This paper presents an empirical investigation of the health impacts of grandparents who are co-residing with their grandchildren at the year their grandchildren were born. Using the PSID (The Panel Study of Income Dynamics) dataset from 2001 to 2011, I identified 2,418 grandparents and 2,268 grandchildren, which comprises 2,656 pairs of grandparent-grandchild. I use Differences-in-Differences methodology and Regression Discontinuity design to examine (1) whether there are any changes in both self-rated health conditions and trajectories of grandparents before and after the birth of their grandchildren; (2) whether there are any difference in both self-rated health conditions and trajectories between grandparents who are living with their grandchild and not living with their grandchild right after their grandchild was born. The results show that there is a statistically significant negative impact on grandparent’s health condition and health deterioration rate if they co-residing with their grandchild. But this effect is very heterogeneous: grandfathers who are living with their grandchildren have a better health condition (by 0.14 units) to begin with but getting sicker faster at the rate of 0.11 units of health for every ten years, comparing to grandmothers; African-American grandparents experience a faster health decline once they co-residing with their grandchildren; those grandparents in families with no adult child (skipped generations) have a worse health condition to begin with and they are also getting sicker at a faster rate and those grandparents with higher SES have a better health condition to start with but experience faster health decline once they co-residing with their grandchildren at the time of their grandchildren were born than those grandparents with lower SES.