OBJECTIVE: To determine if the source from which food is obtained has contributed to the increased obesity of the US population, while controlling for demographic, lifestyle and regional factors. METHODS: Multiple regression was used to estimate the effect of food source on body mass index (BMI) while accounting for other factors which have been shown to affect obesity in a nationally representative sample of the US population. SAMPLE: This study used secondary data from the 1994-1996 Continuing Survey of Food Intake by Individuals (CSFII). The CSFII is a nationally representative sample of 16,103 individuals, obtaining for each respondent 24 h recalls of all food intake on two nonconsecutive days as well as demographics and information on lifestyle choices. RESULTS: For a large number of demographic and lifestyle factors, our results support those which have previously been found to contribute to increased overweight. Our contribution is to examine whether the source from which food is obtained also contributes to increased overweight. Our evidence suggests that this is the case. The average height for males in our sample was 1.77 m. For two such males, one who ate food away from home (FAFH) during the previous 24 h period and the other who did not, results suggest that the first will be about 1 kg heavier, all other factors being equal. For two females of average height (1.63 m) the same is true for those who ate fast food, but not at restaurants. In all cases, except females who ate at restaurants, the effects are significant in the regression (P<0.05). CONCLUSION: The trends in both increased US obesity and in increased consumption of FAFH are unlikely to be coincidental. FAFH, and particularly fast food consumption, are likely to be contributing factors to increased obesity. International Journal of Obesity (2000) 24, 1032-1039 | 
