High parity has been hypothesised to lead to a shorter and less healthy life. Using the 2007 Taft Ageing Health and Fertility Survey consisting of 696 women aged 50-79, this paper examines the extent to which women's health in middle and older ages is affected by their childbearing histories. The results show that high parity (>8) is associated with a reduction of GP-rated health by 0.094 points on a scale from 1 to 10. These health reductions are four times as large as those of an extra year of age, and are robust to controlling for birth interval, age, area of residence, education, marital status, work history, economic satisfaction and surviving daughters. There is a positive but curvilinear relationship between shorter birth intervals (<2 years) and GP-rated health accounting for socio-demographic factors. Our analysis suggests that parity and birth intervals, along with socio-demographic characteristics, affect women's well-being in later life.