Puri, Shikha
(2024)
Predicting the Impact and Trends of SARS-CoV-2 on the Respiratory Viral Season in Pittsburgh Using Interpretable Machine Learning Forecast Models: A Quality Improvement (QI) Retrospective Study.
Master Essay, University of Pittsburgh.
Abstract
This Quality Improvement (QI) project utilizes predictive modeling to understand the dynamics of the COVID-19 pandemic, particularly examining the interaction with the respiratory virus season (RVS) encompassing Respiratory Syncytial Virus (RSV), Influenza, and SARS-CoV-2. This project seeks to determine whether COVID-19 will remain an additional burden on laboratories or diminish, making it another respiratory virus in the RVS. The analysis is from October 2015 to December 2023, examining incidence and ICD-10 cases from UPMC Shadyside and Presbyterian hospitals in Pittsburgh. This analysis compared pre- and post-COVID-19 periods, revealing evolving burdens on laboratories and hospitals. Our exploratory data analysis (EDA) visualizes the seasonal trends of the respiratory viruses, highlighting a shift in typical RVS patterns coinciding with the onset of SARS-CoV-2. Simple and Seasonal Naïve forecasting models provide baseline insights, while ARIMA and SARIMA models offer more advanced prediction techniques, acknowledging data complexities post-COVID-19. Despite SARIMA's superior performance, challenges arise due to limited post-pandemic data, emphasizing the need for continued data collection. The public health implications for our research are for proactive healthcare planning and understanding COVID-19's trajectory as a potentially endemic virus. Future endeavors will focus on continued data collection to refine the predictive models, create effective resource allocation strategies, and relieve the healthcare burden for potential future pandemics.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
Other Thesis, Dissertation, or Long Paper
(Master Essay)
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| Contributors: |
| Contribution | Contributors Name | Email | Pitt Username | ORCID  |
|---|
| Committee Chair | Martinson, Jeremy | jmartins@pitt.edu | jmartins | UNSPECIFIED | | Committee Member | Pasculle, William | pasculleaw@upmc.edu | UNSPECIFIED | UNSPECIFIED |
|
| Date: |
21 May 2024 |
| Date Type: |
Completion |
| Submission Date: |
25 April 2024 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
52 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Public Health > Infectious Diseases and Microbiology |
| Degree: |
MPH - Master of Public Health |
| Thesis Type: |
Master Essay |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
Predictive Model, Forecast, Forecasting model, ARIMA, SARIMA, Naive Forecasting, COVID-19, SARS-Cov-2, RSV, Influenza, Incidence, ICD-10, Pittsburgh, UPMC |
| Date Deposited: |
21 May 2024 14:54 |
| Last Modified: |
21 May 2024 14:54 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/46311 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}