This paper examines the long term impact of exposure to Medicaid in early childhood on health in adulthood. A growing body of evidence from epidemiology, economics and sociology suggests that early life health, health care utilization, and economic conditions, three factors that are plausibly sensitive to Medicaid, have long term consequences.  To study the long term impact of Medicaid, we leverage the program’s gradual adoption across the states, which occurred mainly between 1966 and 1970. This natural experiment created variation in cumulative exposure to Medicaid for birth cohorts that are now in midlife. Results from difference-in-difference models of the Panel Study of Income Dynamics (PSID) suggest that exposure to Medicaid in childhood (age 0-5 years), among subgroups targeted by the program, is associated with statistically significant improvements in health in adulthood (age 18-54). Medicaid improves a composite index of chronic health conditions, that includes information on blood pressure, heart disease, diabetes and obesity, by 0.3 to 0.9 standard deviations (p<0.05). Results are robust to the inclusion of a detailed set of controls measuring the local health care market and public policy environment and to triple difference techniques that net out any state level endogeneity in the timing of Medicaid’s introduction. Results suggest adult health is sensitive to the availability of health insurance during early childhood and that the full value of Medicaid exceeds its immediate benefits.