Gender, elapsed time and BMI contributed to self-report bias when 6101 individuals were asked to recal their body weight years back. Effects of gender and elapsed time were small relative to the effects of BMI, with women tending to underestimate weight by 0.82 kg, men overestimating by 2.27 kg and bias increasing by 0.09 kg for each year of retrospection. Every increase of one unit in BMI was associated with an additional 0.91 kg underestimate in self-reported weight. Conclusion: Accuracy of adult self-reported weight profiles will be greatly underestimated if samples have individuals of current or past high relative weight. Adjusting for underestimation based on the respondent’s weight status alone will considerably improve the validity of weight information.