Improved outcomes following the implementation of a decompensated cirrhosis discharge bundle

IntroductionMortality from liver disease is increasing and management of decompensated cirrhosis (DC) is inconsistent across the UK. Patients with DC have complex medical needs when discharged from hospital and early readmissions are common. Our aims were: (1) to develop a Decompensated Cirrhosis Discharge Bundle (DCDB) to optimise ongoing care and (2) evaluate the impact of the DCDB.MethodsA baseline review of the management of patients with DC was conducted in Newcastle in 2017. The DCCB was developed and implemented in 2018. Impact of the DCDB was evaluated in two cycles, first a paper version (November 2018–October 2019) and then an electronic version (November 2020–March 2021). Key clinical data were collected from the time of discharge.ResultsOverall, 192 patients (62% male; median age 55; median model for end-stage liver disease 17; 72% alcohol related) were reviewed in three cycles. At baseline, management was suboptimal, particularly ascites/diuretic management and provision of follow-up for alcohol misuse and 12% of patients had a potentially avoidable readmission within 30 days. After DCDB introduction, care improved across most domains, particularly electrolyte monitoring (p=0.012) and provision of community alcohol follow-up (p=0.026). Potentially preventable readmissions fell to 5% (p=0.055).ConclusionsUse of a care bundle for patients with DC can standardise care and improve patient management. If used more widely this could improve outcomes and reduce variability in care for patients with DC.

A Case Study on Hospital Readmission Prediction Using Deep Learning Algorithms on EHRs

In this paper, focus is made on predicting the patients who are going to be re-admitted back in the hospital before discharge using latest deep-learning algorithms is applied on the electronic health records of patients which is a time-series data. To begin with the study of the data and its analysis this project deployed the conventional supervised ML algorithms like the Logistic Regression, Naïve Bayes, Random Forest and SVM and compared their performances on different portion sizes of dataset. The final model built uses deep-learning architectures such as RNN and LSTM to improve the prediction results taking advantage of the time series data. Another feature added has been of low dimensional descriptions of medical concepts as the input to the model. Ultimately, this work tests, validates, and explains the developed system using the MIMIC-III dataset, which contains around 38000 patient’s information and about 61,155 patient’s data who admitted in ICU, duration of 10 years. The support from this exhaustive dataset is used to train the models that provide healthcare workers with proper information regarding their discharge and readmission in hospitals. These ML and deep learning models are used to know about the patient who is getting to be readmitted in the ICU before his discharge will help the hospital to allocate resources properly and also reduce the financial risk of patients. In order to reduce ICU readmission that can be avoided, hospitals have to be able to recognize patients who have a higher risk of ICU readmission. Those patients can then continue to stay in the ICU so that they will not have the risk of getting admit back to the hospital. Also, the resources of hospitals that were required for avoidable readmission can be re-allocated to more critical areas in the hospital that need them. A more effective model of predicting readmission system can play an important role in helping hospitals and ICU doctors to find the patients who are going to be readmitted before discharge. To build this system here we use different ML and deep-learning algorithms. Predictive models based on huge amounts of data are made to predict the patients who are going to be admitted back in the hospital after discharge.

Development of a risk prediction model of potentially avoidable readmission for patients hospitalised with community-acquired pneumonia: study protocol and population

Introduction30-day readmission rate is considered an adverse outcome reflecting suboptimal quality of care during index hospitalisation for community-acquired pneumonia (CAP). However, potentially avoidable readmission would be a more relevant metric than all-cause readmission for tracking quality of hospital care for CAP. The objectives of this study are (1) to estimate potentially avoidable 30-day readmission rate and (2) to develop a risk prediction model intended to identify potentially avoidable readmissions for CAP.Methods and analysisThe study population consists of consecutive patients admitted in two hospitals from the community or nursing home setting with pneumonia. To qualify for inclusion, patients must have a primary or secondary discharge diagnosis code of pneumonia. Data sources include routinely collected administrative claims data as part of diagnosis-related group prospective payment system and structured chart reviews. The main outcome measure is potentially avoidable readmission within 30 days of discharge from index hospitalisation. The likelihood that a readmission is potentially avoidable will be quantified using latent class analysis based on independent structured reviews performed by four panellists. We will use a two-stage approach to develop a claims data-based model intended to identify potentially avoidable readmissions. The first stage implies deriving a clinical model based on data collected through retrospective chart review only. In the second stage, the predictors comprising the medical record model will be translated into International Classification of Diseases, 10th revision discharge diagnosis codes in order to obtain a claim data-based risk model.The study sample consists of 1150 hospital stays with a diagnosis of CAP. 30-day index hospital readmission rate is 17.5%.Ethics and disseminationThe protocol was reviewed by the Comité de Protection des Personnes Sud Est V (IRB#6705). Efforts will be made to release the primary study results within 6 months of data collection completion.Trial registration numberClinicalTrials.gov Registry (NCT02833259).

Original and refitted HOSPITAL scores as predictors of 30-day potentially avoidable hospital readmissions: retrospective cohort study (Preprint)

BACKGROUND Hospital readmissions are associated with several negative health outcomes and higher hospital costs. The HOSPITAL score is one of the tools developed to identify patients at high risk of hospital readmission, but its predictive capacity in more heterogeneous populations involving different diagnoses and clinical contexts is poorly understood. OBJECTIVE The aim of this study was to propose a refitted HOSPITAL score to predict the risk of potentially avoidable readmission in 30 days and compare the predictive capacity of the original and refitted HOSPITAL score. METHODS Retrospective cohort study was carried out in a tertiary university hospital with patients over the age of 18 years. We developed a refitted HOSPITAL score with the same definitions and predictive variables included in the original HOSPITAL score and compared the predictive capacity of both. The receiver operating characteristic was constructed by comparing the performance risk forecasting tools measuring the area under the curve (AUC). RESULTS Of the 47,464 patients 50.9% were over 60 years and 58.4% were male. The frequency of 30-day potentially avoidable readmission is 7.70% (3638). The accuracy of HOSPITAL score in readmission was AUC: 0.733 (CI 95%: 0.718, 0.748) and the accuracy of HOSPITAL score refitted was AUC: 0.7401 (CI 95%: 0.7256, 0.7547). The frequency of 60, 90, 180, and 365-days readmissions ranged from 10.60% (5,033) to 18.30% (8693). Discussion: Readmission prediction tools have been developed in recent years, but its predictive capacity in more population with different diagnoses is poorly understood. CONCLUSIONS The refitted HOSPITAL score have similar discrimination to predict 30-day potentially avoidable readmission, in patients with different diagnoses. In this sense, our study expands and reinforces the usefulness of the HOSPITAL score as a tool that can be used as part of intervention strategies to reduce the rate of hospital readmission.

Assessing 30-day avoidable readmission rates: Is it an appropriate tool to manage emergency department quality of care?

Objective: Quality indicators, based on administrative data, are being increasingly used to assess avoidable hospital readmission rates. Their potential to identify areas for improvement at low cost is attractive, but their performance in emergency departments (EDs) has been criticised.Methods: Hospital readmissions were categorised as potentially avoidable or non-avoidable, by a computerised algorithm (SQLape®, version 2016 - Striving for Quality Level and analysing of patient expenditures). Half-yearly rates were reported between July 2015 and June 2016. Two senior physicians conducted a medical record review on 100 randomly selected cases from an ED, flagged as potentially avoidable readmissions (PAR). Results were then discussed with the algorithm’s designer.Results: The algorithm screened 2,182 eligible emergency visits - 105 cases (4.8%), were deemed potentially avoidable by the algorithm. Among 100 randomly selected cases, nine exclusions were due to coding issues and four due to false positives. Overall (N = 87), 20/87 (23%) of readmissions were directly related to sole emergency care, 31/87 (36%) related to healthcare providers other than the ED, and 23/87 (26%) were of mixed provision, while 13/87 (15%) were attributed to the course of the disease.Conclusions: The study confirms the need for a better understanding of the algorithm’s measurement and of its reported results. Careful interpretation is required before a sound conclusion can be made. Indeed, it is apparent that the 30-day PAR quality indicator rate reflects a wider parameter of care than hospitals alone, who understandably tend to concentrate on their own, direct liability of care. In particular the 30-day PAR quality indicator is not well-suited to evaluate ED performance.

Risk Factors and Patterns of Potentially Avoidable Readmission in Patients With Cancer

Purpose: Patients with cancer are particularly at risk for readmission within 30-days after discharge. To identify the patients who might benefit from more-intensive discharge interventions, we identified the risk factors associated with 30-day potentially avoidable readmissions. Methods and Materials: We included all consecutive discharges from the oncology division of an academic tertiary medical center in Boston, Massachusetts, between July 1, 2009, and June 30, 2010. Potentially avoidable 30-day readmissions to the index hospital and two other hospitals within its network were identified. We performed a multivariable logistic regression in which the final model included variables found in bivariable testing to be significantly associated with the outcome. Results: Among the 2,916 patients discharged during the study period, 1,086 (37.3%) were readmitted within 30 days. Of these, 341 (31.4% of all readmissions, 11.7% of all discharges) were identified as potentially avoidable. In the multivariable analysis, the following patient factors were associated with a significantly higher risk of a potentially avoidable readmission: total number of medications at discharge, liver disease, last sodium level, and last hemoglobin level before discharge. In addition, potentially avoidable readmissions occurred significantly earlier than unavoidable readmissions (median, 10 v 13 days; P < .001). Conclusion: Almost 40% of patients with cancer had a 30-day readmission, and almost one third of these were deemed potentially avoidable, and several risk factors for this were identified. Interventions at discharge may be prioritized to patients with these risk factors.