Inpatient Events Research Articles

Predicting Factors for Requiring Routine Postoperative Blood Analysis in Primary Hip and Knee Arthroplasty: A Prospective Study.

Background Minimising unnecessary expenditure is essential to cope with high demands on the health sector. A set of full blood count, electrolyte, creatinine and urea tests cost £12 in the National Health Service (NHS). Identifying selected patientsrequiring postoperative blood tests following primary knee and hip arthroplasty will avoid unnecessarytests and help to reduce expenditure. The aim of our study is to propose criteria for requesting postoperative blood tests that are safe and do not miss patients. Materials and methods We prospectively evaluated 126 patients (72 in the total knee replacement (TKR) group and 54 in the total hip replacement (THR) group) who underwent either an elective primary THR or a TKR. The mean patient age was 71 years. Patient demographics as well as in-patient events throughout each patient's hospital stay were recorded. Hospital readmissions were also monitored for up to 90 days postoperatively. Statistical analysis was performed using SPSS Statistics software (IBM Corp., Armonk, NY) with paired t-tests / Wilcox and mixed measures analysis of variance. Binary logistic regression was used to identify predictors of patients requiring a postoperative blood test. Results Analysis of our data identified the following as risk factors for requiring postoperative full blood count tests, including pre-operative Hb of ≤ 110 g/L, cardiac disease, clinical features of anaemia postoperatively and intraoperative blood loss of > 500 mL. The additional risk factors identified for requiring postoperative electrolyte and urea tests are deranged pre-operative electrolytes and clinical signs or symptoms of electrolyte/renal disturbance such as anuria. No patient was readmitted within 90 days of discharge. Conclusion Overall, applying the criteria we have devised would have saved 74 blood tests in the cohort of 126 patients. This provides an odds ratio of 14.0 (95% confidence interval: 1.77-110, p=0.012) of an abnormal result in the patients that would have been tested, compared to those that would not have been tested.

Impact of an emergency department rapid response system on inpatient clinical deterioration: A controlled pre-post study

AimTo determine the impact implementation of Emergency Department Clinical Emergency Response System (EDCERS) on inpatient deterioration events and identify contributing causal factors. MethodsEDCERS was implemented in an Australian regional hospital, integrating a single parameter track and trigger criteria for escalation of care, and emergency, specialty and critical care clinician response to patient deterioration. In this controlled pre-post study, electronic medical records of patients who experienced a deterioration event (rapid response call, cardiac arrest or unplanned intensive care admission) on the ward within 72 h of admission from the emergency department (ED) were reviewed. Causal factors contributing to the deteriorating event were assessed using a validated human factors framework. ResultsImplementation of EDCERS reduced the number of inpatient deterioration events within 72 h of emergency admission with failure or delayed response to ED patient deterioration as a causal factor. There was no change in the overall rate of inpatient deterioration events. ConclusionThis study supports wider implementation of rapid response systems in the ED to improve management of deteriorating patients. Tailored implementation strategies should be used to achieve successful and sustainable uptake of ED rapid response systems and improve outcomes in deteriorating patients.

Developing a Decision Pathway for Family Presence During Resuscitation.

The standard of care for family presence during resuscitation (FPDR) is evolving, and leading organizations collectively recommend establishing institutional policy for its practice. Although FPDR is supported at this single institution, the process was not standardized. An interprofessional group authored a decision pathway to standardize the care of families during inpatient code blue events at one institution. The pathway was reviewed and applied in code blue simulation events to highlight the role of the family facilitator and the importance of interprofessional teamwork skills. The decision pathway is a patient-centered algorithm that promotes safety and family autonomy. Pathway recommendations are shaped by current literature, expert consensus, and existing institutional regulations. An on-call chaplain responds to all code blue events as the family facilitator and conducts assessments and decision making per the pathway. Clinical considerations include patient prioritization, family safety, sterility, and team consensus. One year after implementation, staff felt that it positively affected patient and family care. The frequency of inpatient FPDR did not increase after implementation. As a result of the decision pathway implementation, FPDR is consistently a safe and coordinated option for patients' family members.

Comparison of the Aotearoa New Zealand Early Warning Score and National Early Warning Score to predict adverse inpatient events in a vital sign dataset.

Aotearoa New Zealand uses a single early warning score (EWS) across all public and private hospitals to detect adult inpatient physiological deterioration. This combines the aggregate weighted scoring of the UK National Early Warning Score with single parameter activation from Australian medical emergency team systems. We conducted a retrospective analysis of a large vital sign dataset to validate the predictive performance of the New Zealand EWS in discriminating between patients at risk of serious adverse events and compared this with the UK EWS. We also compared predictive performance for patients admitted under medical vs. surgical specialties. A total of 1,738,787 aggregate scores (13,910,296 individual vital signs) were obtained from 102,394 hospital admissions to six hospitals within the Canterbury District Health Board of New Zealand's South Island. Predictive performance of each scoring system was determined using area under the receiver operating characteristic curve. Analysis showed that the New Zealand EWS is equivalent to the UK EWS in predicting patients at risk of serious adverse events (cardiac arrest, death and/or unanticipated ICU admission). Area under the receiver operating characteristic curve for both EWSs for any adverse outcome was 0.874 (95%CI 0.871-0.878) and 0.874 (95%CI 0.870-0.877), respectively. Both EWSs showed superior predictive value for cardiac arrest and/or death in patients admitted under surgical rather than medical specialties. Our study is the first validation of the New Zealand EWS in predicting serious adverse events in a broad dataset and supports previous work showing the UK EWS has superior predictive performance in surgical rather than medical patients.

Nephrotoxin Exposure and Acute Kidney Injury in Adults.

Rates of nephrotoxic AKI are not well described in adults due to lack of a clear definition, debate over which drugs should be considered nephrotoxins, and illness-related confounding. Nephrotoxic Injury Negated by Just-in Time Action (NINJA), a program that reduces rates of nephrotoxic AKI in pediatric populations, may be able to address these concerns, but whether NINJA can be effectively applied to adults remains unclear. In this retrospective cohort study conducted at the University of Iowa Hospital, we included adult patients admitted to a general hospital floor for ≥48 hours during 2019. The NINJA algorithm screened charts for high nephrotoxin exposure and AKI. After propensity score matching, Cox proportional hazard modeling was used to evaluate the relationship between nephrotoxic exposure and all-stage AKI, stage 2-3 AKI, or death. Additional analyses evaluated the most frequent nephrotoxins used in this population. Of 11,311 patients, 1527 (16%) had ≥1 day of high nephrotoxin exposure. Patients with nephrotoxic exposures subsequently developed AKI in 29% of cases, and 22% of all inpatient AKI events met nephrotoxic AKI criteria. Common nephrotoxins were vancomycin, iodinated contrast dye, piperacillin-tazobactam, acyclovir, and lisinopril. After propensity score matching, Cox proportional hazard models for high nephrotoxin exposure were significantly associated with all AKI (hazard ratio [HR] 1.43, 1.19-1.72, P<0.001), stage 2-3 AKI (HR 1.78, 1.18-2.67, P=0.006), and mortality (HR 2.12, 1.09-4.11, P=0.03). Nephrotoxin exposure in adults is common and is significantly associated with AKI development, including stage 2-3 AKI.

317 Comparative risk of venous thromboembolism (VTE) among patients with atopic dermatitis (AD) or other immune-mediated diseases: a cohort study using United States claims data

Abstract Patients with rheumatoid arthritis (RA) or inflammatory bowel disease (IBD) are at increased risk for thrombotic events, but less is known about the underlying risk of VTE with immune-mediated skin conditions. The objective of our study was to describe the incidence of VTE among patients diagnosed with AD and compare VTE incidence among patients with other immune-mediated inflammatory diseases, as well as among individuals without such conditions. This retrospective, observational, comparative cohort study used a large national administrative claims database Optum® (Clinformatics® Data Mart). Adults (aged ≥18 years) diagnosed with AD, RA, IBD (Crohn’s disease [CD] or ulcerative colitis [UC]) or other immune-mediated diseases during 2010–2019 were included. Disease cohorts were not mutually exclusive. A non-diseased control population with no such diagnoses was matched 1 : 1 to patients in the AD cohort by age, sex and calendar time of cohort entry. All patients had to have at least 12 months of enrolment (baseline) before a confirmed diagnosis (or matched cohort entry date for controls) and at least 1 day of follow-up data following diagnosis. All patients received normal clinical care and were followed until the earliest VTE event, end of health plan enrolment, death or end of data availability (30 June 2021). The primary outcome was the incidence of VTE (deep vein thrombosis [DVT] or pulmonary embolism [PE]) defined as an inpatient hospitalization with a VTE diagnosis and length of stay ≥1 day or an outpatient VTE diagnosis followed by initiation of an anticoagulant within 7 days. Incidence per 100 person-years (PY) and 95% CI were calculated. Crude and adjusted hazard ratios (HRs) were determined using Cox proportional hazards regression models and adjusted for demographic and comorbid disease factors. The secondary outcome was the incidence of VTE limited to inpatient events. A total of 1,098,633 adults with AD were included in the analysis; the mean (SD) age was 54.1 (18.5) years, 65.3% were female and 1.1% had a history of VTE. A total of 15,456 VTE events (primary outcome) were reported among patients with AD during 2,113,668.9 PY for a crude incidence (95%CI) of 0.73 (0.72, 0.74) cases/100 PY. The incidence of VTE was 1.33 (1.30, 1.36) cases/100 PY among patients with RA or IBD (n = 578,021) and 0.59 (0.58, 0.60) cases/100 PY among AD-matched controls. The adjusted risk for VTE was higher among patients with RA or IBD compared with patients with AD with an adjusted HR (95% CI) of 1.46 (1.43, 1.49). Patients with AD had a slightly higher risk for VTE compared with AD-matched controls with an adjusted HR of 1.09 (1.06, 1.12). Similar results were observed when DVT and PE were assessed separately. Patients with CD or UC had the highest risk for VTE relative to patients with AD with an adjusted HR of 1.61 (1.54, 1.69) and 1.57 (1.5, 1.63), respectively. Patients with RA also had a higher risk for VTE relative to patients with AD with an adjusted HR of 1.42 (1.38–1.46). Similar results were also observed when the assessment of the incidence and risk of VTE was limited to inpatient events. Patients with RA or IBD had a higher risk of VTE compared with patients with AD. The risk of VTE was slightly higher among patients with AD compared with AD-matched controls. Because VTE risk is increased in RA and IBD during periods of active disease, observed rates may depend on the proportion of patients with uncontrolled disease. Characterization of baseline VTE risk among patients with AD may better inform treatment benefit–risk assessments and shared decision-making.

Evaluating the cost of managing patients with cellulitis in Wales, UK: A 20-year population-scale study.

This study aimed to estimate costs associated with managing patients with cellulitis from the UK National Health Service (NHS) perspective. The analysis was undertaken through the Secure Anonymised Information Linkage Databank, which brings together population-scale, individual-level anonymised linked data from a wide range of sources, including 80% of primary care general practices within Wales (population coverage ~3.2 million). The data covered a 20-year period from 1999 to 2019. All patients linked to the relevant codes were tracked through primary care settings, recording the number of general practice visits (number of days with an event recorded) and number of in-patient stays. Resources were valued in monetary terms (£ sterling), with costs determined from national published sources of unit costs. These resources were then extrapolated out to reflect UK NHS costs. This is the first attempt to estimate the financial burden of cellulitis using routine data sources on a national scale. The estimated direct annual costs to the Welsh NHS (£28 554 338) are considerable. In-Patient events and length of stay costs are the main cost drivers, with annual Welsh NHS estimates of £19 664 126 with primary care events costing £8 890 212. Initiatives to support patients and healthcare professionals in identifying early signs/risks of cellulitis, improve the accuracy of initial diagnosis, prevent cellulitis recurrence, and improve evidence-based treatment pathways would result in major financial savings, to both the Welsh and UK NHS. In light of these findings, Wales has developed the innovative National Lymphoedema cellulitis Improvement Programme to address these burdens; providing a proactive model of cellulitis care.

Association of vitamin and/or nutritional supplements with fall among patients with diabetes: A prospective study based on ACCORD and UK Biobank.

To assess the associations of vitamin and/or nutritional supplements (VNS) with falls among patients with diabetes. 9,141 and 21,489 middle-aged participants with diabetes from Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial and UK Biobank were included. Use of VNS was collected at baseline, and fall events were recorded using annual questionnaires in ACCORD and electric records in UK Biobank during follow-up. The associations of VNS use with fall risk were analyzed using logistic regression models in ACCORD and Fine-Gray sub-distribution hazard models in UK Biobank. The role of specific supplements was also estimated in UK Biobank, adjusting for confounding factors and multiple comparisons. 45.9% (4,193/9,141, 5.5 median follow-up years) patients in ACCORD and 10.5% (2,251/21,489, 11.9 median follow-up years) in UK Biobank experienced fall and in-patient events during follow-up, respectively. In ACCORD, VNS using was associated with an increased risk of fall (full-adjusted odds ratio [OR]: 1.26, P < 0.05). In UK Biobank, despite no significant association between VNS overall and in-patient fall, vitamin B, calcium, and iron using increased the risk of falls significantly (full-adjusted hazard ratio range: 1.31-1.37, P < 0.05). Use of specific VNS increased the risk of fall among patients with diabetes. The non-indicative use of nutritional supplements for patients with diabetes might be inadvisable.

A descriptive study of the multidisciplinary healthcare experiences of inpatient resuscitation events

BackgroundIn-hospital resuscitation events have complex and enduring effects on clinicians, with implications for job satisfaction, performance, and burnout. Ethically ambiguous cases are associated with increased moral distress. We aim to quantitatively describe the multidisciplinary resuscitation experience. MethodsMultidisciplinary in-hospital healthcare professionals at an adult academic health center in the Midwestern United States completed surveys one and six weeks after a resuscitation event. Surveys included demographic data, task load (NASA-TLX), overall and moral distress, anxiety, depression, and spiritual peace. Spearman’s rank correlation was computed to assess task load and distress. ResultsDuring the 5-month study period, the study included 12 resuscitation events across six inpatient units. Of 82 in-hospital healthcare professionals eligible for recruitment, 44 (53.7%) completed the one-week post-resuscitation event survey. Of those, 37 (84.1%) completed the six-week survey. Highest median task load burden at one week was seen for temporal demand, effort, and mental demand. Median moral distress scores were low, while “at peace” median scores tended to be high. There were no significant non-zero changes in task load or distress scores from weeks 1–6. Mental demand (r = 0.545, p < 0.001), physical demand (r = 0.464, p = 0.005), performance (r = −0.539, p < 0.001), and frustration (r = 0.545, p < 0.001) significantly correlated with overall distress. Performance (r = −0.371, p = 0.028) and frustration (r = 0.480, p = 0.004) also significantly correlated with moral distress. ConclusionsIn-hospital healthcare professionals’ experiences of resuscitation events are varied and complex. Aspects of task load burden including mental and physical demand, performance, and frustration contribute to overall and moral distress, deserving greater attention in clinical contexts.

A Multisite Assessment of Inpatient Safety Event Rates During the Coronavirus Disease 2019 Pandemic

To date, there has been a notable lack of peer-reviewed or publicly available data documenting rates of hospital quality outcomes and patient safety events during the coronavirus disease 2019 pandemic era. The dearth of evidence is perhaps related to the US health care system triaging resources toward patient care and away from reporting and research and also reflects that data used in publicly reported hospital quality rankings and ratings typically lag 2-5 years. At our institution, a learning health system assessment is underway to evaluate how patient safety was affected by the pandemic. Here we share and discuss early findings, noting the limitations of self-reported safety event reporting, and suggest the need for further widespread investigations at other US hospitals. During the 2-year study period from January 1, 2020, through December 31, 2021 across 3 large US academic medical centers at our institution, we documented an overall rate of 25.8 safety events per 1000 inpatient days. The rate of events meeting “harm” criteria was 12.4 per 1000 inpatient days, the rate of nonharm events was 11.1 per 1000 inpatient days, and the fall rate was 2.3 per 1000 inpatient days. This descriptive exploratory analysis suggests that patient safety event rates at our institution did not increase over the course of the pandemic. However, increasing health care worker absences were nonlinearly and strongly associated with patient safety event rates, which raises questions regarding the mechanisms by which patient safety event rates may be affected by staff absences during pandemic peaks.

137: AUTOMATE TO RESUSCITATE: OUTCOMES OF MECHANICAL CPR TWO YEARS AFTER ACUTE CARE IMPLEMENTATION

Introduction: Expert Guidelines endorsed Mechanical Cardiopulmonary Resuscitation (mCPR) devices in response to the COVID-19 pandemic. In April, 2020, a regional hospital received grant-funding for mCPR devices to reduce team exposure during CPR events. Methods: We aimed to evaluate patient outcomes and teammate perceptions associated with mCPR. The primary clinical outcome measured was 1 hour survival post CPR event among inpatients experiencing a cardiopulmonary arrest before & after mCPR implementation, including subset analysis of CPR events where mCPR was specifically noted as being used. Teammate perceptions comparing resuscitation events with and without the mCPR device, barriers to use, and impact on communication and teamwork were assessed using an electronic survey approximately one year post-implementation. Results: Respondents to the team survey (n=22) were primarily nurses (81%) in the critical care unit or the critical care code team, Respiratory Therapists (9.5%) & physicians or APPs (9.5%). One third of participants self-reported using the mCPR device 60% or more of the time. Most (72%) indicated a reduction in teammates required in the room for CPR events in the ICU. Participants rated improvement in communication and ability to focus on their role during a Code Blue. Most (76%) “Agreed” or “Strongly Agreed” that the quality of resuscitation is improved when the mCPR is in use. We defined event survival as return of spontaneous circulation (ROSC) plus survival >1 hour after the CPR event to align with historic institutional data. Baseline survival (2019) at 1 hour across all inpatient areas was 53.1%. During the 2 years post-implementation, inpatient event survival was similar at 52.9% with confirmed mCPR. In critical care, post implementation non-mCPR codes had 54.2% survival compared with 57.14% confirmed mCPR ICU Codes. mCPR was documented only in about 8% of ICU and non-ICU codes. Data are limited by onconsistent documentation of mCPR Utilization. Measures were implemented May 1 2022 and following to better capture utilization. Conclusions: Staff report benefits with mCPR. mCPR outcomes should be monitored as a variable in ongoing quality improvement. Further study of quantitative outcomes with subgroup analysis & staff perception of mCPR devices in inpatient CPR events is needed.

1416. The Economic Burden of Adverse Events Requiring Acute Care Services from Outpatient Parenteral Antibiotic Therapy (OPAT) Treatment

Abstract Background Antimicrobial resistance (AMR) is a growing threat. ESBL-producing Enterobacterales infections are rising, especially within the community setting. Patients requiring OPAT services will increase based on AMR to oral antibiotic (ABX) options. OPAT adverse events (AEs) are linked to the use of IV catheters, followed by adverse drug events. Complications that arise from OPAT, often necessitate acute care services. We sought to quantify costs associated with OPAT AEs. Methods A multicenter retrospective claims analysis from the state of Utah’s (UT) Public IBIS database was performed for 2020. OPAT AEs as described in the literature were used to query charges. All UT hospitals and common OPAT AE principal diagnosis (PDx) codes were included in the analysis. Estimated inpatient (IP) costs associated with common OPAT AEs were calculated from a charge-to-cost ratio (22.5%) using publicly available data from the Centers for Medicare &amp; Medicaid Services. Event counts reported for UT were scaled proportionally to estimate total events for the U.S. population. Emergency department (ED) incidence rates for OPAT AEs from 2016 to 2020 were also examined. Results During the study period, 248,843 patients met study inclusion for an OPAT AE PDx. Among IV-related complications, catheter phlebitis accounted for highest median cost per IP event at $14,051. Other PDx, included catheter blockage and central line-associated bloodstream infections at $11,237 and $10,103, respectively, followed by $9,371 for complications post injection. Thrombotic events equated to total of $11,915 for the combined costs of deep venous thrombosis and pulmonary embolism. Lastly, C. difficile infections accounted for a median cost of $5,284 (Figure 1). Age-adjusted rates of ED activity related to AEs rose to 17.6 per 10,000 in 2020; this marked an 18% increase from 2016 (Figure 2). Figure 1.Figure 2. Conclusion Give that viable oral ABX treatment options in the community setting are limited, patients will require additional OPAT services as AMR rates continue to escalate. OPAT services are not without added risks of complications, as the average median cost for an OPAT AE was $8,852. These costs may be minimized by the addition of new oral ABXs that overcome AMR, thus improving patient outcomes. Disclosures Mauricio Rodriguez, PharmD, MS-HEOR, BCPS, BCCCP, BCIDP, Spero Therapeutics: Employee Georgia Buscaglia, PhD, Spero Therapeutics: Advisor/Consultant Steven Tolle, BFA, Spero Therapeutics: Advisor/Consultant Darren Michael, PhD, CC, SC, Spero Therapeutics: Advisor/Consultant|Spero Therapeutics: Grant/Research Support.

Abstract 11592: Real-World Performance of Heartlogic Index in Subjects With CRT and ICD Devices

Introduction: The HeartLogic heart failure (HF) diagnostic was developed from data in subjects implanted with CRT-D devices. Diagnostic performance in subjects with ICDs (without CRT) is less well characterized. Objective: Compare HeartLogic performance in ICD and CRT subgroups. Methods: Data from US post-market HeartLogic CRT/ICD devices were directly linked with Medicare claims and subjects with continuous Medicare Part A&amp;B fee-for-service coverage were analyzed. Claims HF events (HFE) were defined using primary HF diagnosis codes (DRG 291-3, I50.xx) and included acute inpatient events, and outpatient events with IV diuretic therapy. Baseline characteristics included patient demographics, implant data, and claims chronic conditions flags (CCFs). Diagnostic performance was defined as the sensitivity to detect HFEs and the false-positive alert rate (FPR), from which a modified receiver operator characteristic (ROC) curve was generated. Continuous variables were expressed as mean±SD and categorical variables as proportions. Two-sided 95% confidence intervals (CI) were calculated for sensitivity (binomial distribution) and FPR (negative binomial model). Results: Out of 4190 subjects,1567 linked subjects were included in the analysis with an average follow-up of 1.4±0.5 years. The CRT group (N=1078, 69%) was more likely to be Caucasian (88% vs 82%), less likely to have de novo devices (62% vs 80%), more likely to have CCFs of AF (43% vs 33%) and congestive HF (89% vs 75%), and a HF hospitalization in the year prior to implant (23% vs 16%). CRT and ICD groups provided 296 and 139 HFEs for performance evaluation. HeartLogic ROCs were similar in both groups with no difference in the nominal sensitivity (83% [CI: 78, 87] vs 80% [72, 86%], p =0.41) or FPR (1.386 [CI: 1.337,1.511] vs 1.322 [1.238,1.488], p =0.41). Conclusions: HeartLogic diagnostic performance determined using real-world claims data was not different in subjects implanted with ICDs and CRT-Ds.

Timing and Type of Venous Thromboembolic Chemoprophylaxis Is Associated with Acute Traumatic Brain Injury Outcomes.

Venous thromboembolic (VTE) prophylaxis in acute traumatic brain injury (TBI) is a controversial topic with wide practice variations. This study examined the association of VTE chemoprophylaxis with inpatient mortality and VTE events among isolated TBI patients. This was a retrospective cohort study of 87 trauma centers within a large hospital system in the United States analyzing 23,548 patients with isolated TBI, 7977 of whom had moderate-to-severe TBI. Primary outcomes were inpatient mortality and VTE events. The control group received no chemoprophylaxis. Other groups received low-molecular-weight heparin (LMWH), unfractionated heparin (UFH), and combined LMWH and UFH chemoprophylaxis. Multi-variable regression accounted for confounders. Outcomes were stratified by timing of administration, body mass index (BMI), and TBI type. Patients without VTE prophylaxis had the least VTE events. LMWH had the lowest mortality for both all-isolated and moderate-to-severe isolated TBI populations at adjusted odds ratio (aOR) 0.24 (95% confidence interval [CI], 0.14-0.43) and aOR 0.25 (95% CI, 0.14-0.44), respectively. Clinically significant progression of TBI was lowest among the LMWH group (0.1%; p value, 0.001). After stratifying by timing of VTE chemoprophylaxis, only patients with subdural hematoma and LMWH between 6 and 24 h (N = 62), as well as patients with ≥35 BMI and LMWH between 6 and 24 h (N = 65) or >24-48 h (N = 54), had no VTE events. VTE chemoprophylaxis timing may have prevented VTE in certain subgroups of isolated TBI patients. Though VTE chemoprophylaxis did not prevent VTE for most TBI patients, LMWH VTE chemoprophylaxis was associated with reduced mortality.

Inpatient long-term video-electroencephalographic monitoring event capture audiovisual diagnostic quality

ObjectivesLong-term video-electroencephalographic monitoring (LTVEM) represents the gold-standard method to evaluate whether events represent electrographic seizures, but limited work has evaluated the quality of inpatient event capture. We evaluated the frequency of audiovisual factors impairing the ideal electroclinical correlation of seizure-like episodes during LTVEM. MethodsWe retrospectively reviewed consecutive inpatient LTVEM studies (11/2019–12/2019) from three academic epilepsy centers. We evaluated all pushbutton events for audiovisual characteristics such as whether the event was narrated, whether the patient was blocked on camera, and what diagnostic challenges impaired the electroencephalographer’s ability to understand either the reason the event button was pushed or clinical semiology (“electroclinical correlation”). We determined the percent of events and studies with each outcome. ResultsThere were 154 studies with 520 pushbutton events. The pushbutton was most commonly activated by patients (41%), followed by nurses (31%) or family (17%). Twenty-nine percent of events represented electrographic seizures, and 78% occurred in the Epilepsy Monitoring Unit. The reason for the push was not stated in 45% of events, and inadequate narration impaired electroclinical correlation in 19% of events. At least one relevant part of the patient’s body was blocked during 12% of events, but this impaired electroclinical correlation in only 1% of events. There was at least one factor impairing electroclinical correlation in 21% of events, most commonly due to incomplete narration (N = 99), lights off (N = 15), or blankets covering the patient (N = 15). At least one factor impaired electroclinical correlation for any event in 36% of studies. ConclusionAudiovisual factors impairing the electroencephalographer’s ability to render an electroclinical correlation were common, particularly related to inadequate narration from bedside observers to explain the reason for pushing the button or event semiology. Future efforts to develop targeted countermeasures should address narration challenges and improve inpatient seizure monitoring quality metrics.

Impact of a subcutaneous casirivimab and imdevimab clinic in outpatients with symptomatic COVID-19: A single-center, propensity-matched cohort study.

To evaluate the success of a clinic for subcutaneous administration of casirivmab and imdevimab (REGEN-COV; Regeneron) for treatment of patients with symptomatic mild to moderate coronavirus disease 2019 (COVID-19) in terms of preventing disease progression and healthcare utilization. This retrospective single-center, propensity-matched cohort study examined healthcare utilization outcomes for patients who received subcutaneous casirivimab and imdevimab at a pharmacist-led clinic of an academic health system. Eligible patients were treated between August 1, 2021, and January 5, 2022, and were at high risk for COVID-19 disease progression. Treatment patients were propensity matched with high-risk control patients with a diagnosis of COVID-19 in the same timeframe who did not receive casirivimab and imdevimab. Patients were followed for 30 days for collection of data on inpatient admissions, emergency department visits, and mortality. Risk of a 30-day healthcare utilization event was assessed and tested for statistical significance utilizing McNemar's test. A total of 585 patients who received treatment with subcutaneous casirivimab and imdevimab were matched with 585 patients who did not receive casirivimab and imdevimab therapy. Patients who received casirivimab and imdevimab had significantly lower risk of a 30-day all-cause inpatient admission event than untreated patients (relative risk reduction, 62.4%; P < 0.0001). Treated patients also had a significantly lower risk of 30-day all-cause emergency department visit than untreated subjects (relative risk reduction, 36.5%; P = 0.0021). There were 6 mortality events in the untreated group and no mortality events in the treatment group. This study provides evidence for the effectiveness of a subcutaneous casirivimab and imdevimab clinic in preventing progression of symptomatic mild to moderate COVID-19.

Inpatient Constipation Among Migraine Patients Prescribed Anti-calcitonin Gene-Related Peptide Monoclonal Antibodies and Standard of Care Antiepileptic Drugs: A Retrospective Cohort Study in a United States Electronic Health Record Database.

Erenumab, an anti-calcitonin gene-related peptide (CGRP) receptor monoclonal antibody (mAb), was approved by the US Food and Drug Administration in May 2018. Constipation with serious complications was added to the Warning and Precautions section in the erenumab Prescribing Information in October 2019 after events were observed during post-marketing surveillance. We aimed to assess and compare the risk of inpatient constipation, and, separately, inpatient constipation with serious complications, among patients with migraine treated with CGRP mAbs and standard of care antiepileptic drugs (AEDs). Within Optum's Electronic Health Record Research Database, patients with migraine who initiated erenumab, other CGRP mAbs, and AEDs were identified from May 2018 through March 2020. Erenumab initiators were propensity score-matched separately to initiators of other CGRP mAbs and AEDs. Incident inpatient constipation events, and serious complications, were identified using multiple risk windows for outcome assessment (30-, 60-, 90-day risk windows, and all available follow-up). Odds ratios (ORs) were calculated comparing inpatient constipation risk among matched erenumab initiators relative to comparators. We identified 17,902 erenumab, 13,404 other CGRP mAb, and 49,497 AED initiators who met study criteria. Among matched initiators, the risk of inpatient constipation was 0.46% (95% confidence interval (CI) 0.35-0.60) for erenumab and 0.44% (95%CI 0.33-0.58) for other CGRP mAbs within the 90-day risk window, with a corresponding OR of 1.06 (95%CI 0.72-1.55). Among matched erenumab and AED initiators, inpatient constipation risk was 0.53% (95%CI 0.42-0.66) and 0.76% (95%CI 0.62-0.92), respectively, and the OR was 0.69 (95%CI 0.51-0.94). Few serious complications were observed. Patients initiating erenumab had similar risk of inpatient constipation within 90days of treatment initiation versus patients initiating other CGRP mAbs, and lower risk versus patients initiating AEDs. These findings provide context to events observed during post-marketing surveillance.

Development and validation of interpretable machine learning models for inpatient fall events and electronic medical record integration.

Falls are one of the most frequently occurring adverse events among hospitalized patients. The Morse Fall Scale, which has been widely used for fall risk assessment, has the two limitations of low specificity and difficulty in practical implementation. The aim of this study was to develop and validate an interpretable machine learning model for prediction of falls to be integrated in an electronic medical record (EMR) system. This was a retrospective study involving a tertiary teaching hospital in Seoul, Korea. Based on the literature, 83 known predictors were grouped into seven categories. Interpretable fall event prediction models were developed using multiple machine learning models including gradient boosting and Shapley values. Overall, 191,778 cases with 272 fall events (0.1%) were included in the analysis. With the validation cohort of 2020, the area under the receiver operating curve (AUROC) of the gradient boosting model was 0.817 (95% confidence interval [CI], 0.720-0.904), better performance than random forest (AUROC, 0.801; 95% CI, 0.708-0.890), logistic regression (AUROC, 0.802; 95% CI, 0.721-0.878), artificial neural net (AUROC, 0.736; 95% CI, 0.650-0.821), and conventional Morse fall score (AUROC, 0.652; 95% CI, 0.570-0.715). The model's interpretability was enhanced at both the population and patient levels. The algorithm was later integrated into the current EMR system. We developed an interpretable machine learning prediction model for inpatient fall events using EMR integration formats.

Modification by Multiple Social Stressors in Air Pollution-CVD Associations

BACKGROUND AND AIM Susceptibility to urban air pollution is shown to vary by chronic psychosocial stressors (e.g., poverty, violence), though it remains unclear whether a confluence of multiple stressors may compound susceptibility, or confound observed modification by any one stressor. METHODS We examined modification by multiple community stressors on relationships between spatio-temporal pollutant exposures and risk of cardiovascular (CVD) event in New York City, using conditional logistic regression in a case-crossover design. We used exhaustive data on all inpatient (admitted) CVD events presenting at NYC emergency departments from 2005-2011 (n = 837,523), fine-scale spatial maps for PM2.5, NO2, SO2, and O3 from NYC Community Air Survey, and daily EPA pollution and NOAA weather data. A range of stressor indicators were drawn from Census and administrative databases, examined against focus group and survey data to confirm relevance to residents’ stress experience. Due to concern about stressors clustering by race and ethnicity (i.e., non-random assignment into tracts), we adjusted year-round models (NO2 and PM2.5) for modification by tract composition. RESULTS Testing stressors individually, we found significant associations between NO2 and CVD risk only in the highest quintile for violent crime, felony assault, poverty, or socioeconomic deprivation (SDI). Associations in all other quintiles were non-significant, as were associations for all other pollutants. In models including separate interaction terms for violent crime and SDI, NO2 was associated with significant increases in CVD risk in all quintiles. Associations for NO2 modestly increased across quintiles of increasing violence, but not SDI. CONCLUSIONS Observed patterns of modification differed substantially when including multiple stressor modifiers, increasing observed NO2-CVD associations overall. These results suggest a need to capture community stress experience more holistically, to reduce confounding across stressors which may dampen observed pollutant-health associations and modification. KEYWORDS Social susceptibility, chronic stress, urban health, air pollution, cardiovascular disease

Piloting an Electronic Sepsis Alerting System to Reduce Inpatient Sepsis-related Serious Safety Events.

Background: Electronic surveillance systems that enable continuous patient screening for emerging signs of sepsis may improve sepsis recognition and patient outcomes. Objectives: An electronic sepsis alerting system was piloted on Pediatric Medicine units at a tertiary care hospital. The pilot served to refine the system and accompanying workflow before hospital-wide implementation, aiming to reduce inpatient serious safety events related to delayed sepsis recognition. Methods: The multitiered intervention included (1) an electronic sepsis screening and alerting system; (2) an interdisciplinary huddle workflow; (3) risk stratification nomenclature (continue routine care, sepsis watch, code sepsis) to reconcile alerts; and (4) streamlined documentation and ordering tools. Iterative changes were made to the tools and workflow using plan, do, study, act cycle methodology. Measures of clinician adherence to the workflow were tracked monthly during the pilot (% alerts reconciled, % alerts with documented huddle note). Alert to huddle time and huddle to reconciliation time (target: 10 minutes) served as outcome measures for the pilot. The frequency of consults to the hospital’s critical care response team for the indication of query sepsis was monitored as a balancing measure. Alert frequency, rate of corresponding clinical concern for sepsis, and missed sepsis cases were tracked as measures of screening tool utility. Results: During the 3-month pilot, 86% of alerts were reconciled and 67% had a documented huddle note. The median alert to huddle time decreased from 9 to 3 minutes over the course of the pilot. The median huddle to reconciliation time decreased from 14 to 8 minutes. The frequency of CCRT consults during the pilot was 1 per month, which was increased slightly from the baseline frequency of 0.67 consults per month over the previous year. An alert was triggered for 4.1% of opportunities (average 2.2 alerts per day). Of the alerts reconciled, 29% indicated a concern for sepsis. There were four instances of clinical concern for sepsis where no alert was triggered. The sepsis huddle workflow was followed in three of these instances. Conclusions/implications: Pilot results suggest clinician adherence to a new streamlined approach to care for inpatients at risk for sepsis. Involving clinicians in system design and embedding electronic tools within existing care processes supported workflow uptake. Future efforts will focus on optimizing the alerting system’s clinical accuracy and examining its impact on clinical outcomes.