Harvey, Tucker J
(2022)
Modeling Patient Satisfaction After Arthroscopic Partial Meniscectomy Using Knee Injury and Osteoarthritis Scores (KOOS) Measured with Error.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
Pain and reduced function resulting from tears of the meniscus can cause disability, compounded by a risk of osteoarthritis. The Knee Injury and Osteoarthritis Score (KOOS) is a tool for quantifying pain and functional deficits, but scores can vary even without underlying biological change. Arthroscopic Partial Meniscectomy (APM) is a minimally invasive surgery used to repair meniscal tears, but it's hypothesized that using improvements in KOOS pain or function sub-scores alone is too simplistic to determine whether patients are satisfied with their post-operative results. This project modelled post-operative satisfaction using KOOS and baseline demographic characteristics, while accounting for the intra-subject variability of KOOS. Logistic regression was used to model satisfaction with KOOS and demographic covariates. A backwards-selection technique with bootstrapping was used to quantify variable importance. Age, education, race, and mental health were identified as important covariates for a reduced model. A multiple imputation technique was used to simulate KOOS uncertainty, in which other covariates were used to impute potentially true values of the erroneously-measured variables using background information about the distribution of errors. This was followed by regression of satisfaction on these imputed values and computation of corrected regression coefficient estimates. Minor changes to regression coefficient estimates and odds ratios resulted, but the associated confidence intervals generally overlapped with the uncorrected estimates. While the effect of KOOS pain on satisfaction marginally decreased, the effect of KOOS function increased. Individuals who were worse off
iv
(high pain and low function) at baseline, and those whose pain and function improved the most, had the highest probability of satisfaction. More research is needed to exactly explain the effect estimates for demographic predictors. Additionally, simulation studies would be useful to determine the performance of this measurement error correction method, as would a study of KOOS using validation data. This project has public health implications in educating clinicians and patients about what factors are important in determining satisfaction after APM, specifically how KOOS should be used. There are considerable ethical and financial benefits to more effectively identifying ideal candidates for surgery, while ruling out those unlikely to benefit.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
11 May 2022 |
| Date Type: |
Publication |
| Defense Date: |
25 April 2022 |
| Approval Date: |
11 May 2022 |
| Submission Date: |
27 April 2022 |
| Access Restriction: |
1 year -- Restrict access to University of Pittsburgh for a period of 1 year. |
| Number of Pages: |
94 |
| 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: |
Knee
Arthroscopic Partial Meniscectomy
KOOS
Logistic Regression
Modelling
Measurement Error |
| Date Deposited: |
11 May 2022 20:18 |
| Last Modified: |
11 May 2023 05:15 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/42792 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}