Development and internal validation of machine learning-based models and external validation of existing risk scores for outcome prediction in patients with ischaemic stroke.

Aims: We developed new machine learning (ML) models and externally validated existing statistical models [ischaemic stroke predictive risk score (iScore) and totalled health risks in vascular events (THRIVE) scores] for predicting the composite of recurrent stroke or all-cause mortality at 90 days and at 3 years after hospitalization for first acute ischaemic stroke (AIS). Methods and results: In adults hospitalized with AIS from January 2005 to November 2016, with follow-up until November 2019, we developed three ML models [random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBOOST)] and externally validated the iScore and THRIVE scores for predicting the composite outcomes after AIS hospitalization, using data from 721 patients and 90 potential predictor variables. At 90 days and 3 years, 11 and 34% of patients, respectively, reached the composite outcome. For the 90-day prediction, the area under the receiver operating characteristic curve (AUC) was 0.779 for RF, 0.771 for SVM, 0.772 for XGBOOST, 0.720 for iScore, and 0.664 for THRIVE. For 3-year prediction, the AUC was 0.743 for RF, 0.777 for SVM, 0.773 for XGBOOST, 0.710 for iScore, and 0.675 for THRIVE. Conclusion: The study provided three ML-based predictive models that achieved good discrimination and clinical usefulness in outcome prediction after AIS and broadened the application of the iScore and THRIVE scoring system for long-term outcome prediction. Our findings warrant comparative analyses of ML and existing statistical method-based risk prediction tools for outcome prediction after AIS in new data sets.

Hospital-Wide Intervention in Billing and Coding to Capture Complexity of Care at an Academic Referral Center.

GOAL: Downcoding at nonprofit healthcare institutions can account for significant revenue losses that, in turn, can affect the amount and quality of care they provide. Using the inpatient medical note to assess the complexity of care, we wanted to quantify the visit coding distribution at the largest tertiary care center in West Virginia and to improve the documentation and coding if found to be below national benchmarks. METHODS: We measured the number of encounters and associated documentation of level 1, 2, and 3 visits among hospitalists. We compared our data to national benchmark data. We then implemented a multifaceted, multidisciplinary intervention to improve documentation and coding. PRINCIPAL FINDINGS: We found a significant average increase of level 3 admission history and physical visits of 76% ( p < .0001) and 112% ( p < .001) for subsequent encounters compared with baseline preintervention visit types. With team-based coding interventions in place, documentation accurately now reflects the complexity of care delivered. Based on Medicare reimbursement rates, this new accuracy has led to an increase in revenue of $233, 988.79 per 10,000 encounters. APPLICATIONS TO PRACTICE: Provider knowledge of medical billing and coding guidelines is essential. In particular, large academic institutions typically operate on small margins, so even simple adjustments and quality improvement efforts in billing and coding can help immensely by accurately representing the amount and quality of medical services. An institution can markedly improve revenues by coding notes to reflect the true complexity of care that is delivered.

30-Day Readmission Reduction in a Skilled Facility Population Through Pharmacist-Driven Medication Reconciliation.

BACKGROUND: Transitions of care can be difficult to manage and if not performed properly, can lead to increased readmissions and poor outcomes. Transitions are more complex when patients are discharged to skilled nursing facilities. PURPOSE: We assessed the impact of pharmacist-led initiatives, including medication reconciliation, on readmission rates between an academic medical center and a local skilled nursing facility (SNF). METHODS: We conducted a two-phase quality improvement project focusing on pharmacist-led medication reconciliation at different points in the transition process. All-cause 30-day readmission rates, medication reconciliation completion rates, and total pharmacist interventions were compared between the 2 groups. RESULTS: The combined intervention and baseline cohorts resulted in a 29.8% relative reduction (14.5% vs. 20.6%) in readmission rates. Medication reconciliation was completed on 93.8% of SNF admitted patients in the first phase and 97.7% of patients in the second phase. Pharmacist interventions per reconciliation were 2.39 in the first phase compared with 1.82 in the second phase. CONCLUSION: Pharmacist-led medication reconciliation can contribute to reduction of hospital readmissions from SNFs and is an essential part of the SNF transition process.

Trends in Characteristics and Outcomes of Hospitalized Young Patients Undergoing Coronary Artery Bypass Grafting in the United States, 2004 to 2018.

Background Data are limited about young adults' characteristics and outcomes undergoing coronary artery bypass grafting (CABG). Methods and Results We used the National Inpatient Sample database to identify adults aged 18 to 45 years who underwent CABG between 2004 and 2018. The data were weighted to generate national estimates of the entire US hospitalized population. We identified 110 463 CABG cases, equivalent to 62.2 per 1 000 000 person-years; 27.1% were women, and 70.2% were White adults. Overall, annual CABG volume per 1 000 000 significantly decreased from 87.3 in 2004 to 45.7 in 2018. The prevalence of obesity, diabetes mellitus, hypertension, drug abuse, and chronic medical conditions increased over time. Overall, inpatient mortality was 1.76%; ST-segment-elevation myocardial infarction, non-ST-segment-elevation myocardial infarction, heart failure, peripheral vascular disease, renal failure, and valvular surgery were associated with higher inpatient mortality. Women had higher inpatient mortality than men (2.29% versus 1.57%), and Black patients had higher deaths than White patients (2.86% versus 1.58%). Inpatient mortality remained stable overall, according to sex, race, or clinical indication of CABG. However, the mean length of stay (8.4 days in 2004 to 9.5 days in 2018) and inflation-adjusted cost of care ($40 522.8 in 2004 to $52 434.2 in 2018) significantly increased during the study period. Conclusions Despite the increased burden of cardiometabolic risk factors, the inpatient mortality in young adults undergoing CABG remained stable during the last 15 years. However, CABG volumes have decreased, but length of stay and inflation-adjusted costs have increased over time.

Transcatheter versus surgical aortic valve replacement in patients with end stage renal disease.

OBJECTIVE: To assess contemporary national trends of comorbidities, outcomes, and health care resource utilization in patients with aortic stenosis (AS) and end-stage renal disease (ESRD) undergoing transcatheter and surgical aortic valve replacement (TAVR and SAVR). METHODS AND RESULTS: The National-Inpatient-Sample was used to study trends in patients with AS and ESRD undergoing TAVR and SAVR between January 2012 and December 2017. Of 12,550 patients, 5,735 underwent TAVR and 6,815 underwent SAVR. Over the years, the utilization of SAVR declined (from 82.0 to 37.7%); and increased for TAVR (from 18.0 to 62.3%; p < .001). Patients receiving TAVR were older (74.6 [9.1] vs. 66.8 years [9.1]), had a higher proportion of females (37.1 vs. 32.5%), Caucasians (68.7 vs. 60.9%) and Asian /Pacific Islanders (3.1 vs. 2.7%; p < .001 for all). The TAVR patients, despite having higher comorbidity burden (anemia, coronary artery disease, chronic pulmonary disease, congestive heart failure, cerebrovascular disease, and peripheral vascular disease) had lower inpatient mortality and complications (ST-elevation myocardial infarction, pneumonia, pneumothorax, pulmonary embolism, cardiogenic shock, cardiac arrest, and need for mechanical ventilators and vasopressors). The median length of stay (13.9-6.5 days; p < .001) and cost of stay ($311,538.16 to $255,693.40; p < .001) reduced with TAVR; but remained unchanged with SAVR. Higher proportion of patients was discharged home after TAVR vs. SAVR. CONCLUSION: Among patients with AS and ESRD, despite providing therapy to subjects with higher comorbidity burden, TAVR was associated with lower inpatient mortality, complications, length of stay, cost of care, and higher home disposition rates when compared with SAVR.

Fungal Peptic Ulcer Disease in an Immunocompetent Patient.

The lifetime prevalence of peptic ulcer disease (PUD) is 5-10%. While PUD in immunocompetent patients is most commonly associated with Helicobacter pylori infection or the use of non-steroidal anti-inflammatory drugs (NSAIDs), other common causes of PUD must also be considered in the differential diagnosis. We describe a case of endoscopic and histological resolution of PUD related to Candida infection in a healthy, immunocompetent woman. LEARNING POINTS: Peptic ulcer disease (PUD) can be secondary to fungal infections, even in immunocompetent patients.A higher index of suspicion needs to be maintained for fungal causes of PUD, particularly if symptoms do not improve.Recognizing fungal causes of PUD may lead to faster diagnosis and treatment.

Guidance for Healthcare Providers Managing COVID-19 in Rural and Underserved Areas.

Coronavirus disease 2019 (COVID-19), which is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has ravaged many urban and high-density areas in the USA. However, rural areas (despite their low population density) may be especially vulnerable to poor outcomes from COVID-19, owing to limited healthcare infrastructure, long distances to advanced health care, and population characteristics (e.g., high tobacco use, hypertension, obesity, older age). A panel of experts who are actively engaged in treating and managing COVID-19 at a rural academic center was convened to address this topic. In this commentary, we provide readers with some specific issues faced by rural healthcare providers and offer guidance in overcoming these challenges. This guidance includes alternative ventilator strategies, personal protective equipment (PPE), and common therapeutic options.

Efficacy and safety of low dose rivaroxaban in patients with coronary heart disease: a systematic review and meta-analysis.

The mortality effects and risk-benefit profile of low dose rivaroxaban (2.5 mg twice daily) in patients with coronary heart disease are not completely understood. Five randomized controlled trials (26,110 patients) were selected using PubMed and Cochrane library till April 2019. The background antiplatelet therapy was aspirin in 3 trials, P2Y12 inhibitor in 1 trial, and in 1 trial 65% patients received aspirin and 35% were on dual antiplatelet therapy (DAPT). The outcomes of interest were cardiovascular mortality, all-cause mortality, myocardial infarction (MI), stroke and major bleeding events. Random effects hazard ratios (HR) with 95% confidence intervals (CI) were calculated. Low dose rivaroxaban did not reduce the risk of cardiovascular mortality (HR 0.90, 95% CI 0.73-1.11, P = 0.34) or all-cause mortality (HR 0.91, 95% CI 0.74-1.12, P = 0.38) compared with control. However, low dose rivaroxaban was associated with reduction in MI (HR 0.85, 95% CI 0.73-0.99, P = 0.04), and stroke (HR 0.59, 95%CI 0.48-0.73, P < 0.001) at the expense of major bleeding (HR 1.64, 95% CI 1.39-1.94, P < 0.001) compared with control. These effects did not vary according to acute coronary syndrome or stable coronary heart disease (P-interaction > 0.05). The use of low dose rivaroxaban in patients with coronary heart disease predominantly receiving antiplatelet monotherapy did not reduce cardiovascular or all-cause mortality. The benefits of preventing MI and stroke were balanced by increased risk of major bleeding.

Meta-analysis of the Relation of Body Mass Index to Cardiovascular Outcomes in Patients Receiving Intensive Low-Density Lipoprotein Cholesterol Lowering Therapy.

The impact of body mass index (BMI) on cardiovascular outcomes in patients receiving intensive low-density lipoprotein cholesterol (LDL-C) lowering therapy is uncertain. We performed meta-analysis of 29 randomized controlled trials using PubMed, Embase, and CENTRAL through April 2019. Therapies were grouped as more intensive LDL-C lowering therapy (statins, ezetimibe + statin or PCSK9 inhibitors) and less intensive LDL-C lowering therapy (less potent active control or placebo). Random effects meta-regressions and meta-analyses were performed to evaluate association of BMI with cardiovascular endpoints. In 265,766 patients, for every 1 kg/m2 increase in BMI, more intensive therapy compared with less intensive therapy was associated with hazard ratio (HR) of 1.07 for cardiovascular mortality (95% confidence interval 1.02 to 1.13); HR of 1.03 for all-cause mortality (0.99 to 1.06) HR of 1.06 for myocardial infarction (1.02 to 1.09), HR of 1.08 (1.03 to 1.12) for revascularization and HR of 1.04 for MACE (1.01 to 1.07). Meta-analysis showed that patients with BMI <25 kg/m2 had the highest risk reduction in mortality and cardiovascular outcomes compared with patients with BMI ≥30 kg/m2 (p-interaction ≤0.05). In conclusion, patients with normal BMI treated with intensive LDL-C lowering regimens may derive a larger clinical benefit compared with patients with larger BMI. The results could be due to the higher mortality rate of obese patients that may artificially lower the efficacy of therapy, or due to a true therapeutic limitation in these patients.