All-cause 30-day Hospital Readmission Research Articles

Homeward Stroke Recovery: Results of a Home Rehabilitation Program

Evaluate the feasibility of a homeward stroke recovery (HSR) program by measuring functional outcomes, intervention fidelity, and Profit & Loss. A retrospective observational analysis of 23 stroke survivors discharged to a home healthcare (HHC) program offering early and intensive rehabilitative therapy within 24 hours of index hospitalization. Functional outcomes include timed up-and-go, transfer functional independent measures, and Barthel Index. Clinical outcomes are mortality, all-cause 30-day hospital readmission, and fall rates. Process measures consist of time-to-first HHC visit, number of physical therapy (PT) and occupational therapy (OT) visits from initial evaluation, and total number of visits by clinician type. Financial outcomes include average daily gross HSR revenue compared with an inpatient rehabilitation facility (IRF) stay and average daily net HSR profit. Functional outcomes significantly improved across all measures ( p < .001) without reported falls, deaths, and only one unrelated readmission. On average, HSR began within 1.6 days from hospital discharge including 4.2 PT and 3.7 OT visits within 5 days of respective initial evaluation. Patients received an average of 9.4 nursing, 9.2 PT, 6.4 OT, and 2.2 speech therapy visits. Exploratory analyses demonstrated similar functional recovery despite longer hospitalization (rs = 0.43; p = .039). Revenue per day was lower for HSR than IRF services but costs exceeded the reimbursement rates for this program. The HSR program is feasible given that patients realized significant functional improvement with negligible consequences. Reimbursement reform is necessary to sustain and scale the program. Further studies with larger sample sizes are warranted.

Analysis of 30-day hospital readmissions and related risk factors for COVID-19 patients with myocarditis hospitalized in the United States during 2020

Background Despite extensive research on COVID-19 and its association with myocarditis, limited data are available on readmission rates for this subset of patients. Thirty-day hospital readmission rate is an established quality metric that is associated with increased mortality and cost. Methods This retrospective analysis utilized the Nationwide Readmission Database for the year 2020 to evaluate 30-day hospital readmission rates, risk factors, and clinical outcomes among COVID-19 patients who presented with myocarditis at their index hospitalization. Results Our analysis revealed that 1) the 30-day all-cause hospital readmission rate for patients initially hospitalized with COVID-19 and myocarditis was 11.7%; 2) after multivariate adjustment, the primary predictor of readmission for COVID-19 patients with myocarditis was discharge against medical advice; 3) COVID-19 patients with myocarditis who required readmission had a higher proportion of older patients and Medicare beneficiaries; 4) the most common diagnoses at readmission were COVID-19, sepsis, congestive heart failure, acute myocardial infarction, and pneumonia; and 5) readmitted patients were more likely to require renal replacement therapy during their index hospitalization. Conclusion This study underscores the importance of optimizing discharge plans, preventing irregular discharges through shared decision-making, and ensuring robust post-hospital follow-up for patients with COVID-19 and myocarditis at index admission.

Community-Acquired Staphylococcus aureus Bacteremia Among People Who Inject Drugs: A National Cohort Study in England, 2017-2020.

People who inject drugs (PWID) are at increased risk of community-acquired Staphylococcus aureus bacteremia (CA-SAB), but little is known about clinical outcomes of CA-SAB in PWID compared with the wider population of patients with CA-SAB. Three national datasets were linked to provide clinical and mortality data on patients hospitalized with CA-SAB in England between 1 January 2017 and 31 December 2020. PWID were identified using the International Classification of Diseases, Tenth Revision code for "mental health and behavioral disorder due to opioid use" (F11). Multivariable logistic regression was used to estimate adjusted odds ratios (aORs) for associations of PWID with 30-day all-cause mortality and 90-day hospital readmission. In 10 045 cases of CA-SAB, 1612 (16.0%) were PWID. Overall, 796 (7.9%) patients died within 30 days of CA-SAB admission and 1189 (11.8%) patients were readmitted to hospital within 90 days of CA-SAB. In those without infective endocarditis, there was strong evidence of lower odds of mortality among PWID compared with non-PWID (aOR, 0.47 [95% confidence interval {CI}: .33-.68]; P < .001), whereas there was no association in CA-SAB case fatality with endocarditis (aOR, 1.40 [95% CI: .87-2.25]; P = .163). PWID were less likely to be readmitted within 90 days of CA-SAB (aOR, 0.79 [95% CI: .65-.95]; P = .011). In this large cohort study of patients with CA-SAB in England, PWID had lower odds of death in the absence of endocarditis and lower odds of readmission within 90 days compared to non-PWID patients. This study highlights the overrepresentation of PWID among patients with CA-SAB nationally.

Evaluation of pharmacist-led transition of care program in patients with acute coronary syndrome

BackgroundGuideline-directed medical therapies (GDMTs), initiated in-hospital and continued during the transition to outpatient care, are paramount to successful outcomes for patients with acute coronary syndrome (ACS). Incomplete discharge medication prescribing and delayed follow-up lead to worse cardiovascular outcomes. ObjectivesWe investigated a system of care using inpatient and outpatient clinical pharmacists to close GDMT gaps, ensure seamless transition to outpatient care, improve patient education, and optimize therapies. MethodsWe conducted a pre-post cohort analysis of patients with ACS pre- versus post-intervention to compare process metrics and key outcomes using electronic health record data. ResultsThere were 181 and 135 patients in the pre- and post-intervention cohorts, respectively. Patients post-intervention were significantly more likely to have appropriately-timed follow-up visits scheduled with cardiology (79% vs. 51%, P < 0.0001) and primary care (57% vs. 43%, P = 0.01), to be discharged with prescriptions for P2Y12 inhibitors (87% vs. 64%, P < 0.0001), high dose statins (86% vs. 70%, P = 0.001), and beta blockers (87% vs. 76%, P = 0.01), and significantly less likely to have 30-day all-cause hospital readmissions (4% vs. 12%, P = 0.02) and emergency department (ED) visits (10% vs. 18%, P = 0.04). ConclusionsThe integration of advanced practicing pharmacists into a cardiology team at transition and post-hospitalization resulted in improved rates of posthospital follow-up visits, optimization of GDMT medications, and significantly lower 30-day hospital readmission and ED utilization.

Effectiveness of empiric carbapenem versus non-carbapenem therapy for extended-spectrum β-lactamase producing Enterobacterales infections in non-intensive care unit patients: a real-world investigation in a hospital with high-prevalence of extended-spectrum β-lactamase producing Enterobacterales.

To investigate whether empiric carbapenem therapy, compared to empiric non-carbapenem therapy, was associated with improved clinical outcomes among hospitalized, non-intensive care unit (ICU) patients with extended-spectrum β-lactamase (ESBL)-producing Enterobacterales infections. We performed a retrospective cohort study of adult, non-ICU patients admitted with ESBL-producing Enterobacterales infections. Primary outcome was time to clinical stability from the first empiric antibiotic dose. Secondary outcomes were early clinical response and 30-day all-cause hospital readmission. We used multivariate regression methods to examine time to clinical stability. Of the 142 patients, 59 (42%) received empiric carbapenems and 83 (58%) received empiric non-carbapenems, most commonly ceftriaxone (49/83, 59%). Median age was 59 years. The most common infection source was urinary (71%). The carbapenem group had a higher proportion of patients who received antibiotics within 6 months of admission (55% vs 28%, P < .01) and history of ESBL (57% vs 17%, P < .01). There were no significant differences in hours until clinical stability between the carbapenem and non-carbapenem groups (22 (IQR: 0, 85) vs 19 (IQR: 0, 69), P = .54). Early clinical response (88% vs 90%, P = .79) and 30-day all-cause hospital readmission (17% vs 8%, P = .13) were similar between groups. Among hospitalized non-ICU patients with ESBL-producing Enterobacterales infection, we found no difference in time to clinical stability after the first empiric antibiotic dose between those receiving carbapenems and those who did not. Our data suggest that empiric carbapenem use may not be an important driver of clinical response in patients with less severe ESBL-producing Enterobacterales infection.

Relationship of health-related social needs and hospital readmissions in patients following a hospitalization for atrial fibrillation

Atrial fibrillation (AF) has a high economic burden on the healthcare system with rehospitalizations as the most significant contributing factor necessitating an understanding of aspects related to hospitalizations to minimize economic costs and improve patient outcomes. Our study aims to assess whether all-cause 30-day hospital readmission following AF-specific hospitalization is associated with health-related social needs (HRSN) using the Nationwide Readmissions Database (NRD). All hospitalization data were abstracted from the 2015–2019 NRD, including hospitalizations for patients at least 18 years of age with a primary discharge diagnosis of AF. For each hospitalization, we identified secondary diagnoses for five HRSN domains including employment, family, housing, psychosocial, and socioeconomic status. Primary outcomes included all-cause 30-day readmission rates. Secondary outcomes included all-cause 90-day readmissions and diagnosis on readmissions. An estimated 1,807,460 index hospitalizations in the United States included a primary discharge diagnosis of AF. Of these, 97.3 % included a diagnosis in only one HRSN domain with the most frequently diagnosed HRSN domain being housing (54.5 %) followed by socioeconomic (29.4 %), family (10.0 %), employment (6.1 %), and psychosocial (2.8 %). Index hospitalizations that included any HRSN diagnosis had 2.2-times greater unadjusted odds of all-cause 30-day readmission (95 % CI: 2.1 to 2.3-times greater, p < .001). Index hospitalizations that included an HRSN diagnosis were associated with higher rates of 90-day readmission due to conduction disorder and COPD. In conclusion, there is a significant association between HRSN and hospital readmissions in patients with AF. Further research is required to explain the true nature of this relationship with a specific emphasis on housing insecurity.

Benefit of prompt initiation of single-inhaler fluticasone furoate, umeclidinium, and vilanterol (FF/UMEC/VI) in patients with COPD in England following an exacerbation: a retrospective cohort study

BackgroundTriple therapy is recommended for patients with chronic obstructive pulmonary disease (COPD) who remain symptomatic despite dual therapy. The optimal timing of triple therapy following an exacerbation of COPD is unknown. The outcomes of prompt (≤ 30 days) vs. delayed (31–180 days) initiation of single-inhaler triple therapy with fluticasone furoate, umeclidinium, and vilanterol (FF/UMEC/VI) following an exacerbation of COPD were examined.MethodsThis was a retrospective cohort study of linked English primary (Clinical Practice Research Datalink) and secondary (Hospital Episode Statistics) care data. Patients aged ≥ 35 years with COPD were indexed on the first and/or earliest date of exacerbation between November 15, 2017 and March 31, 2019 with subsequent FF/UMEC/VI initiation within 180 days. Patients were required to be continuously registered with a general practitioner for ≥ 12 months prior to and following index. Subsequent exacerbations, direct medical costs, and hospital readmissions were compared between prompt and delayed initiators. Inverse probability of treatment weighting was used to adjust for measured confounders between cohorts.ResultsOverall, 1599 patients were included (prompt: 393, delayed: 1206). After weighting, prompt initiators had numerically lower moderate/severe exacerbations compared with delayed initiators (rate ratio: 0.87, 95% confidence interval [CI]: 0.76–1.01, p = 0.0587). Both all-cause and COPD-related 30-day hospital readmissions were significantly lower among patients with prompt initiation compared with delayed initiators (all-cause: 23.6% vs. 34.6%, odds ratio [95% CI]: 0.58 [0.36–0.95], p = 0.0293; COPD-related: 20.3% vs. 30.6%, odds ratio [95% CI]: 0.58 [0.35–0.96], p = 0.0347). Prompt initiators also had numerically lower all-cause total costs and significantly lower COPD-related costs per-person-per year compared with delayed initiators (COPD-related: £742 vs. £801, p = 0.0016).ConclusionPrompt initiation of FF/UMEC/VI following a moderate/severe exacerbation was associated with fewer subsequent exacerbations, fewer hospital readmissions, and lower COPD-related medical costs compared with delayed initiation.

Performance of advanced machine learning algorithms overlogistic regression in predicting hospital readmissions: A meta-analysis

ObjectivesMachine learning algorithms are being increasingly used for predicting hospital readmissions. This meta-analysis evaluated the performance of logistic regression (LR) and machine learning (ML) models for the prediction of 30-day hospital readmission among patients in the US. MethodsElectronic databases (i.e., Medline, PubMed, and Embase) were searched from January 2015 to December 2019. Only studies in the English language were included. Two reviewers performed studies screening, quality appraisal, and data collection. The quality of the studies was assessed using the Quality in Prognosis Studies (QUIPS) tool. Model performance was evaluated using the Area Under the Curve (AUC). A random-effects meta-analysis was performed using STATA 16. ResultsNine studies were included based on the selection criteria. The most common ML techniques were tree-based methods such as boosting and random forest. Most of the studies had a low risk of bias (8/9). The AUC was greater with ML to predict 30-day all-cause hospital readmission compared with LR [Mean Difference (MD): 0.03; 95% Confidence Interval (CI) 0.01–0.05]. Subgroup analyses found that deep-learning methods had a better performance compared with LR (MD 0.06; 95% CI, 0.04–0.09), followed by neural networks (MD: 0.03; 95% CI, 0.03–0.03), while the AUCs of the tree-based (MD: 0.02; 95% CI -0.00-0.04) and kernel-based (MD: 0.02; 95% CI 0.02 (−0.13–0.16) methods were no different compared to LR. More than half of the studies evaluated heart failure-related rehospitalization (N = 5). For the readmission prediction among heart failure patients, ML performed better compared with LR, with a mean difference in AUC of 0.04 (95% CI, 0.01–0.07). The leave-one-out sensitivity analysis confirmed the robustness of the findings. ConclusionMultiple ML methods were used to predict 30-day all-cause hospital readmission. Performance varied across the ML methods, with deep-learning methods showing the best performance over the LR.

Impact of Student Pharmacist-Led Transition-of-Care Services on 30-Day Hospital Readmissions at a University Call Center.

Background: Transition of care (TOC) refers to the movement of patients between different health care settings due to changes in medical conditions and needs. Pharmacists can play an important role in TOC services as polypharmacy is a common reason for hospital readmission that costs the US taxpayers an average of $17 billion annually. Objective: The purpose of this study is to evaluate the impact of TOC telehealth services provided by pharmacy students at a university-based call center on 30-day hospital readmission. Methods: In this retrospective observational study, an electronic chart review was conducted for patients who were discharged from the hospital and received a telephone call from pharmacy students. Patients were referred to the pharmacy team from a primary care provider office. The co-primary endpoints were the number of 30-day all-cause hospital readmissions (including emergency department visits) and 30-day readmission due to initial admission diagnosis in patients who received a telephonic TOC call from a pharmacy student compared with patients who declined or were unable to be reached. Types of pharmacy-related TOC interventions provided by students were also collected. Results: A total of 84 patients were included in this study. All-cause 30-day readmission was similar between groups (13% vs 15.8%), whereas 30-day readmission due to initial admission diagnosis was much lower in the intervention group (5.9% vs 11.1%). Although a positive trend was observed in favor of the intervention group, a statistically significant difference was not observed for both 30-day all-cause readmission and 30-day readmission due to initial admission diagnosis. Medication reconciliation, adherence counseling, and lifestyle education (diet, exercise) are the most common topics discussed with the patients during TOC interventions. Conclusion: Using student pharmacists to provide postdischarge TOC calls can be a benefit to the patient and the health care team while offering pharmacy students valuable learning experience prior to graduation.

High-performing primary care clinics across high-need, high-cost Medicare populations

BackgroundTo reduce spending and improve quality, some primary care clinics in the USA have focused on high-need, high-cost (HNHC) Medicare beneficiaries, which include clinically distinct subpopulations: older adults with frailty,...

Impact of sleep apnoea on 30day hospital readmission rate and cost in heart failure with reduced ejection fraction.

In this study, we estimated the 30day all-cause and heart failure-specific readmission rates, predictors, mortality, and hospitalization costs in patients with obstructive sleep apnoea admitted with acute decompensated heart failure with reduced ejection fraction. This is a retrospective cohort study using the Agency of Healthcare Research and Quality's National Readmission Database for the year 2019. The primary outcome was the 30day all-cause hospital readmission rate. The secondary outcomes were (i) in-hospital mortality rate for index admissions; (ii) 30day mortality rate for index hospitalizations; (iii) the five most common principal diagnosis for readmission; (iv) readmission in-hospital mortality rate; (v) length of hospital stay; (vi) independent risk factors for readmission; and (vii) hospitalization costs. We identified 6908 hospitalizations that met our study definition. The mean patient age was 62.8years, and women comprised only 27.6% of patients. The 30day all-cause readmission rate was 23.4%. 48.9% of readmissions were due to decompensated heart failure. The in-hospital mortality rate during readmissions was significantly higher than that of the index admission (5.6% vs. 2.4%; P<0.05). The mean length of stay for patients during index admissions was 6.5days (6.06-7.02), while during readmissions, it was 8.5days (7.4-9.6; P<0.05). The mean total hospitalization charges at index admissions were $78438 (68053-88824), while during readmissions, they were higher at $124282 (90906-157659; P<0.05). The mean total cost of hospitalization during index admissions was $20535 (18311-22758), while at readmissions, it was higher at $29954 (24041-35867; P<0.05). The total hospital charges for all 30day readmissions were $195 million, and total hospital costs was $46.9 million. The variables found to be associated with increased rate of readmissions were patients with Medicaid insurance, higher Charlson co-morbidity Index, and longer length of stay. The variables associated with lower rate of readmissions were prior percutaneous coronary intervention and patients with private insurance. In patients with obstructive sleep apnoea admitted with heart failure with reduced ejection fraction, we found a substantial all-cause readmission rate of 23.4% with heart failure readmission constituting about 48.9% of readmissions. Readmissions were associated with higher mortality and resource use.

Multidimensional Interventions on Supporting Disease Management for Hospitalized Patients with Heart Failure: The Role of Clinical and Community Pharmacists.

existing trials on the role of clinical pharmacists in managing chronic disease patients have focused on variety of interventions, including preparing patients for the transition from hospital to home. However, little quantitative evidence is available regarding the effect of multidimensional interventions on supporting disease management for hospitalized patients with heart failure (HF). The present paper reviews the effects of inpatient, discharge and/or after-discharge interventions performed on hospitalized HF patients by multidisciplinary teams, including pharmacists. articles were identified through search engines in three electronic databases following the PRISMA Protocol. Randomized controlled trials (RCTs) or non-randomized intervention studies conducted in the period 1992-2022 were included. In all studies, baseline characteristics of patients as well as study end-points were described in relation to a control group i.e., usual care and a group of subjects that received care from a clinical and/or community pharmacist, as well as other health professionals (Intervention). Study outcomes included all-cause hospital 30-day re-admission or emergency room (ER) visits, all-cause hospitalization within >30 days after discharge, specific-cause hospitalization rates, medication adherence and mortality. The secondary outcomes included adverse events and quality of life. Quality assessment was carried out using RoB 2 Risk of Bias Tool. Publication bias across studies was determined using the funnel plot and Egger's regression test. a total of 34 protocols were included in the review, while the data from 33 trials were included in further quantitative analyses. The heterogeneity between studies was high. Pharmacist-led interventions, usually performed within interprofessional care teams, reduced the rates of 30-day all-cause hospital re-admission (odds ratio, OR = 0.78; 95% CI 0.62-0.98; p = 0.03) and all-cause hospitalization >30 days after discharge (OR = 0.73; 95% CI 0.63-0.86; p = 0.0001). Subjects hospitalized primarily due to heart failure demonstrated reduced risk of hospital admission within longer periods, i.e., from 60 to 365 days after discharge (OR = 0.64; 95% CI 0.51-0.81; p = 0.0002). The rate of all-cause hospitalization was reduced by multidimensional interventions taken by pharmacists: reviews of medicine lists and/or their reconciliation at discharge (OR = 0.63; 95% CI 0.43-0.91; p = 0.014), as well as interventions that were based mainly on patient education and counseling (OR = 0.65; 95% CI 0.49-0.88; p = 0.0047). In conclusion, given that HF patients often have complex treatment regimens and multiple comorbid conditions, our findings highlight the need for greater involvement from skilled clinical and community pharmacists in disease management.

Associations Between Long-Term Air Pollutant Exposure and 30-Day All-Cause Hospital Readmissions in US Patients With Stroke.

Long-term exposure to air pollutants is associated with increased stroke incidence, morbidity, and mortality; however, research on the association of pollutant exposure with poststroke hospital readmissions is lacking. We assessed associations between average annual carbon monoxide (CO), nitrogen dioxide (NO2), ozone (O3), particulate matter 2.5, and sulfur dioxide (SO2) exposure and 30-day all-cause hospital readmission in US fee-for-service Medicare beneficiaries age ≥65 years hospitalized for ischemic stroke in 2014 to 2015. We fit Cox models to assess 30-day readmissions as a function of these pollutants, adjusted for patient and hospital characteristics and ambient temperature. Analyses were then stratified by treating hospital performance on the Centers for Medicare and Medicaid Services risk-standardized 30-day poststroke all-cause readmission measure to determine if the results were independent of performance: low (Centers for Medicare and Medicaid Services rate for hospital <25th percentile of national rate), high (>75th percentile), and intermediate (all others). Of 448 148 patients with stroke, 12.5% were readmitted within 30 days. Except for tropospheric NO2 (no national standard), average 2-year CO, O3, particulate matter 2.5, and SO2 values were below national limits. Each one SD increase in average annual CO, NO2, particulate matter 2.5, and SO2 exposure was associated with an adjusted 1.1% (95% CI, 0.4-1.9%), 3.6% (95% CI, 2.9%-4.4%), 1.2% (95% CI, 0.2%-2.3%), and 2.0% (95% CI, 1.1%-3.0%) increased risk of 30-day readmission, respectively, and O3 with a 0.7% (95% CI, 0.0%-1.5%) decrease. Associations between long-term air pollutant exposure and increased readmissions persisted across hospital performance categories. Long-term air pollutant exposure below national limits was associated with increased 30-day readmissions after stroke, regardless of hospital performance category. Whether air quality improvements lead to reductions in poststroke readmissions requires further research.

Abstract WMP98: Associations Between Long-term Co, No 2 , O 3 , Pm 2.5 , And So 2 Exposure And 30-day All-cause Hospital Readmission Among Stroke Patients In The Us: 2014-2015

Objective: There are known associations between long-term air pollution exposure and greater stroke incidence, morbidity, and mortality; however, there is little research on the association of pollutant exposure with poststroke hospital readmission. Methods: We assessed associations of average annual CO, NO 2 , O 3 , PM 2.5 , and SO 2 exposure with 30-day all-cause hospital readmission in US fee-for-service Medicare beneficiaries age ≥65 years hospitalized for ischemic stroke in 2014-2015. We fit Cox models to assess 30-day readmission as a function of these pollutants, adjusted for patient and hospital factors and temperature. We repeated the models stratified by performance of the treating hospital on the Centers for Medicare &amp; Medicaid Services (CMS) risk-standardized 30-day all-cause readmission for stroke measure to determine if the results were independent of performance: Low (CMS rate for hospital &lt;25th percentile of national rate), High (&gt;75 th percentile), and Middle (all others). Results: There were 448148 patients discharged with stroke in 2014-2015 of whom 12.5% were readmitted. Average 2-year CO, O 3 , PM 2.5 , and SO 2 values during the study were below national standards. In adjusted analyses, each 1 standard deviation increase in average annual NO 2 and SO 2 exposure was associated with a 3.6% (95% CI 2.9%-4.4%) and a 2.0% (95% CI 1.1%-3.0%) increased readmission risk within 30-days, respectively ( Table ). Associations between long-term air pollution exposure and increased readmissions persisted across performance categories. Conclusion: Long-term air pollution exposure was associated with increased 30-day readmissions after stroke at pollutant levels below national standards across hospital performance categories. Additional research is needed to determine whether improvements in air quality lead to reductions in poststroke hospital readmissions.

Effect of Variation in Early Rehabilitation on Hospital Readmission After Hip Fracture.

Provision of early rehabilitation services during acute hospitalization after a hip fracture is vital for improving patient outcomes. The purpose of this study was to examine the association between the amount of rehabilitation services received during the acute care stay and hospital readmission in older patients after a hip fracture. Medicare claims data (2016-2017) for older adults admitted to acute hospitals for a hip fracture (n = 131,127) were used. Hospital-based rehabilitation (physical therapy, occupational therapy, or both) was categorized into tertiles by minutes per day as low (median = 17.5), middle (median = 30.0), and high (median = 48.8). The study outcome was risk-adjusted 7-day and 30-day all-cause hospital readmission. The median hospital stay was 5days (interquartile range [IQR] = 4-6 days). The median rehabilitation minutes per day was 30 (IQR = 21-42.5 minutes), with 17 (IQR = 12.6-20.6 minutes) in the low tertile, 30 (IQR = 12.6-20.6 minutes) in the middle tertile, and 48.8 (IQR = 42.8-60.0 minutes) in the high tertile. Compared with high therapy minutes groups, those in the low and middle tertiles had higher odds of a 30-day readmission (low tertile: odds ratio [OR] = 1.11, 95% CI = 1.06-1.17; middle tertile: OR = 1.07, 95% CI = 1.02-1.12). In addition, patients who received low rehabilitation volume had higher odds of a 7-day readmission (OR = 1.20; 95% CI = 1.10-1.30) compared with high volume. Elderly patients with hip fractures who received less rehabilitation were at higher risk of readmission within 7 and 30days. These findings confirm the need to update clinical guidelines in the provision of early rehabilitation services to improve patient outcomes during acute hospital stays for individuals with hip fracture. There is significant individual- and hospital-level variation in the amount of hospital-based rehabilitation delivered to older adults during hip fracture hospitalization. Higher intensity of hospital-based rehabilitation care was associated with a lower risk of hospital readmission within 7 and 30days.

Predicting 30-day all-cause hospital readmission using multimodal spatiotemporal graph neural networks.

Reduction in 30-day readmission rate is an important quality factor for hospitals as it can reduce the overall cost of care and improve patient post-discharge outcomes. While deep-learning-based studies have shown promising empirical results, several limitations exist in prior models for hospital readmission prediction, such as: (a) only patients with certain conditions are considered, (b) do not leverage data temporality, (c) individual admissions are assumed independent of each other, which ignores patient similarity, (d) limited to single modality or single center data. In this study, we propose a multimodal, spatiotemporal graph neural network (MM-STGNN) for prediction of 30-day all-cause hospital readmission, which fuses in-patient multimodal, longitudinal data and models patient similarity using a graph. Using longitudinal chest radiographs and electronic health records from two independent centers, we show that MM-STGNN achieved an area under the receiver operating characteristic curve (AUROC) of 0.79 on both datasets. Furthermore, MM-STGNN significantly outperformed the current clinical reference standard, LACE+ (AUROC=0.61), on the internal dataset. For subset populations of patients with heart disease, our model significantly outperformed baselines, such as gradient-boosting and Long Short-Term Memory models (e.g., AUROC improved by 3.7 points in patients with heart disease). Qualitative interpretability analysis indicated that while patients' primary diagnoses were not explicitly used to train the model, features crucial for model prediction may reflect patients' diagnoses. Our model could be utilized as an additional clinical decision aid during discharge disposition and triaging high-risk patients for closer post-discharge follow-up for potential preventive measures.

Development of a 30-Day Readmission Risk Calculator for the Inpatient Rehabilitation Setting.

Readmission to acute care from the inpatient rehabilitation facility (IRF) setting is potentially preventable and an important target of quality improvement and cost savings. The objective of this study was to develop a risk calculator to predict 30-day all-cause readmissions from the IRF setting. Retrospective database analysis using the Uniform Data System for Medical Rehabilitation (UDSMR) from 2015 through2019. In total, 956 US inpatient rehabilitation facilities and 1,849,768 IRF discharges comprising patients from 14 impairment groups. Logistic regression models were developed to calculate risk-standardized 30-day all-cause hospital readmission rates for patients admitted to an IRF. Models for each impairment group were assessed using 12 common clinical and demographic variables and all but 4 models included various special variables. Models were assessed for discrimination (c-statistics), calibration (calibration plots), and internal validation (bootstrapping). A readmission risk scoring system was created for each impairment group population and was graphically validated. The mean age of the cohort was 68.7 (15.2) years, 50.7% were women, and 78.3% were Caucasian. Medicare was the primary payer for 73.1% of the study population. The final models for each impairment group included between 4 and 13 total predictor variables. Model c-statistics ranged from 0.65 to 0.70. There was good calibration represented for most models up to a readmission risk of 30%. Internal validation of the models using bootstrap samples revealed little bias. Point systems for determining risk of 30-day readmission were developed for each impairment group. Multivariable risk factor algorithms based upon administrative data were developed to assess 30-day readmission risk for patients admitted from IRF. This report represents the development of a readmission risk calculator for the IRF setting, which could be instrumental in identifying high risk populations for readmission and targeting resources towards a diverse group of IRF impairment groups.

Effective hospital readmission prediction models using machine-learned features

Background:Hospital readmissions are one of the costliest challenges facing healthcare systems, but conventional models fail to predict readmissions well. Many existing models use exclusively manually-engineered features, which are labor intensive and dataset-specific. Our objective was to develop and evaluate models to predict hospital readmissions using derived features that are automatically generated from longitudinal data using machine learning techniques.Methods:We studied patients discharged from acute care facilities in 2015 and 2016 in Alberta, Canada, excluding those who were hospitalized to give birth or for a psychiatric condition. We used population-level linked administrative hospital data from 2011 to 2017 to train prediction models using both manually derived features and features generated automatically from observational data. The target value of interest was 30-day all-cause hospital readmissions, with the success of prediction measured using the area under the curve (AUC) statistic.Results:Data from 428,669 patients (62% female, 38% male, 27% 65 years or older) were used for training and evaluating models: 24,974 (5.83%) were readmitted within 30 days of discharge for any reason. Patients were more likely to be readmitted if they utilized hospital care more, had more physician office visits, had more prescriptions, had a chronic condition, or were 65 years old or older. The LACE readmission prediction model had an AUC of 0.66 ± 0.0064 while the machine learning model’s test set AUC was 0.83 ± 0.0045, based on learning a gradient boosting machine on a combination of machine-learned and manually-derived features.Conclusion:Applying a machine learning model to the computer-generated and manual features improved prediction accuracy over the LACE model and a model that used only manually-derived features. Our model can be used to identify high-risk patients, for whom targeted interventions may potentially prevent readmissions.

Single-dose intravenous ketamine or intramuscular naltrexone for high-utilization inpatients with alcohol use disorder: pilot trial feasibility and readmission rates

BackgroundAlcohol use disorder (AUD) accounts for millions of acute care encounters annually in the United States. Hospitalization represents a vital opportunity to intervene pharmacologically, but low medication adherence is a significant barrier. Two single-dose, adherence-independent interventions are well suited for pre-discharge administration: intravenous (IV) ketamine and intramuscular (IM) naltrexone. Their feasibility and readmission-reducing efficacy in hospital settings are not well-established.MethodsA 3-arm, open-label randomized trial was conducted at our safety-net medical hospital among high-utilization inpatients with severe AUD. Consented adults (age 18–65) were randomized to (1) IV ketamine (KET) 0.5 mg/kg over 40 min, (2) IM naltrexone (NTX) 380 mg once, or (3) linkage alone (LA). The primary clinical outcome was 30-day all-cause hospital readmission rate. All were provided enhanced linkage to outpatient addiction clinic.ResultsWe consented and randomized 44 participants (n = 13, 14, 17 for KET, NTX, LA, respectively), with a mean of 3.2 past-year hospitalizations. Compared to the LA arm, both the KET arm (RR 0.37, p = 0.17) and NTX arm (RR 0.52, p = 0.27) had a lower 30-day readmission rate, though the differences were nonsignificant. Immediate acceptability ratings of KET and NTX were 9.50 and 9.17 out of 10, respectively. No serious adverse events or illicit ketamine use was reported.ConclusionsBoth interventions are feasible and showed promise in reducing readmissions for high-utilization AUD inpatients. Despite randomization, baseline characteristics may have differed in ways that biased against the control arm. Additional pragmatic studies—with larger sample size, blinding, and robust follow-up data collection—are needed to verify findings and better understand mediating factors.ClinicalTrials.gov Identifier NCT04562779. Registered 24 September 2020. https://clinicaltrials.gov/ct2/show/NCT04562779

Associations Between Inpatient Induction on Medications for Opioid Use Disorder and Postdischarge Medications for Opioid Use Disorder Adherence, Overdose, and Service Use.

This study aimed to examine outcomes of a pilot program designed to increase inpatient medications for opioid use disorder (MOUD) induction and to support MOUD adherence after discharge. This retrospective cohort analysis examined Medicaid adults diagnosed with opioid use disorder discharged from 2 freestanding inpatient withdrawal management facilities between October 1, 2018, and December 31, 2019. Participants had ≥90 days of continuous Medicaid enrollment before and after admission. Odds ratios (ORs) examined associations of inpatient MOUD induction with discharge against medical advice, 7- and 30-day all-cause hospital readmission, and postdischarge MOUD adherence. Mixed-effect models examined changes associated with MOUD induction and postdischarge MOUD adherence in acute service utilization and opioid overdose in the 90-day postdischarge period. Of the 2332 patients discharged, 493 started MOUD inpatient care (21.1%), with most initiating buprenorphine (76.5%). Induction of MOUD was associated with a lower likelihood of discharge against medical advice (OR, 0.49; 95% confidence interval [CI], 0.37-0.64), 30-day all-cause hospital readmission (OR, 0.61; 95% CI, 0.47-0.80), and higher odds of postdischarge MOUD adherence (OR, 3.83; 95% CI, 3.06-4.81). In the 90 days after discharge, MOUD adherent patients had significant reductions in emergency department visits for behavioral health, inpatient days, withdrawal management episodes, and opioid overdoses compared with the 90-day preadmission period. Inpatient MOUD induction is associated with a higher likelihood of short-term MOUD adherence after discharge, which in turn is associated with significant reductions in short-term service utilization and opioid overdose after discharge.