Date of Award

12-2024

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Applied Health Research and Evaluation

Committee Chair/Advisor

Dr. Sarah Bauer Floyd

Committee Member

Dr. Sarah Griffin

Committee Member

Dr. Xia Jing

Committee Member

Dr. Lu Shi

Committee Member

Dr. John Brooks

Abstract

Proximal humerus fractures (PHF) are a common injury affecting older adults, though clinical decision making for these fractures remains challenging for physicians. Over the past two decades, clinical decision making has evolved with the emergence of new evidence like the ProFHER trial and the dissemination of reverse total shoulder arthroplasty (RTSA). There is a need to characterize the effects of these changes across geographic areas and evaluate the impact of differences in surgical procedure utilization rates on patient outcomes. Therefore, the goal of this dissertation was to gain a better understanding of the utilization patterns and outcomes associated with different surgical procedure options for PHF. We utilize 100% nationwide Medicare fee-for-service (FFS) administrative claims data, allowing us to obtain a comprehensive understanding of how treatment utilization patterns have evolved across all geographic areas.

We first focused on the decision of operative or non-operative management. We found that early surgery utilization for PHF varied widely across local practice areas in 2011 and 2017, with hospital referral region (HRR) risk-adjusted surgery rates ranging from 2.1% to 32.5% and 1.6% to 28.8%, respectively. However, HRRs maintained relatively consistent surgery utilization patterns across time, providing evidence for the concept of surgical signatures. These results are even more surprising given the changes in clinical decision making for PHF during this time period. The relative stability of these practice patterns implies that the treatment a patient receives is partially dependent on where they live.

Examining the use of different surgical procedures, we found that RTSA dissemination from 2011 to 2017 varied widely across local practice areas. This variation was unattributable to differences in measured local area characteristics such as medical device manufacturer payments and orthopaedic physician supply. However, higher RTSA use in 2011 was significantly associated with higher RTSA use in 2017, consistent with the concept of surgical signatures. These findings suggest that areas with higher RTSA use in 2011 continued to use RTSA at higher rates in 2017, though persistent inequities in access to RTSA may exist that are attributable to factors difficult to measure at the population level.

Using instrumental variables (IV) analysis, we found that for patients whose treatment choice is influenced by their local area practice style, higher RTSA use rates were significantly associated with higher treatment benefit rates and lower major shoulder complication rates. Collectively, the results of this dissertation provide evidence of treatment signatures in a proclivity for both surgical treatment and surgical procedure choice. Further, in terms of clinical outcomes, our results suggest that greater diffusion of RTSA use may improve patient outcomes. These findings serve as a first step in investigating the potential value of expanding RTSA use and, ultimately, addressing strategies to mitigate disparities in access to effective care for PHF patients.

Author ORCID Identifier

0000-0001-6744-2821

Available for download on Wednesday, December 31, 2025

Share

COinS