Zhou, Jerry
(2024)
Impact of COVID-19 on Adverse Outcomes for Congestive Heart Failure Inpatients in the Northeast Mid-Atlantic Using NIS 2020 Database.
Master's Thesis, University of Pittsburgh.
(Unpublished)
![[img]](https://d-scholarship.pitt.edu/style/images/fileicons/application_pdf.png) |
PDF (RStudio code for creating a new dataset with only CHF patients)
Supplemental Material
Restricted to University of Pittsburgh users only until 16 May 2025.
Download (347kB)
| Request a Copy
|
|
PDF (RStudio code for univariate analysis)
Supplemental Material
Restricted to University of Pittsburgh users only until 16 May 2025.
Download (377kB)
| Request a Copy
|
|
PDF (RStudio code for 30-day in-hospital mortality for multivariable logistic regression stepwise selection)
Supplemental Material
Restricted to University of Pittsburgh users only until 16 May 2025.
Download (324kB)
| Request a Copy
|
|
PDF (RStudio code for Random forest on 30-day in-hospital mortality, with discharge-level variables and without discharge-level variables)
Supplemental Material
Restricted to University of Pittsburgh users only until 16 May 2025.
Download (366kB)
| Request a Copy
|
|
PDF (RStudio code for Kaplan-Meier, log-rank, and cox proportional hazards model, along with multivariable cox proportional hazards model for NIS stratum and several stratifiers)
Supplemental Material
Restricted to University of Pittsburgh users only until 16 May 2025.
Download (435kB)
| Request a Copy
|
|
PDF
Restricted to University of Pittsburgh users only until 16 May 2025.
Download (1MB)
| Request a Copy
|
Abstract
Background: The Coronavirus disease 2019 (COVID-19) is an infectious disease created by the SARS-CoV-2 virus. Although COVID-19 now remains an endemic disease, its adverse impacts on patients with congestive heart failure (CHF) have not been fully studied. This research uses the National Inpatient Sample (NIS) 2020 database to identify in-hospital outcomes of 109,667 CHF patients (104,015 without COVID-19, 5,652 with COVID-19) admitted to Northeast Mid-Atlantic hospitals (New York, Pennsylvania, New Jersey) in that year.
Methods: We utilized multivariable logistic regression models in analyzing 30-day in-hospital mortality (our primary outcome of interest) for CHF patients with and without COVID-19, adjusting for covariates. We also performed similar multivariable logistic regression models for several secondary in-hospital outcomes. We applied random forest in analyzing the importance of COVID-19, together with other baseline covariates, for predicting 30-day in-hospital mortality. Survival analysis was used to analyze time-to-in-hospital death in CHF patients with and without COVID-19.
Results: Our multivariable logistic regression analyses found that COVID-19 is significantly associated with an increased odds ratio for 30-day in-hospital mortality (OR = 7.20, 95% CI 6.67 - 7.77). Random forest analyses also found COVID-19 as the most important variable for the mean decrease accuracy metric of 30-day in-hospital mortality prediction. Our survival analysis reported that the survival curves of in-hospital mortality are significantly different for CHF patients with COVID-19 and those without COVID-19 (log-rank p < 0.001). The median time-to-in-hospital death for CHF patients with COVID-19 is 25 days (95% CI 23 - 27 days), while the median time-to-in-hospital death for those without COVID-19 is more than 30 days. Our secondary analyses also show significant differences in vasopressor use, sudden cardiac arrest, acute kidney injury on hemodialysis (HD), cardiogenic shock, and mechanical circulatory support (MCS) for CHF patients with and without COVID-19.
Public Health Significance: Understanding the severity of COVID-19 on CHF patients' in-hospital outcomes and how deadly COVID-19 ranked as one of the top risk factors for in-hospital mortality is critical for the clinical care of CHF patients during the pandemic. This also provides insights into the mitigation process for vulnerably chronic individuals during future disruptive public health crises.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
16 May 2024 |
| Date Type: |
Publication |
| Defense Date: |
19 April 2024 |
| Approval Date: |
16 May 2024 |
| Submission Date: |
26 April 2024 |
| Access Restriction: |
1 year -- Restrict access to University of Pittsburgh for a period of 1 year. |
| Number of Pages: |
55 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Public Health > Biostatistics |
| Degree: |
MS - Master of Science |
| Thesis Type: |
Master's Thesis |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
COVID-19, congestive heart failure, multivariable logistic regression, stepwise selection, random forest, Cox proportional hazards model |
| Date Deposited: |
16 May 2024 19:34 |
| Last Modified: |
16 May 2024 19:34 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/46312 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}