We follow Medicare cohorts over time and space to estimate Hurricane Katrina's long-run mortality effects on elderly and disabled victims initially living in New Orleans. Inclusive of the initial shock, the hurricane improved survival eight years past the storm by 1.74 percentage points. Migration to lower-mortality regions explains most of this survival increase. Migrants to low- versus high-mortality regions look similar at baseline, but migrants’ subsequent mortality is 0.83-0.90 percentage points lower for each percentage-point reduction in local mortality, quantifying causal effects of place on mortality among this population. By contrast, migrants’ mortality is unrelated to local Medicare spending.