Tools
Tools
Waters, Michelle (2021) Which geographical, socioeconomic and facility attributes predict hospital readmission from Skilled Nursing Facilities MSc Research Project MSc in Science of Data Analytics Michelle Waters Student ID: X17100020 School of Computing National College of Ireland Supervisor: Jorge Basilio. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration manual)
Download (92kB) | Preview |
Abstract
There are more than 15,000 Skilled Nursing Facilities (SNFs) in the US which play a key role in the rehabilitation of patients post acute hospital care. Hospital readmissions are a significant problem for Health Service and are high at over 20% from SNFs and other post acute care facilities. Understanding the factors that contribute to readmissions is the aim of this study. This research builds on previous work but extends the factors that have been previously modelled to understand. What makes one Skilled Nursing Facility more likely than another for patients to be readmitted to hospital? and secondly, What are the types of socio-economic and demographic attributes of the geographic location of a nursing home that might influence the likelihood of being readmitted to hospital?
The scale and ambition of the work is unique and important and requires integration and examination of data from across states and levels to improve the generalisability. Support Vector Machine and Random Forest Classifiers were utilised to model predictive factors with Random Forest performing best with 73% Accuracy and 80% Sensitivity for 6 features. In agreement with the literature characteristics of the SNF and Patients are important predictors of readmission, but also identified as important are the local area patterns of hospitalisation and health services utilisation suggesting that norms within jurisdictions need also to be evaluated. This points to an opportunity for Federal Health Administrators to examine utilisation policies and practices across Geographies. Socio- demographic were not identified as the most important factors in this research.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | E History America > E151 United States (General) Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software R Medicine > Healthcare Industry |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Clara Chan |
| Date Deposited: | 15 Dec 2021 12:06 |
| Last Modified: | 15 Dec 2021 12:06 |
| URI: | https://norma.ncirl.ie/id/eprint/5235 |
Actions (login required)
 |
View Item |