In Reply We appreciate the comments of Manchia and colleagues on our study1 on lithium treatment for prevention of repeated suicide-related outcomes in veterans with major depression or bipolar disorder. The study reported a nonsignificant hazard ratio for repeated suicide-related behaviors (ie, nonfatal suicide attempts, interrupted attempts, hospitalizations specifically to prevent suicidal behavior, and deaths by suicide) for lithium vs placebo of 1.10 (95% CI, 0.77-1.55), indicating that a treatment effect greater than 23% is unlikely. Manchia and colleagues state, “The findings contrast with a bulk of naturalistic and epidemiological data showing an evident suicide-protective effect with lithium.”