Placing full-service supermarkets in food deserts--areas with limited access to healthy food--has been promoted as a way to reduce inequalities in access to healthy food, improve diet, and reduce the risk of obesity. However, previous studies provide scant evidence of such impacts. We surveyed households in two Pittsburgh, Pennsylvania, neighborhoods in 2011 and 2014, one of which received a new supermarket in 2013. Comparing trends in the two neighborhoods, we obtained evidence of multiple positive impacts from new supermarket placement. In the new supermarket neighborhood we found net positive changes in overall dietary quality; average daily intakes of kilocalories and added sugars; and percentage of kilocalories from solid fats, added sugars, and alcohol. However, the only positive outcome in the recipient neighborhood specifically associated with regular use of the new supermarket was improved perceived access to healthy food. We did not observe differential improvement between the neighborhoods in fruit and vegetable intake, whole grain consumption, or body mass index. Incentivizing supermarkets to locate in food deserts is appropriate. However, efforts should proceed with caution, until the mechanisms by which the stores affect diet and their ability to influence weight status are better understood.