Optimize the management of urological tube-related emergencies during the coronavirus disease 2019 (COVID-19) pandemic

BackgroundTo introduce and determine the value of optimized strategies for the management of urological tube-related emergencies with increased incidence, complexity and operational risk during the global spread of coronavirus disease 2019 (COVID-19).MethodsAll emergent urological patients at Tongji Hospital, Wuhan, during the period of January 23 (the beginning of lockdown in Wuhan) to March 23, 2020, and the corresponding period in 2019 were recruited to form this study’s COVID-19 group and control group, respectively. Tongji Hospital has the most concentrated and strongest Chinese medical teams to treat the largest number of severe COVID-19 patients. Patients in the control group were routinely treated, while patients in the COVID-19 group were managed following the optimized principles and strategies. The case incidence for each type of tube-related emergency was recorded. Baseline characteristics and management outcomes (surgery time, secondary complex operation rate, readmission rate, COVID-19 infection rate) were analyzed and compared across the control and COVID-19 periods.ResultsThe total emergent urological patients during the COVID-19 period was 42, whereas during the control period, it was 124. The incidence of tube-related emergencies increased from 53% to 88% (P<0.001) during the COVID-19 period. In particular, the incidence of nephrostomy tube-related (31% vs. 15%, P=0.027) and single-J stent-related problems (19% vs. 6%, P=0.009) increased significantly. The mean surgery times across the two periods were comparable. The number of secondary complex operations increased from 12 (18%) to 14 (38%) (P=0.028) during the COVID 19-period. The number of 2-week postoperative readmission decreased from 10 (15%) to 1 (3%) (P=0.049). No participants contracted during the COVID-19 period.ConclusionsUrological tube-related emergencies have been found to have a higher incidence and require more complicated and dangerous operations during the COVID-19 pandemic. However, the optimized management strategies introduced in this study are efficient, and safe for both urologists and patients.     

Readmissions of Elder Patients Presenting to Hospital for a Fall (RELIEF): A Systematic Review

ObjectivesFalls are an important issue in older adults as they are frequent, deleterious, and often lead to repeated consultations at the emergency department (ED) and unplanned hospitalizations. Our principal objective was to provide an inventory of interventions designed to prevent unplanned readmissions or ED visits of older patients presenting to hospital with a fall.DesignSystematic review performed on February 11, 2019 in MEDLINE via PubMed, EMBASE, Cochrane Central Register of Controlled Trials, and Web of Science, without date or language restriction. We manually updated this search in August 1, 2019. Study selection, data extraction and risk of bias assessment were conducted independently by 2 reviewers.Setting and ParticipantsWe included studies reporting interventions to prevent unplanned readmissions or ED visits of older patients (aged 65 years or over) presenting to hospital because of a fall.ResultsWe identified 475 unique citations after removing duplicates and included 6 studies (2 observational and 4 interventional studies). The studies were published between 2012 and 2019; they evaluated heterogeneous interventions that were frequently multifaceted and multidisciplinary. The interventions were shown effective in reducing readmissions or ED revisits compared with control groups in 3 studies (relative risk reductions between 30% and 65%), all of which were multifaceted and 2/3 multidisciplinary.Conclusions and ImplicationsWith 6 articles showing inconsistent results, our study highlights the need to adequately design and evaluate interventions to reduce the burden of hospital readmissions among older fallers. Retrieved studies are recent, which underlines that hospital readmissions are a current concern for researchers and public health authorities [PROSPERO registration number: CRD42019131965].     

How Good Is Machine Learning in Predicting All-Cause 30-Day Hospital Readmission? Evidence From Administrative Data

ObjectivesHospital readmission is a main cost driver for healthcare systems, but existing works often had poor or moderate predictive results. Although the available information differs in different studies, improving prediction is different from the search for important explanatory variables. With large sample size and abundant information, this study explores state-of-the-art machine-learning algorithms and shows their performance in prediction.MethodsUsing administrative data on 1 631 611 hospital stays from Quebec between 1995 and 2012, we predict the probability of 30-day readmission at hospital admission and discharge. We compare the performance between traditional logistic regression, logistic regression with penalization, and more recent machine-learning algorithms such as random forest, deep learning, and extreme gradient boosting.ResultsAfter a 10-fold cross-validation on the training set (80% of the data), machine learning produced very good results on a separate hold-out test set (20% of the data). The importance of explanatory variables is not the same for different algorithms. The area under receiver operating characteristic curve (AUC) reached above 0.79 at hospital admission and above 0.88 at hospital discharge. Diagnostic codes, which include many different categories, are among the most predictive variables. Logistic regression with penalization also produced good results, but a standard logistic regression failed without penalization. The good results are confirmed by calibration curves.ConclusionAlthough the identification of those at highest risk of readmission is just 1 step to preventing hospital readmissions, 30-day readmission is highly predictable with machine learning.     

The Association of Race/Ethnicity and Total Knee Arthroplasty Outcomes in a Universally Insured Population

BackgroundPrior studies have documented racial/ethnic disparities in the United States for total knee arthroplasty (TKA) outcomes. One factor cited as a potential mediator is unequal access to care. We sought to assess whether racial/ethnic disparities persist in a universally insured TKA population.MethodsA US integrated health system’s total joint replacement registry was used to identify elective primary TKA (2000-2016). Racial/ethnic differences in revision and 90-day postoperative events (readmission, emergency department [ED] visit, infection, venous thromboembolism, and mortality) were analyzed using Cox proportional hazard and logistic regression with adjustment for confounders.ResultsOf 129,402 TKA, 68.8% were white, 16.2% were Hispanic, 8.4% were black, and 6.6% were Asian. Compared to white patients, Hispanic patients had lower risks of septic revision (hazard ratio [HR] = 0.69, 95% confidence interval [CI] = 0.57-0.83) and infection (odds ratio [OR] = 0.42, 95% CI = 0.30-0.59), but a higher likelihood of ED visit (OR = 1.28, 95% CI = 1.22-1.34). Black patients had higher risks of aseptic revision (HR = 1.61, 95% CI = 1.42-1.83), readmission (OR = 1.13, 95% CI = 1.02-1.24), and ED visit (OR = 1.31, 95% CI = 1.23-1.39). Asian patients had lower risks of aseptic revision (HR = 0.67, 95% CI = 0.54-0.83), septic revision (HR = 0.78, 95% CI = 0.60-0.99), readmission (OR = 0.89, 95% CI = 0.79-1.00), and venous thromboembolism (OR = 0.59, 95% CI = 0.45-0.78).ConclusionWe observed differences in TKA outcome, even within a universally insured population. While lower risks in some outcomes were observed for Asian and Hispanic patients, the higher risks of aseptic revision and readmission for black patients and ED visit for black and Hispanic patients warrant further research to determine reasons for these findings to mitigate disparities.Level of EvidenceLevel III.