Risk Assessment of Acute, All-Cause 30-Day Readmission in Patients Aged 65+: a Nationwide, Register-Based Cohort Study

Multimodal Machine Learning for Prediction of 30-Day Readmission Risk in Elderly Population

BACKGROUND

Readmission within 30 days is a prevalent issue among elderly patients, linked to unfavorable health outcomes. Our objective was to develop and validate multimodal machine learning models for predicting 30-day readmission risk in elderly patients discharged from internal medicine departments.

METHODS

This was a retrospective cohort study which included elderly patients aged 75 or older, who were hospitalized at the Hadassah Medical Center internal medicine departments between 2014 and 2020. Three machine learning algorithms were developed and employed to predict 30-day readmission risk. The primary measures were predictive model performance scores, specifically area under the receiver operator curve (AUROC), and average precision.

RESULTS

This study included 19,569 admissions. Of them, 3258 (16.65%) resulted in 30-day readmission. Our 3 proposed models demonstrated high accuracy and precision on an unseen test set, with AUROC values of 0.87, 0.89, and 0.93, respectively, and average precision values of 0.76, 0.78, and 0.81. Feature importance analysis revealed that the number of admissions in the past year, history of 30-day readmission, Charlson score, and admission length were the most influential variables. Notably, the natural language processing score, representing the probability of readmission according to a textual-based model trained on social workers' assessment letters during hospitalization, ranked among the top 10 contributing factors.

CONCLUSIONS

Leveraging multimodal machine learning offers a promising strategy for identifying elderly patients who are at high risk for 30-day readmission. By identifying these patients, machine learning models may facilitate the effective execution of preventive actions to reduce avoidable readmission incidents.

Nutritional risk predicts readmission within 30 and 180 days after discharge among older adult patients across a broad spectrum of diagnoses

BACKGROUND AND AIMS

Hospital readmissions can have negative consequences for older adult patients, their relatives, the hospital, and society. Previous studies indicate that older adult patients who are at nutritional risk during hospital admission are at higher risk of readmission. There is a lack of studies investigating this relationship across different older adult patient groups while using recommended instruments and adjusting for relevant confounders. Thus, the aim of the present study was to investigate whether nutritional status according to the Nutrition Risk Screening 2002 during hospitalization predicted readmission among older adult patients within 30 and 180 days across a broad spectrum of wards and diagnoses when adjusting for age, sex, length-of-stay, diagnosis, and discharge destination.

MATERIALS AND METHODS

The present study is a retrospective cohort study based on registry data and included 21,807 older adult patients (≥65 years) hospitalized during a 5-year period. In order to investigate the relationship between nutritional risk and readmission, hierarchical logistic regression analyses with readmission within 30 days (n = 8371) and 180 days (n = 7981) as the dependent variable were performed.

RESULTS

Older adult patients at nutritional risk during the index admission were 1.44 times more likely to be readmitted within 30 days after discharge (P < 0.001), and 1.47 times more likely to be readmitted within 180 days after discharge (P < 0.001), compared to older adult patients who were not at nutritional risk during index admission when adjusting for age, sex, discharge destination, diagnosis group, and length-of-stay.

CONCLUSIONS

Our results highlight the importance of focusing on nutritional status in older adults as a factor in the prevention of readmissions, including ensuring that practices, resources, and guidelines support appropriate screening procedures. Because nutritional risk predicts readmission both in a 30-days and 180-days perspective, the results point to the importance of ensuring follow-up on the screening result, both in the hospital context and after discharge.

Multimorbidity and prior falls correlate with risk of 30-day hospital readmission in aged 80+: A prospective cohort study

BACKGROUND/PURPOSE

Thirty-day hospital readmission rate significantly raised with advanced age. The performance of existing predictive models for readmission risk remained uncertain in the oldest population. We aimed to examine the effect of geriatric conditions and multimorbidity on readmission risk among older adults aged 80 and over.

METHODS

This prospective cohort study enrolled patients aged 80 and older discharged from a geriatric ward at a tertiary hospital, with phone follow-up for 12 months. Demographics, multimorbidity, and geriatric conditions were assessed before hospital discharge. Logistic regression models were conducted to analyse risk factors for 30-day readmission.

RESULTS

Patients readmitted had higher Charlson comorbidity index scores, and were more likely to have falls, frailty, and longer hospital stay, compared to those without 30-day readmission. Multivariate analysis revealed that higher Charlson comorbidity index score was associated with readmission risk. Older patients with a fall history within 12 months had a near 4-fold increase in readmission risk. Severe frailty status before index admission was associated with a higher 30-day readmission risk. Functional status at discharge was not associated with readmission risk.

CONCLUSION

In addition to multimorbidity, history of falls and frailty were associated with higher hospital readmission risk in the oldest.

Machine learning for predicting readmission risk among the frail: Explainable AI for healthcare

Healthcare costs due to unplanned readmissions are high and negatively affect health and wellness of patients. Hospital readmission is an undesirable outcome for elderly patients. Here, we present readmission risk prediction using five machine learning approaches for predicting 30-day unplanned readmission for elderly patients (age ≥ 50 years). We use a comprehensive and curated set of variables that include frailty, comorbidities, high-risk medications, demographics, hospital, and insurance utilization to build these models. We conduct a large-scale study with electronic health record (her) data with over 145,000 observations from 76,000 patients. Findings indicate that the category boost (CatBoost) model outperforms other models with a mean area under the curve (AUC) of 0.79. We find that prior readmissions, discharge to a rehabilitation facility, length of stay, comorbidities, and frailty indicators were all strong predictors of 30-day readmission. We present in-depth insights using Shapley additive explanations (SHAP), the state of the art in machine learning explainability.

A simple index predicting 30-day readmissions in acutely hospitalized patients

BACKGROUND

There are various models attempting to predict 30-day readmissions of acutely admitted internal medicine patients. However, it is uncertain how to create a parsimonious index that has equivalent predictive ability and can be extrapolated to other settings.

METHODS

We developed a regression equation to predict 30-day readmissions from all acute hospitalizations in internal medicine departments in a regional hospital in 2015-2016 and validated the model in 2019. The independent (predictor) variables were age, past hospitalizations, admission laboratory test results, length of stay in hospital and discharge diagnoses. We compared the predictive value of a logistic regression model and index that included discharge diagnoses and admission laboratory test results with one that included only age, past hospitalizations, and hospital length of stay.

RESULTS

Readmission rates were associated with age, time since last hospitalization, number of previous hospitalizations, and length of stay, as well as with a diagnosis of chronic obstructive lung disease and congestive heart failure and several laboratory data. Logistic regressions of the independent variables for 30-day readmission rates were similar in 2015-2016 and 2019. An index was derived from number of previous admissions to hospitals, time since last admission, age, and length of stay. In 2019, for every unit of the index, the odds of readmission increased by 1.33 (95% CI- 1.30-1.37), and ranged from 2.1% to 37.1%. Addition of discharge diagnoses and laboratory variables did not significantly improve the risk differentiation of the index. The c-statistic for the final parsimonious model was 0.704.

CONCLUSIONS

An index derived from the number of previous hospital admissions, days since last admission, age, and length of stay in days differentiated between the risks of readmission within 30 days without the need for discharge diagnosis and laboratory variables.

Assess the Performance and Cost-Effectiveness of LACE and HOSPITAL Re-Admission Prediction Models as a Risk Management Tool for Home Care Patients: An Evaluation Study of a Medical Center Affiliated Home Care Unit in Taiwan

The LACE index and HOSPITAL score models are the two most commonly used prediction models identifying patients at high risk of readmission with limited information for home care patients. This study compares the effectiveness of these two models in predicting 30-day readmission following acute hospitalization of such patients in Taiwan. A cohort of 57 home care patients were enrolled and followed-up for one year. We compared calibration, discrimination (area under the receiver operating curve, AUC), and net reclassification improvement (NRI) to identify patients at risk of 30-day readmission for both models. Moreover, the cost-effectiveness of the models was evaluated using microsimulation analysis. A total of 22 readmissions occurred after 87 acute hospitalizations during the study period (readmission rate = 25.2%). While the LACE score had poor discrimination (AUC = 0.598, 95% confidence interval (CI) = 0.488–0.702), the HOSPITAL score achieved helpful discrimination (AUC = 0.691, 95% CI = 0.582–0.785). Moreover, the HOSPITAL score had improved the risk prediction in 38.3% of the patients, compared with the LACE index (NRI = 0.383, 95% CI = 0.068–0.697, p = 0.017). Both prediction models effectively reduced readmission rates compared to an attending physician’s model (readmission rate reduction: LACE, 39.2%; HOSPITAL, 43.4%; physician, 10.1%; p < 0.001). The HOSPITAL score provides a better prediction of readmission and has potential as a risk management tool for home care patients.

Medication-Related Factors and Hospital Readmission in Older Adults with Chronic Kidney Disease

This study aimed to examine the association between medication-related factors and risk of hospital readmission in older patients with chronic kidney disease (CKD). A retrospective analysis was conducted targeting older CKD (n = 204) patients admitted to an Australian hospital. Medication appropriateness (Medication Appropriateness Index; MAI), medication regimen complexity (number of medications and Medication Regimen Complexity Index; MRCI) and use of selected medication classes were exposure variables. Outcomes were occurrence of readmission within 30 and 90 days, and time to readmission within 90 days. Logistic and Cox hazards regression were used to identify factors associated with readmission. Overall, 50 patients (24%) were readmitted within 30 days, while 81 (40%) were readmitted within 90 days. Mean time to readmission within 90 days was 66 (SD 34) days. Medication appropriateness and regimen complexity were not independently associated with 30- or 90-day hospital readmissions in older adults with CKD, whereas use of renin‒angiotensin blockers was associated with reduced occurrence of 30-day (adjusted OR 0.39; 95% CI 0.19⁻0.79) and 90-day readmissions (adjusted OR 0.45; 95% CI 0.24⁻0.84) and longer time to readmission within 90 days (adjusted HR 0.52; 95% CI 0.33⁻0.83). This finding highlights the importance of considering the potential benefits of individual medications during medication review in older CKD patients.

DEMMI Scores, Length of Stay, and 30-Day Readmission of Acute Geriatric Patients in Denmark: A Cross-Sectional Observational Study with Longitudinal Follow-Up

The aims of this study are to describe the mobility of acute geriatric patients, the length of stay, and to characterise patients who were readmitted within 30 days based on the De Morton Mobility Index (DEMMI). A cross-sectional observational study with longitudinal follow-up was conducted in the period from 1 March 2016 to 31 August 2016. Inclusion criteria were acute geriatric patients hospitalised for a minimum of 24 h. Of the 418 patients hospitalised during the study period, 246 (59%) participated in this study (44% male, median age 83 years [70; 94]). For patients in an acute geriatric department, the median DEMMI score was 41 and the mean score was 39.95. Patients with a DEMMI score ≤40 show a significantly lower Barthel 100 index, lower 30 s. sit-to-stand scores and were significantly more likely to be bedridden or, amongst those not bedridden, to use a mobility aid. Lower DEMMI scores were associated with longer admissions. DEMMI seems to have the ability to predict discharge within one week. There was no significant association between a lower DEMMI score and higher risk for 30-day readmission. Further research is needed to determine whether the DEMMI is suitable for identifying the patient's need for further rehabilitation following the discharge.