiac surgery patients at high risk for perioperative low cardiac output syndrome
Lorusso, Roberto , Salazar, Leonardo , Nersesian, Gaik , Wiedemann, Dominik , Goldstein, Daniel, Soltesz, Edward G. , Bermúdez, C. , Haft, Jonathan W., Patel, Bhavesh , Moreno, Pedro , Tonna, Joseph E. , Lamarche, Yoan , Rao, V., Meyns, Bart , Takeda, Koji , Whitman, Glenn, Potapov, Evgenij, Group, EACTS Scientific Document, Sartipy, Ulrik , Fux, Thomas , Turina, Marko , Morshuis, Michiel , Knosalla, Christoph , Milojevic, Milan , Zimpfer, Daniel , Wiedemann, Dominik , Takayama, Hiroo , Rao, Vivek , Meyns, Bart , Potapov, Evgenij , Tonna, Joseph E. , Lorusso, Roberto , Schmack, Bastian , Engelman, Daniel T. , Stein, Louis H. , Silvestry, Scott , Zimpfer, Daniel , Moreno, Pedro , MacLaren, Graeme , Takayama, Hiroo , Färber, Gloria , Boeken, Udo , Arora, Rakesh C. , Group, EACTS Scientific Document, Russo, Claudio , Meineri, Massimiliano , Paparella, Domenico

Mayo Clinic

European Journal of Cardio-Thoracic Surgery | October 2, 2025 | Volume: 68 | Issue: 3 | Pages: No page available

10.1093/ejcts/ezaf296 | PMID 41063414 Get It

Abstract Perioperative low cardiac output syndrome (LCOS) remains a significant complication in adult cardiac surgery, contributing to substantial morbidity, prolonged intensive care, and increased mortality. Its incidence is expected to rise further due to the evolving complexity of referred surgical patients, often characterised by advanced age, multiple comorbidities, challenging anatomy, and impaired haemodynamics. Despite advances in pharmacological and perioperative care, outcomes for high-risk patients have not shown significant improvement, prompting interest in temporary mechanical circulatory support (tMCS) as a proactive strategy. This Expert Consensus Document from the European Association for Cardio-Thoracic Surgery (EACTS) presents the first dedicated guidance on the pre-emptive use of tMCS in high-risk adult cardiac surgical patients. Developed by a multidisciplinary task force, it emphasises structured risk stratification, early initiation, and individualised device management informed by interdisciplinary Heart Team discussions. The document proposes clinical pathways for patient selection, defines criteria for tMCS initiation, and provides practical algorithms for various scenarios, including advanced heart failure, cardiogenic shock, and post-cardiotomy LCOS. It reviews the current evidence on available tMCS devices, such as intra-aortic balloon pumps, microaxial flow pumps, veno-arterial extracorporeal life support and hybrid strategies, and addresses perioperative care, intensive care unit protocols, ethical considerations, as well as informed consent and support withdrawal. Despite promising results, substantial knowledge gaps remain, including long-term outcome data, device selection criteria, and cost-effectiveness analyses. This consensus aims to support clinical decision-making, standardise practice, and stimulate research to improve outcomes in a growing population of high-risk surgical patients.
Teaching AI for Radiology Applications: A Multisociety‑Recommended Syllabus from the AAPM, ACR, RSNA, and SIIM
Kitamura, Felipe , Kline, Timothy L. , Warren, Daniel , Santos, Igor , Bialecki, Brian , O'Donnell, Kevin , Morgan, Matthew B. , Safdar, Nabile , Andriole, Katherine P. , Geis, J. Raymond , Dreyer, Keith J. , Wood, Monica J. , Langer, Steve G., Farina, Eduardo Moreno Júdice de Mattos , Reiser, Ingrid , Wald, Christoph , Cook, Tessa S. , Tenenholtz, Neil , Kline, Timothy, Langer, Steve, Kitamura, Felipe C , Reiser, Ingrid , Moy, Linda , Daneshjou, Roxana , Maleki, Farhad , Gichoya, Judy Wawira , Wiggins, Walter F. , Flanders, Adam E , Allen, Bibb , Lungren, Matt , Kohli, Marc , Shih, George , Kahn, Charles E. , Giger, Maryellen , Mongan, John

Mayo Clinic

Medical Physics | October 1, 2025 | Volume: 52 | Issue: 10 | Pages: No page available

10.1002/mp.17779 | PMID 41030223 Get It

Abstract Medical imaging is undergoing a transformation driven by the advent of new, highly effective, machine learning techniques paired with increases in computational capabilities (Cheng et al. 2021; Gilson et al. 2023; Almeida et al. 2024; Krishna et al. 2024). These advanced algorithms have the potential to improve disease detection, diagnosis, prognosis, and treatment outcomes. However, the complexity of machine learning models, the large amounts of curated and annotated data required by some methods, and the potential for bias and error make it challenging for individuals to safely and effectively leverage these methods (Lin et al. 2024; Guo et al. 2024; Xu et al. 2024; Linguraru et al. 2024; Wood et al. 2019). To address these challenges, the American Association of Physicists in Medicine (AAPM), American College of Radiology (ACR), Radiological Society of North America (RSNA), and Society for Imaging Informatics in Medicine (SIIM) have worked together to develop a syllabus detailing a recommended set of competencies for medical imaging professionals interacting with these systems. This guide is aimed at four different personas: users of AI systems, purchasers of AI systems, individuals who provide clinical expertise during the development of AI systems (“clinical collaborators”), and developers of AI systems.1 This is a syllabus, not a curriculum, and is intentional in this scope. Recognizing that individuals may benefit from different presentations of the same material, this work enumerates a series of relevant competencies but does not prescribe, nor offer, a method of instruction (Schuur, Rezazade Mehrizi, and Ranschaert 2021; Garin et al. 2023). By addressing the task‐specific demands of each role, this guide will enable medical imaging professionals to utilize machine learning systems more safely and effectively, ultimately improving patient care and outcomes.
Albumin prevents kidney injury but is underutilized in a cohort of patients undergoing large-volume paracentesis
Rajkumar, Rahul , Duong, Nikki , Kim, W. Ray , Ardiles, Thomas, Coll-Ortega, Cristina , Davis, Elizabeth , Lilley, Jonathan, Zhang, Xuan , Wadhwani, Nisha, Lodaya, Kunal , Viayna, Elisabet , Lilley, Jonathan , Zhang, Xuan , Wadhwani, Nisha , Lodaya, Kunal

Mayo Clinic

Hepatology Communications | September 29, 2025 | Volume: 9 | Issue: 10 | Pages: No page available

10.1097/hc9.0000000000000760 | PMID 41021268 Get It

Background: Cirrhosis and cirrhosis-related deaths have risen in the United States in recent years. Ascites is a common complication, often requiring large-volume paracentesis (LVP). The American Association for the Study of Liver Diseases (AASLD) recommends the administration of albumin in conjunction with LVP to prevent further complications of cirrhosis. Emerging research in cirrhosis care reveals significant variations in outcomes among different demographics. Therefore, we assessed the use of guideline-adherent albumin and outcomes in U.S. patients undergoing LVPs, particularly at the intersection of race, ethnicity, socioeconomic disparities, and cirrhosis. Methods: This retrospective study utilized Cerner Real World Data to identify adults with cirrhosis and ascites undergoing LVP between January 2016 and June 2022. We assessed albumin utilization patterns across racial and ethnic groups and payor types, and their overall impact on acute kidney injury (AKI)-related hospitalization using an adjusted generalized linear model (aGLM). Results: We identified 736 patients: 301 in the LVP + albumin group and 435 in the LVP-only group. Despite clinical recommendations, only 41% undergoing LVPs received albumin. White patients and commercially insured patients received albumin at higher rates (p=0.042 and p=0.009, respectively). The overall rate of AKI-related admissions within the 30-day post-procedure period was 26%. However, patients who received albumin during LVP had a 36% lower risk of short-term AKI-related hospitalization (OR: 0.64; p=0.03). Conclusions: These findings indicate a potential for broader albumin utilization in U.S. patients with refractory ascites undergoing repeated LVPs to reduce AKI-related admissions.
Physical activity and molecular subtypes of colorectal cancer: a pooled observational analysis and Mendelian randomization study
Chalitsios, Christos V. , Markozannes, Georgios , Aglago, Elom K. , Berndt, Sonja I., Buchanan, Daniel D. , Campbell, Peter T. , Chan, Andrew T. , Dimou, Niki , Drew, David A. , French, Amy J. , Georgeson, Peter , Giannakis, Marios , Gruber, Stephen B., Harrison, Tabitha A. , Hullar, Meredith A.J. , Huyghe, Jeroen R. , Newton, Christina C. , Schmit, Stephanie L. , Steinfelder, Robert S. , Ugai, Tomotaka, Zaidi, Syed Hassan Ejaz , Woods, Michael O. , Andreson, Laura, Peters, Ulrike , Phipps, Amanda I , Tsilidis, Konstantinos K, Berndt, Sonja I , Cao, Yin , Chan, Andrew T , Dimou, Niki , Drew, David A , Georgeson, Peter , Giannakis, Marios , Gruber, Stephen B , Harrison, Tabitha A , Hoffmeister, Michael , Hsu, Li , Huang, Wen-Yi , Hullar, Meredith A , Huyghe, Jeroen R , Lynch, Brigid M , Moreno, Victor , Newton, Christina C , Nowak, Jonathan A , Obón-Santacana, Mireia , Ogino, Shuji , Schmit, Stephanie L , Steinfelder, Robert S , Sun, Wei , Thomas, Claire E , Toland, Amanda Ewart , Trinh, Quang M , Ugai, Tomotaka , Um, Caroline Y , Van Guelpen, Bethany , Schoen, Robert E , Woods, Michael O , Brenner, Hermann , Anderson, Laura , Pellatt, Andrew J , Phipps, Amanda I , Tsilidis, Konstantinos K , Buchanan, Daniel D , Campbell, Peter T , Aglago, Elom K , Markozannes, Georgios , Berndt, Sonja I. , Gunter, Marc J. , Hoffmeister, Michael , Huang, Wen‐Yi , Moreno, Vı́ctor , Murphy, Neil , Nowak, Jonathan A. , Obón‐Santacana, Mireia , Qu, Conghui , Sun, Wei , Thomas, Claire E. , Trinh, Quang M. , Um, Caroline Y. , Guelpen, Bethany Van , Schoen, Robert E. , Brenner, Hermann , Pellatt, Andrew J. , Tsilidis, Konstantinos K.

Mayo Clinic

JNCI Cancer Spectrum | September 29, 2025 | Volume: 9 | Issue: 6 | Pages: No page available

10.1093/jncics/pkaf095 | PMID 41031512 Get It

Abstract Background Physical activity is associated with lower colorectal cancer (CRC) risk, but its association with molecular subtypes defined by genetic and epigenetic alterations of the disease is unclear. Such information may enhance the understanding of the mechanisms related to the benefits of physical activity. Methods Pooled observational (cases: n = 5386; controls: n = 6798; studies n = 5) and genome-wide association data (cases: n = 8178; controls: n = 10 472; studies n = 5) were used. We used multivariable logistic regression models and Mendelian randomization to assess the association between physical activity and the risk of CRC subtypes defined by individual tumor markers (and marker combinations), namely microsatellite instability status, CpG island methylator phenotype status, and BRAF and KRAS mutations. We used case-only analysis to test for differences between molecular subtypes. We applied Bonferroni correction to account for multiple tests. Results In the pooled observational analysis, higher levels of physical activity were associated with lower CRC risk (Obs-per 1SD, odds ratio [OR] = 0.94, 95% confidence interval [CI] = 0.90 to 0.97), with an association that was stronger in males (Obs-per 1SD, OR = 0.91, 95% CI = 0.87 to 0.96) than in females (Obs-per 1SD, OR = 0.97, 95% CI = 0.91 to 1.03; Pinteraction = .04). Higher physical activity was associated with a lower risk of CRC across all molecular subtypes, especially in males. There was no difference in the associations by subtypes by pooled observational or Mendelian randomization analyses. The findings did not differ by study design, anatomical site, and early or late age onset of CRC. Conclusions Our findings suggest that physical activity is not differentially associated with the 4 major molecular subtypes involved in colorectal carcinogenesis, indicating that its benefits extend broadly across colorectal cancer pathogenesis.
Prospective Assessment of Changes in Target Vessel Peak Systolic Velocity Measurements After Fenestrated-Branched Endovascular Aortic Repair
Macedo, Thanila A. , Lima, Guilherme Baumgardt Barbosa , Hatz, Heather J., Hesley, Gina K. , Lekah, Alexander, Huang, Ying , Mendes, Bernardo C. , Sulzer, Titia A L , Tenorio, Emanuel R , Baumgardt Barbosa Lima, Guilherme, Hatz, Heather, Tallarita, Tiziano , Huang, Ying , Oderich, Gustavo S , Lima, Guilherme B , Mésnard, Thomas , Tenorio, Emanuel R. , Oderich, Gustavo S.

Mayo Clinic

Journal of Endovascular Therapy | September 28, 2025 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.1177/15266028251368250 | PMID 41015897 Get It

Objective: Target vessel stenosis or occlusion is a common indication for secondary intervention after fenestrated-branched endovascular aortic repair requiring long-term imaging surveillance including duplex ultrasound (DUS). This study aimed to describe and compare longitudinal changes in peak systolic velocity (PSV) measurements for renal and mesenteric arteries targeted by directional branches (DBs) or reinforced fenestrations (RFs). Methods: Patients enrolled in a prospective, nonrandomized study (2013-2020) had DUS of celiac axis, superior mesenteric artery, and renal arteries (RAs) obtained preoperatively, at 6 to 8 weeks, 6 months, and annually. Vessels with preprocedural stenosis were excluded. Outcomes were variations in PSV over time for target vessels incorporated by DBs or RFs, differences in balloon-expandable stent-grafts (BESGs) and self-expandable stent-grafts (SESGs), PSV measurements prior to secondary interventions related to vessel stenosis, and a predictive value of PSV for stenosis requiring secondary intervention. Results: A total of 419 patients (292 male, mean age 74 ± 8 years old) were enrolled, with 1,311 target vessels analyzed preoperatively, including 607 mesenteric and 704 RAs. Over a median follow-up of 23 months (interquartile range [IQR], 7-36), PSV measurements decreased in the first 6 to 8 weeks after DB incorporation, remained stable, followed by a nonsignificant increase at 5 years. PSV increased in the first 6 months using RFs with postoperative velocities significantly higher ( P < .05) for RFs compared to DBs. Branched mesenteric vessels stented with BESGs had higher velocities than SESGs ( P < .05). Of the 23 target vessels treated by secondary intervention for stenosis, 19 (83%) had velocities above the thresholds for native, nonstented vessels. Furthermore, in RAs (n = 20) PSV effectively predicted clinically relevant stenosis, AUC was 0.98, with a 231 cm/s threshold offering 84% sensitivity with 100% specificity. Conclusion: Longitudinal follow-up shows that velocity changes vary depending on the type of incorporation, vessel, and bridging stent. For DBs, PSV decreased initially, stabilized, and showed a nonsignificant increase at 5 years. In contrast, PSV increased with RFs and remained higher than DBs. Despite these variations, velocities remained below established thresholds for clinically significant stenosis in nonstented vessels. Criteria for in-stent stenosis may differ, and PSV alone should not be the sole indicator for reintervention. Clinical Impact This study provides new insight into duplex ultrasound (DUS) surveillance after fenestrated and branched endovascular aneurysm repair (FB-EVAR). Peak systolic velocity (PSV) trends differed between fenestrations and branches, and stent type influenced flow dynamics. Importantly, most vessels with secondary interventions had elevated PSVs, but many others exceeded native stenosis thresholds without clinical consequence. Despite these variations, velocities remained below established thresholds for clinically significant stenosis in non-stented vessels. Criteria for in-stent stenosis may differ, and PSV alone should not be the sole indicator for reintervention.The reported PSV measurements served as a benchmark for DUS surveillance following FB-EVAR.
Multiparametric MRI for Diagnosing Clinically Significant Portal Hypertension and Predicting Liver Decompensation
Bhuiyan, Enamul , Geahchan, Amine , Altınmakas, Emre , Abboud, Ghadi , Pavuluri, Swathi, Ehman, Richard L. , Lewis, Sara , Bansal, Meena B. , Thung, Swan N. , Schiano, Thomas D. , Taouli, Bachir, Bhuiyan, Enamul , Kennedy, Paul , Ehman, Richard , Ozkaya, Efe , Bane, Octavia , Hectors, Stefanie J. C. G. , Pavuluri, Swathi, Yin, Meng , Fischman, Aaron M. , Taouli, Bachir

Mayo Clinic

Liver International | September 27, 2025 | Volume: 45 | Issue: 11 | Pages: No page available

10.1111/liv.70368 | PMID 41014103 Get It

ABSTRACT Background and Aims Clinically significant portal hypertension (CSPH; hepatic venous pressure gradient (HVPG) ≥ 10 mmHg) is a severe complication of chronic liver disease, with increased risk of variceal bleeding and liver decompensation (at HVPG ≥ 12 mmHg). This study evaluated the performance of multiparametric (mp)MRI for diagnosing CSPH and predicting liver decompensation, compared to ultrasound‐based shear wave elastography (SWE) and blood tests. Methods In this prospective single‐centre study (2018–2022), 59 patients (M 30, mean age 52.7 years) underwent mpMRI and HVPG measurement, with an average interval of 32 ± 31 days. The mpMRI protocol included 2D/3D MR elastography (MRE), T 1 /proprietary iron‐compensated T 1 (cT 1 )/T 1ρ mapping, gadoxetate‐enhanced dynamic contrast‐enhanced MRI (DCE‐MRI) of the liver and spleen, SWE and blood tests. Statistical analyses included Mann–Whitney U tests, ROC analysis, logistic regression and Cox proportional hazards models. Results CSPH was present in 13 patients (22%). Several MRI parameters showed high diagnostic performance for CSPH (AUC ≥ 0.8), with spleen stiffness‐2D MRE (AUC 0.88, 95% confidence interval (CI) 0.78–0.98) and liver uptake on DCE‐MRI (AUC 0.83, 95% CI 0.70–0.96) performing best, while SWE had AUCs of 0.63 (95% CI 0.45–0.81) for liver and 0.67 (95% CI 0.44–0.89) for spleen. Combining spleen stiffness‐2D MRE and liver uptake achieved an AUC of 0.93 (95% CI 0.86–1.00) for diagnosing CSPH. For predicting decompensation, spleen stiffness‐3D MRE and liver uptake rate had AUCs of 0.83 (95% CI 0.68–0.99) and 0.83 (95% CI 0.70–0.95), respectively, outperforming SWE. Conclusions Spleen stiffness measured with MRE combined with gadoxetate liver uptake (measured with DCE‐MRI) can diagnose CSPH and predict liver decompensation, with superior performance compared to SWE. Trial Registration ClinicalTrials.gov identifier: NCT03436550
Advancing delirium detection through the Open Health Natural Language Processing Consortium and the Evolve to Next-Gen Accrual to Clinical Trials Network
Fu, Sunyang , Kwak, Min Ji , Ahn, Jaerong, Yue, Zhiyi , Ranganath, Shreyas, Applegate, Joseph R., Wang, Liwei , Li, Chenyu , Toth, Kelly M, Girard, Timothy D., Osborne, John D. , Kennedy, Richard , Garduno-Rapp, Nelly-Estefanie, Reeder, Phillip , Murphy, Thomas J. , Heling, Jia, Sohn, Sunghwan , Visweswaran, Shyam, Wang, Yanshan , Girard, Timothy D. , Li, Chenyu , Rousseau, Justin F , Yan, Chao , Murphy, Tyler J , Park, Chan Mi , Bernstam, Elmer V , Visweswaran, Shyam , Liu, Hongfang , Girard, Timothy D , Fu, Sunyang , Kennedy, Richard E , Osborne, John D , Wang, Yanshan , Ahn, Jaerong, Applegate, Joseph R, Wen, Andrew , Morris, Michele M. , Toth, Kelly M, Rousseau, Justin F. , Chen, You , Patel, Mayur Bipin , Malin, Bradley , Jia, Heling, Pagali, Sandeep R. , Palmer, Allyson K. , Sauver, Jennifer L. St. , Liu, Hongfang

Mayo Clinic

The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences | September 26, 2025 | Volume: 80 | Issue: 12 | Pages: No page available

10.1093/gerona/glaf207 | PMID 41017643 Get It

Abstract Background Delirium is often underdiagnosed in clinical practice and is not routinely coded for billing. While manual chart review can identify delirium, it is labor-intensive and impractical for large-scale studies. Natural language processing (NLP) can analyze unstructured text in electronic health records (EHRs) to extract meaningful clinical information. Methods To support national integration of NLP for EHR-based delirium identification across different institutions, we launched the Delirium Interest Group within the national Evolve to Next-Gen Accrual to Clinical Trials (ENACT) NLP Working Group. This paper outlines our initial efforts to standardize, evaluate, and translate an NLP-based delirium detection model into the i2b2/ENACT platform. Results Multisite contextual inquiry identified several key challenges, including variations in local screening practices (eg, tools used, documentation frequency, and quality control), the need for harmonized definitions in the context of EHRs, and the complexity of modeling temporal logic. Multisite NLP evaluation revealed variable performance degradation driven by differences in delirium screening practices, clinical documentation patterns and semantics, and note syntactic structures. Conclusion Our work represents an important first step toward enabling scalable and standardized NLP-based delirium detection across institutions. By engaging diverse institutions through the ENACT NLP Working Group, we identified shared challenges and site-specific variations that impact model implementation and performance. Our collaborative approach enabled the development of a more robust framework for delirium identification across heterogeneous EHR systems. Future efforts will build on this foundation to enhance the validity, usability, and translational impact of delirium detection.