Social support buffers the negative impact of stressful events. Less, however, is known about the characteristics of this association in the context of disaster and findings have been discrepant regarding direct and buffering effects. This study tested whether the protective effects of social support differed across levels of exposure severity (i.e., buffered distress) and assessed whether the buffering effect differed between event-specific and general distress. Participants were 4,600 adult Swedish tourists (44% of invited; 55% women) repatriated within 3 weeks after the 2004 Indian Ocean tsunami. A survey 14 months after the disaster included the Crisis Support Scale, the Impact of Event Scale-Revised (IES-R), and the General Health Questionnaire (GHQ-12). Social support buffered the negative impact of exposure on both outcomes. The support and distress association ranged from very small in participants with low exposure to moderate in those with high exposure (ηp2 = .004 to .053). The buffering effect was not found to differ between the IES-R and GHQ-12, F(2, 4589) = 0.87, p = .42. The findings suggest that social support moderates the stressor-distress relationship after disasters. This study might help explain discrepant findings and point to refinements of postdisaster interventions.