Huddleston, Jeanne M.

Background Improving safety in healthcare has been paramount for decades, yet despite major attention and investment, improvement has remained incremental. Patient safety is a major concern in US healthcare, leading to significant harm and economic losses annually. Accurately identifying safety events remains difficult due to methodological discrepancies and lack of standardisation. This study evaluated the feasibility of implementing a standardised case-review methodology and safety-event taxonomy across diverse hospital settings to assess opportunities for improvement (OFIs) and compare findings with traditional definitions. Methods This multicentre retrospective cohort study reports data from 103 hospitals across the USA and Canada between 2016 and 2023. A multivariable logistic regression was performed to test case reviews for differences in the presence of one or more OFIs across several hospital types (bed size, academic status, urban setting, trauma level and Centres for Medicare and Medicaid Services overall star rating) and patient characteristics (age, gender, length of stay, admission and discharge code status and mortality). Results 19 181 cases were reviewed across the Learning Health System Collaborative, with a median of 107 reviews per hospital. Mortality was the most common cohort selection, studied by 91 hospitals (88%). At least one OFI was identified in 12 714 cases (66.3%). The logistic regression analysis found that all hospital characteristics and patient age, length of stay, code status and discharge disposition were significantly associated with at least one OFI. Of the 46 444 OFIs identified, 41 439 (89%) were from categories focused on omissions of care. The categories of end-of-life, documentation and treatment/care alone accounted for 25 980 OFIs (56%). Conclusion The highest volumes of safety-related OFIs were associated with omissions of care, as opposed to the traditional definition of patient safety, which primarily includes outcomes from acts of commission.

Background: Patients suffering post operative myocardial infarction (PMI) have higher mortality and morbidity rates than observed with non-PMI subsets. Little is known about the true incidence of PMI and how PMI differs from non-PMI with regard to clinical presentation, management and outcome. Methods: We examined the incidence and outcomes of patients with PMI in 1212 subjects (≥65 years) undergoing hip fracture surgery from 1988 – 2002 using a population-based retrospective cohort study from the Rochester Epidemiology Project in Olmstead County, Minnesota. PMI was defined on the basis of cardiac biomarker and/or electrocardiographic evidence of definitive MI within 7 days post operatively. Results: PMI occurred in 170 (14.0%) patients yet only 36 (21.2 %) patients had one or more clinical symptoms suggesting coronary ischemia. One or more classic symptoms of myocardial infarction were present in 26 (15.3%) including shortness of breath (SOB) in 18 (10.6%), chest/arm pain in 11 (6.5%), nausea/vomiting in 8 (4.7%) and diaphoresis in 1 (0.6%). Delirium was present in 15 (8.8%) patients. Median hospital length of stay was longer (10.0 vs. 8 days, p<0.01) and the in-hospital mortality rate was significantly higher (14.1% vs. 1.6%, p<0.001) among patients who had PMI within 7 days. Conclusion: The clinical presentation of PMI differs from non-PMI in the elderly undergoing hip fracture surgery. Most patients were asymptomatic or had atypical clinical presentations. We observed a significantly higher rate of hospital mortality among those experiencing a PMI. Further studies are necessary to better understand the presentation, optimal management and prevention of PMI. Table 1. Study Population Table 2. Clinical Presentation and Outcomes in Patients with and without Post Operative MI This research has received full or partial funding support from the American Heart Association, National Center.

Sepsis is considered a medical emergency where delays in initial treatment are associated with increased morbidity and mortality, yet there is no gold standard for identifying sepsis onset and thus treatment timing. We leverage electronic health record (EHR) data with clinical expertise to develop a continuous-time Markov decision process (MDP) optimal stopping model that identifies the optimal first intervention action (anti-infective, fluid, or wait). To study the impact of initial treatment of patients at risk for developing sepsis, we define the delayed treatment population who received delayed treatment upon admission or during hospitalization and serves as an approximation of the natural history of sepsis. We apply the optimal first treatment policy to sample patient visits from the nondelayed treatment population. This analysis indicates the average risk of death could be reduced by approximately 2.2%, the average time until treatment could be reduced by 106 minutes, and the average severity of the treatment state could be reduced by 15.5% compared with the treatment they received in the hospital. We study the properties of the optimal policy to define an easily interpretable initial treatment heuristic that considers a patient’s organ dysfunction, location, and septic shock status. This generalizable framework can inform personalized treatment of patients at risk for sepsis. History: This paper has been accepted for the Decision Analysis Special Issue on Emerging Topics in Health Decision Analysis. Funding: This material is based upon work supported by the National Science Foundation [Grant 1522107 (North Carolina State University), 1522106 (Mayo Clinic), and 1833538 (Drexel University)].

Abstract Objective We aimed to develop a model for accurate prediction of general care inpatient deterioration. Materials and Methods Training and internal validation datasets were built using 2-year data from a quaternary hospital in the Midwest. Model training used gradient boosting and feature engineering (clinically relevant interactions, time-series information) to predict general care inpatient deterioration (resuscitation call, intensive care unit transfer, or rapid response team call) in 24 hours. Data from a tertiary care hospital in the Southwest were used for external validation. C-statistic, sensitivity, positive predictive value, and alert rate were calculated for different cutoffs and compared with the National Early Warning Score. Sensitivity analysis evaluated prediction of intensive care unit transfer or resuscitation call. Results Training, internal validation, and external validation datasets included 24 500, 25 784 and 53 956 hospitalizations, respectively. The Mayo Clinic Early Warning Score (MC-EWS) demonstrated excellent discrimination in both the internal and external validation datasets (C-statistic = 0.913, 0.937, respectively), and results were consistent in the sensitivity analysis (C-statistic = 0.932 in external validation). At a sensitivity of 73%, MC-EWS would generate 0.7 alerts per day per 10 patients, 45% less than the National Early Warning Score. Discussion Low alert rates are important for implementation of an alert system. Other early warning scores developed for the general care ward have achieved lower discrimination overall compared with MC-EWS, likely because MC-EWS includes both nursing assessments and extensive feature engineering. Conclusions MC-EWS achieved superior prediction of general care inpatient deterioration using sophisticated feature engineering and a machine learning approach, reducing alert rate.

Abstract Introduction Identification of hospitalized patients with suddenly unfavorable clinical course remains challenging. Models using objective data elements from the electronic health record may miss important sources of information available to nurses. Methods We recorded nurses’ perception of patient potential for deterioration in 2 medical and 2 surgical adult hospital units using a 5-point score at the start of the shift (the Worry Factor [WF]), and any time a change or an increase was noted by the nurse. Cases were evaluated by three reviewers. Intensive care unit (ICU) transfers were also tracked. Results 31 159 patient-shifts were recorded for 3185 unique patients during 3551 hospitalizations, with 169 total outcome events. Out of 492 potential deterioration events identified, 380 (77%) were confirmed by reviewers as true deterioration events. Likelihood ratios for ICU transfer were 17.8 (15.2–20.9) in the 24 hours following a WF &gt; 2, and 40.4 (27.1–60.1) following a WF &gt; 3. Accuracy rates were significantly higher in nurses with over a year of experience (68% vs 79%, P = 0.04). The area under the receiver operator characteristic curve (AUROC) was 0.92 for the prediction of ICU transfer within 24 hours. Discussion This is a higher accuracy than most published early warning scores. Conclusion Nurses’ pattern recognition and sense of worry can provide important information for the detection of acute physiological deterioration and should be included in the electronic medical record.

Objective Create a score to identify patients at risk of death or hospice placement who may benefit from goals of care discussion earlier in the hospitalisation. Design Retrospective cohort study to develop a risk index using multivariable logistic regression. Setting Two tertiary care hospitals in Southeastern Minnesota. Participants 92 879 adult general care admissions (50% male, average age 60 years). Primary and secondary outcome measures Our outcome measure was an aggregate of inhospital death or discharge to hospice. Predictor variables for the model encompassed comorbidities, nutrition status, functional status, demographics, fall risk, mental status, Charlson Comorbidity Index and acuity of illness on admission. Resuscitation status, race, geographic area of residence and marital status were added as covariates to account for confounding. Results Inhospital mortality and discharge to hospice were rare, with incidences of 1.2% and 0.8%, respectively. The Hospital End-of-Life Prognostic Score (HELPS) demonstrated good discrimination (C-statistic=0.866 in derivation set and 0.834 in validation set). The patients with the highest 5% of scores had an 8% risk of the outcome measure, relative risk 12.9 (10.9–15.4) when compared to the bottom 95%. Conclusions HELPS is able to identify patients with a high risk of inhospital death or need for hospice at discharge. These patients may benefit from early goals of care discussions.

BACKGROUNDThe use of rapid response systems (RRS), which were designed to bring clinicians with critical care expertise to the bedside to prevent unnecessary deaths, has increased. RRS rely on accurate detection of acute deterioration events. Early warning scores (EWS) have been used for this purpose but were developed using heterogeneous populations. Predictive performance may differ in medical vs surgical patients.OBJECTIVETo evaluate the performance of published EWS in medical vs surgical patient populations.DESIGNRetrospective cohort study.SETTINGTwo tertiary care academic medical center hospitals in the Midwest totaling more than 1500 beds.PATIENTSAll patients discharged from January to December 2011.INTERVENTIONNone.MEASUREMENTSTime‐stamped longitudinal database of patient variables and outcomes, categorized as surgical or medical. Outcomes included unscheduled transfers to the intensive care unit, activation of the RRS, and calls for cardiorespiratory resuscitation (“resuscitation call”). The EWS were calculated and updated with every new patient variable entry over time. Scores were considered accurate if they predicted an outcome in the following 24 hours.RESULTSAll EWS demonstrated higher performance within the medical population as compared to surgical: higher positive predictive value (P < .0001 for all scores) and sensitivity (P < .0001 for all scores). All EWS had positive predictive values below 25%.CONCLUSIONSThe overall poor performance of the evaluated EWS was marginally better in medical patients when compared to surgical patients.

60 Background: A significant number of patients with advanced cancer die in the hospital. Examination of patterns of care and palliative care (PC) involvement may identify opportunities for process of care improvements for this vulnerable population. Methods: Patients were identified using the institutional mortality review system registry (Mayo Clinic hospitals from July, 2013-June, 2014). Within this group, patients with a diagnosis of terminal malignancy were identified by chart review and ICD-9 codes. Patient demographics, clinical characteristics, and use of PC resources were characterized in the last 60 days of life. Results: There were 159 of 924 decedents identified whose primary cause of death was advanced malignancy (Table). Ninety-two patients (57%) had PC consultation during the index admission preceding death, while 31 (19%) patients had seen PC as an outpatient or inpatient prior to index admission. Lack of inpatient PC consultation at index admission was associated with hematologic malignancy (p < 0.001), full code status at death (p 0.009), shorter median time between change of code status to DNR and death (13 vs 67 hrs, p < 0.001), death in the ICU (58% vs 31%, p < 0.001), and presence of code blue (22% vs 5%, p-value = 0.003). Patients who received a PC consultation were less likely to have chemotherapy within 14 days of death (15% vs 37%, p-value = 0.001), and had a median composite aggressive end of life score (6 factor composite score based on utilization in last 30 days of life, higher score indicating more aggressive care) of 3 vs 4 (p 0.002). Conclusions: Inpatient PC consultation for cancer patients who die in the hospital is associated with less aggressive cancer care at end of life. The single-institution patient population and small sample size are the limitations of our study. [Table: see text]