Although not a perfect proxy, the USDA has tracked data for "food available for consumption" (an inventory based statistic) per capita per day since 1909. This

Although not a perfect proxy, the USDA has tracked data for "food available for consumption" (an inventory based statistic) per capita per day since 1909. This makes it good for this question simply due to duration. Charts with the data from 1909-2010 can be found here: http://www.cnpp.usda.gov/USFoodSupply-1909-2010 http://www.cnpp.usda.gov/USFoodSupply-1909-2010 The kilocalorie availability per capita per day was pretty consistently in the 3100-3400 range until about 1976 or so. After 1976, the kilocalorie available per capita per day slowly creeps up until it reaches about 4000-4200 calories a day. This metric is not perfect by any means, but I would say there's a high probability that this correlates to the United States consuming more food, for what it's worth. It's honestly the one food metric with this particular dataset that stands out. The percentage of calories from sugars and sweeteners, sodium levels, and grain consumption seems consistent with the 1950s. The only other data metric that stands out with to me (if this dataset can be inferred as a proxy of changes in the food industry) is a large rise in the percentage calories consisting of fat. Specifically, the "salad / cooking oils" rises from being 0.5% of calories available in 1909 to a whopping 15.1% of calories available in 2010.