Background: Currently, 10 - 15% of individuals over age 65 have no living biological or stepchildren, and 41% are unmarried. By 2030, the proportion of those without partners or surviving children will likely exceed 30%. With no close relatives to participate in advance care planning, older persons may be vulnerable to unwanted medical procedures. Only half of those over age 50 have an advance directive. Methods: Data were drawn from the 2012 Health and Retirement Study (HRS) (N = 18,613), and exit interviews from 2000 - 2012 with proxies of HRS decedents (N = 8,658). Multinomial logistic regression was used to predict the odds of advance care planning by parental status, adjusting for covariates, and testing for interaction by marital status and gender. Results: Overall, 14.4% of respondents over age 50 were biologically childless (13.71%), or had outlived all children (0.69%). If biologically childless, 4.85% had living stepchildren. 15.54% of men and 13.43% of women were childless (p < 0.01). Childlessness was highest among never married men, 80.12% (p < 0.0001). Nearly 53% of childless women lived in a nursing home/long-term-care setting at the time of final illness, compared with 39.93% of women with living children, 37.38% of childless men, and 28.50% of men with children living (p < 0.0001). In the unadjusted model, childless persons were 28% less likely than those with living children to engage in end-of-life discussion only, OR = 0.782 (p < 0.05). Advance care planning was not related to childlessness in the adjusted model. Never married childless men were least likely of all family status groups to engage in end-of-life discussion only, OR = 0.358 (p < 0.05); advance directive completion only, OR = 0.613 (p < 0.10, non-significant); or both, OR = 0.537 (p < 0.05). Divorced men with living children were less likely to engage in end-of-life discussion only, OR = 0.733 (p < 0.05). Significance: Never married childless men and divorced men with living children are at risk of failing to discuss end-of-life issues. These findings point toward a need for new approaches to advance care planning relative to gender and family structure.