Kelaher Margaret, Jessop Dorothy Jones
Medical and Health Research Association of New York City, Inc., NY, USA.
Health Policy. 2006 Mar;76(1):80-92. doi: 10.1016/j.healthpol.2005.05.002. Epub 2005 Jun 21.
To examine predictors of losing Medicaid and the impact of losing Medicaid on health service utilization for persons living with HIV/AIDS.
The data are from the Community Health Advisory and Information Network (CHAIN), an on-going longitudinal survey representative of adults with HIV/AIDS in NYC (N = 698) (1994-1997 data). Change of Medicaid coverage between survey waves was considered a "transition" yielding three groups: "no transition", "transition to insurance", and "transition to no insurance".
To determine predictors of transitions and the impact of transitions on health service utilization, multinomial logistic regression was used to compare the three groups.
There were 114 transitions and 792 cases without transitions, with transitions decreasing over time. Thirty percent of transitions were from Medicaid to no insurance. Transitions to insurance were more likely among the employed and those with incomes over 15,000 US Dollar. Transitions to no insurance were more likely among AIDS cases, recent immigrants, and people less than 30-year-old. People in both transition groups were less likely than people who retained Medicaid to have experienced a life event in the pre-transition period. Those with transitions to insurance reported decreased hospital and drug treatment. People who became uninsured reported decreased use of routine and preventive care, decreased health information and advice and decreased use of private doctors and outpatient clinics.
While the rate of transitions from Medicaid was relatively low, such transitions were associated with greater variability in quality of health care and greater difficulty accessing primary care among the uninsured.
