Leptin is believed to play a role in regulating food intake and body weight. The aim of this study was to examine the influence of parental history of obesity on the association between baseline serum leptin concentrations and subsequent 4-y weight changes. Changes in food intake were also considered in the analysis. Middle-aged, obese women with no obese parent (n = 25) or at least one obese parent (n = 24) were included in the analysis. At baseline, women with no parental history of obesity and women with a parental history of obesity did not differ in body mass index (in kg/m2: 41.2 and 40.2, respectively) or median leptin concentrations (40.8 and 38.8 microg/L, respectively). Four-year weight changes varied widely in both groups combined (from -30 to 24 kg). Stratified regression analysis, adjusted for age, weight, and height, revealed that high leptin concentrations predicted less weight gain (or more weight loss) in women with no obese parent (beta = -21.2, P = 0.0006) but played no significant role in predicting weight gain in women with at least one obese parent (beta = -3.8, P = 0.41). Adding changes in energy and fat intakes to the model reduced the association between leptin and weight change to nonsignificance in the women with no obese parent, indicating that the effect of leptin could be explained largely by dietary changes. In conclusion, serum leptin concentrations predict long-term weight change in obese women with no history of parental obesity, an association largely mediated by changes in food intake.