Quality of care and preadmissions for high risk juvenile onset diabetic children: The impact of Medicaid managed care

BACKGROUND: As many as three million individuals had Type I diabetes in the United States in 2011. With Medicaid program variations, there are potentially large differences in access and quality of care for child diabetics. However, no previous studies have holistically assessed the impact of Medicaid managed care on juvenile diabetic treatment, and how it might interact with other patient variables (including race, age and gender) on the quality of care. The focus here is on readmissions, since previous research indicates that readmissions are a quality measure for diabetes, and that specific clinical measures can prevent and/or reduce readmissions. METHODS: This study used a retrospective four year (2008-2011) cross-sectional design with de-identified Children's Hospital Association PHIS data from 43 children's hospitals in 26 states, including 4,762 patients with 12,104 qualifying admissions, to determine whether juvenile onset diabetic patients are more or less likely to be readmitted with uncontrolled diabetes if they are on a Medicaid managed care plan. Study data included inpatient, observation and emergency room readmissions, enabling a greater understanding of not only the readmission rates but also the examination of the type of services provided on the initial visit adjusted for severity and then on the return visit. Multiple factors for readmission were analyzed including age, race, gender, severity of illness (APR-DRG), patient type on discharge, and state. RESULTS: Simple readmission rates by state ranged from 4.08% to 24.82 %. The relationship between state and days between readmissions was significant (p<.0001). Logistic regression revealed that overall readmission was more likely for Medicaid patients on non-managed care plans (Chi-Square 4.3, p = 0.0373), and that those on non-managed care plans were 1.12 times more likely to be readmitted. For the overall diabetic readmission rate, race (p<.0001), age (p<.0001) and gender (p<.0212) were also significant factors. When combined with these, primary insurance type (Medicaid managed vs. non-managed) provided a significant predictive model for readmissions. CONCLUSIONS: This is the first multi-site study to compare readmission rates for children on Medicaid with diabetes across states specifically focusing on the type of health plan and its relationship to readmissions. The results suggest that medical caregivers need to consider the impact of Medicaid program structure in their state on high risk diabetic children. It also appears that using managed care elements in state Medicaid programs may also facilitate efforts to assure equal access to care and reduce disparities in treatment.