OBJECTIVES: To study the association between amount of social contact and mortality after hip fracture in elderly participants. DESIGN: Prospective cohort. SETTING: Community residents of Baltimore, Maryland. PARTICIPANTS: Six hundred seventy-four elderly participants. MEASUREMENTS: Amount of telephone and direct personal contact between participants and their relatives and friends and mortality up to 2 years after fracture. RESULTS: No social contact with friends during the 2 weeks before the fracture was associated with a five times greater risk of death over 2 years than daily contact with friends during the 2 weeks before the fracture (hazard ratio (HR)=5.04, 95% confidence interval (CI)=2.75-9.23). Participants with less than daily contact were also at greater risk of dying, although the CI spanned 1 (HR=1.76, 95% CI=0.99-3.13). Participants who had no contact with family members prefracture were more than twice as likely to die as those who communicated daily during the 2 weeks before fracture (HR=2.26, 95% CI=1.36-3.77). Participants who had less than daily contact were also more than twice as likely to die (HR=2.55, 95% CI=1.65-3.94). CONCLUSION: This study suggests that lower social contact before hip fracture is associated with poorer survival after 2 years.