Context. Re-admission is the return hospitalization within 30 days from the date of original admission or discharge from hospital. Thecosts of the unplanned re-admissions were estimated to $25 billion per year alone in the U.S. Re-admission rate also has a huge impact onquality of care provided to the patients, cost of health care, and utilization of hospital resources and the image of the care provider. Studies indicate huge potential of savings that can be achieved with incremental performance improvements in detecting cases of preventable re-admissions. Objectives. In this study we find the different features that helpin predicting readmissions, compare different machine learning techniques and build a model to predict readmissions using one technique.We also propose a framework for implementation of this model in the real world situations. Methods. To reach the objective, the data of the patients over a period of time were studied to determine the factors that help in identifying re-admissions. Experiments are performed for identifying the features that are more relevant to predict re-admissions and for investigating the most suitable machine learning techniques for this purpose.This model was tested to predict re-admission cases for Acute Myocardial Infarction and Pneumonia. Results. The features that help in predicting re-admissions are determined,and a model was developed using these features and the selected machine learning algorithm. The model showed good results in predicting re-admissions. The model predicted risk of Acute Myocardial Infarction(c-statistic=0.811), and Pneumonia(c=0.76). Conclusions. We conclude that our model showed good results in predicting re-admissions. The developed model is discriminative for specific diseases like Acute Myocardial Infarction and Pneumonia. Itis also generalized as it incorporates the features that can be easily available from all of the patient population over the globe.

In a stroke-certified 500-bed acute care hospital, the 30-day readmission rates for patients discharged to rehabilitation centers or skilled nursing facilities were higher than the rates for patients discharged to home. A review of data by the stroke team showed 44 patients readmitted within 30 days of initial stroke discharge between October 2016 and January 2017. The rate of re-admission for those discharged home was 41% (18 patients), whereas the rate for those discharged to acute inpatient rehabilitation, long-term acute care, or skilled nursing facilities was 59% (26 patients). The practice-focused question for this project assessed whether using a re-admission risk-assessment tool and implementing interventions during the initial acute-care admission, would help to identify and improve risk for 30-day re-admissions for patients diagnosed with stroke. The goal of this research project was to adopt, test, and recommend the implementation of a readmission risk assessment tool to enable discharge planners to identify stroke patients at risk for readmission and implement interventions to help reduce this risk. Lewin’s theory of change was used to inform the project. A stroke re-admission risk-assessment tool in use at a similar hospital was adopted and tested for 1 week on the hospital’s 28-bed stroke unit by nurse case managers. The test was conducted among 5 patients with confirmed diagnosis of stroke. A re-admission data review was performed 30 days after their discharge, which showed no readmissions for the 5 patients involved in the trial. The tool helped to improve case manager awareness of increased risk for readmissions, guide interventions, and improve patient transition and outcomes. The implications of this project for positive change include the potential to improve risk for patients with stroke in the acute-care facility.

Despite several investigations and interventions aimed at decreasing the time from symptom onset to medical care seeking in acute myocardial infarction patients, the delay time is still too long for best treatment outcomes. In this thesis, investigations aimed at improving our understanding of the factors influencing delay time are evaluated, as well as attitudes to medical care seeking in patients, relatives and the general public. Additionally, an evaluation was performed to examine whether health-related quality of life had any influence on delay time and re-admissions. Participating patients, relatives and representatives of the general public were generally knowledgeable about acute myocardial infarction (AMI) and its symptomatology. The majority of participants knew about the importance of receiving fast treatment when an AMI occurs. Despite people’s knowledge, several patients and relatives felt uncertain of symptom origin and how to act at symptom onset. Patients commonly consulted an additional person when symptoms did not disappear. However, people appeared to act more appropriately if someone else had chest pain compared to self-experienced symptoms. In patients who had suffered from more than one AMI, poor total health status increased the risk of delaying for more than two hours, but no independent association was found between total health status and re-admissions within the first year post-AMI.

Background: Heart Failure (HF) is a frequent cause of mortality and recurrent hospitalization. Although HF databases are assembled based on left ventricular (LV) ejection fraction, patients without LV ejection fraction determination are not further analyzed. Objective: The purpose of this study is to characterize patient attributes and outcomes in this group-HF with unknown Ejection Fraction (HFunEF). Methods: We queried the electronic medical record from a community-based university practice for patients with a HF diagnosis. We included patients with >60 days follow-up and had interpretable Doppler-echocardiograms. We recorded demographic, Doppler-echocardiographic, and outcome variables (up to 2083 days). Results: There were 820 patients: 269 with HF with preserved Ejection Fraction (HFpEF), 364 with HF with reduced Ejection Fraction (HFrEF), of which 231 had a LV ejection fraction=40-49% and 133 had a LV ejection fraction<40%, and 187 with HFunEF. As compared to patients with HFunEF, HFpEF patients were younger, had a higher coronary disease and hyperlipidemia prevalence. Patients with HFrEF had more prevalent coronary disease, myocardial infarction, and hyperlipidemia. Patients with HFunEF were more likely to be seen by non-cardiology providers. All-cause mortality (ACM) was greater in HFunEF patients than patients with HFpEF (Hazard Ratio (HR)=1.60 (1.16-2.29), p=0.004). Furthermore, HF readmission rates were lower in HFunEF as compared to HFpEF (HR=0.33 (0.27-0.54), p<0.0001) and HFrEF (HR=0.30 (0.028-0.50), p<0.0001). Conclusion: Patients with HFunEF have greater ACM and lower HF re-admission than other HF phenotypes. Adherence to core measures, including LV ejection fraction assessment, may improve outcomes in this cohort of patients.

No description available.