Clinician Responses Research Articles

The effect of visual acuity measurement on triage effectiveness in an ophthalmic emergency department.

The effect of pre-triage visual acuity (VA) measurement on triage accuracy in a busy ophthalmic casualty department was investigated as a possible means to improve triage quality. All 576 patients attending the accident and emergency department (A&E) at the Birmingham and Midland Eye Centre (BMEC) over a period of 4days were included in this prospective cross-sectional study. Patients were assigned to two groups: those who underwent a VA measurement prior to triage (n = 242) and a control group who did not have a VA measurement (n = 234). Clinicians who were masked from the allocation also assessed whether they agreed with the triage decision after assessing each patient. Triage outcomes were recorded for 469 (81%) patients. Those with a pre-triage VA measurement were more likely to be assessed as more urgent (p = 0.005) and less likely to be discharged (p = 0.04). 248 (43%) patients had clinician response with corresponding triage records, of which 136 (55%) had prior VA measurement and 112 (45%) were in the control group. Clinicians responded that patients with VA measurement prior to triage were more accurately triaged than the control group (66% and 54%; p = 0.03). VA measurement prior to triage can help improve triage accuracy and allow better allocation of resources at an overcrowded eye-dedicated emergency department. However, waiting times may increase due to longer triage duration and more patients to be triaged into urgent pathway. Greater staffing resources may be necessary to complement the proposed change to avoid undermining triage efficiency.

System-Wide, Electronic Health Record-Based Medication Alerts for Appropriate Prescribing of Direct Oral Anticoagulants: Pilot Randomized Controlled Trial.

While direct oral anticoagulants (DOACs) have improved oral anticoagulation management, inappropriate prescribing remains prevalent and leads to adverse drug events. Antithrombotic stewardship programs seek to enhance DOAC prescribing but require scalable and sustainable strategies. We present a pilot, prescriber-level randomized controlled trial to assess the effectiveness of electronic health record (EHR)-based medication alerts in a large health system. The pilot assessed prescriber responses to alerts for initial DOAC prescription errors (apixaban and rivaroxaban). A user-centered, multistage design process informed alert development, emphasizing clear indication, appropriate dosing based on renal function, and drug-drug interactions. Alerts appeared whenever a DOAC was being prescribed in a way that did not follow package label instructions. Clinician responses measured acceptability, accuracy, feasibility, and utilization of the alerts. The study ran from August 1, 2022, through April 30, 2023. Only 1 prescriber requested trial exclusion, demonstrating acceptability. The error rate for false alerts due to incomplete data was 6.6% (16/243). Two scenarios with alert design and/or execution errors occurred but were quickly identified and resolved, underlining the importance of a responsive quality assurance process in EHR-based interventions. Trial feasibility issues related to alert-data capture were identified and resolved. Trial feasibility was also assessed with balanced randomization of prescribers and the inclusion of various alerts across both medications. Assessing utilization, 34.2% (83/243) of the encounters (with 134 prescribers) led to a prescription change. The pilot implementation study demonstrated the acceptability, accuracy, feasibility, and estimates of the utilization of EHR-based medication alerts for DOAC prescriptions and successfully established just-in-time randomization of prescribing clinicians. This pilot study sets the stage for large-scale, randomized implementation evaluations of EHR-based alerts to improve medication safety. ClinicalTrials.gov NCT05351749; https://clinicaltrials.gov/study/NCT05351749.

Adolescent, caregivers, and therapists’ experiences of youth and family suicide intervention: A qualitative study

Objective This study explores the experience of an individual and family therapy youth suicide intervention from the perspectives of seven psychotherapy triads (young people aged 12–18, their parents/caregivers and therapists). Method Data were collected through semi-structured individual interviews and analyzed using consensual qualitative research methods. Results Four domains were identified: Focusing on the youth-parent relationship, individual therapy for young people and their parents, conjoint therapy, and public service structures and systemic practice that facilitated tailored treatment. All participant groups valued intervention that improved the youth-parent relationship; however, they held different views about its influence on suicidality and recovery. Separate youth and parent therapy was crucial for facilitating the tailoring of treatment and enhancing the reparative potential of conjoint therapy. Barriers to productive intervention included inaccessible, fragmented, and siloed treatment that excluded parents, stigmatizing clinician responses, and acute care that was not attuned to need or developmental context. Helpful intervention was systemically and attachment-informed, multi-disciplinary, individually tailored, and integrated freely available specialist therapy with crisis and inpatient care. Conclusion A relational, nuanced, and flexible approach is needed to tailor youth suicide treatment in the context of the complexity of youth-parent relationships.

Monetary Incentives in Clinician Surveys: An Analysis and Systematic Review With a Focus on Establishing Best Practices.

Surveys involving health care providers continue to be characterized by low and declining response rates (RRs), and researchers have utilized various strategies to increase survey participation. An important approach is to employ monetary incentives to improve survey response. Using a systematic review and analyses of 100 randomized comparisons (published in 48 papers) between monetary incentives and a non-incentive condition, this paper seeks to advance the understanding of best practices for using monetary incentives in clinician surveys. These analyses show even small incentives (≤$2) to be effective in improving clinician response relative to non-incentive subgroups, with diminished returns associated with serial incremental increases above that amount up to amounts greater than $25, at which point there is an appreciable improvement, supporting the use of higher incentives in this population. Cash and direct cash equivalents (e.g., cash cards and checks) produced greater odds of survey participation compared to vouchers, lotteries and charitable contributions, with lotteries and charities being the least effective forms of monetary incentive. Survey mode, timing and ethical considerations are also addressed. Noting the challenges associated with surveying clinicians, researchers must make every effort to improve access to this difficult-to-reach population by implementing appropriate incentive-based strategies designed to improve participation rates.

Perspectives on Artificial Intelligence–Generated Responses to Patient Messages

This cross-sectional study of patient queries in US electronic health records examines laypersons’ satisfaction with answers generated with artificial intelligence (AI) compared with clinician responses, and whether results were concordant with clinician-determined quality of AI responses.

Clinician response to the 2021 USPSTF recommendation for colorectal cancer screening in average risk adults aged 45-49 years

IntroductionIn 2021, the USPSTF lowered the recommended age of colorectal cancer (CRC) screening initiation from 50 to 45 years. We assessed clinician response to the updated guideline in a major health system. MethodsThis was a retrospective cohort study of average-risk, CRC screening-naïve adults aged 45-50 years with a primary care appointment between July 2018 and February 2023. We defined the pre-guideline change period as July 2018-February 2020 (pre-period) and the post-guideline change period as July 2021-February 2023 (post-period). Clinician ordering of any CRC screening type was assessed. Mixed effects Poisson regression was used to model the incidence rate ratio (IRR) of a patient receiving a screening order, including an interaction between age (45-49 years versus 50 years) and time-period (pre- versus post-guideline change.) Variation in screening orders were also described by calendar quarter and clinician. ResultsThere were 28,114 patients in the pre-period and 22,509 in the post-period. Compared to patients aged 40-49 years in the pre-period, those in the post-period were more likely to have screening ordered (IRR:12.1; 95%CI:11.3-13.0). The screening ordering rate increased for 50-year-olds from the pre- to the post-period (IRR:1.08;95%CI:1.01-1.16) and was slightly higher than that of 45-49-year-olds in the post-period (IRR:1.08; 95%CI:1.02-1.14). All clinicians increased their ordering rate for patients aged 45-49 years. Within five months of the guideline change, the ordering rate for 45-49-year-olds and 50-year-olds was nearly the same. ConclusionsRapidly following the guideline change, clinicians increased their screening ordering rate for 45-49-year-olds, indicating almost complete uptake of the recommendation.

Expert-Guided Large Language Models for Clinical Decision Support in Precision Oncology.

Rapidly expanding medical literature challenges oncologists seeking targeted cancer therapies. General-purpose large language models (LLMs) lack domain-specific knowledge, limiting their clinical utility. This study introduces the LLM system Medical Evidence Retrieval and Data Integration for Tailored Healthcare (MEREDITH), designed to support treatment recommendations in precision oncology. Built on Google's Gemini Pro LLM, MEREDITH uses retrieval-augmented generation and chain of thought. We evaluated MEREDITH on 10 publicly available fictional oncology cases with iterative feedback from a molecular tumor board (MTB) at a major German cancer center. Initially limited to PubMed-indexed literature (draft system), MEREDITH was enhanced to incorporate clinical studies on drug response within the specific tumor type, trial databases, drug approval status, and oncologic guidelines. The MTB provided a benchmark with manually curated treatment recommendations and assessed the clinical relevance of LLM-generated options (qualitative assessment). We measured semantic cosine similarity between LLM suggestions and clinician responses (quantitative assessment). MEREDITH identified a broader range of treatment options (median 4) compared with MTB experts (median 2). These options included therapies on the basis of preclinical data and combination treatments, expanding the treatment possibilities for consideration by the MTB. This broader approach was achieved by incorporating a curated medical data set that contextualized molecular targetability. Mirroring the approach MTB experts use to evaluate MTB cases improved the LLM's ability to generate relevant suggestions. This is supported by high concordance between LLM suggestions and expert recommendations (94.7% for the enhanced system) and a significant increase in semantic similarity from the draft to the enhanced system (from 0.71 to 0.76, P = .01). Expert feedback and domain-specific data augment LLM performance. Future research should investigate responsible LLM integration into real-world clinical workflows.

Crying wolf, alarm safety and management in paediatrics: A scoping review.

To provide a contemporaneous evidentiary overview of neonatal and paediatric studies investigating alarm-related patient safety and alarm system management. Furthermore, to describe how clinical alarm burden is captured and reported, to identify clinical devices that contribute to alarm burden, to explore alarm-related and patient safety measures and terminologies and to review alarm management initiatives. Scoping review. A systematic search of PubMed, Embase, Cumulative Index to Nursing and Allied Health Literature, Scopus and EBSCOhost was conducted from 2013 to 2023 using predetermined search terms, index terms, medical subject headings and truncation. Observational and qualitative studies with neonatal and paediatric populations reporting monitoring and alarm practices; and interventional studies reporting the success of alarm safety interventions were included. The quality of the included studies was assessed using the mixed methods appraisal tool. The search yielded 37 studies of acceptable quality. The majority explored alarm burden associated with physiological monitoring (n = 35; 95%). Alarm definitions were reported in 46% (n = 17) of studies, and commonly included what constituted actionable and non-actionable alarms. While 32% (n = 12) of studies considered alarms in relation to clinical outcomes surrounding patient safety, clinician response to alarms was only reported in 19% (n = 7) of studies. Alarm and monitoring interventions were assessed in 51% (n = 19) of included studies, with categorization into six domains: changing alarm parameters, clinician education, communication and planning, technology, alarm ordering and standardization or guidelines. This review has demonstrated the enormity of alarms in clinical settings, heterogeneity of alarm definitions and outlined interventions associated with alarm burden and patient safety. Strategies to ensure appropriate alarm limits are set and clinicians are empowered through education to recognize and respond appropriately to alarms can maximize patient safety. This review adheres to the preferred reporting items for systematic reviews and meta-analysis protocols extension for scoping reviews. No patient or public contribution.

Improving Patient Understanding of Emergency Department Discharge Instructions.

Previous studies have shown that patients in the emergency department (ED) are frequently given incomplete discharge instructions that are written at least four grade levels above the recommended sixth-grade reading level, leading to poor understanding. Our aims in this study were to implement standardized discharge instructions containing six key components written at a more appropriate reading level for common emergency department (ED) diagnoses to improve patient understanding. We conducted this study in a 20-bed ED at an urban Veteran's Administration hospital. Data was collected via in-person patient and clinician interviews. Patient interviews were conducted after patients received their discharge instructions. We compared patient responses to clinician responses and marked them as incorrect, partially correct, or correct with a score of 0, 0.5, or 1, respectively. The maximum possible score for each interview was six. Six key components of discharge instructions were asked about: diagnosis; new medications; at-home care; duration of illness; reasons to return; and follow-up. There were 25 patients in the pre-intervention group and 20 in the intervention group with the standardized set of instructions. We performed a Mann-Whitney U test on the total interview scores in the control and intervention groups and conducted a sub-analysis on the individual scores for each of the six key components. The patients in the intervention demonstrated a statistically significant increase in patient-clinician correlation when compared to the patients in the pre-intervention group overall (P < 0.05), and two of the six key components of the discharge instructions individually showed statistically significant increase in patient-clinician correlation when standardized discharge instructions were used. Patients who received the standardized discharge instructions had improved understanding of their discharge instructions. Future opportunities extending off this pilot study include expanding the number of diagnoses for which standardized instructions are used and investigating patient-centered outcomes related to these instructions.

Clinician actions in response to Adverse Childhood Experience (ACE) screening

Background and ObjectiveAdverse Childhood Experiences(ACEs) have a powerful influence on mental health, physical health, and life expectancy. Screening for ACEs and the clinician response to ACEs are critical to addressing the health and well-being of children; however, little is known about the actions clinicians take in response to ACE screening. Therefore, we aimed to examine clinician responses to ACE screening at five California pediatric clinics in a large public health care system. MethodsPatient demographics, indicators of social and behavioral determinants of health (e.g., housing insecurity), the number of ACEs endorsed on a screening instrument, and the actions clinicians took in response to each ACE screen were collected. Data was collected from May to October 2021. These data were used to examine the association between number of ACEs reported and clinician response, controlling for patient demographics and their social and behavioral determinants of health using multiple logistic regression. ResultsFive participating pediatric clinics conducted 2,652 ACE screens in six-months. Clinicians documented an action twice as often when ACEs were present, after controlling for patient demographics and their social and behavioral determinants of health (odds ratio(OR) = 2.2, 95 % confidence interval(CI): 1.5–3.3, p-value < 0.0001). Clinicians were three times more likely to record referrals to mental health clinicians, social workers, and community organizations relative to anticipatory guidance when the number of ACEs increased from one to three to four or more (OR=3.2, 95 %CI: 1.6–6.5, p < 0.0001). ConclusionFindings provide early information that ACE screening results are associated with patient care.

Utilization of Real-World Data in the Intervention to Reduce Early Peanut Allergy in Children (iREACH) Trial

ObjectiveTo demonstrate the use of linked, real-world data in the context of a pediatric practice-based clinical trial. ApproachThe Intervention to Reduce Early Peanut Allergy in Children (iREACH) is a two-arm, cluster randomized, controlled trial. The intervention included clinician education and clinical decision support to increase pediatric clinician adherence to the NIAID’s 2017 Prevention of Peanut Allergy guidelines during infants’ 4- and 6-month well child care visits. Electronic health record (EHR) data were shared with the data coordinating center under a Health Insurance Portability and Accountability Act (HIPAA) waiver. All caregivers of eligible infants were invited to complete two surveys via REDCap, an online survey platform. The consents included an optional HIPAA authorization to link EHR data with survey responses and authorization to collect and link administrative data from the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) program, if applicable. ResultsOverall, 30 Illinois pediatric practices participated. EHR data was collected for 18,480 infants of whom 4,331 had any caregiver survey data (23.4% response rate). Permission to link survey responses to EHR data was obtained for 3,950 infants among those with any survey data (91.2%). Consent for linking WIC data with study data was obtained for 229 infants and 194 mothers. ConclusionsAuthorization for data linkage in the consent forms allowed investigators to amass data to compare caregiver and clinician responses, generate hypotheses, and propose future research outside the scope of the main trial. ImplicationsMultiple real-world data sources can be effectively combined to extend the findings of clinical trials.

Domestic and Family Violence Screening and Response: A Prospective, Cross-Sectional, Mixed Methods Survey in Private Mental Health Clients.

Most domestic and family violence (DFV) research has focused on establishing prevalence and screening rates in public health and community samples. This study sought to address a gap in the literature by evaluating DFV screening and response practices in a private mental healthcare inpatient service and determining if clients of the service had unmet DFV needs. A prospective, convenience sample, mixed methods, cross-sectional survey of adult inpatient mental health consumers was employed. Sixty-two participants completed the Royal Melbourne Hospital Patient Family Violence Survey. Quantitative Likert-type and categorical responses were collated and analysed descriptively (count and percentage). Free-text responses were analysed using qualitative description within a content analysis framework. Sixty-five percent of participants had been screened for at least one DFV issue, on at least one occasion, with 35% not being screened, to their recall. Twenty-three percent reported disclosing DFV concerns, 82% felt very supported by the clinician's response to their disclosure, and 86% were provided with information they found helpful. Unmet needs were identified in 13% of participants, who had wanted to disclose DFV concerns but not feel comfortable to do so. No unscreened respondents disclosed DFV concerns, highlighting the need to uphold best practice guidelines for direct enquiry. Most disclosing clients were positive about the support they received. Indicated areas for improvement were screening rates, active follow-up, increasing psychology support levels and safety planning.

Clinician Experiences of and Responses to the Challenges of Working with Patients in the Australian Compensation Setting.

Previous research indicates that the compensation process is stressful for people with a compensable injury, contributing to poorer recovery. However, little is known about the challenges faced by clinicians who work in this setting. This study aims to qualitatively explore the experiences of clinicians delivering care to patients with compensable injuries in Australia. Semi-structured interviews were conducted with 26 experienced clinicians providing care to compensable patients in Australia. The interviews were transcribed, and data were analysed using reflective thematic analysis. Participants described their work as challenging due to factors such as high patient distress, poor clinical outcomes, and high administrative load. However, they responded to these challenges in varying ways. While some reported feelings of vicarious injustice, emotional exhaustion, and self-doubt, others derived a sense of meaning, purpose, and mastery from these challenges. Clinician responses to the challenge of working with people with a compensation claim were associated with access to mentoring, continuous education and training, and a supportive workplace culture. Clinicians reported both positive and negative responses to the challenges of working with people with a compensable injury. Coping strategies that were associated with more positive reactions included seeking further education, mentoring, peer support. Prioritising these support systems is important for clinician wellbeing and patient outcomes.

Comparison of Use Rates of Telehealth Services for Substance Use Disorder During and Following COVID-19 Safety Distancing Recommendations: Two Cross-Sectional Surveys.

The COVID-19 social distancing guidelines resulted in a dramatic transition to telephone and video technologies to deliver substance use disorder (SUD) treatment. Before COVID-19, the question was "Will telehealth ever take hold for SUD services?" Now that social distancing guidelines have been lifted, the question is "Will telehealth remain a commonly used care modality?" The principal purpose of this investigation was to examine the extent to which telehealth use in SUD service settings persisted following the lifting of COVID-19 safety distancing recommendations. Additionally, the study aimed to explore practitioners' perceptions of telehealth convenience and value after its regular implementation during the pandemic. Specifically, the goal of this study was to compare telehealth activity between time intervals: May-August 2020 (during peak COVID-19 safety distancing recommendations) and October-December 2022 (following discontinuation of distancing recommendations). Specifically, we compared (1) telehealth technologies and services, (2) perceived usefulness of telehealth, (3) ease of use of telephone- and video-based telehealth services, and (4) organizational readiness to use telehealth. An online cross-sectional survey consisting of 108 items was conducted to measure the use of telehealth technologies for delivering a specific set of SUD services in the United States and to explore the perceived readiness for use and satisfaction with telephonic and video services. The survey took approximately 25-35 minutes to complete and used the same 3 sets of questions and 2 theory-driven scales as in a previous cross-sectional survey conducted in 2020. Six of 10 Regional Addiction Technology Transfer Centers funded by the Substance Abuse and Mental Health Services Administration distributed the survey in their respective regions, collectively spanning 37 states. Responses of administrators and clinicians (hereafter referred to as staff) from this 2022 survey were compared to those obtained in the 2020 survey. Responses in 2020 and 2022 were anonymous and comprised two separate samples; therefore, an accurate longitudinal model could not be analyzed. A total of 375 staff responded to the 2022 survey (vs 457 in 2020). Baseline organizational characteristics of the 2022 sample were similar to those of the 2020 sample. Phone and video telehealth utilization rates remained greater than 50% in 2022 for screening and assessment, case management, peer recovery support services, and regular outpatient services. The perceived usefulness of phone-based telehealth was higher in 2022 than in 2020 (mean difference [MD] -0.23; P=.002), but not for video-based telehealth (MD -0.12; P=.13). Ease of use of video-based telehealth was perceived as higher in 2022 than in 2020 (MD-0.35; P<.001), but no difference was found for phone-based telehealth (MD -0.12; P=.11). From the staff's perspective, patients had greater readiness for using telehealth via phone than video, but the staff perceived their personal and organizational readiness for using telehealth as greater for video-based than for phone-based telehealth. Despite lower telephone and video use in 2022 for telehealth services than in 2020, both modalities continue to be perceived positively. Future research may further determine the relative cost and clinical effectiveness of video-based services and thereby help to address some sources of the noted challenges to implementation by SUD organizations.

Continuous pulse oximetry monitoring in children hospitalized with bronchiolitis: A qualitative analysis of clinicians'justifications.

Continuous pulse oximetry (cSpO2) monitoring use outside established guidelines is common in children hospitalized with bronchiolitis. We analyzed clinicians' real-time rationale for continuous monitoring in stable children with bronchiolitis not requiring supplemental oxygen. Data for this study were collected as part a multicenter deimplementation trial for cSpO2in children hospitalized with bronchiolitis. We analyzed 371 clinician responses across 36 hospitals; 258 (70%) responses did not include a clinical reason for monitoring ("nonclinical"; e.g., respondent forgot to discontinue monitoring, did not know why the patient was monitored, or was following an order). The remaining 113 (30%) responses contained a clinical reason for monitoring ("clinical"; e.g., recently requiring oxygen, physical exam concerns, or concerns relating to patient condition or history). Strategies to reduce unnecessary monitoring should include changes in workflow to facilitate shared understanding of monitoring goals and timely discontinuation of monitoring.

Evaluation of Real-World Tumor Response Derived From Electronic Health Record Data Sources: A Feasibility Analysis in Patients With Metastatic Non-Small Cell Lung Cancer Treated With Chemotherapy.

Real-world data (RWD) holds promise for ascribing a real-world (rw) outcome to a drug intervention; however, ascertaining rw-response to treatment from RWD can be challenging. Friends of Cancer Research formed a collaboration to assess available data attributes related to rw-response across RWD sources to inform methods for capturing, defining, and evaluating rw-response. This retrospective noninterventional (observational) study included seven electronic health record data companies (data providers) providing summary-level deidentified data from 200 patients diagnosed with metastatic non-small cell lung cancer (mNSCLC) and treated with first-line platinum doublet chemotherapy following a common protocol. Data providers reviewed the availability and frequency of data components to assess rw-response (ie, images, radiology imaging reports, and clinician response assessments). A common protocol was used to assess and report rw-response end points, including rw-response rate (rwRR), rw-duration of response (rwDOR), and the association of rw-response with rw-overall survival (rwOS), rw-time to treatment discontinuation (rwTTD), and rw-time to next treatment (rwTTNT). The availability and timing of clinician assessments was relatively consistent across data sets in contrast to images and image reports. Real-world response was analyzed using clinician response assessments (median proportion of patients evaluable, 77.5%), which had the highest consistency in the timing of assessments. Relative consistency was observed across data sets for rwRR (median 46.5%), as well as the median and directionality of rwOS, rwTTD, and rwTTNT. There was variability in rwDOR across data sets. This collaborative effort demonstrated the feasibility of aligning disparate data sources to evaluate rw-response end points using clinician-documented responses in patients with mNSCLC. Heterogeneity exists in the availability of data components to assess response and related rw-end points, and further work is needed to inform drug effectiveness evaluation within RWD sources.

Morality and emotion in therapy clients’ accounts for suicidality

ABSTRACT Suicide is a substantial public health problem, yet little is understood about the ways in which clinicians and clients discuss suicide attempts. This study investigates commonalities in psychology clients’ accounts of suicide attempts that were survived, or aborted after being enacted or planned and prepared. We used conversation analysis to investigate conversations within psychosocial (risk-assessment) sequences during four initial psychology consultations (interviews) recorded in an Australian voluntary outpatient multidisciplinary healthcare clinic. Analysis showed that clients’ suicidality accounts routinely included reports of morally problematic treatment by others, with clients typically seeking clinician responses to these elements of their accounts. Clinicians were not, however, observed to engage the moral dimensions of clients’ accounts and instead routinely oriented to the impact of these behaviours on clients. We conclude by discussing the implications and consequences of these differing orientations for psychologists, clients, and future suicide-risk research.

Large language models for precision oncology: Clinical decision support through expert-guided learning.

e13609 Background: Precision oncology revolutionized cancer treatment by identifying molecular biomarkers to guide personalized care. The ever-growing body of medical literature presents a challenge for oncologists researching targeted therapies. While recent studies investigated large language models (LLMs) to streamline this process, LLM reliance on general rather than medical knowledge limits clinical relevance and trustworthiness. To address these limitations, we developed a retrieval augmented generation (RAG) system that integrates PubMed clinical studies, trial databases and oncological guidelines with LLMs to support targeted treatment recommendations. The Molecular Tumor Board (MTB) at the Center of Personalized Medicine (ZPMTUM) guided and evaluated treatment options proposed by the LLM to assess their applicability for clinical decision support. Methods: We used 10 publicly accessible fictional patient cases with 7 tumor types and 59 distinct molecular alterations. Our LLM system MEREDITH (Medical Evidence Retrieval and Data Integration for Tailored Healthcare) consists of Google's Gemini Pro, enhanced with RAG and Chain-of-Thought (CoT) prompting. To establish a benchmark, clinical experts at ZPMTUM manually annotated the cases. Informed by MTB expert feedback, we iteratively improved our LLM system from a draft system relying on PubMed-indexed data to an enhanced system, which replicated expert annotation processes by incorporating oncology guidelines, drug availability and trial databases (ClinicalTrials.gov, QuickQueck.de). ZPMTUM assessed credibility and clinical relevance of manually annotated and LLM-generated recommendations. Patient-level data on (likely) pathogenic molecular alterations and recommended treatment options were summarized using median and interquartile range (IQR). Semantic similarity between LLM and clinician responses was assessed using cosine similarity of text vector embeddings; paired t-test evaluated significance. Results: The median of (likely) pathogenic molecular alterations per patient was 2.5 (IQR: 2-3). ZPMTUM identified a median of 2 treatment options per patient (IQR: 1-3), while the enhanced LLM identified a median of 4 (IQR: 3-5). MEREDITH proposed multiple relevant treatment suggestions, including therapies based on preclinical studies, and molecular interactions, for further assessment by the MTB. ZPMTUM prioritized the most suitable clinical option. The mean semantic textual similarity of LLM responses increased significantly from 0.69 in the draft system to 0.76 in the enhanced system (p &lt;0.001). Thus, feedback from ZPMTUM enhanced the model's ability to align its responses with clinician thought processes. Conclusions: Leveraging expert thought processes to instruct LLMs holds promise as a novel decision support tool for precision oncology.

Clinician response after receipt of abnormal pediatric ambulatory blood pressure monitoring - characteristics associated with inertia and action.

Ambulatory Blood Pressure Monitoring (ABPM) is recommended for diagnosis and management of hypertension. We aimed to identify characteristics associated with physician action after receipt of abnormal findings. This was a retrospective cross-sectional analysis of patients 5-22years old who underwent 24-h ABPM between 2003-2022, met criteria for masked or ambulatory hypertension, and had a pediatric nephrology clinic visit within 2weeks of ABPM. "Action" was defined as medication change/initiation, lifestyle or adherence counseling, evaluation ordered, or interpretation with no change. Characteristics of children with/without 1 or more actions were compared using Student t-tests and Chi-square. Regression analyses explored the independent association of patient characteristics with physician action. 115 patients with masked (n = 53) and ambulatory (n = 62) hypertension were included: mean age 13.0years, 48% female, 38% Black race, 21% with chronic kidney disease, and 25% overweight/obesity. 97 (84%) encounters had a documented physician action. Medication change (52%), evaluation ordered (40%), and prescribed lifestyle change (35%) were the most common actions. Adherence counseling for medication and lifestyle recommendations were documented in 3% of encounters. 24-h, wake SBP load, and sleep DBP load were significantly higher among those with physician action. Patients with > 1 action had greater adiposity, SBP, and dipping. Neither age, obesity, nor kidney disease were independently associated with physician action. While most abnormal ABPMs were acted upon, 16% did not have a documented action. Greater BP load was one of the few characteristics associated with physician action. Of potential actions, adherence counseling was underutilized.

Informal Peer Support and Intentional Acts of Kindness May Attenuate the Impact of Work-Related Stressors on Compassion Satisfaction, Secondary Traumatic Stress, and Burnout of Emergency Medical Services Clinicians

ObjectiveEmergency medical services (EMS) Code Lavender was developed to support EMS clinicians after stressful events via consistent recognition of events, informal peer support, and intentional acts of kindness. This study evaluated changes in burnout screening tool responses of EMS clinicians in response to program implementation and the coincidental start of coronavirus disease 2019. MethodsAnonymous surveys with demographic questions and 2 burnout screening tools were distributed before program implementation (spring 2020) and 20 months later (fall 2021). Analysis included t-tests, Fisher exact tests, and multivariable linear regression. ResultsSeventy-seven preprogram (59% response rate) and 108 intraprogram (88% response rate) survey responses were included. No changes existed between preprogram and intraprogram responses across all subscale scores. Sex was associated with depersonalization subscale scores, with men having scores 1.53 (95% confidence interval [CI] 0.11-2.95) higher than women. Compared with emergency medical technicians, paramedics had higher compassion satisfaction (OR 3.50; 95% CI 1.79-5.70) and personal accomplishment scores (OR 2.40; 95% CI 1.08-3.71). Transport nurses had higher personal accomplishment (OR 3.29; 95% CI 1.18-5.40), depersonalization (OR 3.73; 95% CI 1.19-6.26), and rates of burnout symptoms (OR 0.54; 95% CI 0.09-0.98) than emergency medical technicians. ConclusionThe organizational commitment, peer support, and authentic leadership of EMS Code Lavender may attenuate work-related stressors among EMS clinicians.