The optimal level of government-provided disability insurance (DI) should incorporate benefits and costs to social welfare and interactions with private insurance. I investigate how DI interacts with private insurance provided by the family. I examine changes in family-provided monetary and in-kind transfers after the onset of a disability and test whether DI increases or decreases these private family transfers. Theoretically, if recipients and their families are uncertain about disability severity or DI receipt, and if families and recipients face different budget constraints, then DI could have spillover effects on the family. I take advantage of rich information on disability, health and transfers in the Health and Retirement Study to analyze this question empirically. Using a fixed effects, difference in differences research design, I compare transfers between DI recipients and two control groups: rejected applicants and a propensity-score weighted sample of disabled individuals who do not apply for DI. I reject the hypothesis that DI reduces the probability of receiving a transfer by more than 3 percentage points, and find that DI could increase the probability of receiving a transfer. These findings suggest that crowd out of family transfers in response to DI is lower than crowd out in response to other social insurance programs, and that DI could send a welfare-improving information signal.