Assessing the relationship between health insurance plan type and length of stay among HIV inpatients with Medicaid: A multi-level approach

Background: Current literature indicates differential hospital use among HIV inpatients due to non-medical factors (e.g., health insurance status). However, it is unclear how health insurance plan type (i.e., managed care vs. fee-for-service) may affect outcomes like length of stay among vulnerable populations, such as low-income inpatients with HIV. This study aims to determine whether enrollment in a managed care plan is associated with shorter lengths of stay compared to enrollment in a traditional fee-for-service plan among HIV inpatients with Medicaid. Methods: Hospital discharge data collected by the California Office of Statewide Health Planning and Development (OSHPD) was used to identify 1,283 HIV inpatients with Medi-Cal (i.e., California’s Medicaid program) who had been discharged from 186 hospitals in 39 counties in 2011. A fixed-effects negative binomial regression of length of stay on Medi-Cal plan type was used to account for potential unobserved heterogeneity and the highly skewed distribution of length of stay. Results: Among HIV inpatients with Medi-Cal, being enrolled in a managed care plan significantly decreased the expected count of days spent in the hospital in 2011 by a factor of 0.85 on average compared to those enrolled in a fee-for-service plan, even after controlling for potential person- and hospital-level confounders and clustering at the county level (p=.03). Conclusions: Results from this study lend support in favor of current efforts across California to transition Medi-Cal beneficiaries, including those with HIV, out of fee-for-service plans into managed care plans in order to improve access to care and control health care spending.