Unplanned readmissions within 30 days after discharge: improving quality through easy prediction

Casalini, Francesca; Salvetti, S.; Memmini, Silvia; Lucaccini, Elena; Massimetti, Gabriele; Lopalco, Pietro Luigi; Privitera, Gaetano Pierpaolo

doi:10.1093/intqhc/mzx011

Cited by 17 publications

(16 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…According to our estimates, women were at higher risk of readmission and this finding was in contrast to other studies using single-center settings, where no significant effect of gender was identified [9,37,38,39]. Moreover, according to a systematic review on risk prediction models for readmission, the gender variable in most studies “did not contribute enough to be included in the final model” [6].…”

Section: Discussioncontrasting

confidence: 84%

“…An American study based on data from a university hospital in California analyzed 10,359 admissions (2006–2008) discharged from general medicine service with RaR of 17.0% [37]. Another recent Italian study identified the rate of unplanned readmissions as 11.6% at Pisa University Hospital (data for 5388 admissions in 2012) [38]. A prospective cohort study from a single Belgian university hospital (data for years 2011–2012) reported a RaR of 18.6% for emergency department patients aged >75 years [39].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Prevalence, Reasons, and Predisposing Factors Associated with 30-day Hospital Readmissions in Poland

Kryś

Łyszczarz

Wyszkowska

et al. 2019

IJERPH

View full text Add to dashboard Cite

There is a growing interest in quality issues associated with hospital care, with readmissions (rehospitalizations) being one of the main areas of interest. Retrospective data from a 914-bed university hospital in Bydgoszcz, Poland, was used to identify 30-day readmissions in 2015. We developed a catalogue of reasons for rehospitalization and differentiated between planned and unplanned readmissions, as well as those related and unrelated to index (initial) hospitalization. Multilevel logistic regression was used to determine factors associated with readmission risk. A total of 12.5% of patients were readmitted within 30 days of being discharged. The highest readmission rates were identified in pediatric, transplantation, and urology patients. The highest share of readmissions was due to the specific nature of a disease and its routine treatment practice. Almost two-thirds of readmission cases were classified as unplanned and related to the index hospitalization. The following characteristics were associated with a higher risk of rehospitalization: female gender, residing >35 km from the hospital, longer than average and very short stays at index admission, higher comorbidity score, and admission to a high-volume hospital sector. Due to the importance of quality issues in health policy, the topic should be further pursued to identify evidence-based practices that would improve hospitals’ performance.

show abstract

Section: Discussioncontrasting

confidence: 84%

Section: Discussionmentioning

confidence: 99%

Prevalence, Reasons, and Predisposing Factors Associated with 30-day Hospital Readmissions in Poland

Kryś

Łyszczarz

Wyszkowska

et al. 2019

IJERPH

View full text Add to dashboard Cite

show abstract

“…Interestingly, the underlying deep learning logic and its findings are in line with the existing clinical literature. Prior research has found that the presence of comorbidities, such as diabetes, heart failure, renal failure, and pneumonia, are the main risk factors resulting in unplanned readmissions [41,42]. These disorders are shown to have strong correlations with abnormal features identified by our model [17].…”

Section: Discussionmentioning

confidence: 77%

Analysis and prediction of unplanned intensive care unit readmission using recurrent neural networks with long short-term memory

et al. 2019

View full text Add to dashboard Cite

Background Unplanned readmission of a hospitalized patient is an indicator of patients’ exposure to risk and an avoidable waste of medical resources. In addition to hospital readmission, intensive care unit (ICU) readmission brings further financial risk, along with morbidity and mortality risks. Identification of high-risk patients who are likely to be readmitted can provide significant benefits for both patients and medical providers. The emergence of machine learning solutions to detect hidden patterns in complex, multi-dimensional datasets provides unparalleled opportunities for developing an efficient discharge decision-making support system for physicians and ICU specialists. Methods and findings We used supervised machine learning approaches for ICU readmission prediction. We used machine learning methods on comprehensive, longitudinal clinical data from the MIMIC-III to predict the ICU readmission of patients within 30 days of their discharge. We incorporate multiple types of features including chart events, demographic, and ICD-9 embeddings. We have utilized recent machine learning techniques such as Recurrent Neural Networks (RNN) with Long Short-Term Memory (LSTM), by this we have been able to incorporate the multivariate features of EHRs and capture sudden fluctuations in chart event features (e.g. glucose and heart rate). We show that our LSTM-based solution can better capture high volatility and unstable status in ICU patients, an important factor in ICU readmission. Our machine learning models identify ICU readmissions at a higher sensitivity rate of 0.742 (95% CI, 0.718–0.766) and an improved Area Under the Curve of 0.791 (95% CI, 0.782–0.800) compared with traditional methods. We perform in-depth deep learning performance analysis, as well as the analysis of each feature contribution to the predictive model. Conclusion Our manuscript highlights the ability of machine learning models to improve our ICU decision-making accuracy and is a real-world example of precision medicine in hospitals. These data-driven solutions hold the potential for substantial clinical impact by augmenting clinical decision-making for physicians and ICU specialists. We anticipate that machine learning models will improve patient counseling, hospital administration, allocation of healthcare resources and ultimately individualized clinical care.

show abstract

“…Second, over recent years, the lengths of hospital admissions have reduced [35] and so patients are increasingly going home with ongoing care needs such as wound care management or medication monitoring. Collecting data during the initial post-discharge period when patients may still be unwell and/or particularly vulnerable to hospital readmissions [36] may impact our CRF return rates. Third, retention and attrition will need to be monitored as follow-ups are being conducted up to 90-days post-discharge.…”

Section: Post-discharge Follow-upsmentioning

confidence: 99%

Improving patient experience and safety at transitions of care through the Your Care Needs You (YCNY) intervention: a study protocol for a cluster randomised controlled feasibility trial

Baxter¹,

Murray²,

O’Hara

et al. 2020

Pilot Feasibility Stud

View full text Add to dashboard Cite

Background: Patients, particularly older people, often experience safety issues when transitioning from hospital to home. Although the evidence is currently equivocal as to how we can improve this transition of care, interventions that support patient involvement may be more effective. The 'Your Care Needs You' (YCNY) intervention supports patients to 'know more' and 'do more' whilst in hospital in order that they better understand their health condition and medications, maintain their daily activities, and can seek help at home if required. The intervention aims to reduce emergency hospital readmissions and improve safety and experience during the transition to home. Methods: As part of the Partners At Care Transitions (PACT) programme of research, a multi-centred cluster randomised controlled trial (cRCT) will be conducted to explore the feasibility of the YCNY intervention and trial methodology. Data will be used to refine the intervention and develop a protocol for a definitive cRCT. Ten acute hospital wards (the clusters) from varying medical specialties including older peoples' medicine, trauma and orthopaedics, cardiology, intermediate care, and stroke will be randomised to deliver YCNY or usual care on a 3:2 basis. Up to 200 patients aged 75 years and over and discharged to their own homes will be recruited to the study. Patients will complete follow-up questionnaires at 5-, 30-, and 90-days post-discharge and readmission data up to 90-days post-discharge will be extracted from their medical records. Study outcomes will include measures of feasibility (e.g. screening, recruitment, and retention data) and processes required to collect routine data at a patient and ward level. In addition, interviews and observations involving up to 24 patients/carers and 28 staff will be conducted to qualitatively assess the acceptability, usefulness, and feasibility of the intervention and implementation package to patients and staff. A separate sub-study will be conducted to explore how accurately primary outcome data (30-day emergency hospital readmissions) can be gathered for the definitive cRCT.

show abstract

Unplanned readmissions within 30 days after discharge: improving quality through easy prediction

Cited by 17 publications

References 8 publications

Prevalence, Reasons, and Predisposing Factors Associated with 30-day Hospital Readmissions in Poland

Prevalence, Reasons, and Predisposing Factors Associated with 30-day Hospital Readmissions in Poland

Analysis and prediction of unplanned intensive care unit readmission using recurrent neural networks with long short-term memory

Improving patient experience and safety at transitions of care through the Your Care Needs You (YCNY) intervention: a study protocol for a cluster randomised controlled feasibility trial

Contact Info

Product

Resources

About