In this paper we estimate the effect of early life health on home care use later in life, and we analyse whether this effect is mediated through household composition. We use Dutch administrative data on men born in 1944-1947 who were examined for military service between 1961-1965 and we link them to national data on non-residential care in the period 2004-2013 and data on house- hold status information for the years 1999-2014. We account for confounding factors that influence both early life health and later life home care use. We also account for selective attrition. Our empirical findings show that general health problems in youth is an important health predictor for later life home care use. A large portion of this effect is an indirect effect running through changes in partnership status.