Studies relating to the adverse effect of monosodium glutamate (MSG) on body weight in rodent models were cited in our paper(Reference Shi, Luscombe-Marsh and Wittert1). Obesity in these models is the result of hypothalamic damage due to MSG administration at a critical developmental time point; this is not relevant to the human population in our study. The absence of an association between MSG and obesity does not exclude the possibility of other adverse effects of MSG in the population, but this was not the point of our paper. In Table 1 of our paper(Reference Shi, Luscombe-Marsh and Wittert1), we presented data on weight, waist circumference, central obesity, and BMI across quartiles of MSG intake at baseline. The prevalence of central obesity was also shown in Table 2. The OR for obesity across quartiles of MSG intake at baseline was presented in the text. The inability to account for any change in MSG consumption during the 5-year follow-up is a limitation of the study and this was acknowledged; however, we do not agree that our conclusion based on baseline MSG is unacceptable. Dietary intakes based on FFQ reflect long-term food habits(Reference Willett2). With regard to the analysis strategy, in Table 2 we presented different models. We believe these models give a clear picture of the association between MSG intake and weight change. In the text, we also presented results from a rural population. Adjusting for physical activity and alcohol intake is reasonable since these are known factors related to obesity.
This letter was written in response to comments by Samuels(Reference Samuels3).
The authors have no conflict of interest and there was no funding associated with this letter.
