Higher Risk Of Readmission Research Articles

Contemporary Trends of Clinical Outcomes in Primary Left Ventricular Assist Device Implantation and Postprocedure High-Risk Categories

ObjectiveWe aimed to analyze trends of 30-day readmission and find high-risk patients associated with increased risk of mortality, resource use, and readmission after primary left ventricular assist device (LVAD) implantation. Limited data exist on the contemporary trends of readmission rates and patients at a higher risk of worse outcomes after LVAD implantation. Methods and ResultsThis is a retrospective study of adults from the Nationwide Readmission Database who underwent primary durable LVAD implantation from 2010 to 2018. The main outcomes were 30-day readmission rates and their trends in patients with primary durable LVAD implantation from 2010 to 2018. This study also sought to identify patients at the highest risk for readmission, in-hospital mortality, and resource use. A total of 31,002 adults with primary durable LVAD implantation were included in the present analysis. Overall, 3808 patients (12.3%) died and 27,168 (87.6%) were discharged alive. Of those discharged alive, 8303 patients (30.6%) were readmitted within 30 days. The trend of 30-day all-cause readmission among LVAD implantation patients remained similar from 2010 to 2018 (P = .809). The in-hospital mortality rate during the index hospitalization decreased significantly (P = .014), and the mean cost of an index hospitalization increased (P = .031) during the study period. The patients with post-LVAD in-hospital cardiac, vascular, and thromboembolic complications (ie, high-risk patients) had the highest mortality, resource use, and readmission rates compared with patients without major complications. ConclusionsThis study found that the readmission rates associated with LVAD implantation did not change from 2010 to 2018 and identified high-risk patients who may benefit from closer monitoring after primary LVAD implantation.

EP.TH.649Identifying Factors Associated with Readmissions after acute biliary admission. Can it help prioritising patients in COVID recovery plans?

Abstract Aim We conducted a retrospective analysis of readmissions after Acute admission with biliary pathology managed conservatively under the Surgical Team in a teaching London Hospital from 01/03/2019-29/02/2020 Methods We obtained records of patients admitted with Acute Cholecystitis, Biliary Colic, Cholelithiasis, Choledocholithiasis and Gallstone Pancreatitis from the Audit Department between 01/03/2019-29/02/2020, and analysed these regarding patient demographics, comorbidities, duration of index admission, method of diagnosis and management and identified patients’ readmissions Differences in readmission rates based on before mentioned characteristics were studied. Results 157 patients presented between 01/03/2019-29/02/2020, 76 acute cholecystitis, 22 Biliary Colic, 24 Gallstone Pancreatitis, 6 Ascending Cholangitis and 29 Choledocholithiasis The highest representation rate was for patients with choledocholithiasis (41.3%) followed by Acute Cholecystitis (31.5%), 3 patients required cholecystostomies. Baseline characteristics and differences in these characteristics based on occurrence of readmission were studied. 45-60 age group, increasing comorbidity, and biliary obstruction were all associated with increased risk of readmission. Conclusion Readmissions is a substantial burden on the health care services and patient’s safety and QoL We propose adherence to the NICE/BSG Guidelines for management of Acute Biliary Disease, to alleviate this pressure which is already sometimes challenging due to the logistics and resources and would be more challenging with the COVID situation and limited emergency and elective theatre availability so the group at high risk of readmission should be prioritised in the recovery plans.

Evaluation of Incident 7-Day Infection and Sepsis Hospitalizations in an Integrated Health System.

Rationale: Prehospital opportunities to predict infection and sepsis hospitalization may exist, but little is known about their incidence following common healthcare encounters. Objectives: To evaluate the incidence and timing of infection and sepsis hospitalization within 7 days of living hospital discharge, emergency department discharge, and ambulatory visit settings. Methods: In each setting, we identified patients in clinical strata based on the presence of infection and severity of illness. We estimated number needed to evaluate values with hypothetical predictive model operating characteristics. Results: We identified 97,614,228 encounters, including 1,117,702 (1.1%) hospital discharges, 4,635,517 (4.7%) emergency department discharges, and 91,861,009 (94.1%) ambulatory visits between 2012 and 2017. The incidence of 7-day infection hospitalization varied from 37,140 (3.3%) following inpatient discharge to 50,315 (1.1%) following emergency department discharge and 277,034 (0.3%) following ambulatory visits. The incidence of 7-day infection hospitalization was increased for inpatient discharges with high readmission risk (10.0%), emergency department discharges with increased acute or chronic severity of illness (3.5% and 4.7%, respectively), and ambulatory visits with acute infection (0.7%). The timing of 7-day infection and sepsis hospitalizations differed across settings with an early rise following ambulatory visits, a later peak following emergency department discharges, and a delayed peak following inpatient discharge. Theoretical number needed to evaluate values varied by strata, but following hospital and emergency department discharge, were as low as 15-25. Conclusions: Incident 7-day infection and sepsis hospitalizations following encounters in routine healthcare settings were surprisingly common and may be amenable to clinical predictive models.

Role of frailty in prediction of hospitalized older adult patient’s outcomes: a prospective study

Background/aim Frailty is associated with an increased risk of negative short-term and long-term hospital outcomes. This study aimed to evaluate the role of frailty in predicting readmission, length of stay, and quality of life in the hospitalized older adults.Materials and methods This observational study was conducted at Ziaiyan Hospital, Tehran, Iran. In total, 304 participants (65–85 years), were enrolled through the inclusion criteria from August to December 2019. The frailty index (FI) was assessed by the minimum data set-home care. Readmission was obtained through telephone interviews. The length of stay was gathered by the patient’s hospital records, and the EuroQol questionnaire was used for assessing the quality of life. Data were collected by a researcher nurse at the admission time, 30, 60, and 90 days after discharge. The logistic regression model and repeated measures ANOVA were employed to analyze the association between frailty and outcomes.ResultsAccording to FI, 102 (33.55%) participants were pre-frail, whereas 35 (11.51%) were frail. In the fully-adjusted model for readmission, the pre-frail participants had a higher risk of readmission at the hospital in comparison with the nonfrail and frail groups (OR = 1.88, 95% CI = 1.90–3.26), and also for GP visits, frail patients showed nearly significant differences (OR = 2.45, 95% CI = 0.99–6.06) but there were no differences between frail and pre-frail patients in readmissions in the emergency ward. In a fully-adjusted prolonged stay model, pre-frail patients had a higher probability to stay longer in hospital (OR = 2.28, 95% CI: 1.24–4.18). The fully-adjusted model for QoL showed, frail patients were more prone to the declined levels of QoL in comparison with pre-frail patients (OR = 10.77, 95% CI: 3.97–29.18).ConclusionsThe findings indicated that frailty worsened negative outcomes and declined QoL. Early diagnosis in hospital settings could be beneficial for designing optimal care plans for the frail and pre-frail patients.

Early and short-term intensive management after discharge for patients hospitalized with acute heart failure: a randomized study (ECAD-HF).

Hospitalization for acute heart failure (HF) is followed by a vulnerable time with increased risk of readmission or death, thus requiring particular attention after discharge. In this study, we examined the impact of intensive, early follow-up among patients at high readmission risk at discharge after treatment for acute HF. Hospitalized acute HF patients were included with at least one of the following: previous acute HF < 6 months, systolic blood pressure ≤ 110 mmHg, creatininaemia ≥ 180 µmol/L, or B-type natriuretic peptide ≥ 350 pg/mL or N-terminal pro B-type natriuretic peptide ≥ 2200 pg/mL. Patients were randomized to either optimized care and education with serial consultations with HF specialist and dietician during the first 2-3 weeks, or to standard post-discharge care according to guidelines. The primary endpoint was all-cause death or first unplanned hospitalization during 6-month follow-up. Among 482 randomized patients (median age 77 and median left ventricular ejection fraction 35%), 224 were hospitalized or died. In the intensive group, loop diuretics (46%), beta-blockers (49%), angiotensin-converting enzyme inhibitors or angiotensin receptor blockers (39%) and mineralocorticoid receptor antagonists (47%) were titrated. No difference was observed between groups for the primary endpoint (hazard ratio 0.97; 95% confidence interval 0.74-1.26), nor for mortality at 6 or 12 months or unplanned HF rehospitalization. Additionally, no difference between groups according to age, previous HF and left ventricular ejection fraction was found. In high-risk HF, intensive follow-up early post-discharge did not improve outcomes. This vulnerable post-discharge time requires further studies to clarify useful transitional care services.

Predictive modeling for 14-day unplanned hospital readmission risk by using machine learning algorithms

BackgroundEarly unplanned hospital readmissions are associated with increased harm to patients, increased medical costs, and negative hospital reputation. With the identification of at-risk patients, a crucial step toward improving care, appropriate interventions can be adopted to prevent readmission. This study aimed to build machine learning models to predict 14-day unplanned readmissions.MethodsWe conducted a retrospective cohort study on 37,091 consecutive hospitalized adult patients with 55,933 discharges between September 1, 2018, and August 31, 2019, in an 1193-bed university hospital. Patients who were aged < 20 years, were admitted for cancer-related treatment, participated in clinical trial, were discharged against medical advice, died during admission, or lived abroad were excluded. Predictors for analysis included 7 categories of variables extracted from hospital’s medical record dataset. In total, four machine learning algorithms, namely logistic regression, random forest, extreme gradient boosting, and categorical boosting, were used to build classifiers for prediction. The performance of prediction models for 14-day unplanned readmission risk was evaluated using precision, recall, F1-score, area under the receiver operating characteristic curve (AUROC), and area under the precision–recall curve (AUPRC).ResultsIn total, 24,722 patients were included for the analysis. The mean age of the cohort was 57.34 ± 18.13 years. The 14-day unplanned readmission rate was 1.22%. Among the 4 machine learning algorithms selected, Catboost had the best average performance in fivefold cross-validation (precision: 0.9377, recall: 0.5333, F1-score: 0.6780, AUROC: 0.9903, and AUPRC: 0.7515). After incorporating 21 most influential features in the Catboost model, its performance improved (precision: 0.9470, recall: 0.5600, F1-score: 0.7010, AUROC: 0.9909, and AUPRC: 0.7711).ConclusionsOur models reliably predicted 14-day unplanned readmissions and were explainable. They can be used to identify patients with a high risk of unplanned readmission based on influential features, particularly features related to diagnoses. The operation of the models with physiological indicators also corresponded to clinical experience and literature. Identifying patients at high risk with these models can enable early discharge planning and transitional care to prevent readmissions. Further studies should include additional features that may enable further sensitivity in identifying patients at a risk of early unplanned readmissions.

'Time is prognosis' in heart failure: time-to-treatment initiation as a modifiable risk factor.

In heart failure (HF), acute decompensation can occur quickly and unexpectedly because of worsening of chronic HF or to new‐onset HF diagnosed for the first time (‘de novo’). Patients presenting with acute HF (AHF) have a poor prognosis comparable with those with acute myocardial infarction, and any delay of treatment initiation is associated with worse outcomes. Recent HF guidelines and recommendations have highlighted the importance of a timely diagnosis and immediate treatment for patients presenting with AHF to decrease disease progression and improve prognosis. However, based on the available data, there is still uncertainty regarding the optimal ‘time‐to‐treatment’ effect in AHF. Furthermore, the immediate post‐worsening HF period plays an important role in clinical outcomes in HF patients after hospitalization and is known as the ‘vulnerable phase’ characterized by high risk of readmission and early death. Early and intensive treatment for HF patients in the ‘vulnerable phase’ might be associated with lower rates of early readmission and mortality. Additionally, in the chronic stable HF outpatient, treatments are often delayed or not initiated when symptoms are stable, ignoring the risk for adverse outcomes such as sudden death. Consequently, there is a dire need to better identify HF patients during hospitalization and after discharge and treating them adequately to improve their prognosis. HF is an urgent clinical scenario along all its stages and disease conditions. Therefore, time plays a significant role throughout the entire patient's journey. Therapy should be optimized as soon as possible, because this is beneficial regardless of severity or duration of HF. Time lavished before treatment initiation is recognized as important modifiable risk factor in HF.

Prediction of readmission in patients with acute exacerbation of chronic obstructive pulmonary disease within one year after treatment and discharge

BackgroundTo investigate the risk factors and construct a logistic model and an extreme gradient boosting (XGBoost) model to compare the predictive performances for readmission in acute exacerbation of chronic obstructive pulmonary disease (AECOPD) patients within one year.MethodsIn total, 636 patients with AECOPD were recruited and divided into readmission group (n = 449) and non-readmission group (n = 187). Backward stepwise regression method was used to analyze the risk factors for readmission. Data were divided into training set and testing set at a ratio of 7:3. Variables with statistical significance were included in the logistic model and variables with P < 0.1 were included in the XGBoost model, and receiver operator characteristic (ROC) curves were plotted.ResultsPatients with acute exacerbations within the previous 1 year [odds ratio (OR) = 4.086, 95% confidence interval (CI) 2.723–6.133, P < 0.001), long-acting β agonist (LABA) application (OR = 4.550, 95% CI 1.587–13.042, P = 0.005), inhaled corticosteroids (ICS) application (OR = 0.227, 95% CI 0.076–0.672, P = 0.007), glutamic-pyruvic transaminase (ALT) level (OR = 0.985, 95% CI 0.971–0.999, P = 0.042), and total CAT score (OR = 1.091, 95% CI 1.048–1.136, P < 0.001) were associated with the risk of readmission. The AUC value of the logistic model was 0.743 (95% CI 0.692–0.795) in the training set and 0.699 (95% CI 0.617–0.780) in the testing set. The AUC value of XGBoost model was 0.814 (95% CI 0.812–0.815) in the training set and 0.722 (95% CI 0.720–0.725) in the testing set.ConclusionsThe XGBoost model showed a better predictive value in predicting the risk of readmission within one year in the AECOPD patients than the logistic regression model. The findings of our study might help identify patients with a high risk of readmission within one year and provide timely treatment to prevent the reoccurrence of AECOPD.

What happens to patients undergoing cancer surgery with psychiatric comorbidities?: A nationwide retrospective cohort study.

We aimed to describe the surgical outcomes and readmissions rates in patients with psychological comorbidities who underwent surgical intervention for cancer. The Hospital Readmission Reduction Program, a campaign by the Affordable Care Act, penalizes hospitals for greater readmission rates. One prevalent patient risk factor associated with hospital readmission is a comorbid psychiatric condition. Currently, no studies exist identifying psychiatric comorbidities and their effects on patients with cancer undergoing surgical intervention. A cross-sectional analysis was performed using the Nationwide Readmissions Database for 2010-2014 and included adult patients undergoing cancer resection operations. Multivariable logistic regression analyses were utilized to assess determinants of surgical outcomes and readmissions. Of 1 016 636 eligible patients, 11.6% had psychiatric comorbidities. Patients with psychiatric disorders had greater risks of mortality (odds ratio [OR] = 1.21, 95% confidence interval [CI] = 1.18-1.25), complications (OR = 1.51, 95%CI = 1.50-1.52), extended length of stay (OR = 1.21, 95%CI = 1.20-1.22), and incurred healthcare expenditure at primary admission (OR = 1.16, 95% CI = 1.15-1.18). Psychiatric comorbidities were associated with an increased likelihood of readmission (OR = 1.13, 95%CI = 1.11-1.15). Concomitant psychiatric conditions were attributed to poor surgical outcomes and a higher risk of readmissions. Multidisciplinary care within the perioperative setting should be encouraged for patients with psychiatric disorders.

One-Year Readmission Following Undifferentiated Acute Hypercapnic Respiratory Failure

Patients with acute hypercapnic respiratory failure (AHRF) often require hospitalization and respiratory support. Early identification of patients at risk of readmission would be helpful. We evaluated 1-y readmission and mortality rates of patients admitted for undifferentiated AHRF and identified the impact of initial severity on clinically important outcomes. We retrospectively analyzed patients who presented with AHRF to the emergency department of St Michael’s Hospital in 2017. We collected data about patients’ characteristics, hospital admission, readmission and mortality one year after the index admission. We analyzed predictors of readmission and mortality and conducted a survival analysis comparing patients who did and did not receive ventilatory support. A cohort of 212 patients with AHRF who survived their hospital admission were analyzed. At one year, 150 patients (70.8%) were readmitted and 19 (9%) had died. Main diagnoses included chronic obstructive pulmonary disease (60%), congestive heart failure (36%), asthma (22%) and obesity (19%), and these categories of patients had similar 1 y readmission rates. One third had more than one coexisting chronic illness. Although comorbidities were more frequent in readmitted patients, only a history of previous hospital admissions remained associated with 1 y readmission and mortality in multivariate analysis. Need for ventilatory support at admission was not associated with higher 1 y probability of readmission or death. Undifferentiated AHRF is the presentation of multiple chronic illnesses. Patients who survive one episode of AHRF and with previous history of admission have the highest risk of readmission and death regardless of whether they receive ventilatory support during index admission.

Acute kidney injury is a common and significant complication following ileostomy formation.

The aim was to characterize the incidence and short-term prognostic value of an acute kidney injury (AKI) during the admission where an ileostomy is formed. Adults with a baseline serum creatinine measurement discharged alive after ileostomy formation from 2014 to 2016 were included.All patients had daily basic metabolicpanels and the Kidney Diseases Improving Global Outcomes criteria were used to determine the presence and severity of any AKI. Dehydration was defined by a single urine abnormality or clinical criteria combined with an objective abnormality in vitals or basic metabolic panels. Of 262 patients, 19.4% sustained an AKI (74.5%Stage I, 15.7% Stage II, 9.8% Stage III) during the index admission. Predictors of incident AKI were increasing age, male sex, higher baseline creatinine and open surgery. Patients with AKI had significantly longer length of stay and 45% had creatinine <1.0mg/dl at discharge. Of the total cohort, 11% were readmitted with dehydration and the independent predictors were AKI during the index admission, high ileostomy output, age >65 years, male sex and prior ileostomy. Of those readmitted with dehydration, 79% had AKI at readmission. Nearly 20% of patients with ileostomies develop an AKI during the index admission with almost half resolving by discharge. Patients with AKIs are at high risk for 30-day dehydration-related readmission and AKI is present in nearly 80% of those readmitted with dehydration. SinceAKI is objective, based on routine laboratory measures, and has known prognostic value it is probably amorerobust outcome than dehydration for researchers, surgeons and patients.

ARTIFICIAL INTELLIGENCE AS A PREDICTIVE TOOL TO IDENTIFY PATIENTS AT HIGH-RISK FOR NON ST ELEVATION MYOCARDIAL INFARCTION READMISSION: A RETROSPECTIVE STUDY

TOPIC: Education, Research, and Quality Improvement TYPE: Original Investigations PURPOSE: Cardiovascular death is the number one cause of death in the US. Over half of these deaths are due to myocardial infarction. While high costs and hospital admissions due to myocardial infarction are common, there is additional stress to the healthcare system with regards to readmission. In an effort to curb the cost of readmission and to improve patient outcomes, the Centers for Medicare and Medicaid Services have used 30-day hospital readmission rates as a metric to evaluate quality of care provided by hospitals across the country and to determine reimbursement. A commercially available AI machine learning tool was deployed through the electronic health record (EHR) at our institution to predict risk of readmission. The aim of our work was to assess the accuracy of an artificial intelligence-based tool in predicting which patients are at high-risk for readmission at 30 days and comparing it with a validated prediction tool, the LACE index for patients admitted with non ST elevation myocardial infarction (NSTEMI). Prospectively identifying patients hospitalized with NSTEMI who are at high risk for readmission could help prevent readmissions. METHODS: Data was collected retrospectively on patients 18 years or older with a primary diagnosis of NSTEMI and discharged from a tertiary care center from 11/26/2018 to 02/01/2020. We evaluated the effectiveness and accuracy of the proprietary AI product to identify patients at high risk for hospital readmission. This was compared with the risk predictions made by the LACE Index. The primary evaluated outcome was all-cause readmissions for HF within 30-days from the index hospitalization. The statistical method for comparison was logistic regression with LACE score as a single predictor. Sensitivities were compared after choosing a threshold that most closely matched the proprietary AI product specificity. RESULTS: A logistic regression with LACE Index score as a single predictor was used. Sensitivities were compared after choosing a threshold which resulted in the specificity most closely matching AI product specificity. After matching for specificity, the proprietary AI product had a greater sensitivity (4.7%) as compared to LACE Index score (2.7%) in detecting all-cause readmissions for NSTEMI. CONCLUSIONS: This study concludes that a proprietary AI product was a more useful predictor of readmission rates for NSTEMI as compared to prior conventional prediction tools, specifically the LACE index score. Models using AI should be implicated in clinical settings to predict readmission rates for all medical conditions to reduce readmission and overall improve patient care. We recommend that future studies should be performed to include AI models in everyday practice. CLINICAL IMPLICATIONS: Artificial intelligence can be an important asset in the future of medicine. It can be used to improve overall patient care and clinical outcomes. DISCLOSURES: No relevant relationships by Asad Cheema, source=Web Response No relevant relationships by radhika deshpande, source=Web Response No relevant relationships by Michael Gleason, source=Web Response No relevant relationships by Daniel Holtz, source=Web Response No relevant relationships by Cameron Koester, source=Web Response No relevant relationships by Abhishek Kalidas Kulkarni, source=Web Response No relevant relationships by Zurain Niaz, source=Web Response No relevant relationships by vivek prakash, source=Web Response No relevant relationships by Prashanth Singanallur, source=Web Response

Malnutrition Universal Screening Tool and Patient-Generated Subjective Global Assessment Short Form and their predictive validity in hospitalized patients.

Malnutrition screening is a first step in the nutrition care process for hospitalized patients, to identify those at risk of malnutrition and associated worse outcome, preceding further assessment and intervention. Frequently used malnutrition screening tools including the Malnutrition Universal Screening Tool (MUST) mainly screen for characteristics of malnutrition, while the Patient-Generated Subjective Global Assessment Short Form (PG-SGA SF) additionally includes risk factors for development of malnutrition, yielding a higher percentage of patients at risk. To investigate whether this translates into higher risk of worse outcome, we aimed to determine the predictive validity of MUST and PG-SGA SF for prolonged hospitalization >8 days, readmission, and mortality <6 months after hospital discharge. In this observational study, MUST was performed according to university hospital protocol. Additional screening using PG-SGA SF was performed within 24h of hospital admission (high risk: MUST≥2, PG_SGA SF≥9). Associations of MUST and PG-SGA SF with outcomes were analyzed by logistic- and Cox PH-regression. Of 430 patients analyzed (age 58±16 years, 53% male, BMI 26.9±5.5kg/m2), MUST and PG-SGA SF identified 32 and 80at high risk, respectively. One-hundred-eight patients had prolonged hospitalization, 109 were readmitted and 20 died. High risk by MUST was associated with mortality (HR=3.9; 95% CI 1.3-12.2, P=0.02), but not with other endpoints. High risk by PG-SGA SF was associated with prolonged hospitalization (OR=2.5; 95% CI 1.3-5.0, P=0.009), readmission (HR=1.9; 95% CI 1.1-3.2, P=0.03), and mortality (HR=34.8; 95% CI 4.2-289.3, P=0.001), independent of age, sex, hospital ward and previous hospitalization <6 months. In the 363/430 patients classified as low risk by MUST, high risk by PG-SGA SF was independently associated with higher risk of readmission (HR=1.9; 95% CI 1.0-3.5, P=0.04) and mortality (HR=19.5; 95% CI 2.0-189.4, P=0.01). Whereas high malnutrition risk by MUST was only associated with mortality, PG-SGA SF was associated with higher risk of prolonged hospitalization, readmission, and mortality. In patients considered as low risk by MUST, high malnutrition risk by PG-SGA SF was also predictive of worse outcome. Our findings support the use of PG-SGA SF in routine care to identify patients at risk of malnutrition and worse outcome, and enable proactive interventions.

21.1 Building Remote Bridges: Mental Health Care Coordination in the COVID-19 Pandemic

We will describe the impact of a pilot clinical program providing time-limited therapeutic and case management support for patients discharging from the hospital. The Community Advocacy, Resource, and Engagement (CARE) Team is a 30-day bridge service (with the option to extend if indicated) offering care coordination, short-term therapy and medication management, phone coaching, and community advocacy for patients identified as being high risk for readmission. A sample of 98 patients were referred to the CARE Team. Data were obtained via a semi-structured phone survey and chart review. A primary outcome measure was the number of emergency department (ED) visits and mental health hospitalizations in the 60 days post–CARE Team involvement; due to the absence of a control group, we compared these events with the 60-day period before CARE Team involvement for the same population. Other outcome measures were family satisfaction and linkage to longer term treatment. The IRB at Seattle Children’s Hospital approved this study. The average age was 13 years (range = 6-17 years) and the majority of patients lived with a caregiver, although a few were in state custody. This sample presented with a broad range of diagnoses. All CARE Team patients were linked with ongoing mental health treatment; the average patient and family satisfaction rating was 4.1 on a scale of 1 to 5. There was a 31% decrease in the number of mental health hospitalizations in the 60 days post–CARE Team compared to 60 days pre-index hospitalization (from 29 to 20). We also noticed a 52% decrease in ED visits for mental health reasons not resulting in hospitalization (from 33 to 16). The average duration of services was 6.4 weeks; the COVID-19 pandemic impacted availability of services, thus extending many patients’ involvement with the team. We saw a decrease in the number of hospitalizations and ED visits after patients discharged from the CARE Team compared to before their linkage with the team. Families were satisfied with the service. Given recent trends of increasing suicidality and hospitalizations in youth, there is an enormous need for novel clinical services like the CARE Team to provide case management and therapeutic support as patients transition out of the hospital.

Incidence and Association of Patient Safety Indicator 9 (Perioperative hemorrhage or hematoma) and clinical outcomes in Cardio-thoracic surgery hospital, Ain Shams University, Egypt

Abstract Background Patient Safety Indicators (PSIs) were developed as a tool for hospitals to identify potentially preventable complications and improve patient safety performance. Aim the study aimed to measure the association between the AHRQ patient safety indicator PSI9 (Perioperative hemorrhage or hematoma) and the clinical outcome including death, readmission within 30 days and length of stay at the cardiothoracic surgery hospital Ain Shams University. Methods exploratory prospective cohort study was conducted to follow up patients from admission till 1 month after discharge at the cardiothoracic surgery hospital who fulfills the inclusion criteria. Data were collected for 330 patients through basic information sheet and follow-up sheet. Results the incidence rate of PSI9 was 49.54 per 1000 discharges. Demographic data was not significantly associated with increased incidence of PSI9. The risk of development of PSI9 was significantly higher in patients admitted directly to ICU [relative risk (RR) =5.6]. The risk of death and readmission was higher in cases developed PSI9 than the cases without PSI9 [RR = 2.40 (0.60-9.55) and 2.43 (0.636 - 9.48) respectively]. Conclusion high incidence rate of PSI9 and the incidence is higher in male gender and 60 years old and more patients. Those patients developed PSI9 were at high risk for readmission and death. Recommendations the hospital administration should consider strategies and policies to decrease the rate of PSI9 and subsequent unfavorable clinical outcomes.

Targeted Clinical Interventions for Reducing Pediatric Readmissions.

In this interventional study, we addressed the selection and application of clinical interventions on pediatric patients identified as at risk by a predictive model for readmissions. A predictive model for readmissions was implemented, and a team of providers expanded corresponding clinical interventions for at-risk patients at a freestanding children's hospital. Interventions encompassed social determinants of health, outpatient care, medication reconciliation, inpatient and discharge planning, and postdischarge calls and/or follow-up. Statistical process control charts were used to compare readmission rates for the 3-year period preceding adoption of the model and clinical interventions with those for the 2-year period after adoption of the model and clinical interventions. Potential financial savings were estimated by using national estimates of the cost of pediatric inpatient readmissions. The 30-day all-cause readmission rates during the periods before and after predictive modeling (and corresponding 95% confidence intervals [CI]) were 12.5% (95% CI: 12.2%-12.8%) and 11.1% (95% CI: 10.8%-11.5%), respectively. More modest but similar improvements were observed for 7-day readmissions. Statistical process control charts indicated nonrandom reductions in readmissions after predictive model adoption. The national estimate of the cost of pediatric readmissions indicates an associated health care savings due to reduced 30-day readmission during the 2-year predictive modeling period at $2 673 264 (95% CI: $2 612 431-$2 735 364). A combination of predictive modeling and targeted clinical interventions to improve the management of pediatric patients at high risk for readmission was successful in reducing the rate of readmission and reducing overall health care costs. The continued prioritization of patients with potentially modifiable outcomes is key to improving patient outcomes.

ARTIFICIAL INTELLIGENCE AS A PREDICTIVE TOOL TO IDENTIFY PATIENTS WITH PULMONARY EMBOLISM AT HIGH RISK FOR 30-DAY READMISSION

TOPIC: Education, Research, and Quality Improvement TYPE: Original Investigations PURPOSE: Pulmonary Embolism (PE) is the third leading cause of cardiovascular death in the United States [1]. Venous thromboembolism (VTE) has an annual incidence of 1-2 people per 1000. The 1-year mortality can be as high as 21.6% [2]. The greatest risk factor for VTE, is a prior episode of VTE. Bikdeli et al (2019), found that though hospitalization rates for PE increased from the years 1999 to 2015, the 30-day readmission rates decreased.30-day hospital readmission rates have been used as a metric to evaluate the quality of care provided by hospitals and to determine reimbursement[5]. Risk prediction models have been developed to help identify patients at high risk of readmission with goals to reduce hospital readmissions and healthcare costs. Results with these various models have varied. Machine learning models using artificial intelligence (AI) have been developed to predict hospital readmissions. A commercially available AI/ machine learning tool was deployed through the electronic health record to predict risk of readmission. The aim of this study was to assess the accuracy of artificial intelligence-based tool in predicting which patients are at high-risk for readmission at 30 days by comparing it with a validated prediction tool, the LACE index for patients admitted with PE. METHODS: Data was collected retrospectively on patients 18 years or older with a primary diagnosis of PE and discharged from a tertiary care center from 11/26/2018 to 02/01/2020. We evaluated the effectiveness and accuracy of a proprietary AI product, to identify patients at high risk for hospital readmission and compared the effectiveness with the LACE Index. The statistical method for comparison was logistic regression with LACE score as a single predictor. Sensitivities were compared after choosing a threshold that most closely matched the proprietary AI product specificity. The primary evaluated outcome was readmissions for PE within 30-days from the index hospitalization. RESULTS: Logistic regression with LACE Index score as a single predictor was used. Sensitivities were compared after choosing a threshold which resulted in the specificity most closely matching AI product specificity. At specificity of 96.3% the AI product model had sensitivity of 19.4%. The LACE model had either 2.8 point higher (22.2%) or 5.5 point lower (13.9%) sensitivity depending on the chosen specificities of 94.7% and 97.9% respectively. We conclude that the LACE model performs roughly on par with proprietary AI tool in detecting all-cause readmission for PE. CONCLUSIONS: The LACE model performs roughly on par with proprietary AI tool in detecting 30-day readmission for PE. Proprietary AI tools may be helpful for predicting readmission rates for other diseases, in the case of PE, it did not provide a greater advantage. CLINICAL IMPLICATIONS: The LACE model performs roughly on par with proprietary AI tool in detecting all-cause readmission for PE. DISCLOSURES: No relevant relationships by Asad Cheema, source=Web Response No relevant relationships by radhika deshpande, source=Web Response No relevant relationships by Michael Gleason, source=Web Response No relevant relationships by Daniel Holtz, source=Web Response No relevant relationships by Cameron Koester, source=Web Response No relevant relationships by Abhishek Kalidas Kulkarni, source=Web Response No relevant relationships by Zurain Niaz, source=Web Response No relevant relationships by vivek prakash, source=Web Response No relevant relationships by Prashanth Singanallur, source=Web Response

361 Analyzing Risk Factors for Readmission in a Mobile Integrated Health: Community Paramedicine Population

High readmission rates are a persistent challenge for healthcare systems across the country, as they suggest poor quality of care resulting in penalties and higher costs from potentially avoidable hospital utilization. Some systems have implemented mobile integrated health - community paramedicine (MIH-CP) programs, which seek to improve patients’ transitions from hospital to home. The aim of this study is to improve MIH-CP services in a Baltimore City program by investigating risk factors associated with higher risk of readmission.

Time to readmission in psychiatric inpatients with a therapeutic leave

AimsInterventions to decrease readmissions in psychiatric patients are urgently needed. In Switzerland therapeutic leave (TL) composes a cornerstone of inpatient treatment. TL is a planned time-limited absence from the inpatient ward giving patients the opportunity to test their resilience in their usual environment. Evidence of its applicability as an intervention reducing readmissions is lacking. Therefore, our objective was to examine the association between TL and readmission risk. MethodsUsing the Kaplan-Meier curve we compared the time to readmission of 3′302 inpatients at the UPK Basel with and without TL. Cox regression was applied, integrating other covariates associated with readmission. ResultsThe Kaplan-Meier curve indicated longer cumulative survival in patients with TL. The log-rank test implied statistical significance (χ2(1) = 18.8, p < .05). The Cox regression showed a reduced hazard for patients with TL (HR = 0.735, CI 95% = [0.639, 0.846], p < .001) and for involuntarily hospitalized patients (HR = 0.760, CI 95% = [0.618, 0.934], p < .01). A higher readmission risk was found for a history of psychiatric admissions (HR = 1.005, CI 95% = [1.004, 1.005], p < .001), higher severity of symptoms at admission (HR = 1.029, CI 95% = [1.018, 1.040], p < .001), comorbidity (HR = 1.178, CI 95% [1.024, 1.355], p = .022), and a diagnosis with schizophrenia-spectrum disorders (HR = 1.401, CI 95% [1.164, 1.687], p = .001). ConclusionLinking TL with readmission risk, our results imply an easy way to improve quality of care, with possible implications for practice, policies and quality interventions. TL might be suitable to enhance recovery, reduce readmissions and health care costs. RCTs are needed for validation.

Impact of a Pharmaceutical Care Program at Discharge on Patients at High Risk of Readmission According to the Hospital Score

Background: A significant percentage of hospital readmissions within 30 days of discharge are a result of avoidable drug-related problems. Stratifying patients according to readmission risk is key to pharmaceutical intervention (PI) design strategies to improve treatment outcomes. Objective: To assess whether a pharmaceutical care (PC) program at discharge in polymedicated patients at high potentially avoidable readmission (PAR) risk, according to the HOSPITAL score, improves 30-day readmission rate (30-dRR). Methods: This prospective controlled, quasi-experimental, 11-month study included 163 chronic polymedicated patients (>5 medications) at high PAR risk according to the HOSPITAL score. We calculated the 30-dRR and number of medication variations and Medication Regimen Complexity Index-E (MRCI-E) after PI. Results were compared with a retrospective cohort of chronic patients at high PAR risk. Results: The 30-dRR was 18.4% in the intervention group and 25.6% in the control group (odds ratio [OR] = 0.66; 95% CI = 0.38 to 1.14). Total medication reduction (-1.28; 95% CI = -1.88 to -0.68), number of high-risk medications in chronic patients (-0.58; 95% CI = -0.9 to -0.26), and MRCI-E (-6.42; 95% CI = -8.07 to -4.76) were statistically significant (P < .001). The number of medications at discharge was associated with an increased readmission risk (OR = 1.07; 95% CI = 1.01 to 1.14). Conclusions: The degree of polypharmacy and patients' treatment complexity after hospital discharge significantly reduced as a result of the PC program compared with the control group. This highlights the need for patient selection and prioritization strategies for implementing PIs focused on reducing polypharmacy and preventing drug-related problems that may cause PAR.