Search

Background Intensive Care Unit (ICU) readmissions in patients with heart failure (HF) result in a significant risk of death and financial burden for patients and healthcare systems. Prediction of at-risk patients for readmission allows for targeted interventions that reduce morbidity and mortality. Methods and results We presented a process mining/deep learning approach for the prediction of unplanned 30-day readmission of ICU patients with HF. A patient’s health records can be understood as a sequence of observations called event logs; used to discover a process model. Time information was extracted using the DREAM (Decay Replay Mining) algorithm. Demographic information and severity scores upon admission were then combined with the time information and fed to a neural network (NN) model to further enhance the prediction efficiency. Additionally, several machine learning (ML) algorithms were developed to be used as the baseline models for the comparison of the results. Results By using the Medical Information Mart for Intensive Care III (MIMIC-III) dataset of 3411 ICU patients with HF, our proposed model yielded an area under the receiver operating characteristics (AUROC) of 0.930, 95% confidence interval of [0.898–0.960], the precision of 0.886, sensitivity of 0.805, accuracy of 0.841, and F-score of 0.800 which were far better than the results of the best baseline model and the existing literature. Conclusions The proposed approach was capable of modeling the time-related variables and incorporating the medical history of patients from prior hospital visits for prediction. Thus, our approach significantly improved the outcome prediction compared to that of other ML-based models and health calculators.

Uncontrolled hypertension contributes significantly to cardiovascular morbidity and mortality and is more frequently encountered among patients presenting to the emergency department (ED).1,2 Approximately 50% of patients with previously undiagnosed hypertension who present with elevated blood pressures in the ED will continue to have BP elevations one week after being discharged to home.3 Hypertensive patients presenting to the ED are a particularly high-risk group with >50% with stage II or higher hypertension (SBP ≥160 or DBP ≥100).1 Significant percentages of elevated blood pressures in the ED are incidental findings and not related to the patient's chief complaints. Incidental hypertension represents a quandary for the emergency physician who cannot guarantee follow-up treatment.4,5 Several studies have found that only 7–25% of ED patients with elevated BP are given instructions for outpatient BP follow-up.3–5 Currently, there is no risk assessment or stratification based upon blood pressure assessment performed on hypertensive patients prior to discharge from the emergency department. The emergency medicine literature does not provide definitive recommendations on screening tests for patients with asymptomatic elevated blood pressures.6 Published guidelines regarding the management of these patients only recommend referral of these patients for close primary care follow-up within 48–72 h.6 These same guidelines also urge future research related to the optimal screening for this patient population in the ED.6 The purpose of this study was to determine the progression of subclinical heart disease in an urban ED population with asymptomatic elevated blood pressure and examine the correlation with blood pressure changes after a brief ED hypertension education intervention.

Background and Purpose —Several studies observed that people from lower socioeconomic groups have higher chances of dying of stroke. There are reasons to expect that these differences are relatively small in southern European countries or in Nordic welfare states. This report therefore presents an international overview of socioeconomic differences in stroke mortality. Methods —Unpublished data on mortality by occupational class were obtained from national longitudinal studies or cross-sectional studies. The data refer to deaths among men aged 30 to 64 years in the 1980s. A common occupational class scheme was applied to most countries. The mortality difference between manual classes and nonmanual classes was measured in relative terms (by rate ratios) and in absolute terms (by rate differences). Results —In all countries, manual classes had higher stroke mortality rates than nonmanual classes. This difference was relatively large in England and Wales, Ireland, and Finland and relatively small in Sweden, Norway, Denmark, Italy, and Spain. Differences were intermediate in the United States, France, and Switzerland. In Portugal, mortality differences were intermediate in relative terms but large in absolute terms. In most countries, inequalities were much larger for stroke mortality than for ischemic heart disease mortality. Conclusions —Socioeconomic differences in stroke mortality are a problem common to all countries studied. There are probably large variations, however, in the contribution that different risk factors, such as tobacco and alcohol consumption, make to the stroke mortality excess of lower socioeconomic groups. Medical services can contribute to reducing socioeconomic differences in stroke mortality.