Background: Social support is consistently associated with reduced risk of depression. Few studies have investigated how this relationship may be modified by alcohol use, the effects of which may be particularly relevant in traumatized populations in which rates of alcohol use are known to be high. Methods: In 2008 a representative sample of 1622 Jewish and Palestinian citizens in Israel were interviewed by phone at two time points during a period of ongoing terrorism and war threat. Two multivariable mixed effects regression models were estimated to measure the longitudinal association of social support from family and friends on depression symptoms. Three-way interaction terms between social support, alcohol use and time were entered into the models to test for effect modification. Results: Findings indicated that increased family social support was associated with less depression symptomatology (p= <.01); this relationship was modified by alcohol use and time (p= <.01). Social support from friends was also associated with fewer depression symptoms (p= <.01) and this relationship was modified by alcohol use and time as well (p= <.01). Stratified analyses in both models revealed that the effect of social support was stronger for those who drank alcohol regularly than those who did not drink or drank rarely. Conclusions: These findings suggest that social support is a more important protective factor for depression among regular drinkers than among those who do not drink or drink rarely in the context of political violence. Additional research is warranted to determine whether these findings are stable in other populations and settings.