Electroconvulsive therapy for the acute management of severe agitation in dementia (ECT-AD): A modified study protocol.

OBJECTIVE: This study began as a single-blind randomized controlled trial (RCT) to investigate the efficacy and safety of electroconvulsive therapy (ECT) for severe treatment-refractory agitation in advanced dementia. The aims are to assess agitation reduction using the Cohen-Mansfield Agitation Inventory (CMAI), evaluate tolerability and safety outcomes, and explore the long-term stability of agitation reduction and global functioning. Due to challenges encountered during implementation, including recruitment obstacles and operational difficulties, the study design was modified to an open-label format and other protocol amendments were implemented. METHODS: Initially, the RCT randomized participants 1:1 to either ECT plus usual care or simulated ECT plus usual care (S-ECT) groups. As patients were enrolled, data were collected from both ECT and simulated ECT (S-ECT) patients. The study now continues in an open-label study design where all patients receive actual ECT, reducing the targeted sample size from 200 to 50 participants. RESULTS: Study is ongoing and open to enrollment. CONCLUSION: The transition of the ECT-AD study design from an RCT to open-label design exemplifies adaptive research methodologies in response to real-world challenges. Data from both the RCT and open-label phases of the study will provide a unique perspective on the role of ECT in managing severe treatment-refractory agitation in dementia, potentially influencing future clinical practices and research approaches.

Patient Outcomes Compared Between Admissions Coordinated by the Transfer Center and Emergency Department at a U.S. Tertiary Care Hospital.

BACKGROUND: Patient admissions at a U.S. tertiary care hospital occur via the emergency department (ED), or transfer center. We aim to compare the clinical outcomes of patients admitted from the ED to admissions coordinated by the transfer center. METHODS: Admissions to Mayo Clinic Hospital, Rochester, MN, between July 2019 to June 2021 were identified in this retrospective study and categorized into two cohorts-transfer center and ED. The two cohorts were then matched for age, sex, admitting service, and Charlson Comorbidity Index. Univariate and multivariate analyses were performed to compare hospital length of stay (LOS), mortality, 30-day mortality, and 30-day readmissions between the two cohorts. RESULTS: 73,685 admissions were identified, of which 24,262 (33%) were transfer center admissions. In the matched cohorts (n = 19,093, each), in-hospital mortality (2.4% versus 1.9%), 30-day mortality (5.4% versus 3.9%), 30-day readmission (12.7% versus 7.2%), and LOS (6.4 days versus 5.1 days) were significantly higher ( P < 0.001) among the admissions coordinated by transfer center. A higher palliative care consultation rate (9.4% versus 6.2%, P < 0.001), and a lower proportion of home discharges home (76.2% versus 82.5%, P < 0.001) among transfer center admissions was observed. Similar findings were noted in multivariate analysis, even when adjusting for LOS. CONCLUSIONS: Transfer center admissions had higher in-hospital mortality, LOS, 30-day mortality, and 30-day readmission compared to ED admissions. This study also highlights new considerations for palliative care consultation before transfer acceptance, especially to avoid futile transfers. Additional studies analyzing factors behind the outcomes of transfer center admissions are required.

Impact of inpatient geriatrics consultation on hospital outcomes in older adults with trauma.

BACKGROUND: Older adults presenting with trauma have worse outcomes than younger adults. Starting in 2016, we provided geriatrics consultation (GC) to older adults admitted to the trauma service. We aimed to analyze the impact of GC on patient outcomes. METHODS: We performed a retrospective pre-post study and year-matched cohort study. We identified patients from the trauma registry at our level 1 trauma center. In the pre-post study, we compared patients who received GC (2016-2022) with controls (2011-2015). In the cohort study (2016-2022), we compared patients who received GC with controls. We matched for age, race, sex, and injury severity score (ISS) in both studies, as well as admission year in the cohort study. Outcome variables included mortality (in-hospital, 30-day, 90-day), length of stay (LOS), discharge disposition, and hospital readmission rates (30-day, 90-day). RESULTS: We analyzed 1968 patients in the pre-post study and 2544 patients in the cohort study. Patients were similar in age, race, and sex. GC patients had a slightly higher ISS score and a higher rate of ICU stay. Delirium occurrence was lower among GC patients. GC patients had lower in-hospital mortality compared to controls (pre-post OR 0.27, p < 0.001; cohort OR 0.31, p < 0.001) and increased LOS (6 days vs 4 days, p < 0.001; both studies). GC patients in the cohort study also had lower 30- and 90-day mortality (OR 0.52 and 0.65, p < 0.01) and were less likely to return home (OR 0.81, p < 0.01); similar trends, though not statistically significant, were noted in the pre-post study. Lower readmission rates (statistically non-significant) were noted in the GC group across both studies. CONCLUSIONS: GC in older adults with trauma has proven benefit with reduced mortality and a trend toward lower readmission rates but was associated with increased LOS and higher rates of discharge to skilled facility.

Efficacy and safety of transcranial magnetic stimulation on cognition in mild cognitive impairment, Alzheimer's disease, Alzheimer's disease-related dementias, and other cognitive disorders: a systematic review and meta-analysis.

OBJECTIVE: We aim to analyze the efficacy and safety of TMS on cognition in mild cognitive impairment (MCI), Alzheimer's disease (AD), AD-related dementias, and nondementia conditions with comorbid cognitive impairment. DESIGN: Systematic review, Meta-Analysis. SETTING: We searched MEDLINE, Embase, Cochrane database, APA PsycINFO, Web of Science, and Scopus from January 1, 2000, to February 9, 2023. PARTICIPANTS AND INTERVENTIONS: RCTs, open-label, and case series studies reporting cognitive outcomes following TMS intervention were included. MEASUREMENT: Cognitive and safety outcomes were measured. Cochrane Risk of Bias for RCTs and MINORS (Methodological Index for Non-Randomized Studies) criteria were used to evaluate study quality. This study was registered with PROSPERO (CRD42022326423). RESULTS: The systematic review included 143 studies (n = 5,800 participants) worldwide, encompassing 94 RCTs, 43 open-label prospective, 3 open-label retrospective, and 3 case series. The meta-analysis included 25 RCTs in MCI and AD. Collectively, these studies provide evidence of improved global and specific cognitive measures with TMS across diagnostic groups. Only 2 studies (among 143) reported 4 adverse events of seizures: 3 were deemed TMS unrelated and another resolved with coil repositioning. Meta-analysis showed large effect sizes on global cognition (Mini-Mental State Examination (SMD = 0.80 [0.26, 1.33], p = 0.003), Montreal Cognitive Assessment (SMD = 0.85 [0.26, 1.44], p = 0.005), Alzheimer's Disease Assessment Scale-Cognitive Subscale (SMD = -0.96 [-1.32, -0.60], p < 0.001)) in MCI and AD, although with significant heterogeneity. CONCLUSION: The reviewed studies provide favorable evidence of improved cognition with TMS across all groups with cognitive impairment. TMS was safe and well tolerated with infrequent serious adverse events.

Identification of delirium from real-world electronic health record clinical notes.

Introduction: We tested the ability of our natural language processing (NLP) algorithm to identify delirium episodes in a large-scale study using real-world clinical notes. Methods: We used the Rochester Epidemiology Project to identify persons ≥ 65 years who were hospitalized between 2011 and 2017. We identified all persons with an International Classification of Diseases code for delirium within ±14 days of a hospitalization. We independently applied our NLP algorithm to all clinical notes for this same population. We calculated rates using number of delirium episodes as the numerator and number of hospitalizations as the denominator. Rates were estimated overall, by demographic characteristics, and by year of episode, and differences were tested using Poisson regression. Results: In total, 14,255 persons had 37,554 hospitalizations between 2011 and 2017. The code-based delirium rate was 3.02 per 100 hospitalizations (95% CI: 2.85, 3.20). The NLP-based rate was 7.36 per 100 (95% CI: 7.09, 7.64). Rates increased with age (both p < 0.0001). Code-based rates were higher in men compared to women (p = 0.03), but NLP-based rates were similar by sex (p = 0.89). Code-based rates were similar by race and ethnicity, but NLP-based rates were higher in the White population compared to the Black and Asian populations (p = 0.001). Both types of rates increased significantly over time (both p values < 0.001). Conclusions: The NLP algorithm identified more delirium episodes compared to the ICD code method. However, NLP may still underestimate delirium cases because of limitations in real-world clinical notes, including incomplete documentation, practice changes over time, and missing clinical notes in some time periods.

The Macro Impact of Microtransitions in Post-acute and Long-Term Care.

The care transitions concept emerged in medical literature more than 40 years ago, with an exponential rise in publications dedicated to its exploration since that time. It is generally accepted that older patients are particularly vulnerable during care transitions because of complex medical comorbidity, frailty, cognitive dysfunction, and the fragmented nature of health care. A care transition is defined as the movement of patients from one health care setting to another as their care needs change during acute or chronic illness. Easily recognizable examples include the discharge of a patient from the hospital to a skilled nursing facility or an admission to the hospital after a patient is evaluated in the emergency department. These macrotransitions are marked by major changes in clinical condition and span days to weeks. This discussion examines a new term coined by the authors: microtransitions, which are care transitions characterized by movement of a patient between health care settings or within a given setting, usually over shorter periods (less than 24 hours) and accompanied by changes in clinical or custodial responsibility for a patient. Although often unrecognized as formal care transitions, these microtransitions, if not handled appropriately, can lead to poor outcomes, including clinical deterioration and the need for macrotransition. The authors propose formal recognition of microtransitions, standardization of processes related to them, and practical considerations for implementation.

Postdischarge Video Visits for Adherence to Hospital Discharge Recommendations: A Randomized Clinical Trial.

Objective: To determine whether a postdischarge video visit with patients, conducted by hospital medicine advanced practice providers, improves adherence to hospital discharge recommendations. Patients and Methods: We conducted a single-institution 2-site randomized clinical trial with 1:1 assignment to intervention vs control, with enrollment from August 10, 2020, to June 23, 2022. Hospital medicine patients discharged home or to an assisted living facility were randomized to a video visit 2-5 days postdischarge in addition to usual care (intervention) vs usual care (control). During the video visit, advanced practice providers reviewed discharge recommendations. Both intervention and control groups received telephone follow-up 3-6 days postdischarge to ascertain the primary outcome of adherence to all discharge recommendations for new and chronic medication management, self-management and action plan, and home support. Results: Among 1190 participants (594 intervention; 596 control), the primary outcome was ascertained in 768 participants (314 intervention; 454 control). In intervention vs control, there was no difference in the proportion of participants with the primary outcome (76.7% vs 72.5%; P=.19) or in the individual domains of the primary outcome: new and chronic medication management (94.1% vs 92.8%; P=.50), self-management and action plan (76.5% vs 71.5%; P=.18), and home support (94.1% vs 94.3%; P=.94). Women receiving intervention vs control had higher adherence to recommendations (odds ratio, 1.77; 95% CI, 1.08-2.91). Conclusion: In hospital medicine patients, a postdischarge video visit did not improve adherence to discharge recommendations. Potential gender differences in adherence require further investigation.Clinicaltrials.gov number, NCT04547803.

Hospitalist perspective on pandemic related clinical and administrative changes: a cross sectional survey study.

OBJECTIVE: Hospitalists have played a leading role in caring for hospitalized COVID-19 patients. Many clinical and administrative changes occurred in hospitals to meet the varied pandemic needs. We surveyed hospitalists to understand their perspective on pandemic-related changes in technology, models of care, administration and leadership, impact on personal lives, and which of these changes should be continued versus reverting to pre-pandemic practices. METHODS: A 30-question survey was distributed to hospitalists working across the United States between 6 April 2022 to 16 May 2022. Baseline demographics were measured, and post-pandemic perspectives related to changes were analyzed. Perspectives were measured using a 5-point Likert scale and responses were categorized into 'agree' and 'did not agree' for analysis. Variation was assessed using Chi-square or Fisher exact tests. Open-ended questions were reported following qualitative content analysis organized into themes and reported as frequency. RESULTS: 177 respondents (39%) completed the survey. Nearly three-fourths favored hybrid meetings, and two-thirds preferred to continue new models of care. Nearly 90% desired more family and leisure time, continued wellness, and support services, and resumption of social gatherings. No major differences in perspectives were noted between hospitalists at teaching facilities and non-teaching facilities except for resuming protected time for non-clinical activities in those from teaching facilities (83.0% vs 62.5%). Respondents less than age 50 were more likely to prefer virtual meetings (59.0% vs 31.3%). Content analysis of open-ended questions resulted in different themes for each question. Respondents favored more work-life balance and less administrative and logistical work burden. CONCLUSIONS: Hospitalists preferred to continue the use of technology and new models of care even in the post-pandemic period and express a desire for more work-life balance and less administrative and logistical work burden.

An artificial intelligence approach for predicting death or organ failure after hospitalization for COVID-19: development of a novel risk prediction tool and comparisons with ISARIC-4C, CURB-65, qSOFA, and MEWS scoring systems.

BACKGROUND: We applied machine learning (ML) algorithms to generate a risk prediction tool [Collaboration for Risk Evaluation in COVID-19 (CORE-COVID-19)] for predicting the composite of 30-day endotracheal intubation, intravenous administration of vasopressors, or death after COVID-19 hospitalization and compared it with the existing risk scores. METHODS: This is a retrospective study of adults hospitalized with COVID-19 from March 2020 to February 2021. Patients, each with 92 variables, and one composite outcome underwent feature selection process to identify the most predictive variables. Selected variables were modeled to build four ML algorithms (artificial neural network, support vector machine, gradient boosting machine, and Logistic regression) and an ensemble model to generate a CORE-COVID-19 model to predict the composite outcome and compared with existing risk prediction scores. The net benefit for clinical use of each model was assessed by decision curve analysis. RESULTS: Of 1796 patients, 278 (15%) patients reached primary outcome. Six most predictive features were identified. Four ML algorithms achieved comparable discrimination (P > 0.827) with c-statistics ranged 0.849-0.856, calibration slopes 0.911-1.173, and Hosmer-Lemeshow P > 0.141 in validation dataset. These 6-variable fitted CORE-COVID-19 model revealed a c-statistic of 0.880, which was significantly (P < 0.04) higher than ISARIC-4C (0.751), CURB-65 (0.735), qSOFA (0.676), and MEWS (0.674) for outcome prediction. The net benefit of the CORE-COVID-19 model was greater than that of the existing risk scores. CONCLUSION: The CORE-COVID-19 model accurately assigned 88% of patients who potentially progressed to 30-day composite events and revealed improved performance over existing risk scores, indicating its potential utility in clinical practice.

Characteristics, Treatment Patterns, and Clinical Outcomes After Heart Failure Hospitalizations During the COVID-19 Pandemic, March to October 2020.

OBJECTIVE: To compare clinical characteristics, treatment patterns, and 30-day all-cause readmission and mortality between patients hospitalized for heart failure (HF) before and during the coronavirus disease 2019 (COVID-19) pandemic. PATIENTS AND METHODS: The study was conducted at 16 hospitals across 3 geographically dispersed US states. The study included 6769 adults (mean age, 74 years; 56% [5033 of 8989] men) with cumulative 8989 HF hospitalizations: 2341 hospitalizations during the COVID-19 pandemic (March 1 through October 30, 2020) and 6648 in the pre-COVID-19 (October 1, 2018, through February 28, 2020) comparator group. We used Poisson regression, Kaplan-Meier estimates, multivariable logistic, and Cox regression analysis to determine whether prespecified study outcomes varied by time frames. RESULTS: The adjusted 30-day readmission rate decreased from 13.1% (872 of 6648) in the pre-COVID-19 period to 10.0% (234 of 2341) in the COVID-19 pandemic period (relative risk reduction, 23%; hazard ratio, 0.77; 95% CI, 0.66 to 0.89). Conversely, all-cause mortality increased from 9.7% (645 of 6648) in the pre-COVID-19 period to 11.3% (264 of 2341) in the COVID-19 pandemic period (relative risk increase, 16%; number of admissions needed for one additional death, 62.5; hazard ratio, 1.19; 95% CI, 1.02 to 1.39). Despite significant differences in rates of index hospitalization, readmission, and mortality across the study time frames, the disease severity, HF subtypes, and treatment patterns remained unchanged (P>0.05). CONCLUSION: The findings of this large tristate multicenter cohort study of HF hospitalizations suggest lower rates of index hospitalizations and 30-day readmissions but higher incidence of 30-day mortality with broadly similar use of HF medication, surgical interventions, and devices during the COVID-19 pandemic compared with the pre-COVID-19 time frame.

Natural Language Processing CAM Algorithm Improves Delirium Detection Compared With Conventional Methods.

Delirium is known to be underdiagnosed and underdocumented. Delirium detection in retrospective studies occurs mostly by clinician diagnosis or nursing documentation. This study aims to assess the effectiveness of natural language processing-confusion assessment method (NLP-CAM) algorithm when compared to conventional modalities of delirium detection. A multicenter retrospective study analyzed 4351 COVID-19 hospitalized patient records to identify delirium occurrence utilizing three different delirium detection modalities namely clinician diagnosis, nursing documentation, and the NLP-CAM algorithm. Delirium detection by any of the 3 methods is considered positive for delirium occurrence as a comparison. NLP-CAM captured 80% of overall delirium, followed by clinician diagnosis at 55%, and nursing flowsheet documentation at 43%. Increase in age, Charlson comorbidity score, and length of hospitalization had increased delirium detection odds regardless of the detection method. Artificial intelligence-based NLP-CAM algorithm, compared to conventional methods, improved delirium detection from electronic health records and holds promise in delirium diagnostics.

Within-Person Blood Pressure Variability During Hospitalization and Clinical Outcomes Following First-Ever Acute Ischemic Stroke.

BACKGROUND: Uncertainty remains over the relationship between blood pressure (BP) variability (BPV), measured in hospital settings, and clinical outcomes following acute ischemic stroke (AIS). We examined the association between within-person systolic blood pressure (SBP) variability (SBPV) during hospitalization and readmission-free survival, all-cause readmission, or all-cause mortality 1 year after AIS. METHODS: In a cohort of 862 consecutive patients (age [mean ± SD] 75 ± 15 years, 55% women) with AIS (2005-2018, follow-up through 2019), we measured SBPV as quartiles of standard deviations (SD) and coefficient of variation (CV) from a median of 16 SBP readings obtained throughout hospitalization. RESULTS: In the cumulative cohort, the measured SD and CV of SBP in mmHg were 16 ± 6 and 10 ± 5, respectively. The hazard ratios (HR) for readmission-free survival between the highest vs. lowest quartiles were 1.44 (95% confidence interval [CI] 1.04-1.81) for SD and 1.29 (95% CI 0.94-1.78) for CV after adjustment for demographics and comorbidities. Similarly, incident readmission or mortality remained consistent between the highest vs. lowest quartiles of SD and CV (readmission: HR 1.29 [95% CI 0.90-1.78] for SD, HR 1.29 [95% CI 0.94-1.78] for CV; mortality: HR 1.15 [95% CI 0.71-1.87] for SD, HR 0.86 [95% CI 0.55-1.36] for CV). CONCULSIONS: In patients with first AIS, SBPV measured as quartiles of SD or CV based on multiple readings throughout hospitalization has no independent prognostic implications for the readmission-free survival, readmission, or mortality. This underscores the importance of overall patient care rather than a specific focus on BP parameters during hospitalization for AIS.

Quality assessment of functional status documentation in EHRs across different healthcare institutions.

The secondary use of electronic health records (EHRs) faces challenges in the form of varying data quality-related issues. To address that, we retrospectively assessed the quality of functional status documentation in EHRs of persons participating in Mayo Clinic Study of Aging (MCSA). We used a convergent parallel design to collect quantitative and qualitative data and independently analyzed the findings. We discovered a heterogeneous documentation process, where the care practice teams, institutions, and EHR systems all play an important role in how text data is documented and organized. Four prevalent instrument-assisted documentation (iDoc) expressions were identified based on three distinct instruments: Epic smart form, questionnaire, and occupational therapy and physical therapy templates. We found strong differences in the usage, information quality (intrinsic and contextual), and naturality of language among different type of iDoc expressions. These variations can be caused by different source instruments, information providers, practice settings, care events and institutions. In addition, iDoc expressions are context specific and thus shall not be viewed and processed uniformly. We recommend conducting data quality assessment of unstructured EHR text prior to using the information.

Post-discharge early assessment with remote video link (PEARL) initiative for patients discharged from hospital medicine services.

OBJECTIVES: The COVID-19 pandemic impacted the availability and accessibility of outpatient care following hospital discharge. Hospitalists (physicians) and hospital medicine advanced practice providers (HM-APPs) coordinate discharge care of hospitalized patients; however, it is unknown if they can deliver post-discharge virtual care and overcome barriers to outpatient care. The objective was to develop and provide post-discharge virtual care for patients discharged from hospital medicine services. METHODS: We developed the Post-discharge Early Assessment with Remote video Link (PEARL) initiative for HM-APPs to conduct a post-discharge video visit (to review recommendations) and telephone follow-up (to evaluate adherence) with patients 2-6 days following hospital discharge. Participants included patients discharged from hospital medicine services at an institution's hospitals in Rochester (May 2020-August 2020) and Austin (November 2020-February 2021) in Minnesota, US. HM-APPs also interviewed patients about their experience with the video visit and completed a survey on their experience with PEARL. RESULTS: Of 386 eligible patients, 61.4% were enrolled (n = 237/386) including 48.1% women (n = 114/237). In patients with complete video visit and telephone follow-up (n = 141/237), most were prescribed new medications (83.7%) and took them as prescribed (93.2%). Among five classes of chronic medications, patient-reported adherence ranged from 59.2% (narcotics) to 91.5% (anti-hypertensives). Patient-reported self-management of 12 discharge recommendations ranged from 40% (smoking cessation) to 100% (checking rashes). Patients reported benefit from the video visit (agree: 77.3%) with an equivocal preference for video visits over clinic visits. Among HM-APPs who responded to the survey (88.2%; n = 15/17), 73.3% reported benefit from visual contact with patients but were uncertain if video visits would reduce emergency department visits. CONCLUSION: In this novel initiative, HM-APPs used video visits to provide care beyond their hospital role, reinforce discharge recommendations for patients, and reduce barriers to outpatient care. The effect of this initiative is under evaluation in a randomized controlled trial.

A Review Study on the Trends of Psychological Challenges, Coping Ways, and Public Support During the COVID-19 Pandemic in the Vulnerable Populations in the United States.

Background: The COVID-19 pandemic resulted in significant mortality and morbidity in the United States. The mental health impact during the pandemic was huge and affected all age groups and population types. We reviewed the existing literature to understand the present trends of psychological challenges and different coping strategies documented across different vulnerable sections of the United States population. This rapid review was carried out to investigate the trends in psychological impacts, coping ways, and public support during the COVID-19 pandemic crisis in the United States. Materials and Methods: We undertook a rapid review of the literature following the COVID-19 pandemic in the United States. We searched PubMed as it is a widely available database for observational and experimental studies that reported the psychological effects, coping ways, and public support on different age groups and healthcare workers (HCWs) during the COVID-19 pandemic. Results: We included thirty-five studies in our review and reported data predominantly from the vulnerable United States population. Our review findings indicate that COVID-19 has a considerable impact on the psychological wellbeing of various age groups differently, especially in the elderly population and HCWs. Review findings suggest that factors like children, elderly population, female gender, overconcern about family, fear of getting an infection, personality, low spirituality, and lower resilience levels were at a higher risk of adverse mental health outcomes during this pandemic. Systemic support, higher resilience levels, and adequate knowledge were identified as protecting and preventing factors. There is a paucity of similar studies among the general population, and we restricted our review specifically to vulnerable subgroups of the population. All the included studies in our review investigated and surveyed the psychological impacts, coping skills, and public support system during the COVID-19 pandemic. Conclusion: The evidence to date suggests that female gender, child and elderly population, and racial factors have been affected by a lack of support for psychological wellbeing. Further, research using our hypothesized framework might help any population group to deal with a pandemic-associated mental health crisis, and in that regard, analysis of wider societal structural factors is recommended.

The Implication of Latent Information Quality to the Reproducibility of Secondary Use of Electronic Health Records.

Reproducibility is an important quality criterion for the secondary use of electronic health records (EHRs). However, multiple barriers to reproducibility are embedded in the heterogeneous EHR environment. These barriers include complex processes for collecting and organizing EHR data and dynamic multi-level interactions occurring during information use (e.g., inter-personal, inter-system, and cross-institutional). To ensure reproducible use of EHRs, we investigated four information quality dimensions and examine the implications for reproducibility based on a real-world EHR study. Four types of IQ measurements suggested that barriers to reproducibility occurred for all stages of secondary use of EHR data. We discussed our recommendations and emphasized the importance of promoting transparent, high-throughput, and accessible data infrastructures and implementation best practices (e.g., data quality assessment, reporting standard).

Validation and Recalibration of Modified Mayo Delirium Prediction Tool in a Hospitalized Cohort.

BACKGROUND: Delirium prediction can augment and optimize care of older adults. Mayo Delirium Prediction (MDP) tool is a robust tool, developed from a large retrospective data set. The MDP tool predicts delirium risk for hospitalized older adults, within 24 hours of hospital admission, based on risk factor information available from electronic health record. OBJECTIVE: We intend to validate the prediction performance of this tool and optimize the tool for clinical use. METHODS: This is an observational cohort study conducted at Mayo Clinic Hospitals, Rochester, MN. All hospitalized older adults (age >50 years) from December 2019 to June 2020 were included. Patients with an admitting diagnosis of substance use disorder were excluded. The original MDP tool was modified to adjust for the fall risk variable as a binary variable that will facilitate broader applicability across different fall risk tools. The modified MDP tool was validated in the retrospective derivation and validation data set which yielded similar prediction capability (area under the receiver operating curve = 0.85 and 0.83, respectively). Diagnosis of delirium was captured by flowsheet diagnosis of delirium documented by nursing staff in the medical record. Predictive variable data were collected daily. RESULTS: A total of 8055 patients were included in the study (median age 71 y). Delirium prediction of the modified MDP tool compared to delirium occurrence was 4% in the low-risk group, 17.8% in the medium-risk group, and 45.3% in the high-risk group (area under receiver operating curve of 0.80). Recalibration of the tool was attempted to further optimize the tool which resulted in both simplification and increased performance (area under receiver operating curve 0.82). The simplified tool was able to predict delirium in hospitalized patients admitted to both medical and surgical services. CONCLUSIONS: Validation of the modified MDP tool revealed good prediction capabilities. Recalibration resulted in simplification with increased performance of the tool in both medical and surgical hospitalized patients.

Ascertainment of Delirium Status Using Natural Language Processing From Electronic Health Records.

BACKGROUND: Delirium is underdiagnosed in clinical practice and is not routinely coded for billing. Manual chart review can be used to identify the occurrence of delirium; however, it is labor-intensive and impractical for large-scale studies. Natural language processing (NLP) has the capability to process raw text in electronic health records (EHRs) and determine the meaning of the information. We developed and validated NLP algorithms to automatically identify the occurrence of delirium from EHRs. METHODS: This study used a randomly selected cohort from the population-based Mayo Clinic Biobank (N = 300, age ≥65). We adopted the standardized evidence-based framework confusion assessment method (CAM) to develop and evaluate NLP algorithms to identify the occurrence of delirium using clinical notes in EHRs. Two NLP algorithms were developed based on CAM criteria: one based on the original CAM (NLP-CAM; delirium vs no delirium) and another based on our modified CAM (NLP-mCAM; definite, possible, and no delirium). The sensitivity, specificity, and accuracy were used for concordance in delirium status between NLP algorithms and manual chart review as the gold standard. The prevalence of delirium cases was examined using International Classification of Diseases, 9th Revision (ICD-9), NLP-CAM, and NLP-mCAM. RESULTS: NLP-CAM demonstrated a sensitivity, specificity, and accuracy of 0.919, 1.000, and 0.967, respectively. NLP-mCAM demonstrated sensitivity, specificity, and accuracy of 0.827, 0.913, and 0.827, respectively. The prevalence analysis of delirium showed that the NLP-CAM algorithm identified 12 651 (9.4%) delirium patients, the NLP-mCAM algorithm identified 20 611 (15.3%) definite delirium cases, and 10 762 (8.0%) possible cases. CONCLUSIONS: NLP algorithms based on the standardized evidence-based CAM framework demonstrated high performance in delineating delirium status in an expeditious and cost-effective manner.

A hybrid model to identify fall occurrence from electronic health records.

INTRODUCTION: Falls are a leading cause of unintentional injury in the elderly. Electronic health records (EHRs) offer the unique opportunity to develop models that can identify fall events. However, identifying fall events in clinical notes requires advanced natural language processing (NLP) to simultaneously address multiple issues because the word "fall" is a typical homonym. METHODS: We implemented a context-aware language model, Bidirectional Encoder Representations from Transformers (BERT) to identify falls from the EHR text and further fused the BERT model into a hybrid architecture coupled with post-hoc heuristic rules to enhance the performance. The models were evaluated on real world EHR data and were compared to conventional rule-based and deep learning models (CNN and Bi-LSTM). To better understand the ability of each approach to identify falls, we further categorize fall-related concepts (i.e., risk of fall, prevention of fall, homonym) and performed a detailed error analysis. RESULTS: The hybrid model achieved the highest f1-score on sentence (0.971), document (0.985), and patient (0.954) level. At the sentence level (basic data unit in the model), the hybrid model had 0.954, 1.000, 0.988, and 0.999 in sensitivity, specificity, positive predictive value, and negative predictive value, respectively. The error analysis showed that that machine learning-based approaches demonstrated higher performance than a rule-based approach in challenging cases that required contextual understanding. The context-aware language model (BERT) slightly outperformed the word embedding approach trained on Bi-LSTM. No single model yielded the best performance for all fall-related semantic categories. CONCLUSION: A context-aware language model (BERT) was able to identify challenging fall events that requires context understanding in EHR free text. The hybrid model combined with post-hoc rules allowed a custom fix on the BERT outcomes and further improved the performance of fall detection.

Quality Performance of a Transfer Center Reduces Interhospital Transfer and Direct Admission-Related ED Evaluations.

Transfer centers play a vital role in the efficient triage of hospital admission requests that generate outside the emergency department (ED) of the given facility. This cohort study includes all the calls processed through the transfer center requesting an admission to Mayo Clinic, Rochester, from January 2016 to December 2018. More than 116,000 transfer request calls were processed. Of these, about 65% (75,000) were accepted for ED evaluation or direct admission. Of the 75,000 patients, >50% were accepted as direct admits. Among patients accepted for direct admission, a trend toward reduced utilization of ED reevaluation at the receiving facility was noted from 2016 to 2018. A temporal trend of overall reduced ED utilization reflects the adeptness of the transfer center. An effective transfer center promotes value-based care, optimizes the workflow in a hospital, and augments hospital administrative decisions to allocate resources.