Date of Award

Fall 2019

Document Type

Open Access Dissertation


Health Services and Policy Management

First Advisor

Melanie J. Cozad


The Patient Protection and Affordable Care Act (ACA) served as a paradigm shift to reimburse physicians based on health outcomes or quality of care patients receive in relation to and conscious of the cost to provide care, rather than the traditional fee-for- service (FFS) system. To implement value-based care under the ACA, value-based care models (VBCM), such as patient-centered medical homes (PCMH) and accountable care organizations (ACO), were formed with the ultimate goal to advance quality of care. Among commercially insured populations, clinically integrated networks (CIN) have emerged as another type of VBCM.

Since CINs are the newest type of VBCM, the current literature explains their formation and intended goals, yet no studies examine a CIN’s ability to improve quality. This dissertation fills that knowledge gap by examining a large and advanced CIN in the Midwestern US to evaluate the effect the CIN’s formation has on the improvement of the quality of care. In particular, this study focuses on evaluating whether a physician becoming a participating member of the CIN improves performance outcomes in readmissions and cardiovascular disease (CVD).

I developed a framework using Donabedian theory to explain why a change in structure through physicians becoming members of the CIN may have an effect on process and outcome quality metrics. To empirically investigate the framework, this dissertation uses a retrospective, longitudinal study design. To estimate the effect of a physician becoming a participating member of the CIN on quality improvement in readmissions and CVD, a regression discontinuity in time (RDiT) empirical strategy is deployed. Using the CINs own data collection and analytics platform, quality metrics were collected across approximately 3.1 million patients and 180 million patient encounters from 2016-2018.

There was no observed effect between the formation of the CIN and the quality of care delivered. This was explained in the data by the near optimal performance of participating physicians within the CIN. For example, the national average for 30-day readmissions is approximately 20%; yet, the CINs average is around 2.2%. These findings suggest that a strategy targeted directly toward physicians within the CIN could more clearly enhance quality outcomes; implementing a strategy that disseminates these quality metrics to each individual physician is the logical next step for quality improvement. Taking the additional step to unblind these results allows physicians to see their own performance and how they compare to their peers. This holds the potential for an even greater effect on quality outcomes.


© 2019, Kaitlyn Ann Crosby