The Utility of ICU Readmission as a Quality Indicator and the Effect of Selection*:

Quality Indicators in Adult Critical Care Medicine

Quality indicators are increasingly used in the intensive care unit (ICU) to compare and improve the quality of delivered healthcare. Numerous indicators have been developed and are related to multiple domains, most importantly patient safety, care timeliness and effectiveness, staff well-being, and patient/family-centered outcomes and satisfaction. In this review, we describe pertinent ICU quality indicators that are related to organizational structure (such as the availability of an intensivist 24/7 and the nurse-to-patient ratio), processes of care (such as ventilator care bundle), and outcomes (such as ICU-acquired infections and standardized mortality rate). We also present an example of a quality improvement project in an ICU indicating the steps taken to attain the desired changes in quality measures.

Developing a reflection and analysis tool (We-ReAlyse) for readmissions to the intensive care unit: A quality improvement project

BACKGROUND

Readmissions to the intensive care unit are associated with poorer patient outcomes and health prognoses, alongside increased lengths of stay and mortality risk. To improve quality of care and patients' safety, it is essential to understand influencing factors relevant to specific patient populations and settings. A standardized tool for systematic retrospective analysis of readmissions would help healthcare professionals understand risks and reasons affecting readmissions; however, no such tool exists.

PURPOSE

This study's purpose was to develop a tool (We-ReAlyse) to analyze readmissions to the intensive care unit from general units by reflecting on affected patients' pathways from intensive care discharge to readmission. The results will highlight case-specific causes of readmission and potential areas for departmental- and institutional-level improvements.

METHOD

A root cause analysis approach guided this quality improvement project. The tool's iterative development process included a literature search, a clinical expert panel, and a testing in January and February 2021.

RESULTS

The We-ReAlyse tool guides healthcare professionals to identify areas for quality improvement by reflecting the patient's pathway from the initial intensive care stay to readmission. Ten readmissions were analyzed by using the We-ReAlyse tool, resulting in key insights about possible root causes like the handover process, patient's care needs, the resources on the general unit and the use of different electronic healthcare record systems.

CONCLUSIONS

The We-ReAlyse tool provides a visualization/objectification of issues related to intensive care readmissions, gathering data upon which to base quality improvement interventions. Based on the information on how multi-level risk profiles and knowledge deficits contribute to readmission rates, nurses can target specific quality improvements to reduce those rates.

IMPLICATIONS FOR CLINICAL PRACTICE AND RESEARCH

With the We-ReAlyse tool, we have the opportunity to collect detailed information about ICU readmissions for an in-depth analysis. This will allow health professionals in all involved departments to discuss and either correct or cope with the identified issues. In the long term, this will allow continuous, concerted efforts to reduce and prevent ICU readmissions. To obtain more data for analysis and to further refine and simplify the tool, it may be applied to larger samples of ICU readmissions. Furthermore, to test its generalizability, the tool should be applied to patients from other departments and other hospitals. Adapting it to an electronic version would facilitate the timely and comprehensive collection of necessary information. Finally, the tool's emphasis comprises reflecting on and analyzing ICU readmissions, allowing clinicians to develop interventions targeting the identified problems. Therefore, future research in this area will require the development and evaluation of potential interventions.

Association between nursing cost and patient outcomes in intensive care units: A retrospective cohort study of Belgian hospitals

INTRODUCTION

Hospitals with better nursing resources report more favourable patient outcomes with almost no difference in cost as compared to those with worse nursing resources. The aim of this study was to assess the association between nursing cost per intensive care unit bed and patient outcomes (mortality, readmission, and length of stay).

METHODOLOGY

This was a retrospective cohort study using data collected from the intensive care units of 17 Belgian hospitals from January 01 to December 31, 2018. Hospitals were dichotomized using median annual nursing cost per bed. A total of 18,235 intensive care unit stays were included in the study with 5,664 stays in the low-cost nursing group and 12,571 in the high-cost nursing group.

RESULTS

The rate of high length of stay outliers in the intensive care unit was significantly lower in the high-cost nursing group (9.2% vs 14.4%) compared to the low-cost nursing group. Intensive care unit readmission was not significantly different in the two groups. Mortality was lower in the high-cost nursing group for intensive care unit (9.9% vs 11.3%) and hospital (13.1% vs 14.6%) mortality. The nursing cost per intensive care bed was different in the two groups, with a median [IQR] cost of 159,387€ [140,307-166,690] for the low-cost nursing group and 214,032€ [198,094-230,058] for the high-cost group. In multivariate analysis, intensive care unit mortality (OR = 0.80, 95% CI: 0.69-0.92, p < 0.0001), in-hospital mortality (OR = 0.82, 95% CI: 0.72-0.93, p < 0.0001), and high length of stay outliers (OR = 0.48, 95% CI: 0.42-0.55, p < 0.0001) were lower in the high-cost nursing group. However, there was no significant effect on intensive care readmission between the two groups (OR = 1.24, 95% CI: 0.97-1.51, p > 0.05).

CONCLUSIONS

This study found that higher-cost nursing per bed was associated with significantly lower intensive care unit and in-hospital mortality rates, as well as fewer high length of stay outliers, but had no significant effect on readmission to the intensive care unit. .

Explainable Machine-Learning Model for Prediction of In-Hospital Mortality in Septic Patients Requiring Intensive Care Unit Readmission

INTRODUCTION

Septic patients requiring intensive care unit (ICU) readmission are at high risk of mortality, but research focusing on the association of ICU readmission due to sepsis and mortality is limited. The aim of this study was to develop and validate a machine learning (ML) model for predicting in-hospital mortality in septic patients readmitted to the ICU using routinely available clinical data.

METHODS

The data used in this study were obtained from the Medical Information Mart for Intensive Care (MIMIC-IV, v1.0) database, between 2008 and 2019. The study cohort included patients with sepsis requiring ICU readmission. The data were randomly split into a training (75%) data set and a validation (25%) data set. Nine popular ML models were developed to predict mortality in septic patients readmitted to the ICU. The model with the best accuracy and area under the curve (A.C.) in the validation cohort was defined as the optimal model and was selected for further prediction studies. The SHAPELY Additive explanations (SHAP) values and Local Interpretable Model-Agnostic Explanation (LIME) methods were used to improve the interpretability of the optimal model.

RESULTS

A total of 1117 septic patients who had required ICU readmission during the study period were enrolled in the study. Of these participants, 434 (38.9%) were female, and the median (interquartile range [IQR]) age was 68.6 (58.4-79.2) years. The median (IQR) ICU interval duration was 2.60 (0.64-5.78) days. After feature selection, 31 of 47 clinical factors were ultimately chosen for use in model construction. Of the nine ML models tested, the best performance was achieved with the random forest (RF) model, with an A.C. of 0.81, an accuracy of 85% and a precision of 62% in the validation cohort. The SHAP summary analysis revealed that Glasgow Coma Scale score, urine output, blood urea nitrogen, lactate, platelet count and systolic blood pressure were the top six most important factors contributing to the RF model. Additionally, the LIME method demonstrated how the RF model works in terms of explaining risk of death prediction in septic patients requiring ICU readmission.

CONCLUSION

The ML models reported here showed a good prognostic prediction ability in septic patients requiring ICU readmission. Of the features selected, the parameters related to organ perfusion made the largest contribution to outcome prediction during ICU readmission in septic patients.

Predictive Model for ICU Readmission Based on Discharge Summaries Using Machine Learning and Natural Language Processing

Predicting ICU readmission risk will help physicians make decisions regarding discharge. We used discharge summaries to predict ICU 30-day readmission risk using text mining and machine learning (ML) with data from the Medical Information Mart for Intensive Care III (MIMIC-III). We used Natural Language Processing (NLP) and the Bag-of-Words approach on discharge summaries to build a Document-Term-Matrix with 3000 features. We compared the performance of support vector machines with the radial basis function kernel (SVM-RBF), adaptive boosting (AdaBoost), quadratic discriminant analysis (QDA), least absolute shrinkage and selection operator (LASSO), and Ridge Regression. A total of 4000 patients were used for model training and 6000 were used for validation. Using the bag-of-words determined by NLP, the area under the receiver operating characteristic (AUROC) curve was 0.71, 0.68, 0.65, 0.69, and 0.65 correspondingly for SVM-RBF, AdaBoost, QDA, LASSO, and Ridge Regression. We then used the SVM-RBF model for feature selection by incrementally adding features to the model from 1 to 3000 bag-of-words. Through this exhaustive search approach, only 825 features (words) were dominant. Using those selected features, we trained and validated all ML models. The AUROC curve was 0.74, 0.69, 0.67, 0.70, and 0.71 respectively for SVM-RBF, AdaBoost, QDA, LASSO, and Ridge Regression. Overall, this technique could predict ICU readmission relatively well.

The Impact of an Intensivist-Led Critical Care Transition Program

Objective: Evaluate the impact of a post-discharge critical care transition program (CTP) on intensive care unit (ICU) readmission, in-hospital mortality, and six-month survival.

Methods: This was a prospective observational, single-center study, with a before-after design, in a critical care department in a tertiary hospital in Northern Portugal. Critically ill patients with ICU stay > 48 h or intermediate care stay >72 h or tracheostomized patients were included in the program. Historic controls included critically ill patients admitted in the six months prior to program implementation. The follow-up visit included a medical evaluation by an intensivist and a meeting with the attending physician. The primary outcome was critical care department readmission. Secondary outcomes were mortality at hospital discharge, 28-day, and six-month mortality. The readmission rate was compared between groups. Multivariate analysis and Kaplan-Meyer survival analysis were used to evaluate survival benefits.

Results: Between September 2020 and March 2021, 132 patients were included in the CTP. The Control group included 196 patients. The intensivist’s assessment led to management change in 15.1% of patients. The CTP group had a non-significant lower readmission rate (0.8% vs. 4.1%; p=0.09). Multivariate analysis showed a benefit for the CTP regarding in-hospital, 28-day, and six-month mortality. Kaplan-Meyer survival analysis showed improved survival in the CTP group.

Conclusions: The CTP reduced, non-significantly, the readmission rate, and significantly improved in-hospital and six-month mortality. Further analyses are needed to improve inclusion criteria and better allocate human resources.

Is intensive care unit mortality a valid survival outcome measure related to critical illness?

RATIONALE

Use of death as an outcome of intensive care unit (ICU) admission may be biased by differential discharge decisions.

OBJECTIVE

To determine the validity of ICU survival status as an outcome measure of all cause case-fatality.

METHODS

A retrospective cohort of first admissions among adults to four ICUs in North Brisbane, Australia was assembled. Death in ICU (censored at discharge or 30 days) was compared with 30-day overall case-fatality.

RESULTS

The 30-day overall case-fatality was 8.1% (2436/29,939). One thousand six hundred and thirty-one deaths occurred within the ICU stay and 576 subsequent during hospital post-ICU discharge within 30-days; ICU and hospital case-fatality rates were 5.4% and 7.4%, respectively. An additional 229 patients died after hospital separation within 30 days of ICU admission of which 110 (48.0%) were transferred to another acute care hospital, 80 (34.9%) discharged home, and 39 (17.0%) transferred to an aged care/chronic care/rehabilitation facility. Patients who died after ICU discharge were older, had higher APACHE III scores, were more likely to be elective surgical patients, and were less likely to be out of state residents or managed in a tertiary referral hospital. Limiting determination of case-fatality to ICU information alone would correctly detect 95% (780/821) of all-cause mortality at day 3, 90% (1093/1213) at day 5, 75% (1524/2019) at day 15, 72% (1592/2244) at day 21, and 67% (1631/2436) at day 30 of follow-up.

CONCLUSIONS

Use of ICU case-fatality significantly underestimates the true burden and biases assessment of determinants of critical illness-related mortality in our region.

Nursing workload on the day of discharge from the intensive care unit is associated with readmission

PURPOSE

To verify whether there is an association between the Nursing Activities Score (NAS) on the day of discharge from the intensive care unit and readmission..

MATERIALS AND METHODS

A retrospective cohort study of all patients admitted to the intensive care unit of Hospital Ernesto Dornelles, Porto Alegre, Brazil, who were discharged to the ward from October 2018 to December 2019. We collected demographic and clinical variables of the patients and the Nursing Activities Scoreon the day of discharge. Patients were followed up until the day of hospital discharge or death.

RESULTS

We included 1045 patients in the final sample. One hundred eighty-eight (18.0%) patients were readmitted, in addition there were two (0.2%) unexpected deaths that occurred in the ward. The median NAS was 59.9 (50.9-67.3), which was higher in the bivariate analysis in patients who were readmitted (64.0, 55.7-71.4) than in patients who were not readmitted (58.7, 49.7-66.1) (p < 0.001). Patients with a Nursing Activities Score  ≥ 60.0 and < 60.0 had rates of readmission of 23.4% and 12.7%, respectively (p < 0.001). After multivariable adjustment, the Nursing Activities Score at discharge maintained an association with readmission. In addition, in the Cox regression, the Nursing Activities Score as a dichotomous variable was independently associated with readmission (adjusted HR 1.560; CI 1.146-2.125; p = 0.005).

CONCLUSIONS

We found that the nursing workload, assessed by the Nursing Activities Score at the time of discharge from the intensive care unit, was associated with risk of readmission..

Explainable Machine Learning on AmsterdamUMCdb for ICU Discharge Decision Support: Uniting Intensivists and Data Scientists

Unexpected ICU readmission is associated with longer length of stay and increased mortality. To prevent ICU readmission and death after ICU discharge, our team of intensivists and data scientists aimed to use AmsterdamUMCdb to develop an explainable machine learning–based real-time bedside decision support tool.

Predicting Readmission to Intensive Care After Cardiac Surgery Within Index Hospitalization: A Systematic Review

Readmission to the cardiac intensive care unit after cardiac surgery has significant implications for both patients and healthcare providers. Identifying patients at risk of readmission potentially could improve outcomes. The objective of this systematic review was to identify risk factors and clinical prediction models for readmission within a single hospitalization to intensive care after cardiac surgery. PubMed, MEDLINE, and EMBASE databases were searched to identify candidate articles. Only studies that used multivariate analyses to identify independent predictors were included. There were 25 studies and five risk prediction models identified. The overall rate of readmission pooled across the included studies was 4.9%. In all 25 studies, in-hospital mortality and duration of hospital stay were higher in patients who experienced readmission. Recurring predictors for readmission were preoperative renal failure, age >70, diabetes, chronic obstructive pulmonary disease, preoperative left ventricular ejection fraction <30%, type and urgency of surgery, prolonged cardiopulmonary bypass time, prolonged postoperative ventilation, postoperative anemia, and neurologic dysfunction. The majority of readmissions occurred due to respiratory and cardiac complications. Four models were identified for predicting readmission, with one external validation study. As all models developed to date had limitations, further work on larger datasets is required to develop clinically useful models to identify patients at risk of readmission to the cardiac intensive care unit after cardiac surgery.

Early prediction of ICU readmissions using classification algorithms

CONTEXT

Determining which patients are ready for discharge from an Intensive Care Unit (ICU) presents a huge challenge, as ICU readmissions are associated with several negative outcomes such as increased mortality, length of stay, and cost compared to those patients who are not readmitted during their hospital stay. For these reasons, enhancing risk stratification in order to identify patients at high risk of clinical deterioration might benefit and improve the outcomes of critically ill hospitalized patients. Existing work on predicting ICU readmissions relies on information available at the time of discharge, however, in order to be more useful and to prevent complications, predictions need to be made earlier.

GOALS

In this work, we investigate the hypothesis that the basal characteristics and information collected at the time of the patient's admission can enable accurate predictions of ICU readmission.

MATERIALS AND METHODS

We analyzed an anonymized dataset of 11,805 adult patients from three ICUs in a Brazilian university hospital. After excluding 1879 patients who died during their first ICU admission, our final dataset contained 9,926 patients. Of these, 658 patients (6.6%) had been readmitted to the ICU. The original dataset had 185 attributes, including demographics, length of stay prior to ICU admission, comorbidities, severity indexes, interventions, organ support care during ICU stay and laboratory results. The problem of predicting ICU readmissions was modeled as a binary classification task. We tested eight classification algorithms (including Bayesian algorithms, decision trees, rule-based, and ensemble methods) over different sets of attributes and evaluated their results based on six metrics.

RESULTS

Predictions made solely based on the attributes collected at the admission are highly accurate. Their quality in terms of prediction is no different from predictions made using the complete set of attributes for our dataset and for a subset of attributes selected by a feature selection method. Furthermore, our AUROC score of 0.91 (95% CI [0.89,0.92]) is higher than existing results published in the literature for other datasets.

DISCUSSION AND CONCLUSION

The results confirm our hypothesis. Our findings suggest that early markers can be used to anticipate patients at high risk of clinical deterioration after ICU discharge.

Effects of Lean Healthcare on Patient Flow: A Systematic Review

OBJECTIVES

To assess the effects of lean healthcare (LH) on patient flow in ambulatory care and determine whether waiting time and length of stay (LOS) decrease after LH interventions.

METHODS

A systematic review was performed with close adherence to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). We searched for studies of healthcare organizations applying LH interventions within ambulatory care published between 2002 and 2018. Six databases and grey literature sources were used. Two reviewers independently screened and assessed each study. When consensus was difficult to reach, a third reviewer intervened. Finally, a summary of findings was generated.

RESULTS

Out of 5627 studies, 40 were included. Regarding LOS for all patients, 19 out of 22 studies reported a decrease. LOS for discharged patients decreased in 11 out of 13 studies, whereas LOS for admitted patients was reduced in 6 out of 7 studies. Waiting time for patients before seeing a healthcare professional decreased in 24 out of 26 studies. Waiting time to treatment and waiting time for appointments were minimized in 4 and 2 studies, respectively. Patients who left without being seen by a doctor decreased in 9 out of 12 studies. Finally, patient and staff satisfaction were measured in 8 and 2 studies, respectively, with each reporting improvements.

CONCLUSIONS

According to our findings, LH helped to reduce waiting time and LOS in ambulatory care, mainly owing to its focus on identifying and minimizing non-value added (NVA) activities. Nevertheless, evidence of the impact of LH on patient/staff satisfaction and the translation of the obtained benefits into savings is scarce among studies.

Readmissions to General ICUs in a Geographic Area of Poland Are Seemingly Associated with Better Outcomes

BACKGROUND

Various factors can contribute to high mortality rates in intensive care units (ICUs). Here, we intended to define a population of patients readmitted to general ICUs in Poland and to identify independent predictors of ICU readmission.

METHODS

Data derived from adult ICU admissions from the Silesian region of Poland were analyzed. First-time ICU readmissions (≤30 days from ICU discharge after index admissions) were compared with first-time ICU admissions. Pre-admission and admission variables that independently influenced the need for ICU readmission were identified.

RESULTS

Among the 21,495 ICU admissions, 839 were first-time readmissions (3.9%). Patients readmitted to the ICU had lower mean APACHE II (21.2 ± 8.0 vs. 23.2 ± 8.8, p < 0.001) and TISS-28 scores (33.7 ± 7.4 vs. 35.2 ± 7.8, p < 0.001) in the initial 24 h following ICU admission, compared to first-time admissions. ICU readmissions were associated with lower mortality vs. first-time admissions (39.2% vs. 44.3%, p = 0.004). Independent predictors for ICU readmission included the admission from a surgical ward (among admission sources), chronic respiratory failure, cachexia, previous stroke, chronic neurological diseases (among co-morbidities), and multiple trauma or infection (among primary reasons for ICU admission).

CONCLUSIONS

High mortality associated with first-time ICU admissions is associated with a lower mortality rate during ICU readmissions.

Variation in Case-Mix Adjusted Unplanned Pediatric Cardiac ICU Readmission Rates

OBJECTIVES

To identify modifiable factors leading to unplanned readmission and characterize differences in adjusted unplanned readmission rates across hospitals.

DESIGN

Retrospective cohort study using prospectively collected clinical registry data SETTING:: Pediatric Cardiac Critical Care Consortium clinical registry.

PATIENTS

Patients admitted to a pediatric cardiac ICU at Pediatric Cardiac Critical Care Consortium hospitals.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We examined pediatric cardiac ICU encounters in the Pediatric Cardiac Critical Care Consortium registry from October 2013 to March 2016. The primary outcomes were early (< 48 hr from pediatric cardiac ICU transfer) and late (2-7 d) unplanned readmission. Generalized logit models identified independent predictors of unplanned readmission. We then calculated observed-to-expected ratios of unplanned readmission and identified higher-than or lower-than-expected unplanned readmission rates for those with an observed-to-expected ratios greater than or less than 1, respectively, and a 95% CI that did not cross 1. Of 11,301 pediatric cardiac ICU encounters (16 hospitals), 62% were surgical, and 18% were neonates. There were 175 (1.6%) early unplanned readmission, and 300 (2.7%) late unplanned readmission, most commonly for respiratory (31%), or cardiac (28%) indications. In multivariable analysis, unique modifiable factors were associated with unplanned readmission. Although shorter time between discontinuation of vasoactive infusions and pediatric cardiac ICU transfer was associated with early unplanned readmission, nighttime discharge was independently associated with a greater likelihood of late unplanned readmission. Two hospitals had lower-than-expected unplanned readmission in both the early and late categories, whereas two other hospitals were higher-than-expected in both.

CONCLUSIONS

This analysis demonstrated time from discontinuation of critical care therapies to pediatric cardiac ICU transfer as a significant, modifiable predictor of unplanned readmission. We identified two hospitals with lower-than-expected adjusted rates of both early and late unplanned readmission, suggesting that their systems are well designed to prevent unplanned readmission. This offers the possibility of disseminating best practices to other hospitals through collaborative learning.