Enhancing Early Detection of Cognitive Decline in the Elderly: A Comparative Study Utilizing Large Language Models in Clinical Notes.

Background: Large language models (LLMs) have shown promising performance in various healthcare domains, but their effectiveness in identifying specific clinical conditions in real medical records is less explored. This study evaluates LLMs for detecting signs of cognitive decline in real electronic health record (EHR) clinical notes, comparing their error profiles with traditional models. The insights gained will inform strategies for performance enhancement. Methods: This study, conducted at Mass General Brigham in Boston, MA, analyzed clinical notes from the four years prior to a 2019 diagnosis of mild cognitive impairment in patients aged 50 and older. We used a randomly annotated sample of 4,949 note sections, filtered with keywords related to cognitive functions, for model development. For testing, a random annotated sample of 1,996 note sections without keyword filtering was utilized. We developed prompts for two LLMs, Llama 2 and GPT-4, on HIPAA-compliant cloud-computing platforms using multiple approaches (e.g., both hard and soft prompting and error analysis-based instructions) to select the optimal LLM-based method. Baseline models included a hierarchical attention-based neural network and XGBoost. Subsequently, we constructed an ensemble of the three models using a majority vote approach. Results: GPT-4 demonstrated superior accuracy and efficiency compared to Llama 2, but did not outperform traditional models. The ensemble model outperformed the individual models, achieving a precision of 90.3%, a recall of 94.2%, and an F1-score of 92.2%. Notably, the ensemble model showed a significant improvement in precision, increasing from a range of 70%-79% to above 90%, compared to the best-performing single model. Error analysis revealed that 63 samples were incorrectly predicted by at least one model; however, only 2 cases (3.2%) were mutual errors across all models, indicating diverse error profiles among them. Conclusions: LLMs and traditional machine learning models trained using local EHR data exhibited diverse error profiles. The ensemble of these models was found to be complementary, enhancing diagnostic performance. Future research should investigate integrating LLMs with smaller, localized models and incorporating medical data and domain knowledge to enhance performance on specific tasks.

Multi-Institution Survey of Accepting Physicians' Perception of Appropriate Reasons for Interhospital Transfer: A Mixed-Methods Evaluation.

OBJECTIVES: There is a lack of evidence-based guidelines to direct best practices in interhospital transfers (IHTs). We aimed to identify frontline physicians' current and ideal reasons for accepting IHT patients to inform future IHT research and guidelines. METHODS: We conducted a cross-sectional survey of hospitalist physicians across 11 geographically diverse hospitals. The survey asked respondents how frequently they currently consider and should consider various factors when triaging IHT requests. Responses were dichotomized into "highly considered" and "less considered" factors. Frequencies of the "highly considered" factors (current and ideal) were analyzed. Write-in responses were coded into themes within a priori domains in a qualitative analysis. RESULTS: Of the 666 hospitalists surveyed, 238 (36%) responded. Respondents most frequently identified the need for specialty procedural and nonprocedural care and bed capacity as factors that should be considered when triaging IHT patients in current and ideal practice, whereas the least frequently considered factors were COVID-related care, insurance/financial considerations, and patient/family preference. More experienced respondents considered patient/family preference more frequently in current and ideal practice compared with less experienced respondents (33% versus 11% [ P = 0.0001] and 26% versus 9% [ P = 0.01], respectively). Qualitative analysis identified several themes in the domains of Criteria for Acceptance, Threshold for Acceptance, and Indications for Physician-to-Physician Communication. CONCLUSIONS: This geographically diverse sample of hospitalist physicians responsible for accepting IHT patients showed general agreement between primary factors that are currently and that should be considered for IHT acceptance, with greatest weight placed on patients' need for specialty care.

Investigating racial and ethnic disparities in interhospital transfer within an academic integrated healthcare system: A matched cohort study.

The presence of racial and ethnic disparities in interhospital transfer (IHT) within integrated healthcare systems has not been fully explored. We matched Black and Latinx patients admitted to community hospitals in our integrated healthcare system between June 2015 and December 2019 to White patients by origin hospital, age, time of year, and disease severity. We performed conditional logistic regression models to determine if race or ethnicity was associated with IHT in one of the tertiary academic medical centers in the system, adjusting for covariates. The sample contained 107,895 admissions (82.6% White, 7.8% Black, and 9.6% Latinx). Transfer rates were 2.2% versus 2.2% after the Black/White match and 1.8% versus 1.8% after the Latinx/White match. After adjusting for covariates, there was no association between race or ethnicity and IHT (Black vs. White odds ratio [OR]: 0.87, 95% confidence interval [CI]: 0.72-1.07; Latinx vs. White OR: 1.05, 95% CI: 0.79-1.40). This may be due to reduced barriers to transfer with an integrated healthcare system.

Advancing health information during interhospital transfer: An interrupted time series.

INTRODUCTION: Although the transfer of patients between acute care hospitals (interhospital transfer, IHT) is common, health information exchange (HIE) during IHT remains inadequate, with fragmented communication and unreliable access to clinical information. This study aims to design, implement, and rigorously evaluate the implementation of a HIE platform to improve data access during IHT. METHODS AND ANALYSIS: Study subjects include patients aged >18 transferred to the medical, cardiology, oncology, or intensive care unit (ICU) services at an 800-bed quaternary care hospital; and healthcare workers involved in their care. The first aim of this study is to optimize clinician workflow, data visualization, and interoperability through user-centered design sessions for HIE platform development. The second aim is to evaluate the impact of the intervention on clinician-reported medical errors among 500 pre- and 500 postintervention IHT patients using interrupted time series methodology, adjusting for confounding variables and temporal trends. The third aim is to evaluate intervention fidelity, use and perceived usability of the platform, and barriers and facilitators of implementation from interprofessional stakeholder input, using mixed-methods evaluation. The fourth aim is to consolidate key findings to create a toolkit for spread and sustainability. ETHICS AND DISSEMINATION: We will track patient safety endpoints and clinician workflow burdens and ensure the protection of patient data throughout the study. We will disseminate our findings via the creation of a toolkit for spread and sustainability, partnering with our funder (AHRQ) for dissemination, and communicating our results via abstracts and publications.

Implementation of a standardised accept note to improve communication during inter-hospital transfer: a prospective cohort study.

IMPORTANCE: The transfer of patients between hospitals (interhospital transfer, IHT), exposes patients to communication errors and gaps in information exchange. OBJECTIVE: To design and implement a standardised accept note to improve communication during medical service transfers, and evaluate its impact on patient outcomes. DESIGN: Prospective interventional cohort study. SETTING: A 792-bed tertiary care hospital. PARTICIPANTS: All patient transfers from any acute care hospital to the general medicine, cardiology, oncology and intensive care unit (ICU) services between August 2020 and June 2022. INTERVENTIONS: A standardised accept note template was developed over a 9-month period with key stakeholder input and embedded in the electronic health record, completed by nurses within the hospital's Access Centre. MAIN OUTCOMES AND MEASURES: Primary outcome was clinician-reported medical errors collected via surveys of admitting clinicians within 72 hours after IHT patient admission. Secondary outcomes included clinician-reported failures in communication; presence and 'timeliness' of accept note documentation; patient length of stay (LOS) after transfer; rapid response or ICU transfer within 24 hours and in-hospital mortality. All outcomes were analysed postintervention versus preintervention, adjusting for patient demographics, diagnosis, comorbidity, illness severity, admitting service, time of year, hospital COVID census and census of admitting service and admitting team on date of admission. RESULTS: Of the 1004 and 654 IHT patients during preintervention and postintervention periods, surveys were collected on 735 (73.2%) and 462 (70.6%), respectively. Baseline characteristics were similar among patients in each time period and between survey responders and non-responders. Adjusted analyses demonstrated a 27% reduction in clinician-reported medical error rates postimplementation versus preimplementation (11.5 vs 15.8, adjusted OR (aOR) 0.73, 95% CI 0.53 to 0.99). Secondary outcomes demonstrated lower adjusted odds of clinician-reported failures in communication (aOR 0.88; 0.78 to 0.98) and rapid response/ICU transfer (aOR 0.57; 0.34 to 0.97), and improved presence (aOR 2.30; 1.75 to 3.02) and timeliness (-21.4 hours vs -8.7 hours, p<0.001) of accept note documentation. There were no significant differences in LOS or mortality. CONCLUSIONS AND RELEVANCE: Among 1658 medical patient transfers, implementing a standardised accept note was associated with improved presence and timeliness of accept note documentation, clinician-reported medical errors, failures in communication and clinical decline following transfer, suggesting that improving communication during IHT can improve patient outcomes.

A multilocus genetic risk score for obesity: Association with BMI and metabolic alterations in a cohort with severe obesity.

Genome wide association studies have identified numerous single nucleotide polymorphisms (SNPs) associated with obesity, yet effect sizes of individual SNPs are small. Therefore, the aim of our study was to investigate whether a genetic risk score (GRS) comprising risk alleles of SNPs identified in the GIANT consortium meta-analyses shows association with body mass index (BMI) and other BMI related metabolic alterations in a cohort with an extreme phenotype. Genotyping of 93 SNPs was performed in 314 obese individuals (mean BMI 40.5 ± 7.8 kg/m², aged 45 ± 12 years), participating in a standardized weight reduction program, and in 74 lean controls (mean BMI 24.6 ± 3.3 kg/m², aged 41.7 ± 13.4 years). Allele numbers of all 93 SNPs were added to a GRS. Anthropometric parameters, parameters of glucose/insulin and lipid metabolism were assessed standardized after a 12 hours fast. GRS was significantly different between controls and obese individuals (unweighted GRS: 86.6 vs 89.0, P = .002; weighted GRS: 84.9 vs 88.3, P = .005). Furthermore, linear regression analysis showed significant associations of GRS with BMI ( P < .0001), weight ( P = .0005), waist circumference ( P = .0039), fat mass ( P < .0001) and epicardial fat thickness ( P = .0032), yet with small effect sizes ( r ² < 0.06). In conclusion, in our study GRS could differentiate between extreme obese and lean individuals, and was associated with BMI and its related traits, yet with small effect sizes.

Safety and utility of indwelling pleural catheters in lung transplant recipients.

INTRODUCTION: The safety and efficacy of indwelling pleural catheters (IPCs) in lung allograft recipients is under-reported. METHODS: We performed a multicenter, retrospective analysis between 1/1/2010 and 6/1/2022 of consecutive IPCs placed in lung transplant recipients. Outcomes included incidence of infectious and non-infectious complications and rate of auto-pleurodesis. RESULTS: Seventy-one IPCs placed in 61 lung transplant patients at eight centers were included. The most common indication for IPC placement was recurrent post-operative effusion. IPCs were placed at a median of 59 days (IQR 40-203) post-transplant and remained for 43 days (IQR 25-88). There was a total of eight (11%) complications. Infection occurred in five patients (7%); four had empyema and one had a catheter tract infection. IPCs did not cause death or critical illness in our cohort. Auto-pleurodesis leading to the removal of the IPC occurred in 63 (89%) instances. None of the patients in this cohort required subsequent surgical decortication. CONCLUSIONS: The use of IPCs in lung transplant patients was associated with an infectious complication rate comparable to other populations previously studied. A high rate of auto-pleurodesis was observed. This work suggests that IPCs may be considered for the management of recurrent pleural effusions in lung allograft recipients.

Application of floating wetlands for the improvement of degraded urban waters: Findings from three multi-year pilot-scale installations.

Floating Treatment Wetlands (FTWs) are an emerging ecological engineering technology being applied the restoration of eutrophic urban water bodies. Documented water-quality benefits of FTW include nutrient removal, transformation of pollutants, and reduction in bacterial contamination. However, translating findings from short-duration lab and mesocosm scale experiments, into sizing criteria that might be applied to field installations is not straightforward. This study presents the results of three well established (>3 years) pilot-scale (40-280 m2) FTW installations in Baltimore, Boston, and Chicago. We quantify annual phosphorus removal through harvesting of above-ground vegetation and find an average removal rate of 2 g-P m-2. In our own study and in a review of literature, we find limited evidence of enhanced sedimentation as a pathway for phosphorus removal. In addition to water-quality benefits, FTW planted with native species, provide valuable wetland habitat; and theoretically improve ecological function. We document efforts to quantify the local effect of FTW installations on benthic and sessile macroinvertebrates, zooplankton, bloom-forming cyanobacteria, and fish. Data from these three projects suggest that, even on a small scale, FTW produce localized changes in biotic structure that reflect improving environmental quality. This study provides a simple and defensible method for sizing FTW for nutrient removal in eutrophic waterbodies. We propose several key research pathways which would advance our understanding of the effects FTW have on the ecosystem they are deployed in.

Diagnosis and management of lymphoblastic lymphoma in children, adolescents and young adults.

Lymphoblastic lymphoma (LBL) is the second most common type of non-Hodgkin Lymphoma (NHL) in children, adolescents, and young adults (CAYA), accounting for 25-35% of all cases. T-lymphoblastic lymphoma (T-LBL) comprises 70-80% of cases, while precursor B-lymphoblastic lymphoma (pB-LBL) makes up the remaining 20-25% of cases. Event-free and overall survival (EFS and OS) for paediatric LBL patients both exceed 80% with current therapies. Treatment regimens, especially in T-LBL with large mediastinal tumours, are complex with significant toxicity and long-term complications. Though prognosis overall is good for T-LBL and pB-LBL with upfront therapy, outcomes for patients with relapsed or refractory (r/r) disease remain dismal. Here, we review new understanding about the pathogenesis and biology of LBL, recent clinical results and future directions for therapy, and remaining obstacles to improve outcomes while reducing toxicity.

A scoping review of intimate partner violence in hospitalized patients.

BACKGROUND: Despite the high prevalence and significant health effects of intimate partner violence (IPV), little is known about its associations with hospitalization. OBJECTIVE: To perform a scoping review of how IPV impacts hospitalization rates, characteristics, and outcomes in adult patients. DATA SOURCES: A search of four databases (MEDLINE, Embase, Web of Science, and CINAHL) using a combination of terms including hospitalized patients and IPV revealed 1608 citations. STUDY SELECTION AND DATA EXTRACTION: One reviewer determined eligibility based on inclusion and exclusion criteria, which a second reviewer independently verified. Data were extracted and organized a posteriori into three categories based on research aim: (1) comparative studies of hospitalization risk associated with recent IPV exposure, (2) comparative studies of hospitalization outcomes by IPV exposure, and (3) descriptive studies of hospitalizations for IPV. RESULTS: Of the 12 included studies, 7 were comparative studies of hospitalization risk associated with IPV, 2 were comparative studies of hospitalization outcomes by IPV, and 3 were descriptive studies of hospitalizations for IPV. Nine out of 12 studies focused on specific patient populations. All but one study demonstrated that IPV was associated with increased risk of hospitalization and/or worse hospitalization outcomes. Six of the seven comparative studies showed a positive association between recent IPV and hospitalization risk. CONCLUSION: This review suggests that IPV exposure increases the risk of hospitalization and/or worsens inpatient outcomes in specific patient populations. Additional work is needed to characterize hospitalization rates and outcomes for persons who have experienced IPV in a broader, nontrauma population.

What works in medication reconciliation: an on-treatment and site analysis of the MARQUIS2 study.

BACKGROUND: The second Multicenter Medication Reconciliation Quality Improvement Study demonstrated a marked reduction in medication discrepancies per patient. The aim of the current analysis was to determine the association of patient exposure to each system-level intervention and receipt of each patient-level intervention on these results. METHODS: This study was conducted at 17 North American Hospitals, the study period was 18 months per site, and sites typically adopted interventions after 2-5 months of preintervention data collection. We conducted an on-treatment analysis (ie, an evaluation of outcomes based on patient exposure) of system-level interventions, both at the category level and at the individual component level, based on monthly surveys of implementation site leads at each site (response rate 65%). We then conducted a similar analysis of patient-level interventions, as determined by study pharmacist review of documented activities in the medical record. We analysed the association of each intervention on the adjusted number of medication discrepancies per patient in admission and discharge orders, based on a random sample of up to 22 patients per month per site, using mixed-effects Poisson regression with hospital site as a random effect. We then used a generalised linear mixed-effects model (GLMM) decision tree to determine which patient-level interventions explained the most variance in discrepancy rates. RESULTS: Among 4947 patients, patient exposure to seven of the eight system-level component categories was associated with modest but significant reductions in discrepancy rates (adjusted rate ratios (ARR) 0.75-0.97), as were 15 of the 17 individual system-level intervention components, including hiring, reallocating and training personnel to take a best possible medication history (BPMH) and training personnel to perform discharge medication reconciliation and patient counselling. Receipt of five of seven patient-level interventions was independently associated with large reductions in discrepancy rates, including receipt of a BPMH in the emergency department (ED) by a trained clinician (ARR 0.40, 95% CI 0.37 to 0.43), admission medication reconciliation by a trained clinician (ARR 0.57, 95% CI 0.50 to 0.64) and discharge medication reconciliation by a trained clinician (ARR 0.64, 95% CI 0.57 to 0.73). In GLMM decision tree analyses, patients who received both a BPMH in the ED and discharge medication reconciliation by a trained clinician experienced the lowest discrepancy rates (0.08 per medication per patient). CONCLUSION AND RELEVANCE: Patient-level interventions most associated with reductions in discrepancies were receipt of a BPMH of admitted patients in the ED and admission and discharge medication reconciliation by a trained clinician. System-level interventions were associated with modest reduction in discrepancies for the average patient but are likely important to support patient-level interventions and may reach more patients. These findings can be used to help hospitals and health systems prioritise interventions to improve medication safety during care transitions.

Prevalence and Causes of Diagnostic Errors in Hospitalized Patients Under Investigation for COVID-19.

BACKGROUND: The COVID-19 pandemic required clinicians to care for a disease with evolving characteristics while also adhering to care changes (e.g., physical distancing practices) that might lead to diagnostic errors (DEs). OBJECTIVE: To determine the frequency of DEs and their causes among patients hospitalized under investigation (PUI) for COVID-19. DESIGN: Retrospective cohort. SETTING: Eight medical centers affiliated with the Hospital Medicine ReEngineering Network (HOMERuN). TARGET POPULATION: Adults hospitalized under investigation (PUI) for COVID-19 infection between February and July 2020. MEASUREMENTS: We randomly selected up to 8 cases per site per month for review, with each case reviewed by two clinicians to determine whether a DE (defined as a missed or delayed diagnosis) occurred, and whether any diagnostic process faults took place. We used bivariable statistics to compare patients with and without DE and multivariable models to determine which process faults or patient factors were associated with DEs. RESULTS: Two hundred and fifty-seven patient charts underwent review, of which 36 (14%) had a diagnostic error. Patients with and without DE were statistically similar in terms of socioeconomic factors, comorbidities, risk factors for COVID-19, and COVID-19 test turnaround time and eventual positivity. Most common diagnostic process faults contributing to DE were problems with clinical assessment, testing choices, history taking, and physical examination (all p < 0.01). Diagnostic process faults associated with policies and procedures related to COVID-19 were not associated with DE risk. Fourteen patients (35.9% of patients with errors and 5.4% overall) suffered harm or death due to diagnostic error. LIMITATIONS: Results are limited by available documentation and do not capture communication between providers and patients. CONCLUSION: Among PUI patients, DEs were common and not associated with pandemic-related care changes, suggesting the importance of more general diagnostic process gaps in error propagation.

A Public Health Critical Race Praxis Informed Congestive Heart Failure Quality Improvement Initiative on Inpatient General Medicine.

BACKGROUND: Prior evaluation at our hospital demonstrated that, compared to White patients, Black and Latinx patients with congestive heart failure (CHF) were less likely to be admitted to the cardiology service rather than the general medicine service (GMS). Patients admitted to GMS (compared to cardiology) had inferior rates of cardiology follow-up and 30-day readmission. OBJECTIVE: To develop and test the feasibility and impacts of using quality improvement (QI) methods, in combination with the Public Health Critical Race Praxis (PHCRP) framework, to engage stakeholders in developing an intervention for ensuring guideline-concordant inpatient CHF care across all patient groups. METHODS: We compared measures for all patients admitted with CHF to GMS between September 2019 and March 2020 (intervention group) to CHF patients admitted to GMS in the previous year (pre-intervention group) and those admitted to cardiology during the pre-intervention and intervention periods (cardiology group). Our primary measures were 30-day readmissions and 14- and 30-day post-discharge cardiology follow-up. RESULTS: There were 79 patients admitted with CHF to GMS during the intervention period, all of whom received the intervention. There were similar rates of Black and Latinx patients across the three groups. Compared to pre-intervention, intervention patients had a significantly lower 30-day readmission rate (18.9% vs. 24.8%; p=0.024), though the cardiology group also had a decrease in 30-day readmissions from the pre-intervention to intervention period. Compared to pre-intervention, intervention patients had significantly higher 14-day and 30-day post-discharge follow-up visits scheduled with cardiology (36.7% vs. 24.8%, p=0.005; 55.7% vs. 42.3%, p=0.0029), but no improvement in appointment attendance. CONCLUSION: This study provides a first test of applying the PHCRP framework within a stakeholder-engaged QI initiative for improving CHF care across races and ethnicities. Our study design cannot evaluate causation. However, the improvements in 30-day readmission, as well as in processes of care that may affect it, provide optimism that inclusion of a racism-conscious framework in QI initiatives is feasible and may enhance QI measures.

Implementation and Utilization of a Web-Based Departmental Annual Faculty Review Program.

ABSTRACT: Supporting faculty in their professional development is a central mission of academic medical centers (AMCs), assisted by routine assessment of faculty development. However, AMCs face barriers to standard faculty assessment, particularly among large departments with diverse faculty roles. We report on the development, implementation, and utilization of a standardized, web-based faculty annual career review (FACR) process deployed within the Department of Medicine at a large AMC, composed of 22 divisions and 1400 faculty. The FACR process was developed to achieve the following goals: Creation of a trackable review of professional goals and barriers to professional development; provide longitudinal faculty reflection on professional growth; and provide data to identify faculty development needs to target initiatives. The FACR was initiated in 2008. We provide specific case examples of utilization of the FACR data for targeted faculty development initiatives, including (1) division benchmarking; (2) targeting career opportunities to individual faculty members; and (3) longitudinal data analysis of 5-year trends of FACR response, including subanalysis by gender, to target faculty development initiatives across the DOM. In summary, we demonstrate successful development, implementation, and utilization of a web-based standard faculty annual review program within a large AMC that allows for systematic evaluation and reflection of professional development by faculty members and their supervisors and allows for data collection and analysis to help target faculty development initiatives.