Electronic Theses and Dissertations



Document Type


Degree Name

Doctor of Philosophy


Public Health

Committee Chair

Paige Powell

Committee Member

Meredith Ray

Committee Member

Aram Dobalian

Committee Member

Sandra Richardson


A mammogram follow-up rate investigates the rate of initial mammograms that require a follow-up visit to confirm breast cancer. The ideal mammogram follow-up rate for an organization is 10%. Rates below mean organizations are underperforming the scan and missing possible cancer cases; above means that hospitals are overdoing mammograms. Previous research examined the problem in a micro-level approach focusing on patient and provider level factors. Open systems theory views the problem from a macro-level approach showing the components and how they interact with each other and their environments. The objectives of this study include reviewing how open systems theory can be used to understand the components affecting hospitals mammography follow-up rates, and examining market and organizational factors are associated with hospitals or counties meeting the recommendation, and understand how the guarantor in open systems theory changed the system between 2010 and 2016. We merged three datasets: Centers for Medicare & Medicaid Services, Area Health Resource File, and the Healthcare Cost and Utilization Project. We performed multinomial logistic regressions to analyze what market and organizational factors affect the follow-up rate. We conducted t-tests and sign tests to provide a profile of hospitals that changed recommended meeting status between 2010 and 2016. Our results showed that open systems theory is underutilized and has the capability to be included in future research on public health problems. Our findings showed that counties were less likely to meet the recommendation if they had more deaths per 100,000 population, more radiologists per 100,000 population, a higher percentage of screening hospitals, and higher percent of African Americans. Organization level factors that affected quality of care included hospital ownership and region where hospitals that were private/not-for-profit, government funded, or in the Western region were more likely to meet the recommendation compared to hospitals that were for-profit or in the Northeast. The guarantor results showed no significant findings between 2010 and 2016 for organizational and market level factors. Future research should focus on open systems and viewing the problem in a macro-level approach in market and organization level factors in relation to their effect on quality of care.


Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to ProQuest