This paper uses the longitudinal aspect of the Health and Retirement Survey to explore the characteristics associated with reversals in retirement (referred to here as “unretirement”).  Through the use of survival time analysis, this paper show that health insurance plays a significant role in unretirement decisions.  This role is underestimated when a static probit analysis is used alone.  The results hold up for a number of different retirement identifiers that are based both on self-reports of retirement and actual work levels.  The results are also robust to various definitions of retirement prompted by the difficult question of how to classify partial retirements.  The importance of health insurance provision in a retiree’s decision also remains significant when other “shocks” and the prospect of planned unretirement are introduced.