BACKGROUND It is not known how the relationship between weight change and mortality is influenced by initial body mass index (BMI) or the magnitude of weight change. METHODS We use the nationally representative Health and Retirement Study (n = 13,104; follow-up 1992-2006) and Cox regression analysis to estimate relative mortality risks for 2-year weight change by initial BMI among 50- to-70-year-old Americans. We defined small weight loss or gain as a change of 1-2.9 BMI units and large weight loss or gain as a change of 3-5 BMI units. RESULTS Large and small weight losses were associated with excess mortality for all initial BMI levels below 32 kg/m (eg, hazard ratio [HR] for large weight loss from BMI of 30 = 1.61 [95% confidence interval = 1.31-1.98]; HR for small weight loss from BMI of 30 = 1.19 [1.06-1.28]). Large weight gains were associated with excess mortality only at high BMIs (eg, HR for large weight gain from BMI of 35 = 1.33 [1.00-1.77]). Small weight gains were not associated with excess mortality for any initial BMI level. The weight loss-mortality association was robust to adjustments for health status and to sensitivity analyses considering unobserved confounders. CONCLUSIONS Weight loss is associated with excess mortality among normal, overweight, and mildly obese middle- and older-aged adults. The excess risk increases for larger losses and lower initial BMI. These results suggest that the potential benefits of a lower BMI may be offset by the negative effects associated with weight loss. Weight gain may be associated with excess mortality only among obese people with an initial BMI over 35.