Clinical and demographic predictors of thrombophilia among patients with venous thromboembolism: Data from a large referral center Öffentlichkeit
Wickham, Zoe (Spring 2021)
Published
Abstract
Purpose: To develop a predictive model to identify factors associated with a positive thrombophilia workup.
Background and Rationale: The term ‘thrombophilia’ describes a group of intrinsic conditions that increase likelihood of blood clot formation. Testing is expensive and subject to misinterpretation if reviewed by a physician unfamiliar with hypercoagulable states. There remains considerable uncertainty on who should be tested and how thrombophilia status should inform clinical management of VTE.
Methods: With a sample of 345 patients treated for their first documented VTE event between January 2011 and December 2015 at Emory University Hospital, a series of logistic regression models were constructed to examine factors associated with positive thrombophilia status. We used a hierarchical backwards elimination approach to reduce the model to variables independently associated with a positive outcome, one more stringent with a cut-off of p-value of ≤0.1 and then another less stringent with a p-value cutoff of ≤0.25. Both models were examined for the presence of two-way interactions and collinearity. The results of all final models were expressed as multivariable adjusted odds ratios (OR) and the corresponding 95% confidence intervals (CI). Predictive value of the models was assessed by calculation of a c-statistic.
Results: The following variables were the strongest predictors within both models: sex (OR=2.24), race (specifically African American vs. White, OR =0.46), family history of hypercoagulable state (OR=45.2), active malignancy (OR=0.11), and liver disease (OR=2.49). The calculated c-statistic was 0.758 for model 1 and 0.761 for model 2.
Discussion: In this study, the strongest predictors in both logistic models were sex, race, active malignancy, and liver disease. Current testing guidelines instruct physicians to test patients in situations of unprovoked VTE, VTE is in an unusual site, strong family history of thrombotic disease, or patients’ preference. This study was limited by its small sample size but despite this, the results of this paper suggest that guidelines for testing can be enhanced by considering additional clinical factors.
Table of Contents
1. Chapter I: Introduction and Rationale………………………………………………………......1
2. Definition of terms………………………………………………………………………….....….....3
3. Chapter II: Background & Review of the Literature………………………………………....5
a. Summary of Current Problem and Study Relevance…………………………………….6
4. Chapter III: Methodology…………………………………………………………………....….....8
a. Research Design…………………………………………………………….………...........…..8
b. Study Sample……………………………………………………………………….…..............8
c. Statistical Analysis…………………………………………………………………….............8
5. Chapter IV: Data analysis…………………………………………………………….……….......9
6. Chapter V: Results…………………………………………………………………….…….......…10
7. Chapter VI: Discussion………………………………………………………………………........11
8. References………………………………………………………………………………..…….........13
9. Tables………………………………………………………………………………..…………..........14
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Stichwort | |
| Committee Chair / Thesis Advisor |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Clinical and demographic predictors of thrombophilia among patients with venous thromboembolism: Data from a large referral center () | 2021-04-30 10:07:09 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|