Improving evidence from administrative data to better inform the practice of cardiac surgery
Abstract/Contents
- Abstract
- Success in cardiac surgery is no longer defined by a patient's vital status at hospital discharge, but rather by his/her survival a decade—or more—later. Low early-event rates complicate randomized clinical trial design and make them prohibitively expensive or infeasible. As a result, practice recommendations from cardiovascular societies are seldom supported by strong evidence. Administrative databases contain a wealth of longitudinal patient information, but their lack of granularity and observational nature limit the conclusions drawn from such "big data" research. Herein lies a major opportunity for improvement; rather than dismiss data that chronicles the "real world" care of millions of patients, perhaps we can use this data more effectively to understand and inform clinical practice. In this dissertation, we apply rigorous causal inference methods to statewide and national claims databases to better guide care for patients with valvular heart disease, coronary artery disease, and aortic dissection. We found that: 1) despite an increasing use of biological prostheses, the long-term survival benefit of a mechanical prosthesis persists until 70 years of age in patients undergoing mitral valve replacement, but disappears by age 55 years in those undergoing aortic valve replacement; 2) second arterial conduit use is low and declining, but arterial grafts significantly improve survival and reduce cardiovascular events in patients undergoing multi-vessel coronary bypass; and 3) the majority of patients with an acute type A aortic dissection are treated at low-volume hospitals, yet directed transfer of patients to high-volume hospitals is safe, feasible, and substantially improves survival. Our methods may be broadly applied to improve claims database research in a variety of healthcare fields, and our results have the potential to inform the care of hundreds of thousands of patients undergoing heart surgery annually in the United States.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2017 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Goldstone, Andrew Brooks |
|---|---|
| Associated with | Stanford University, Program in Epidemiology and Clinical Research. |
| Primary advisor | Nelson, Lorene M |
| Thesis advisor | Nelson, Lorene M |
| Thesis advisor | Baiocchi, Michael |
| Thesis advisor | Sainani, Kristin |
| Thesis advisor | Woo, Joseph |
| Advisor | Baiocchi, Michael |
| Advisor | Sainani, Kristin |
| Advisor | Woo, Joseph |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Andrew Brooks Goldstone. |
|---|---|
| Note | Submitted to the Program in Epidemiology and Clinical Research. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2017. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2017 by Andrew Brooks Goldstone
- License
- This work is licensed under a Creative Commons Attribution Non Commercial No Derivatives 3.0 Unported license (CC BY-NC-ND).
Also listed in
Loading usage metrics...