BMJ Health & Care Informatics

ObjectivesTo determine whether single-layer evaluation characterises clinical risk in the final note of a multilingual ambient AI scribe, and where serious errors arise. DesignTwo-arm evaluation on one common clinical-risk scale, using a frozen, reference-aligned synthetic corpus. SettingScripted consultations spanning 77 languages and five complexity levels, from simple general practice to expert multidisciplinary-team handover. Main outcome measuresPer-session incidence of at least one serious (HIGH or CRITICAL) note-layer error in intrinsic (reference script to note, n=385) and end-to-end (automatic speech recognition (ASR) transcript to note, n=2,302) generation; three LLM raters classified discrepancies using a four-mode taxonomy. ResultsIntrinsic note generation had a 3.4% serious-error rate, with no detectable gradient across complexity (1% to 6%; Cochran-Armitage p=0.42) or language resource (high 2%, medium 5%, low 4%). End-to-end serious errors rose steeply with complexity (L1 1% to L4/L5 24%) and language scarcity (high 7%, medium 10%, low 15%; OR 1.54 per tier, 95% CI 1.15-2.07; p=0.004, GEE clustered on language). Of serious in-note errors, 87% were ASR-derived and 7% note-originated; amplification exceeded correction threefold (fate entropy 1.19 bits). Repeated generation showed moderate reproducibility (Fleiss kappa 0.42); word error rate explained 24% of cross-language variance versus [~]1% for transcription risk density. ConclusionsLow serious-error rates in individual layers did not preclude higher end-to-end note risk. Evaluation should therefore include the clinician-facing note, because component-level metrics cannot fully capture risk created or transformed at the transcription-to-generation interface. What is already known on this topicO_LIAmbient AI scribes are being deployed globally, and safety evaluation has characterised the transcription layer in isolation, most commonly with word error rate C_LIO_LIFrequency-based transcription metrics such as word error rate do not directly encode clinical consequence, and component-level performance may not represent the risk of the final generated note C_LIO_LILarge language models produce clinically relevant output stochastically, so identical prompts can yield materially different content C_LI What this study addsO_LINote generation from a perfect transcript had a low observed serious-error rate and no detectable complexity or language-resource gradient, whereas the end-to-end system showed steep gradients on both axes. C_LIO_LISerious note-layer errors were predominantly ASR-derived, but note generation transformed upstream errors variably, amplifying serious errors about three times more often than it corrected them C_LIO_LITranscription-layer metrics had limited predictive value for note-layer risk after accounting for complexity and language-level dependence; word error rate explained approximately one-quarter of cross-language variance C_LI How this study might affect research, practice or policyO_LIEvaluation and procurement of ambient scribes should include end-to-end assessment of the clinician-facing note, with results stratified by consultation complexity and language where relevant C_LIO_LIA three-surface root-cause alignment using a four-mode root-cause taxonomy (note-originated, propagated-ASR, amplified-ASR and silent-correction) can attribute documentation errors to their point of origin or transformation C_LIO_LILanguage-related performance should remain a surveillance target because disparities can emerge after components are composed, even when corresponding gradients are not detectable in an individual layer C_LI

Background: As required by the 21st Century Cures Act, all test results, including final radiology reports, are released to patients as soon as the results are finalized. However, radiology reports are primarily intended for communication between medical professionals and can often be difficult for patients to interpret, particularly in cancer care settings. We developed a tool that leverages a large language model (LLM; ChatGPT-5.2, OpenAI, San Francisco, CA) to draft a patient-friendly plain language version of imaging results (Patient-Friendly Scan Result Generation [PFSG] tool) that oncologists can incorporate into their communication with patients via patient portals. This article describes our mixed-methods study protocol. Objective: Our primary aims are to: 1) assess whether patients who receive patient-friendly versions of scan results report better understanding of their results than patients who do not; 2) assess whether introduction of the PFSG tool affects provider workload; and 3) explain how and under what circumstances the PFSG tool enhances communication and workflows, and affects patient and provider outcomes. Methods: We will use an embedded mixed-methods, prospective cohort design within a quality improvement (QI) initiative. The PFSG tool will be implemented using a modified three-wave staircase trial design, with providers in randomized cohorts of oncology disease groups progressively gaining access to the tool over the course of the study. Approximately 60 providers from 11 oncology disease focus groups will be randomized. Longitudinal provider outcomes will be measured through surveys at one month pre-implementation, one month post-implementation, and three months post-implementation. Patient outcomes will be measured cross-sectionally through surveys at one month pre-implementation (n=350) and at post-implementation result point-of-contact (n=350). Additionally, we will conduct interviews with patients and oncologists to inform the iterative design of the tool and its implementation. Results: This study was funded in September 2025 and data collection began in February 2026. As of September 2026, the PFSG tool has been rolled out to all oncology providers, with Cohort 3 post-implementation data collection beginning in August 2026. A total of 738 patient baseline surveys have been collected and patient post-implementation data collection is in progress. Conclusions: This is the protocol for a study of the implementation of a patient-friendly results tool in oncology provider workflows to facilitate improved communication of radiology results to oncology patients. The results of this study will lend evidence to the utility and acceptability of the study tool, and tools like it, to close an important oncology care gap. If the tool is shown to improve key oncology patient and provider outcomes, it may lead to development of a wider release for more general patient populations.

BackgroundEvaluating fairness in artificial intelligence (AI) is essential before clinical implementation to prevent unintended harm across diverse patient populations. This study assesses performance differences of an image-based surgical site infection (SSI) detection model across age, gender, geography, race, and skin color. MethodsModel development/validation study using retrospective, multicenter data across surgical subspecialties from adults who submitted an image of their incision via the electronic health record within 30 days of their procedure between 2019-2024. Area under receiver operating curve (AUROC) was compared by age (<55 vs. [&ge;]55 years old), gender (male vs. female), geography (Rochester, Minnesota vs. Jacksonville, Florida vs. Scottsdale, Arizona vs. regional Minnesota/Wisconsin areas), race (White/Caucasian vs. non-White/Caucasian), and skin color (Fitpatrick I-III vs. IV-VI and Monk A-E vs. F-J). ResultsTraining data contained 13,702 images from 4,274 patients between 2019-2022 (50.1% [&ge;]55 years old; 63.2% female; 91.6% White/Caucasian; 58.5% Rochester, 17.3% Florida, 10.5% Arizona; 96.3% light skin tone), of which 1,348 images (9.8%) had an SSI. No differences in SSI detection existed for age, gender, race, or skin color (AUROC 0.83-0.87, p>0.09), however geographical differences were significant: lowest AUROC 0.83 (95% CI 0.77-0.88) vs. highest AUROC 0.87 (95% CI 0.84-0.89, p=0.01). Temporal validation using 4,242 images from 1,353 patients between 2022-2023 showed intact fairness across subgroups ({Delta}AUROC [&le;]0.06) except geography ({Delta}AUROC [&le;]0.16). ConclusionsThe model behaved fairly across age, gender, race, and skin color, however geographical differences persisted. These findings have practical implications underscoring the importance of evaluating bias in clinical AI before deployment.

As health AI tools enter clinical spaces, how they are perceived by providers in safety-net environments shapes whether their benefits reach underserved populations. We examined the degree and drivers of healthcare providers' trust in health AI, whether that trust moderates perceived benefits for safety-net populations, and how providers balance benefit against concern across four specific hypothetical AI tools. We used a two-study mixed-methods design: 18 semi-structured interviews with healthcare practitioners and research leaders, followed by a survey of 229 Texas healthcare providers incorporating four vignettes describing hypothetical AI tools, each rated for appeal and concern and followed by open-ended explanation. Responses were thematically coded and stratified by self-reported trust in health AI (47.6% high trust, 52.4% low trust). Low-trust respondents most often cited building trust (50.5%) and data bias and accuracy (33.0%) as drivers of their current trust level, while high-trust respondents more often cited efficient and effective care (23.5% vs. 3.7% low-trust, p<0.001) and accuracy of data (11.8% vs. 1.8% low-trust, p=0.004); concerns about data security, governance, and the need for human oversight were raised at similar rates by both groups. When asked how AI could specifically benefit safety-net populations, both groups converged on expanding access to care (45.9% each), and none of 25 benefit-related codes differed significantly by trust group. Across all four vignettes, mean appeal ratings exceeded mean concern ratings, and high-trust providers rated every tool as more appealing and, for three of four tools, less concerning, without changing which tool ranked where relative to others. Providers exhibit conditional optimism: support is contingent on governance, human oversight, and context-sensitive implementation rather than on the technology itself. Realizing AI's promise for safety-net populations will require human-in-the-loop models, localized implementation strategies, community-informed design, and engaging providers directly in policy design and tool development.

The explosive growth of ambient AI scribes in healthcare has focused almost exclusively on relieving clinician documentation burdens and enhancing clinician efficiency and reducing burnout. While most research has focused on the clinician experience, these same types of tools, such as a patient held AI Personal Health Assistant, can be put in the hands of patients to facilitate communication with healthcare providers, to provide greater autonomy and activation for patients in managing their own healthcare, and to facilitate the task of receiving and retaining complex new information. However, to ensure these tools serve as more than just administrative aids, we partnered with a large technology company to conceptually codesign how such a tool might be studied in real world settings in a way that reflected outcomes of greatest importance to patients and families. We used the Nominal Group Technique and engaged thirty three patients and caregivers and ten multidisciplinary clinicians in structured virtual sessions to identify the outcomes on which subsequent studies should hinge. Our findings reveal a significant misalignment between current industry trends and patient priorities. While the market emphasizes the scribe function for billing, stakeholders overwhelmingly prioritized the AIPHAs role in longitudinal care management and post visit coordination. The most critical outcome identified was the impact on communication and respect, suggesting that patients view AI as a mechanism to improve the human to human dynamic and ensure a shared understanding of the care plan. Our findings further highlighted that the primary hurdle for the use of patient held AI is not technical utility but clinician trust and the need for formal consent processes. Importantly, as this technology is new and integration into the clinical experience is still being explored, patients cannot assume they will have access to the transcript of their own encounter when that record is created and held by the health system. This gap highlights the potential value of a patient held tool, preserving discussions from their visit and follow up steps in clearer detail than they may receive in their clinical note or traditional visit summary. From a system perspective, this transition requires a strategic shift from measuring workflow focused efficiency to measuring patient side advocacy and engagement. By centering the evaluation of AI on these patient defined metrics and fundamentally designing subsequent AI investigations using a partnership model, health systems can move beyond simple automation toward a model of care delivery that empowers patients to navigate complex clinical journeys with greater autonomy and clarity.

Objectives: To assess the feasibility and early performance of an AI-assisted screening workflow that reviews inpatient discharge summaries to identify potential treatment injuries for clinician review and, where appropriate, Accident Compensation Corporation (ACC) treatment injury claim consideration in New Zealand. Design: Single-site, six-month pilot reported as a quality-improvement/service-evaluation activity. A large language model (LLM)-enabled system acted as an index screening test over discharge summaries to generate candidate treatment injury signals for structured clinician adjudication and escalation. Setting: A New Zealand district hospital (Wairarapa). Screening operated on routinely produced inpatient discharge summaries. Participants: All inpatient discharge summaries produced during the pilot period were screened (n = 2,642). The unit of analysis was one discharge summary per inpatient episode. Intervention: An LLM-based screening workflow that outputs structured candidate harm signals with supporting excerpts and a certainty label. Outputs were reviewed by clinicians using a structured adjudication template. Suitable cases could be escalated to senior clinicians and followed through the local ACC treatment injury process. Main outcome measures: Primary outcomes were yield, positive predictive value (PPV) of AI- flagged signals for clinician-adjudicated iatrogenic harm, and the rate of unsupported/hallucinated outputs. Secondary outcomes included overlap with incident reporting and administrative coding, downstream ACC claim outcomes, and clinician review workload. Results: Across 2,642 inpatient discharge summaries screened, the AI-assisted workflow generated 237 candidate signals (yield 9.0%; 95% CI 7.9% to 10.1%). Clinician adjudication classified 102 as confirmed iatrogenic harm, 43 as possible harm requiring additional context, 81 as no harm, and 11 as unsupported/hallucinated outputs. PPV for confirmed harm was 43.0% (95% CI 36.8% to 49.4%). PPV for confirmed-or-possible harm was 61.2% (95% CI 54.9% to 67.1%). Seven ACC treatment injury claims were lodged as a direct result of AI-assisted surfacing; three were accepted, two declined, and two were held/pending at analysis. Organisational ACC data identified 68 treatment injury claims lodged during the same period, of which 17 overlapped with AI-identified cases. Conclusions: AI-assisted screening of discharge summaries can feasibly increase detection of potential treatment injuries for structured clinician review and may support more systematic ACC consideration and safety learning. Safe scaling requires transparent reporting of model and prompt details, robust human adjudication and escalation pathways, and governance controls to mitigate false positives and unsupported outputs. A next-stage evaluation should estimate diagnostic accuracy against a chart-review reference standard and examine implementation outcomes including workload, acceptability, fidelity and sustainability.

Objective: To develop and evaluate a framework for characterizing clinician editing of Artificial Intelligence (AI)-generated[KS1.1] documentation and assess its feasibility for health system-level monitoring of AI scribes. Materials and Methods: We analyzed outpatient encounters in which an AI scribe was used at a single academic health system, examining the History of Present Illness (HPI) and Assessment and Plan (A&P) sections of notes. We characterized edits on three dimensions[KS2.1][AG2.2]: lexical edit intensity based on Levenshtein distance, embedding edit intensity BERT[KS3.1]Score, and clinical edit intensity based on removed and added UMLS[KS4.1][AG4.2] concepts.[KS5.1][AG5.2] We analyzed the Positive Predictive Value (PPV) of clinical edit intensity as a measure of clinically meaningful editing using clinicians as the gold standard, examined correlations among dimensions, and designed exponentially weighted moving-average control charts to monitor longitudinal changes in clinician editing behavior.[KS6.1][AG6.2] Results: 268,379 encounters were included (267,654 HPI, 267,594 A&P). Clinical edit intensity [&ge;]1 had an 88.9% PPV for clinically meaningful editing. Lexical and embedding edit intensity were highly correlated (Spearman {rho} 0.95), while clinical edit intensity was less strongly correlated with both ({rho} 0.77-0.81). Longitudinal monitoring detected changes coinciding with system-wide rollout.[KS7.1][AG7.2] Discussion:[KS8.1][AG8.2] Clinicians edited A&Ps more heavily than HPIs, potentially reflecting greater attention to content involving clinical decision-making. Over one-third of sections involved clinical concept changes, and clinical edit intensity identified clinically meaningful edits while providing information complementary to lexical editing measures. Conclusion: Clinician editing can be characterized at scale using complementary editing dimensions, providing a scalable signal for post-deployment surveillance of the human-AI documentation process.

Conversational systems based on large language models may support point-of-care evidence retrieval, but most evaluations use static benchmarks and few involve practicing physicians in low- and middle-income settings. We assessed the validity of answers produced by physicians using an evidence-traceable conversational system versus usual information-search practice, alongside response efficiency and acceptability. Physicians in Latin America were randomly assigned to answer four simulated clinical cases, four open-ended questions each, with system support or usual search practice. Two blinded specialists per area scored each response with a six-dimension rubric. The main outcome was the composite validity score aggregated per physician, analyzed with logistic regression adjusted for academic degree. Of 202 physicians who began, 71 completed and were analyzed, 26 supported and 45 using usual practice, yielding 1,136 responses. Physicians using the system scored higher, with medians of 2.83 versus 2.46 (difference 0.38; 95% CI 0.17 to 0.54; P < .001). In the adjusted model they were more likely to meet the validity threshold (odds ratio 3.61; 95% CI 1.16 to 11.17; P = .026). In the adjusted models the association held for the four criteria with the higher interrater agreement, and did not reach significance for accuracy alone. Response time and self-reported searches did not differ, though uneven missingness makes that comparison uninformative. Acceptability among supported physicians was high (mean total score 2.92 of 3). An evaluator outside the organization repeated the accuracy-only analysis, which showed no significant difference; the composite was designated the main outcome after that replication, so the three specifications are reported side by side. The analyzed sample comprises volunteers who completed the exercise, so the findings are exploratory.

Background: Electronic health records (EHRs) are central to NHS clinical practice, and the value of the data they hold depends on clinicians entering structured data. Structured data entry (SDE) is nonetheless used inconsistently, and most evidence comes from US settings documenting barriers rather than the mechanisms interventions must target. This is, to our knowledge, the first study to apply the Theory of Planned Behaviour (TPB) to SDE, to examine SDE behaviour among UK NHS doctors, and to apply corpus linguistics to clinicians' free-text accounts of EHR use. Methods: A two-part mixed-methods study was conducted at a large NHS teaching hospital in central London using the Epic EHR. Study 1 was a cross-sectional survey of non-consultant grade doctors comprising the System Usability Scale (SUS), a purpose-designed TPB questionnaire, and two open-text questions analysed using corpus linguistics. Study 2 was an Interpretative Phenomenological Analysis (IPA) of semi-structured interviews with six resident doctors. Results: Eighty-three doctors responded; 65 completed the SUS and 49 the TPB questionnaire. Mean SUS score was 62.1 (95% CI 58.3-65.9), below the benchmark of 68 and equivalent to a grade of "D". Structured features were used markedly less often than narrative notes. TPB constructs explained 56% of the variance in intention (F(3,45) = 18.95, p < .001); attitudes (p = .003) and subjective norms (p = .009) predicted intention independently, whereas perceived behavioural control did not (p = .14). Intention predicted self-reported use of problem lists, event-specific templates and standardised documents, but not narrative notes. Corpus analysis identified time pressure and interface friction as dominant barriers. IPA generated four themes: structured documentation as cognitive relief; as fragile collective infrastructure; as a practice clinicians are expected to perform but not enabled to learn; and as a format that can flatten clinical reasoning. Conclusions: Engagement with SDE is driven principally by attitudes and social norms rather than by perceived control, and the barriers clinicians describe arise from the intersection of interface design, training, team norms and institutional expectation. Interventions addressing usability alone are unlikely to succeed: sustained change requires team-level normative reinforcement, specialty-relevant training that explains both how and why, and systems that preserve space for clinical narrative. Keywords: Electronic health records; structured data entry; Theory of Planned Behaviour; System Usability Scale; Interpretative Phenomenological Analysis; corpus linguistics; clinical documentation; data quality; NHS

Background: Clinical decision support systems (CDSS) are increasingly embedded in national telemedicine services, however how different stakeholders who use them experience these tools, particularly in low- and middle-income countries, is less explored. In the assisted model of India's eSanjeevani service, spoke health workers enter structured symptoms and hub clinicians review a symptom summary and an artificial intelligence (AI)-generated differential diagnosis. Objective: To describe role-specific experiences of eSanjeevani's CDSS functions and to estimate how acceptance coexists with specific reported difficulties within each role. Methods: We analyzed a program feedback web survey completed between 11 July and 29 July, 2025 by hub clinicians and spoke providers. Prespecified item profiles covered review of summaries and suggested diagnoses, the perceived relation of suggestions to the clinician's own diagnosis, ease of use and intention to use (hub), and symptom-capture utility, question quantity, question relevance, summary presentation and satisfaction (spoke). Within-role contrasts were reported as absolute differences in acceptance with Newcombe 95% CIs. Short open comments underwent content analysis. Results: The cohort comprised 190 hub and 662 spoke respondents from 13 and 15 states (21.4% and 19.3% of eligible providers). Among hub respondents, 76.8% regularly reviewed the symptom summary and 58.9% the suggested diagnoses; 72.6% judged suggestions matched (32.6%) or appropriate but different (40.0%), 7.9% mostly inappropriate, and 11.1% reported that suggestions were mostly not generated. Overall, 85.3% intended to keep using the tools. Affirmative intention was 98.4% when suggestions matched, 66.7% when they were mostly inappropriate (difference vs matched or appropriate -23.2 percentage points, 95% CI -48.5 to -4.0) and 61.9% when mostly absent (-25.7, 95% CI -47.1 to -7.3); technical issues, reported by 38.4%, made little difference (-2.8, 95% CI -14.1 to 7.3). Among spoke respondents, 63.4% rated question quantity appropriate, with miscalibration in both directions (15.6% too many; 16.6% too few); 89.9% found the questions relevant and 91.8% found the summary clear. Satisfaction was 75.3% and did not differ materially between those reporting too many questions and those reporting an appropriate number (-2.7, 95% CI -12.8 to 6.0). Comments located spoke difficulties in question burden and missing complaints, and hub difficulties in freezing and in the scope and availability of suggestions. Estimates were stable across cohort, ambiguity and state-exclusion sensitivity analyses. Conclusions: Acceptance of eSanjeevani's CDSS was high in both roles. For hub clinicians it tracked the perceived quality and availability of AI suggestions rather than technical friction; spoke workers accepted the question burden, but its two-directional miscalibration points toward complaint-specific, adaptive questioning. These provider reports do not measure diagnostic accuracy. They identify log-based and audit-based indicators that the next evaluation cycle should measure directly.

Background: Patient Decision aids (PtDAs) are important tools that support shared decision-making (SDM) between clinicians and patients, enabling patients to be more informed and engaged in decisions regarding their care. The use of PtDAs can increase patient knowledge, reduce decisional regret, and engage the clinician and patient in meaningful dialogue. Shared decision-making relies on decision aids, but how patients receive the PtDA can vary. This qualitative study elucidated patient familiarity with PtDAs and the barriers, needs, and requirements of patients to access and confidently use DAs, including when received via a clinician-facing EHR-integrated Decision Aid Navigator System (DEANS). Methods: We conducted six focus groups (FG) with participants who were recruited from two academic medical centers. Focus groups included an introduction to what a PtDA is and an exploration of how patients or family members, i.e., people who had not made a medical decision for themselves using a PtDA, but who have helped a family member, had previously used PtDAs in their healthcare. The FGs focused on PtDA-users, affordances and barriers to using PtDAs, and PtDA-user needs and requirements of a proposed DEAN patient portal. We applied Rapid Qualitative Analysis (RQA) to summarize the data, using a consensus making approach to identify emerging themes. Results: We identified three overarching themes: 1) Patients had varied familiarity with PtDAs and often confused them with general health tools, but found value in SDM, 2) Patients perceived technology accessibility, PtDA credibility, and time as barriers potentially impeding PtDA use, and 3) Patients valued flexible, user-friendly, and accessible PtDAs that support patient-led decisions, particularly when introduced by clinicians and delivered via a patient portal. Conclusion: This study provides patient-derived insights into factors that influence PtDA uptake, highlighting common misconceptions about PtDAs and identifying key patient priorities such as accessibility, flexibility, clear content, and clinician-guided usage. Patient also expected clinicians to be familiar with PtDAs, especially when the PtDA covers a complex decision. These findings can inform the development of more patient-centered SDM infrastructure and strategies to support PtDA use in clinical practice, including clinician-supported delivery, accessibility, and integration into routine care to strengthen shared decision-making.

Background and Objective: The volume of patient portal messages has grown substantially and artificial intelligence (AI) can support clinical staff and improve the timeliness of patient care. Our objective is to test if an AI tool for managing the electronic inbox In-Basket reduces response times to patient portal messages in neurology. Methods: We conducted an implementation study from January to May 2026 in the neurology department at two clinical sites within a regional medical system. A single AI tool was implemented to auto-categorize and auto-route all patient messages received in the electronic inbox. The tool also generates optional draft replies to messages. Primary process measures were the median weekly time (hours) from message creation to the first action and to closure. Impact was analyzed using a controlled interrupted time-series (ITS) design, comparing message-volume-matched control divisions. Results: Among 24,142 messages analyzed, pre-intervention median weekly time to first action was 11.3 hours [IQR: 5.6, 15.1] (Site A) and 16.1 hours [IQR: 13.8, 19.0] (Site B), falling post-intervention to 1.0 hour [IQR: 0, 3.1] (Site A) and 0.6 hours [IQR: 0.2, 2.4] (Site B). Controlled interrupted time-series analysis showed a significant immediate differential level change in median weekly hours to first action at Site A (-92.5%, 95% CI: -97.5% to -78.0%), and a significant downward trend at Site B (-43.5%, 95% CI: -65.2% to -8.2%). Changes in time to closure were directionally favorable but not statistically significant at either site. There was low use of the generated draft replies, only 0.5% at Site A and 3% at Site B. Discussion: Implementation of AI-assisted message categorization and routing was associated with a substantially reduced initial action time for patient In-Basket messages in Epilepsy. The utilization of AI-generated draft-replies to messages was low. These findings support the feasibility and effectiveness of AI-augmented workflows to improve inbox management.

Objective: Retrospective evaluation of clinical pathway adherence using electronic medical record (eMR) data is limited by heterogeneous data structures and statistical measures that do not account for patient-specific context, care sequence, or timing. We designed a modular event-based architecture to dynamically evaluate the New South Wales Adult Sepsis Pathway over time. Method: Data from eMRs were collected from adult emergency department encounters across four hospitals between July 2022 and June 2024 and processed through three sequential layers: event-stream transformation, pathway-relevance filtering, and state-based pathway evaluation. Heterogeneous records were standardized and organized into a chronological timeline for each encounter. The framework then updated the pathway state as new clinical information became available and assessed whether recommended care was delivered within the relevant timeframe and sequence. Result: The architecture processed 451,124 encounters. This included 57,631,084 clinical events extracted and transformed into encounter-level event streams. After filtering for pathway relevance, 43,951,146 events remained (76.3%). A total of 129,907 (34.6%) encounters entered the sepsis pathway, 50,118 (38.6%) entered YELLOW state, indicating progression to suspected sepsis, and 3,349 (2.6%) encounters progressed to RED state from GREEN, indicating the initial shock. Of 50,118 encounters entering YELLOW state, 2,426 (4.8%) progressed to shock, 8,700 (17.4%) encounters were recognized with sepsis through a blood culture order, and 4,287 (8.6%) encounters were recognized with sepsis through an antibiotic order. This demonstrates the ability of the architecture to prioritize higher-acuity states during evaluation rather than going through an intermediate state. Conclusion: The architecture developed shows the feasibility of transforming fragmented routine eMR data into a reusable longitudinal representation that supports context-dependent pathway evaluation. Separating data preparation from pathway rules improves transparency, maintainability, and adaptability and provides a foundation for future interoperable, near-real-time pathway monitoring

Introduction: Hospitals face increasing financial pressures. Hospital reimbursement is tied to the Case Mix Index (CMI). The CMI is a measure of complexity of care provided by the hospital and the total reimbursement is the CMI multiplied by a base pay factor, multiplied by the total number of discharges from the hospital. The few studies in the literature have involved logistically complex attempts to improve documentation. We sought to determine whether limited, structured chart review by surgeons can improve CMI in a multihospital system. Methods: We performed a prospective, observational study in a four-hospital academic medical system on all surgical patients from 10/2022 to 7/2026. Two clinically busy trauma surgeons performed limited structured chart reviews for select surgical patients. In the limited intervention period (1/2024-9/2024) this focused only on the trauma division. In the full intervention period (10/2024-7/2026) expanded to all surgical patients across the system. The main outcome was surgical case mix index (CMI). Results: A total of 35,695 patients were included (13,811 pre-intervention, 9,337 limited intervention, 12,547 full intervention). CMI increased significantly among surgical patients across the whole system (2.68 (95%CI 2.65-2.71) to 2.87 (95%CI 2.82-2.92) to 2.97 (95%CI 2.93-3.01), ANOVA p <0.001). Overall length of stay was unchanged among surgical patients across the whole system. The estimated increase in revenue per patient attributable to this intervention was $3,912 in surgical patients across the system. Conclusion: Having trauma surgeons review targeted surgical patients for high value documentation capture significantly enhanced the CMI and estimated revenue of surgical services across a multi-hospital integrated health system. Such documentation reviews may represent a practical method to improve hospital documentation and revenue.

Background: Ambient artificial-intelligence (AI) scribes are being deployed beyond medicine into allied-health practice, but comparative evidence mainly concerns physician documentation. Objective: To compare quality and transcript-grounded clinical-risk error burden of AI-generated and allied health professional notes from identical simulated consultations. Methods: Paired simulation across dietetics, physiotherapy, occupational therapy and speech and language therapy (five consultations each; 20 pairs; 40 notes). Heidi generated an unedited first-draft note and a professional documented each encounter. Professionals rated PDQI-9 quality. Professional and automated review identified unsupported statements and omissions; two language models graded clinical risk. Co-primary outcomes were PDQI-9 score and Critical+High error density. Results: Mean PDQI-9 scores were 38.45 and 35.05 for AI and professional notes (paired difference +3.40, 95% CI +0.01 to +6.79; Wilcoxon p=0.042; dz=0.47). The Bayesian estimate was +3.47 (94% equal-tailed credible interval +0.66 to +6.30; P[AI>professional]=0.989). Automated-reviewer all-error density was 3.35 vs 8.55 per note (professional-minus-AI difference 5.20, 95% CI 3.85 to 6.55); omission density was 1.45 vs 5.20 (difference 3.75, 2.80 to 4.65). Critical+High density was 0.35 vs 0.75 (difference 0.40, -0.10 to 1.00). The pooled quality result was driven principally by dietetics. Conclusions: AI-generated notes achieved modestly higher pooled quality and fewer transcript-grounded errors, chiefly omissions; serious-error evidence was inconclusive. Small discipline samples and incomplete ground truth preclude safety or discipline-wide superiority claims.

BackgroundPatient education is critical to safe and effective postoperative bariatric surgery care. Large language models (LLMs) may improve access to personalized patient education but evidence regarding direct patient use in clinical settings remains limited. We evaluated BariBot, an LLM-powered educational application for patients undergoing bariatric surgery. MethodsWe conducted a prospective, single-arm mixed-methods feasibility study at a tertiary academic medical center from May 2024 to May 2025. Adult patients undergoing bariatric surgery used BariBot independently on postoperative day 1. The application provided conversational postoperative education and adapted language complexity to participants educational attainment. Outcomes included feasibility, usability, acceptability, perceived trustworthiness, and anticipated future use. Usability was assessed using the System Usability Scale (SUS). Immediately after BariBot use, participants completed a semi-structured qualitative interview assessing acceptability, perceived trustworthiness, and anticipated future use. Interview transcripts underwent thematic analysis. ResultsSixteen participants completed the BariBot use session, SUS assessment, and post-intervention interview. No technical failures prevented application use, and no participant requested early discontinuation. The median SUS score was 91.3 (IQR, 81.3-98.1), consistent with excellent perceived usability. Qualitative analysis showed that participants valued the applications immediacy, structured responses, conversational context, ability to address follow-up questions, judgement-free nature, and future availability between clinical encounters. Trust was strengthened by alignment with prior clinical guidance and perceived institutional association, although participants emphasized the need for transparency, clinician oversight, and use as an adjunct rather than a replacement of clinician guidance. ConclusionsBariBot was feasible to implement and was associated with high perceived usability and acceptability postoperatively after bariatric surgery. Larger, longitudinal, and comparative studies are needed to evaluate safety during unsupervised use and determine whether LLM-based educational applications improve patient understanding, engagement, and clinical outcomes.

Background: Digital systems are central to regulating access to care in Brazil's Unified Health System (SUS), but fragmented data can hinder coordination. Objective: This study examines the database structure of SUSfacilMG and identifies modernization priorities, using CORE Saude MG as a comparative reference. Methods: We assessed metadata from 1,586 SUSfacilMG tables, focusing on naming patterns, relationships, and repeated attributes. We then compared architectural features described for CORE Saude MG. Results: Approximately 40% of the assessed tables repeated descriptive attributes, including patient identifiers, suggesting substantial duplication. Metadata alone cannot establish operational failures, security incidents, or effects on care. Conclusion: A phased review of data models, interfaces, and governance is warranted. Standardized exchange, referential integrity, and controlled access should be evaluated alongside service continuity and data quality.

Background: Identification of in-hospital cardiac arrest (IHCA) through manual chart abstraction is the gold standard, but its time-consuming and resource-intensive, which naturally limits its practicality. Diagnostic codes provide an accessible automated alternative, but prior work has shown this method of extraction has poor sensitivity and poor positive predictive value. More accurate and scalable automated methods are needed to support IHCA surveillance and research. Methods: We developed large language model (LLM)-based systems that incorporated physician-defined clinical knowledge through guideline-enriched prompting, supervised fine-tuning, and a self-correction procedure. The systems identified IHCA events and their locations from electronic health record notes. Performance was evaluated at the encounter level against physician-adjudicated chart review and compared with ICD-code-based identification. Results: Compared with an ICD-code-based algorithm (precision 0.68, recall 0.91, F1 score 0.78), the supervised fine-tuned LLM offered substantially better performance for IHCA identification on a curated evaluation corpus, reaching a precision of 0.88, a recall of 1.00, and an F1 score of 0.94. When tested on a larger, unselected validation cohort of 45,525 encounters, a separate LLM configuration - relying on guideline-enriched prompting and a self-correction procedure rather than fine-tuning - achieved a precision of 0.79 for encounter-level IHCA identification, meaning most encounters flagged by the model were genuine IHCA cases. The fine-tuned LLM underperformed on this cohort due to its inability to self-correct its decisions, an issue that can be mitigated by having a separate model perform the correction step. Conclusion: Explicit integration of clinical knowledge through prompting and supervised fine-tuning enabled accurate identification of IHCA events and locations from routine clinical documentation. In a two-pass workflow, these systems could prioritize encounters for clinician adjudication, reduce manual screening requirements, and support more scalable IHCA surveillance and registry development.

BackgroundClinicians increasingly use AI systems to search the medical literature, but current benchmarks do not jointly test whether a response identifies the originating paper and its supporting passage. LitBench evaluates four dimensions: stated versus interpretation-requiring facts, the number and composition of competing papers, one-versus two-paper evidence, and refusal when evidence is absent. MethodsLitBench comprises 1,980 open-access papers (1,000 epilepsy papers holding the answers, 980 decoys from unrelated fields) and 9,472 human-reviewed facts, giving 2,188 single-paper questions and 170 requiring a fact from each of two papers. Difficulty rose by burying the answer among up to 2,000 distractors, then again over live PubMed Central. The same questions were then asked with the answering paper removed, so that refusal was the correct response. Four systems were tested: Gemma-4B, Gemma-12B, Sonnet-5, and DeepSeek-V4-Flash (refusal conditions only). Three model judges scored each answer by majority; two-paper questions counted only when both facts were found. ResultsAcross fixed-corpus single-paper conditions, accuracy ranged from 55.9% to 71.9% for Gemma-4B, 61.2% to 75.5% for Gemma-12B, and 90.0% to 91.3% for Sonnet-5. Similar epilepsy papers were harder than mixed candidate sets for both Gemma configurations. Two-paper accuracy was 5.7%, 17.1%, and 37.8%, respectively. With evidence absent, Gemma-12B refusal fell from 88.3% to 39.9% as candidate sets grew; Gemma-4B almost never refused. Sonnet-5 refused on 94.9% of sampled single-paper and all sampled two-paper cells. DeepSeek-V4-Flash also refused frequently, but on 80.3% of matched answer-present controls. ConclusionsNo system was reliable across every axis tested: accuracy fell as competing papers were added, interpretation-requiring facts were harder than stated ones, two-paper synthesis was rarely achieved, and only some systems recognized absent evidence. What counts as "successful" literature search carries considerable nuance; LitBench can evaluate proposed tools at a fine-grained level.

Background: Robust evidence linking individual surgeon technical performance to patient outcomes in complex minimally invasive surgery remains scarce. Crowd-sourced video assessment offers a scalable alternative to expert review, but its clinical relevance in complex oncologic procedures such as robot-assisted minimally invasive esophagectomy (RAMIE) is unestablished. Methods: In this exploratory study, video recordings of 60 consecutive RAMIE procedures performed by 7 surgeons (2018-2023) were segmented into five representative surgical steps and rated by 93 blinded crowd-raters per segment using a modified GEARS instrument. Tertiles were used to compare top-performing operations and steps against the remaining two terciles for overall and step-specific performance, and associations with perioperative outcomes were analyzed. Results: Top-tercile overall performance was associated with significantly shorter operative time (abdominal, thoracic, and total) and less intraoperative blood loss (all p<0.05), without differences in postoperative outcomes. In exploratory step-specific analyses, top-tertile performance during infracarinal lymphadenectomy was associated with significantly fewer surgical complications (30.0% vs 60.0%; p=0.028), fewer severe surgical complications (25.0% vs 52.5%; p=0.043), and higher lymph node yield (22.8 vs 18.9; p=0.037), while the purse-string suture step showed the broadest operative-time benefit. These step-specific findings are hypothesis-generating and were not adjusted for multiple testing. Conclusion: Crowd-sourced video assessment enables scalable, step-specific evaluation of surgical performance in RAMIE associated with intraoperative blood loss and operative time. Exploratory analyses suggest that the relationship between technical performance and clinical outcomes might be step-specific, with infracarinal lymphadenectomy emerging as risky step linked to postoperative morbidity. These findings support further investigation of individualized performance assessment for quality improvement and risk stratification.