Date of Graduation


Document Type


Degree Type



School of Pharmacy


Pharmaceutical Systems and Policy

Committee Chair

Usha Sambamoorthi

Committee Co-Chair

Joel Halverson

Committee Member

Michael Hendryx

Committee Member

Xiaoyun Pan

Committee Member

Leonard Pogach

Committee Member

Cindy Tworek.


Objective: The current studies examined the relationship between lapses in quality (ambulatory care sensitive hospitalizations and all cause 30-day readmissions) and patient-level, provider-level, and county-level healthcare resources. Specific attention is paid to the association between patient-level primary care use and provider-level care coordination. One of the studies also evaluated the association between chronic complex illness and lapses in quality after adjusting for individual-level, provider-level, and county-level characteristics within a longitudinal and unified framework. Yet another study evaluated whether avoiding poor quality outcomes can be achieved below an expected expenditure benchmark.;Study Design: The study used a retrospective cross-sectional as well as longitudinal design using observational data in real-world settings.;Data Source: Medicaid administrative claims files from four states, California, Illinois, New York and Texas, for 2008 were used. These states were chosen for their low managed care penetration rates relative to other states, as well as their diverse patient populations. The Medicaid files consisted of personal summary, outpatient, inpatient, skilled nursing home, prescription drugs, and long-term care. Personal Summary file included information on FFS beneficiary demographics (gender, age, race/ethnicity, county of residence), Medicaid enrollment and eligibility status. The Outpatient and Inpatient files included claims for services provided in ambulatory and inpatient settings and contained International Classification of Diseases 9th edition Clinical Modification (ICD-9-CM) codes. Medicaid claims were linked with 2008 Area Resource File through county identifiers to obtain county-level information on socio-economic status, healthcare resources, facilities, providers and utilization.;Study Population: The Study population consisted of fee-for-service Medicaid beneficiaries, aged 18-64 years, with full-year continuous enrollment and not dually enrolled in Medicare.;Statistical Techniques: Chi square tests of independence were used to determine significance between individual, provider and community level characteristics and ACSH or readmissions. Multilevel logistic regression models on likelihood of ACSH and 30-day readmissions were conducted. . Due to the large numbers, logistic regressions were conducted on a 10% random sample of our study population. In these models, county was specified as a random intercept using GLIMMIX procedure in Statistical Analysis Software version 9.3 (SAS Inc., Cary, North Carolina USA).;Findings: In cross-sectional analyses of all Medicaid fee-for-service beneficiaries included in study 1 (N = 2.95 million across all four states), we observed that11% had any ACSH, 9.2% had all cause 30-day readmission, and 2% had combined ACSH+Readmisison. In longitudinal analyses (study 2), among fee-for-service Medicaid beneficiaries with diabetes, 14% (N = 43,753) had cooccurring diabetes and depression. Across three states included in study 3 analyses, approximately 5.5 million beneficiaries did not have any ACSH or 30-day readmission. In all studies, patient complexity in terms of chronic conditions increased the risk of any ACSH and readmissions. Chronic complex illness was associated with increased risk of ACSH and decrease risk of readmissions. County-level variables were generally not associated with ACSH or readmissions. However, some county-level healthcare resources such as access to primary care at the county-level reduced the risk of very poor quality outcomes. However, greater availability of other types of healthcare resource increased the risk of poor quality outcomes (example; presence of mental health centers and greater availability of OBGYNs).;Discussion/Conclusion: Our findings suggest that chronic diseases need to be better managed perhaps within an integrated system. Access to primary alone may not be enough to reduce risk of preventable hospitalizations. There is a need for innovative strategies such as comprehensive primary care for our nation's vulnerable and indigent populations. In the absence of system level restructuring of Medicaid programs, states will need to prioritize interventions for targeted groups of beneficiaries. We propose that cost containment may be maximized by aiming to reduce racial disparities and serve those with mental illness. If programs provide comprehensive primary care services to beneficiaries (especially racial ethnic minorities) and those with severe mental illness or substance abuse we expect to see reductions in poor outcomes and improved expenditure profiles. While county-level variables were generally not associated with ACSH or readmissions, some features such as access to primary care at the county-level may reduce the risk of very poor outcomes such as combined ACSH and hospital readmissions. However, greater availability of other types of healthcare resources may indeed increase the risk of poor quality outcomes. These findings taken together suggest that problems in healthcare quality cannot be solved by investments in more resources alone, but by investing in the value of the care provided. State Medicaid programs should explore models of delivery that support value based provision of care over volume based care.