Panny, Benjamin
(2024)
Continuous Time Structural Equation Modelling for Fatigue and Step Counts in People with Cancer During Chemotherapy.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
The bidirectional and temporal dynamics of fatigue and physical activity in cancer patients during chemotherapy treatment are not well understood. To this end, digital health technologies such as Fitbits and smartphones provide convenient means to collect patient physical activity and symptom data for analysis. However, longitudinal data in this format often involve missingness and uneven time intervals between observations, which pose challenges to common longitudinal modeling approaches. Continuous-time structural equation models (CTSEMs) offer tools for addressing these issues in longitudinal data by leveraging stochastic differential equations. In this thesis, CTSEMs are applied to longitudinal patient-reported fatigue and Fitbit-recorded step count data while accommodating for baseline covariates, chemotherapy treatment days, and hospital stays. These models are applied both with and without multiple imputation via auxiliary baseline variables and then compared. Results suggest that days with high step counts are followed by higher fatigue presence on subsequent days on average, while days with fatigue presence are followed by slightly lower step counts on subsequent days on average. Results also suggest that hospital stays are associated with lower step counts, while chemotherapy treatment days are associated with lower subsequent step counts and higher subsequent fatigue. Findings additionally include that CTSEMs capture the variation in fatigue and step counts well, even without imputing missing data or accommodating for baseline covariates and time-dependent events such as chemotherapy treatments. These results contextualize previous studies demonstrating that exercise interventions decrease cancer-related fatigue from pre- to post-intervention by highlighting that in the short-term, increased physical activity on a given day measured by step counts may temporarily increase fatigue on subsequent days. These results additionally inform the debate on handling missing mobile health data by showing that CTSEM can be an effective approach.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
14 May 2024 |
| Date Type: |
Publication |
| Defense Date: |
16 April 2024 |
| Approval Date: |
14 May 2024 |
| Submission Date: |
24 April 2024 |
| Access Restriction: |
1 year -- Restrict access to University of Pittsburgh for a period of 1 year. |
| Number of Pages: |
102 |
| 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: |
fatigue, cancer, chemotherapy, physical activity, steps, continuous-time structural equation model |
| Date Deposited: |
14 May 2024 18:58 |
| Last Modified: |
14 May 2024 18:58 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/46282 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[img]](http://d-scholarship.pitt.edu/style/images/fileicons/application_pdf.png)
![[feed]](/images/feed-icon-32x32.png){kind=icon}